diff options
Diffstat (limited to 'net')
198 files changed, 10580 insertions, 2207 deletions
diff --git a/net/Kconfig b/net/Kconfig index 57f51a279ad6..3101bfcbdd7a 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -430,6 +430,7 @@ config NET_SOCK_MSG config NET_DEVLINK bool default n + imply NET_DROP_MONITOR config PAGE_POOL bool diff --git a/net/atm/mpoa_caches.c b/net/atm/mpoa_caches.c index 4bb418313720..3286f9d527d3 100644 --- a/net/atm/mpoa_caches.c +++ b/net/atm/mpoa_caches.c @@ -180,8 +180,7 @@ static int cache_hit(in_cache_entry *entry, struct mpoa_client *mpc) static void in_cache_put(in_cache_entry *entry) { if (refcount_dec_and_test(&entry->use)) { - memset(entry, 0, sizeof(in_cache_entry)); - kfree(entry); + kzfree(entry); } } @@ -416,8 +415,7 @@ static eg_cache_entry *eg_cache_get_by_src_ip(__be32 ipaddr, static void eg_cache_put(eg_cache_entry *entry) { if (refcount_dec_and_test(&entry->use)) { - memset(entry, 0, sizeof(eg_cache_entry)); - kfree(entry); + kzfree(entry); } } diff --git a/net/atm/pppoatm.c b/net/atm/pppoatm.c index bd3da9af5ef6..45d8e1d5d033 100644 --- a/net/atm/pppoatm.c +++ b/net/atm/pppoatm.c @@ -216,9 +216,7 @@ static void pppoatm_push(struct atm_vcc *atmvcc, struct sk_buff *skb) pvcc->chan.mtu += LLC_LEN; break; } - pr_debug("Couldn't autodetect yet (skb: %02X %02X %02X %02X %02X %02X)\n", - skb->data[0], skb->data[1], skb->data[2], - skb->data[3], skb->data[4], skb->data[5]); + pr_debug("Couldn't autodetect yet (skb: %6ph)\n", skb->data); goto error; case e_vc: break; diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 240ed70912d6..d78938e3e008 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -277,17 +277,23 @@ static u8 batadv_hop_penalty(u8 tq, const struct batadv_priv *bat_priv) * batadv_iv_ogm_aggr_packet() - checks if there is another OGM attached * @buff_pos: current position in the skb * @packet_len: total length of the skb - * @tvlv_len: tvlv length of the previously considered OGM + * @ogm_packet: potential OGM in buffer * * Return: true if there is enough space for another OGM, false otherwise. */ -static bool batadv_iv_ogm_aggr_packet(int buff_pos, int packet_len, - __be16 tvlv_len) +static bool +batadv_iv_ogm_aggr_packet(int buff_pos, int packet_len, + const struct batadv_ogm_packet *ogm_packet) { int next_buff_pos = 0; - next_buff_pos += buff_pos + BATADV_OGM_HLEN; - next_buff_pos += ntohs(tvlv_len); + /* check if there is enough space for the header */ + next_buff_pos += buff_pos + sizeof(*ogm_packet); + if (next_buff_pos > packet_len) + return false; + + /* check if there is enough space for the optional TVLV */ + next_buff_pos += ntohs(ogm_packet->tvlv_len); return (next_buff_pos <= packet_len) && (next_buff_pos <= BATADV_MAX_AGGREGATION_BYTES); @@ -315,7 +321,7 @@ static void batadv_iv_ogm_send_to_if(struct batadv_forw_packet *forw_packet, /* adjust all flags and log packets */ while (batadv_iv_ogm_aggr_packet(buff_pos, forw_packet->packet_len, - batadv_ogm_packet->tvlv_len)) { + batadv_ogm_packet)) { /* we might have aggregated direct link packets with an * ordinary base packet */ @@ -1704,7 +1710,7 @@ static int batadv_iv_ogm_receive(struct sk_buff *skb, /* unpack the aggregated packets and process them one by one */ while (batadv_iv_ogm_aggr_packet(ogm_offset, skb_headlen(skb), - ogm_packet->tvlv_len)) { + ogm_packet)) { batadv_iv_ogm_process(skb, ogm_offset, if_incoming); ogm_offset += BATADV_OGM_HLEN; diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 319249f0f85f..dc4f7430cb5a 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -806,17 +806,23 @@ batadv_v_ogm_process_per_outif(struct batadv_priv *bat_priv, * batadv_v_ogm_aggr_packet() - checks if there is another OGM aggregated * @buff_pos: current position in the skb * @packet_len: total length of the skb - * @tvlv_len: tvlv length of the previously considered OGM + * @ogm2_packet: potential OGM2 in buffer * * Return: true if there is enough space for another OGM, false otherwise. */ -static bool batadv_v_ogm_aggr_packet(int buff_pos, int packet_len, - __be16 tvlv_len) +static bool +batadv_v_ogm_aggr_packet(int buff_pos, int packet_len, + const struct batadv_ogm2_packet *ogm2_packet) { int next_buff_pos = 0; - next_buff_pos += buff_pos + BATADV_OGM2_HLEN; - next_buff_pos += ntohs(tvlv_len); + /* check if there is enough space for the header */ + next_buff_pos += buff_pos + sizeof(*ogm2_packet); + if (next_buff_pos > packet_len) + return false; + + /* check if there is enough space for the optional TVLV */ + next_buff_pos += ntohs(ogm2_packet->tvlv_len); return (next_buff_pos <= packet_len) && (next_buff_pos <= BATADV_MAX_AGGREGATION_BYTES); @@ -993,7 +999,7 @@ int batadv_v_ogm_packet_recv(struct sk_buff *skb, ogm_packet = (struct batadv_ogm2_packet *)skb->data; while (batadv_v_ogm_aggr_packet(ogm_offset, skb_headlen(skb), - ogm_packet->tvlv_len)) { + ogm_packet)) { batadv_v_ogm_process(skb, ogm_offset, if_incoming); ogm_offset += BATADV_OGM2_HLEN; diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 67d7f83009ae..1d5bdf3a4b65 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -2303,7 +2303,7 @@ __batadv_mcast_flags_dump(struct sk_buff *msg, u32 portid, while (bucket_tmp < hash->size) { if (batadv_mcast_flags_dump_bucket(msg, portid, cb, hash, - *bucket, &idx_tmp)) + bucket_tmp, &idx_tmp)) break; bucket_tmp++; @@ -2420,8 +2420,10 @@ void batadv_mcast_purge_orig(struct batadv_orig_node *orig) batadv_mcast_want_unsnoop_update(bat_priv, orig, BATADV_NO_FLAGS); batadv_mcast_want_ipv4_update(bat_priv, orig, BATADV_NO_FLAGS); batadv_mcast_want_ipv6_update(bat_priv, orig, BATADV_NO_FLAGS); - batadv_mcast_want_rtr4_update(bat_priv, orig, BATADV_NO_FLAGS); - batadv_mcast_want_rtr6_update(bat_priv, orig, BATADV_NO_FLAGS); + batadv_mcast_want_rtr4_update(bat_priv, orig, + BATADV_MCAST_WANT_NO_RTR4); + batadv_mcast_want_rtr6_update(bat_priv, orig, + BATADV_MCAST_WANT_NO_RTR6); spin_unlock_bh(&orig->mcast_handler_lock); } diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 6f08fd122a8d..7e052d6f759b 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -164,7 +164,7 @@ batadv_netlink_get_ifindex(const struct nlmsghdr *nlh, int attrtype) { struct nlattr *attr = nlmsg_find_attr(nlh, GENL_HDRLEN, attrtype); - return attr ? nla_get_u32(attr) : 0; + return (attr && nla_len(attr) == sizeof(u32)) ? nla_get_u32(attr) : 0; } /** diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index b9585e7d9d2e..04bc79359a17 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -3202,6 +3202,7 @@ struct hci_dev *hci_alloc_dev(void) hdev->conn_info_min_age = DEFAULT_CONN_INFO_MIN_AGE; hdev->conn_info_max_age = DEFAULT_CONN_INFO_MAX_AGE; hdev->auth_payload_timeout = DEFAULT_AUTH_PAYLOAD_TIMEOUT; + hdev->min_enc_key_size = HCI_MIN_ENC_KEY_SIZE; mutex_init(&hdev->lock); mutex_init(&hdev->req_lock); diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c index bb67f4a5479a..402e2cc54044 100644 --- a/net/bluetooth/hci_debugfs.c +++ b/net/bluetooth/hci_debugfs.c @@ -433,6 +433,35 @@ static int auto_accept_delay_set(void *data, u64 val) return 0; } +static int min_encrypt_key_size_set(void *data, u64 val) +{ + struct hci_dev *hdev = data; + + if (val < 1 || val > 16) + return -EINVAL; + + hci_dev_lock(hdev); + hdev->min_enc_key_size = val; + hci_dev_unlock(hdev); + + return 0; +} + +static int min_encrypt_key_size_get(void *data, u64 *val) +{ + struct hci_dev *hdev = data; + + hci_dev_lock(hdev); + *val = hdev->min_enc_key_size; + hci_dev_unlock(hdev); + + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(min_encrypt_key_size_fops, + min_encrypt_key_size_get, + min_encrypt_key_size_set, "%llu\n"); + static int auto_accept_delay_get(void *data, u64 *val) { struct hci_dev *hdev = data; @@ -545,6 +574,8 @@ void hci_debugfs_create_bredr(struct hci_dev *hdev) if (lmp_ssp_capable(hdev)) { debugfs_create_file("ssp_debug_mode", 0444, hdev->debugfs, hdev, &ssp_debug_mode_fops); + debugfs_create_file("min_encrypt_key_size", 0644, hdev->debugfs, + hdev, &min_encrypt_key_size_fops); debugfs_create_file("auto_accept_delay", 0644, hdev->debugfs, hdev, &auto_accept_delay_fops); } diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 5abd423b55fa..8d889969ae7e 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -101,6 +101,7 @@ static int hidp_send_message(struct hidp_session *session, struct socket *sock, { struct sk_buff *skb; struct sock *sk = sock->sk; + int ret; BT_DBG("session %p data %p size %d", session, data, size); @@ -114,13 +115,17 @@ static int hidp_send_message(struct hidp_session *session, struct socket *sock, } skb_put_u8(skb, hdr); - if (data && size > 0) + if (data && size > 0) { skb_put_data(skb, data, size); + ret = size; + } else { + ret = 0; + } skb_queue_tail(transmit, skb); wake_up_interruptible(sk_sleep(sk)); - return 0; + return ret; } static int hidp_send_ctrl_message(struct hidp_session *session, diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index cc506fe99b4d..dfc1edb168b7 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -1361,7 +1361,7 @@ static bool l2cap_check_enc_key_size(struct hci_conn *hcon) * actually encrypted before enforcing a key size. */ return (!test_bit(HCI_CONN_ENCRYPT, &hcon->flags) || - hcon->enc_key_size >= HCI_MIN_ENC_KEY_SIZE); + hcon->enc_key_size >= hcon->hdev->min_enc_key_size); } static void l2cap_do_start(struct l2cap_chan *chan) diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index 428af1abf8cc..44594635a972 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -77,6 +77,53 @@ static void __mdb_entry_to_br_ip(struct br_mdb_entry *entry, struct br_ip *ip) #endif } +static int __mdb_fill_info(struct sk_buff *skb, + struct net_bridge_mdb_entry *mp, + struct net_bridge_port_group *p) +{ + struct timer_list *mtimer; + struct nlattr *nest_ent; + struct br_mdb_entry e; + u8 flags = 0; + int ifindex; + + memset(&e, 0, sizeof(e)); + if (p) { + ifindex = p->port->dev->ifindex; + mtimer = &p->timer; + flags = p->flags; + } else { + ifindex = mp->br->dev->ifindex; + mtimer = &mp->timer; + } + + __mdb_entry_fill_flags(&e, flags); + e.ifindex = ifindex; + e.vid = mp->addr.vid; + if (mp->addr.proto == htons(ETH_P_IP)) + e.addr.u.ip4 = mp->addr.u.ip4; +#if IS_ENABLED(CONFIG_IPV6) + if (mp->addr.proto == htons(ETH_P_IPV6)) + e.addr.u.ip6 = mp->addr.u.ip6; +#endif + e.addr.proto = mp->addr.proto; + nest_ent = nla_nest_start_noflag(skb, + MDBA_MDB_ENTRY_INFO); + if (!nest_ent) + return -EMSGSIZE; + + if (nla_put_nohdr(skb, sizeof(e), &e) || + nla_put_u32(skb, + MDBA_MDB_EATTR_TIMER, + br_timer_value(mtimer))) { + nla_nest_cancel(skb, nest_ent); + return -EMSGSIZE; + } + nla_nest_end(skb, nest_ent); + + return 0; +} + static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb, struct net_device *dev) { @@ -95,7 +142,6 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb, hlist_for_each_entry_rcu(mp, &br->mdb_list, mdb_node) { struct net_bridge_port_group *p; struct net_bridge_port_group __rcu **pp; - struct net_bridge_port *port; if (idx < s_idx) goto skip; @@ -106,43 +152,24 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb, break; } + if (mp->host_joined) { + err = __mdb_fill_info(skb, mp, NULL); + if (err) { + nla_nest_cancel(skb, nest2); + break; + } + } + for (pp = &mp->ports; (p = rcu_dereference(*pp)) != NULL; pp = &p->next) { - struct nlattr *nest_ent; - struct br_mdb_entry e; - - port = p->port; - if (!port) + if (!p->port) continue; - memset(&e, 0, sizeof(e)); - e.ifindex = port->dev->ifindex; - e.vid = p->addr.vid; - __mdb_entry_fill_flags(&e, p->flags); - if (p->addr.proto == htons(ETH_P_IP)) - e.addr.u.ip4 = p->addr.u.ip4; -#if IS_ENABLED(CONFIG_IPV6) - if (p->addr.proto == htons(ETH_P_IPV6)) - e.addr.u.ip6 = p->addr.u.ip6; -#endif - e.addr.proto = p->addr.proto; - nest_ent = nla_nest_start_noflag(skb, - MDBA_MDB_ENTRY_INFO); - if (!nest_ent) { - nla_nest_cancel(skb, nest2); - err = -EMSGSIZE; - goto out; - } - if (nla_put_nohdr(skb, sizeof(e), &e) || - nla_put_u32(skb, - MDBA_MDB_EATTR_TIMER, - br_timer_value(&p->timer))) { - nla_nest_cancel(skb, nest_ent); + err = __mdb_fill_info(skb, mp, p); + if (err) { nla_nest_cancel(skb, nest2); - err = -EMSGSIZE; goto out; } - nla_nest_end(skb, nest_ent); } nla_nest_end(skb, nest2); skip: @@ -589,6 +616,19 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port, return err; } + /* host join */ + if (!port) { + /* don't allow any flags for host-joined groups */ + if (state) + return -EINVAL; + if (mp->host_joined) + return -EEXIST; + + br_multicast_host_join(mp, false); + + return 0; + } + for (pp = &mp->ports; (p = mlock_dereference(*pp, br)) != NULL; pp = &p->next) { @@ -613,19 +653,21 @@ static int __br_mdb_add(struct net *net, struct net_bridge *br, { struct br_ip ip; struct net_device *dev; - struct net_bridge_port *p; + struct net_bridge_port *p = NULL; int ret; if (!netif_running(br->dev) || !br_opt_get(br, BROPT_MULTICAST_ENABLED)) return -EINVAL; - dev = __dev_get_by_index(net, entry->ifindex); - if (!dev) - return -ENODEV; + if (entry->ifindex != br->dev->ifindex) { + dev = __dev_get_by_index(net, entry->ifindex); + if (!dev) + return -ENODEV; - p = br_port_get_rtnl(dev); - if (!p || p->br != br || p->state == BR_STATE_DISABLED) - return -EINVAL; + p = br_port_get_rtnl(dev); + if (!p || p->br != br || p->state == BR_STATE_DISABLED) + return -EINVAL; + } __mdb_entry_to_br_ip(entry, &ip); @@ -640,9 +682,9 @@ static int br_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, { struct net *net = sock_net(skb->sk); struct net_bridge_vlan_group *vg; + struct net_bridge_port *p = NULL; struct net_device *dev, *pdev; struct br_mdb_entry *entry; - struct net_bridge_port *p; struct net_bridge_vlan *v; struct net_bridge *br; int err; @@ -653,18 +695,22 @@ static int br_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, br = netdev_priv(dev); + if (entry->ifindex != br->dev->ifindex) { + pdev = __dev_get_by_index(net, entry->ifindex); + if (!pdev) + return -ENODEV; + + p = br_port_get_rtnl(pdev); + if (!p || p->br != br || p->state == BR_STATE_DISABLED) + return -EINVAL; + vg = nbp_vlan_group(p); + } else { + vg = br_vlan_group(br); + } + /* If vlan filtering is enabled and VLAN is not specified * install mdb entry on all vlans configured on the port. */ - pdev = __dev_get_by_index(net, entry->ifindex); - if (!pdev) - return -ENODEV; - - p = br_port_get_rtnl(pdev); - if (!p || p->br != br || p->state == BR_STATE_DISABLED) - return -EINVAL; - - vg = nbp_vlan_group(p); if (br_vlan_enabled(br->dev) && vg && entry->vid == 0) { list_for_each_entry(v, &vg->vlan_list, vlist) { entry->vid = v->vid; @@ -700,6 +746,15 @@ static int __br_mdb_del(struct net_bridge *br, struct br_mdb_entry *entry) if (!mp) goto unlock; + /* host leave */ + if (entry->ifindex == mp->br->dev->ifindex && mp->host_joined) { + br_multicast_host_leave(mp, false); + err = 0; + if (!mp->ports && netif_running(br->dev)) + mod_timer(&mp->timer, jiffies); + goto unlock; + } + for (pp = &mp->ports; (p = mlock_dereference(*pp, br)) != NULL; pp = &p->next) { @@ -732,9 +787,9 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh, { struct net *net = sock_net(skb->sk); struct net_bridge_vlan_group *vg; + struct net_bridge_port *p = NULL; struct net_device *dev, *pdev; struct br_mdb_entry *entry; - struct net_bridge_port *p; struct net_bridge_vlan *v; struct net_bridge *br; int err; @@ -745,18 +800,22 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh, br = netdev_priv(dev); + if (entry->ifindex != br->dev->ifindex) { + pdev = __dev_get_by_index(net, entry->ifindex); + if (!pdev) + return -ENODEV; + + p = br_port_get_rtnl(pdev); + if (!p || p->br != br || p->state == BR_STATE_DISABLED) + return -EINVAL; + vg = nbp_vlan_group(p); + } else { + vg = br_vlan_group(br); + } + /* If vlan filtering is enabled and VLAN is not specified * delete mdb entry on all vlans configured on the port. */ - pdev = __dev_get_by_index(net, entry->ifindex); - if (!pdev) - return -ENODEV; - - p = br_port_get_rtnl(pdev); - if (!p || p->br != br || p->state == BR_STATE_DISABLED) - return -EINVAL; - - vg = nbp_vlan_group(p); if (br_vlan_enabled(br->dev) && vg && entry->vid == 0) { list_for_each_entry(v, &vg->vlan_list, vlist) { entry->vid = v->vid; diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 9b379e110129..ad12fe3fca8c 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -148,8 +148,7 @@ static void br_multicast_group_expired(struct timer_list *t) if (!netif_running(br->dev) || timer_pending(&mp->timer)) goto out; - mp->host_joined = false; - br_mdb_notify(br->dev, NULL, &mp->addr, RTM_DELMDB, 0); + br_multicast_host_leave(mp, true); if (mp->ports) goto out; @@ -512,6 +511,27 @@ static bool br_port_group_equal(struct net_bridge_port_group *p, return ether_addr_equal(src, p->eth_addr); } +void br_multicast_host_join(struct net_bridge_mdb_entry *mp, bool notify) +{ + if (!mp->host_joined) { + mp->host_joined = true; + if (notify) + br_mdb_notify(mp->br->dev, NULL, &mp->addr, + RTM_NEWMDB, 0); + } + mod_timer(&mp->timer, jiffies + mp->br->multicast_membership_interval); +} + +void br_multicast_host_leave(struct net_bridge_mdb_entry *mp, bool notify) +{ + if (!mp->host_joined) + return; + + mp->host_joined = false; + if (notify) + br_mdb_notify(mp->br->dev, NULL, &mp->addr, RTM_DELMDB, 0); +} + static int br_multicast_add_group(struct net_bridge *br, struct net_bridge_port *port, struct br_ip *group, @@ -534,11 +554,7 @@ static int br_multicast_add_group(struct net_bridge *br, goto err; if (!port) { - if (!mp->host_joined) { - mp->host_joined = true; - br_mdb_notify(br->dev, NULL, &mp->addr, RTM_NEWMDB, 0); - } - mod_timer(&mp->timer, now + br->multicast_membership_interval); + br_multicast_host_join(mp, true); goto out; } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index b7a4942ff1b3..ce2ab14ee605 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -702,6 +702,8 @@ void br_multicast_get_stats(const struct net_bridge *br, struct br_mcast_stats *dest); void br_mdb_init(void); void br_mdb_uninit(void); +void br_multicast_host_join(struct net_bridge_mdb_entry *mp, bool notify); +void br_multicast_host_leave(struct net_bridge_mdb_entry *mp, bool notify); #define mlock_dereference(X, br) \ rcu_dereference_protected(X, lockdep_is_held(&br->multicast_lock)) diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index f5b2aeebbfe9..bb98984cd27d 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -1281,6 +1281,8 @@ int br_vlan_get_info(const struct net_device *dev, u16 vid, p_vinfo->vid = vid; p_vinfo->flags = v->flags; + if (vid == br_get_pvid(vg)) + p_vinfo->flags |= BRIDGE_VLAN_INFO_PVID; return 0; } EXPORT_SYMBOL_GPL(br_vlan_get_info); diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index c8177a89f52c..4096d8a74a2b 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -221,7 +221,7 @@ unsigned int ebt_do_table(struct sk_buff *skb, return NF_DROP; } - ADD_COUNTER(*(counter_base + i), 1, skb->len); + ADD_COUNTER(*(counter_base + i), skb->len, 1); /* these should only watch: not modify, nor tell us * what to do with the packet @@ -959,8 +959,8 @@ static void get_counters(const struct ebt_counter *oldcounters, continue; counter_base = COUNTER_BASE(oldcounters, nentries, cpu); for (i = 0; i < nentries; i++) - ADD_COUNTER(counters[i], counter_base[i].pcnt, - counter_base[i].bcnt); + ADD_COUNTER(counters[i], counter_base[i].bcnt, + counter_base[i].pcnt); } } @@ -1280,7 +1280,7 @@ static int do_update_counters(struct net *net, const char *name, /* we add to the counters of the first cpu */ for (i = 0; i < num_counters; i++) - ADD_COUNTER(t->private->counters[i], tmp[i].pcnt, tmp[i].bcnt); + ADD_COUNTER(t->private->counters[i], tmp[i].bcnt, tmp[i].pcnt); write_unlock_bh(&t->lock); ret = 0; diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c index 1804e867f715..7c9e92b2f806 100644 --- a/net/bridge/netfilter/nft_meta_bridge.c +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -53,7 +53,7 @@ static void nft_meta_bridge_get_eval(const struct nft_expr *expr, goto err; br_vlan_get_proto(br_dev, &p_proto); - nft_reg_store16(dest, p_proto); + nft_reg_store16(dest, htons(p_proto)); return; } default: diff --git a/net/can/Kconfig b/net/can/Kconfig index d4319aa3e1b1..d77042752457 100644 --- a/net/can/Kconfig +++ b/net/can/Kconfig @@ -53,6 +53,8 @@ config CAN_GW They can be modified with AND/OR/XOR/SET operations as configured by the netlink configuration interface known e.g. from iptables. +source "net/can/j1939/Kconfig" + source "drivers/net/can/Kconfig" endif diff --git a/net/can/Makefile b/net/can/Makefile index 1242bbbfe57f..08bd217fc051 100644 --- a/net/can/Makefile +++ b/net/can/Makefile @@ -15,3 +15,5 @@ can-bcm-y := bcm.o obj-$(CONFIG_CAN_GW) += can-gw.o can-gw-y := gw.o + +obj-$(CONFIG_CAN_J1939) += j1939/ diff --git a/net/can/af_can.c b/net/can/af_can.c index 76cf83b2bd40..5518a7d9eed9 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -1,6 +1,5 @@ // SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) -/* - * af_can.c - Protocol family CAN core module +/* af_can.c - Protocol family CAN core module * (used by different CAN protocol modules) * * Copyright (c) 2002-2017 Volkswagen Group Electronic Research @@ -59,6 +58,7 @@ #include <linux/can.h> #include <linux/can/core.h> #include <linux/can/skb.h> +#include <linux/can/can-ml.h> #include <linux/ratelimit.h> #include <net/net_namespace.h> #include <net/sock.h> @@ -84,9 +84,7 @@ static DEFINE_MUTEX(proto_tab_lock); static atomic_t skbcounter = ATOMIC_INIT(0); -/* - * af_can socket functions - */ +/* af_can socket functions */ static void can_sock_destruct(struct sock *sk) { @@ -132,14 +130,13 @@ static int can_create(struct net *net, struct socket *sock, int protocol, err = request_module("can-proto-%d", protocol); - /* - * In case of error we only print a message but don't + /* In case of error we only print a message but don't * return the error code immediately. Below we will * return -EPROTONOSUPPORT */ if (err) - printk_ratelimited(KERN_ERR "can: request_module " - "(can-proto-%d) failed.\n", protocol); + pr_err_ratelimited("can: request_module (can-proto-%d) failed.\n", + protocol); cp = can_get_proto(protocol); } @@ -180,9 +177,7 @@ static int can_create(struct net *net, struct socket *sock, int protocol, return err; } -/* - * af_can tx path - */ +/* af_can tx path */ /** * can_send - transmit a CAN frame (optional with local loopback) @@ -204,7 +199,7 @@ int can_send(struct sk_buff *skb, int loop) { struct sk_buff *newskb = NULL; struct canfd_frame *cfd = (struct canfd_frame *)skb->data; - struct s_stats *can_stats = dev_net(skb->dev)->can.can_stats; + struct can_pkg_stats *pkg_stats = dev_net(skb->dev)->can.pkg_stats; int err = -EINVAL; if (skb->len == CAN_MTU) { @@ -215,11 +210,11 @@ int can_send(struct sk_buff *skb, int loop) skb->protocol = htons(ETH_P_CANFD); if (unlikely(cfd->len > CANFD_MAX_DLEN)) goto inval_skb; - } else + } else { goto inval_skb; + } - /* - * Make sure the CAN frame can pass the selected CAN netdevice. + /* Make sure the CAN frame can pass the selected CAN netdevice. * As structs can_frame and canfd_frame are similar, we can provide * CAN FD frames to legacy CAN drivers as long as the length is <= 8 */ @@ -250,8 +245,7 @@ int can_send(struct sk_buff *skb, int loop) /* indication for the CAN driver: do loopback */ skb->pkt_type = PACKET_LOOPBACK; - /* - * The reference to the originating sock may be required + /* The reference to the originating sock may be required * by the receiving socket to check whether the frame is * its own. Example: can_raw sockopt CAN_RAW_RECV_OWN_MSGS * Therefore we have to ensure that skb->sk remains the @@ -260,8 +254,7 @@ int can_send(struct sk_buff *skb, int loop) */ if (!(skb->dev->flags & IFF_ECHO)) { - /* - * If the interface is not capable to do loopback + /* If the interface is not capable to do loopback * itself, we do it here. */ newskb = skb_clone(skb, GFP_ATOMIC); @@ -293,8 +286,8 @@ int can_send(struct sk_buff *skb, int loop) netif_rx_ni(newskb); /* update statistics */ - can_stats->tx_frames++; - can_stats->tx_frames_delta++; + pkg_stats->tx_frames++; + pkg_stats->tx_frames_delta++; return 0; @@ -304,17 +297,17 @@ inval_skb: } EXPORT_SYMBOL(can_send); -/* - * af_can rx path - */ +/* af_can rx path */ -static struct can_dev_rcv_lists *find_dev_rcv_lists(struct net *net, - struct net_device *dev) +static struct can_dev_rcv_lists *can_dev_rcv_lists_find(struct net *net, + struct net_device *dev) { - if (!dev) - return net->can.can_rx_alldev_list; - else - return (struct can_dev_rcv_lists *)dev->ml_priv; + if (dev) { + struct can_ml_priv *ml_priv = dev->ml_priv; + return &ml_priv->dev_rcv_lists; + } else { + return net->can.rx_alldev_list; + } } /** @@ -341,7 +334,7 @@ static unsigned int effhash(canid_t can_id) } /** - * find_rcv_list - determine optimal filterlist inside device filter struct + * can_rcv_list_find - determine optimal filterlist inside device filter struct * @can_id: pointer to CAN identifier of a given can_filter * @mask: pointer to CAN mask of a given can_filter * @d: pointer to the device filter struct @@ -367,8 +360,8 @@ static unsigned int effhash(canid_t can_id) * Constistency checked mask. * Reduced can_id to have a preprocessed filter compare value. */ -static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, - struct can_dev_rcv_lists *d) +static struct hlist_head *can_rcv_list_find(canid_t *can_id, canid_t *mask, + struct can_dev_rcv_lists *dev_rcv_lists) { canid_t inv = *can_id & CAN_INV_FILTER; /* save flag before masking */ @@ -376,7 +369,7 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, if (*mask & CAN_ERR_FLAG) { /* clear CAN_ERR_FLAG in filter entry */ *mask &= CAN_ERR_MASK; - return &d->rx[RX_ERR]; + return &dev_rcv_lists->rx[RX_ERR]; } /* with cleared CAN_ERR_FLAG we have a simple mask/value filterpair */ @@ -392,27 +385,26 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, /* inverse can_id/can_mask filter */ if (inv) - return &d->rx[RX_INV]; + return &dev_rcv_lists->rx[RX_INV]; /* mask == 0 => no condition testing at receive time */ if (!(*mask)) - return &d->rx[RX_ALL]; + return &dev_rcv_lists->rx[RX_ALL]; /* extra filterlists for the subscription of a single non-RTR can_id */ if (((*mask & CAN_EFF_RTR_FLAGS) == CAN_EFF_RTR_FLAGS) && !(*can_id & CAN_RTR_FLAG)) { - if (*can_id & CAN_EFF_FLAG) { if (*mask == (CAN_EFF_MASK | CAN_EFF_RTR_FLAGS)) - return &d->rx_eff[effhash(*can_id)]; + return &dev_rcv_lists->rx_eff[effhash(*can_id)]; } else { if (*mask == (CAN_SFF_MASK | CAN_EFF_RTR_FLAGS)) - return &d->rx_sff[*can_id]; + return &dev_rcv_lists->rx_sff[*can_id]; } } /* default: filter via can_id/can_mask */ - return &d->rx[RX_FIL]; + return &dev_rcv_lists->rx[RX_FIL]; } /** @@ -449,10 +441,10 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, canid_t mask, void (*func)(struct sk_buff *, void *), void *data, char *ident, struct sock *sk) { - struct receiver *r; - struct hlist_head *rl; - struct can_dev_rcv_lists *d; - struct s_pstats *can_pstats = net->can.can_pstats; + struct receiver *rcv; + struct hlist_head *rcv_list; + struct can_dev_rcv_lists *dev_rcv_lists; + struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats; int err = 0; /* insert new receiver (dev,canid,mask) -> (func,data) */ @@ -463,50 +455,42 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, if (dev && !net_eq(net, dev_net(dev))) return -ENODEV; - r = kmem_cache_alloc(rcv_cache, GFP_KERNEL); - if (!r) + rcv = kmem_cache_alloc(rcv_cache, GFP_KERNEL); + if (!rcv) return -ENOMEM; - spin_lock(&net->can.can_rcvlists_lock); - - d = find_dev_rcv_lists(net, dev); - if (d) { - rl = find_rcv_list(&can_id, &mask, d); + spin_lock_bh(&net->can.rcvlists_lock); - r->can_id = can_id; - r->mask = mask; - r->matches = 0; - r->func = func; - r->data = data; - r->ident = ident; - r->sk = sk; + dev_rcv_lists = can_dev_rcv_lists_find(net, dev); + rcv_list = can_rcv_list_find(&can_id, &mask, dev_rcv_lists); - hlist_add_head_rcu(&r->list, rl); - d->entries++; + rcv->can_id = can_id; + rcv->mask = mask; + rcv->matches = 0; + rcv->func = func; + rcv->data = data; + rcv->ident = ident; + rcv->sk = sk; - can_pstats->rcv_entries++; - if (can_pstats->rcv_entries_max < can_pstats->rcv_entries) - can_pstats->rcv_entries_max = can_pstats->rcv_entries; - } else { - kmem_cache_free(rcv_cache, r); - err = -ENODEV; - } + hlist_add_head_rcu(&rcv->list, rcv_list); + dev_rcv_lists->entries++; - spin_unlock(&net->can.can_rcvlists_lock); + rcv_lists_stats->rcv_entries++; + rcv_lists_stats->rcv_entries_max = max(rcv_lists_stats->rcv_entries_max, + rcv_lists_stats->rcv_entries); + spin_unlock_bh(&net->can.rcvlists_lock); return err; } EXPORT_SYMBOL(can_rx_register); -/* - * can_rx_delete_receiver - rcu callback for single receiver entry removal - */ +/* can_rx_delete_receiver - rcu callback for single receiver entry removal */ static void can_rx_delete_receiver(struct rcu_head *rp) { - struct receiver *r = container_of(rp, struct receiver, rcu); - struct sock *sk = r->sk; + struct receiver *rcv = container_of(rp, struct receiver, rcu); + struct sock *sk = rcv->sk; - kmem_cache_free(rcv_cache, r); + kmem_cache_free(rcv_cache, rcv); if (sk) sock_put(sk); } @@ -526,10 +510,10 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id, canid_t mask, void (*func)(struct sk_buff *, void *), void *data) { - struct receiver *r = NULL; - struct hlist_head *rl; - struct s_pstats *can_pstats = net->can.can_pstats; - struct can_dev_rcv_lists *d; + struct receiver *rcv = NULL; + struct hlist_head *rcv_list; + struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats; + struct can_dev_rcv_lists *dev_rcv_lists; if (dev && dev->type != ARPHRD_CAN) return; @@ -537,86 +521,69 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id, if (dev && !net_eq(net, dev_net(dev))) return; - spin_lock(&net->can.can_rcvlists_lock); + spin_lock_bh(&net->can.rcvlists_lock); - d = find_dev_rcv_lists(net, dev); - if (!d) { - pr_err("BUG: receive list not found for " - "dev %s, id %03X, mask %03X\n", - DNAME(dev), can_id, mask); - goto out; - } + dev_rcv_lists = can_dev_rcv_lists_find(net, dev); + rcv_list = can_rcv_list_find(&can_id, &mask, dev_rcv_lists); - rl = find_rcv_list(&can_id, &mask, d); - - /* - * Search the receiver list for the item to delete. This should + /* Search the receiver list for the item to delete. This should * exist, since no receiver may be unregistered that hasn't * been registered before. */ - - hlist_for_each_entry_rcu(r, rl, list) { - if (r->can_id == can_id && r->mask == mask && - r->func == func && r->data == data) + hlist_for_each_entry_rcu(rcv, rcv_list, list) { + if (rcv->can_id == can_id && rcv->mask == mask && + rcv->func == func && rcv->data == data) break; } - /* - * Check for bugs in CAN protocol implementations using af_can.c: - * 'r' will be NULL if no matching list item was found for removal. + /* Check for bugs in CAN protocol implementations using af_can.c: + * 'rcv' will be NULL if no matching list item was found for removal. */ - - if (!r) { - WARN(1, "BUG: receive list entry not found for dev %s, " - "id %03X, mask %03X\n", DNAME(dev), can_id, mask); + if (!rcv) { + WARN(1, "BUG: receive list entry not found for dev %s, id %03X, mask %03X\n", + DNAME(dev), can_id, mask); goto out; } - hlist_del_rcu(&r->list); - d->entries--; - - if (can_pstats->rcv_entries > 0) - can_pstats->rcv_entries--; + hlist_del_rcu(&rcv->list); + dev_rcv_lists->entries--; - /* remove device structure requested by NETDEV_UNREGISTER */ - if (d->remove_on_zero_entries && !d->entries) { - kfree(d); - dev->ml_priv = NULL; - } + if (rcv_lists_stats->rcv_entries > 0) + rcv_lists_stats->rcv_entries--; out: - spin_unlock(&net->can.can_rcvlists_lock); + spin_unlock_bh(&net->can.rcvlists_lock); /* schedule the receiver item for deletion */ - if (r) { - if (r->sk) - sock_hold(r->sk); - call_rcu(&r->rcu, can_rx_delete_receiver); + if (rcv) { + if (rcv->sk) + sock_hold(rcv->sk); + call_rcu(&rcv->rcu, can_rx_delete_receiver); } } EXPORT_SYMBOL(can_rx_unregister); -static inline void deliver(struct sk_buff *skb, struct receiver *r) +static inline void deliver(struct sk_buff *skb, struct receiver *rcv) { - r->func(skb, r->data); - r->matches++; + rcv->func(skb, rcv->data); + rcv->matches++; } -static int can_rcv_filter(struct can_dev_rcv_lists *d, struct sk_buff *skb) +static int can_rcv_filter(struct can_dev_rcv_lists *dev_rcv_lists, struct sk_buff *skb) { - struct receiver *r; + struct receiver *rcv; int matches = 0; struct can_frame *cf = (struct can_frame *)skb->data; canid_t can_id = cf->can_id; - if (d->entries == 0) + if (dev_rcv_lists->entries == 0) return 0; if (can_id & CAN_ERR_FLAG) { /* check for error message frame entries only */ - hlist_for_each_entry_rcu(r, &d->rx[RX_ERR], list) { - if (can_id & r->mask) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx[RX_ERR], list) { + if (can_id & rcv->mask) { + deliver(skb, rcv); matches++; } } @@ -624,23 +591,23 @@ static int can_rcv_filter(struct can_dev_rcv_lists *d, struct sk_buff *skb) } /* check for unfiltered entries */ - hlist_for_each_entry_rcu(r, &d->rx[RX_ALL], list) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx[RX_ALL], list) { + deliver(skb, rcv); matches++; } /* check for can_id/mask entries */ - hlist_for_each_entry_rcu(r, &d->rx[RX_FIL], list) { - if ((can_id & r->mask) == r->can_id) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx[RX_FIL], list) { + if ((can_id & rcv->mask) == rcv->can_id) { + deliver(skb, rcv); matches++; } } /* check for inverted can_id/mask entries */ - hlist_for_each_entry_rcu(r, &d->rx[RX_INV], list) { - if ((can_id & r->mask) != r->can_id) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx[RX_INV], list) { + if ((can_id & rcv->mask) != rcv->can_id) { + deliver(skb, rcv); matches++; } } @@ -650,16 +617,16 @@ static int can_rcv_filter(struct can_dev_rcv_lists *d, struct sk_buff *skb) return matches; if (can_id & CAN_EFF_FLAG) { - hlist_for_each_entry_rcu(r, &d->rx_eff[effhash(can_id)], list) { - if (r->can_id == can_id) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx_eff[effhash(can_id)], list) { + if (rcv->can_id == can_id) { + deliver(skb, rcv); matches++; } } } else { can_id &= CAN_SFF_MASK; - hlist_for_each_entry_rcu(r, &d->rx_sff[can_id], list) { - deliver(skb, r); + hlist_for_each_entry_rcu(rcv, &dev_rcv_lists->rx_sff[can_id], list) { + deliver(skb, rcv); matches++; } } @@ -669,14 +636,14 @@ static int can_rcv_filter(struct can_dev_rcv_lists *d, struct sk_buff *skb) static void can_receive(struct sk_buff *skb, struct net_device *dev) { - struct can_dev_rcv_lists *d; + struct can_dev_rcv_lists *dev_rcv_lists; struct net *net = dev_net(dev); - struct s_stats *can_stats = net->can.can_stats; + struct can_pkg_stats *pkg_stats = net->can.pkg_stats; int matches; /* update statistics */ - can_stats->rx_frames++; - can_stats->rx_frames_delta++; + pkg_stats->rx_frames++; + pkg_stats->rx_frames_delta++; /* create non-zero unique skb identifier together with *skb */ while (!(can_skb_prv(skb)->skbcnt)) @@ -685,12 +652,11 @@ static void can_receive(struct sk_buff *skb, struct net_device *dev) rcu_read_lock(); /* deliver the packet to sockets listening on all devices */ - matches = can_rcv_filter(net->can.can_rx_alldev_list, skb); + matches = can_rcv_filter(net->can.rx_alldev_list, skb); /* find receive list for this device */ - d = find_dev_rcv_lists(net, dev); - if (d) - matches += can_rcv_filter(d, skb); + dev_rcv_lists = can_dev_rcv_lists_find(net, dev); + matches += can_rcv_filter(dev_rcv_lists, skb); rcu_read_unlock(); @@ -698,8 +664,8 @@ static void can_receive(struct sk_buff *skb, struct net_device *dev) consume_skb(skb); if (matches > 0) { - can_stats->matches++; - can_stats->matches_delta++; + pkg_stats->matches++; + pkg_stats->matches_delta++; } } @@ -721,7 +687,7 @@ static int can_rcv(struct sk_buff *skb, struct net_device *dev, } static int canfd_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, struct net_device *orig_dev) + struct packet_type *pt, struct net_device *orig_dev) { struct canfd_frame *cfd = (struct canfd_frame *)skb->data; @@ -737,9 +703,7 @@ static int canfd_rcv(struct sk_buff *skb, struct net_device *dev, return NET_RX_SUCCESS; } -/* - * af_can protocol functions - */ +/* af_can protocol functions */ /** * can_proto_register - register CAN transport protocol @@ -770,8 +734,9 @@ int can_proto_register(const struct can_proto *cp) if (rcu_access_pointer(proto_tab[proto])) { pr_err("can: protocol %d already registered\n", proto); err = -EBUSY; - } else + } else { RCU_INIT_POINTER(proto_tab[proto], cp); + } mutex_unlock(&proto_tab_lock); @@ -801,48 +766,19 @@ void can_proto_unregister(const struct can_proto *cp) } EXPORT_SYMBOL(can_proto_unregister); -/* - * af_can notifier to create/remove CAN netdevice specific structs - */ +/* af_can notifier to create/remove CAN netdevice specific structs */ static int can_notifier(struct notifier_block *nb, unsigned long msg, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct can_dev_rcv_lists *d; if (dev->type != ARPHRD_CAN) return NOTIFY_DONE; switch (msg) { - case NETDEV_REGISTER: - - /* create new dev_rcv_lists for this device */ - d = kzalloc(sizeof(*d), GFP_KERNEL); - if (!d) - return NOTIFY_DONE; - BUG_ON(dev->ml_priv); - dev->ml_priv = d; - - break; - - case NETDEV_UNREGISTER: - spin_lock(&dev_net(dev)->can.can_rcvlists_lock); - - d = dev->ml_priv; - if (d) { - if (d->entries) - d->remove_on_zero_entries = 1; - else { - kfree(d); - dev->ml_priv = NULL; - } - } else - pr_err("can: notifier: receive list not found for dev " - "%s\n", dev->name); - - spin_unlock(&dev_net(dev)->can.can_rcvlists_lock); - + WARN(!dev->ml_priv, + "No CAN mid layer private allocated, please fix your driver and use alloc_candev()!\n"); break; } @@ -851,71 +787,54 @@ static int can_notifier(struct notifier_block *nb, unsigned long msg, static int can_pernet_init(struct net *net) { - spin_lock_init(&net->can.can_rcvlists_lock); - net->can.can_rx_alldev_list = - kzalloc(sizeof(struct can_dev_rcv_lists), GFP_KERNEL); - if (!net->can.can_rx_alldev_list) + spin_lock_init(&net->can.rcvlists_lock); + net->can.rx_alldev_list = + kzalloc(sizeof(*net->can.rx_alldev_list), GFP_KERNEL); + if (!net->can.rx_alldev_list) goto out; - net->can.can_stats = kzalloc(sizeof(struct s_stats), GFP_KERNEL); - if (!net->can.can_stats) - goto out_free_alldev_list; - net->can.can_pstats = kzalloc(sizeof(struct s_pstats), GFP_KERNEL); - if (!net->can.can_pstats) - goto out_free_can_stats; + net->can.pkg_stats = kzalloc(sizeof(*net->can.pkg_stats), GFP_KERNEL); + if (!net->can.pkg_stats) + goto out_free_rx_alldev_list; + net->can.rcv_lists_stats = kzalloc(sizeof(*net->can.rcv_lists_stats), GFP_KERNEL); + if (!net->can.rcv_lists_stats) + goto out_free_pkg_stats; if (IS_ENABLED(CONFIG_PROC_FS)) { /* the statistics are updated every second (timer triggered) */ if (stats_timer) { - timer_setup(&net->can.can_stattimer, can_stat_update, + timer_setup(&net->can.stattimer, can_stat_update, 0); - mod_timer(&net->can.can_stattimer, + mod_timer(&net->can.stattimer, round_jiffies(jiffies + HZ)); } - net->can.can_stats->jiffies_init = jiffies; + net->can.pkg_stats->jiffies_init = jiffies; can_init_proc(net); } return 0; - out_free_can_stats: - kfree(net->can.can_stats); - out_free_alldev_list: - kfree(net->can.can_rx_alldev_list); + out_free_pkg_stats: + kfree(net->can.pkg_stats); + out_free_rx_alldev_list: + kfree(net->can.rx_alldev_list); out: return -ENOMEM; } static void can_pernet_exit(struct net *net) { - struct net_device *dev; - if (IS_ENABLED(CONFIG_PROC_FS)) { can_remove_proc(net); if (stats_timer) - del_timer_sync(&net->can.can_stattimer); - } - - /* remove created dev_rcv_lists from still registered CAN devices */ - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { - if (dev->type == ARPHRD_CAN && dev->ml_priv) { - struct can_dev_rcv_lists *d = dev->ml_priv; - - BUG_ON(d->entries); - kfree(d); - dev->ml_priv = NULL; - } + del_timer_sync(&net->can.stattimer); } - rcu_read_unlock(); - kfree(net->can.can_rx_alldev_list); - kfree(net->can.can_stats); - kfree(net->can.can_pstats); + kfree(net->can.rx_alldev_list); + kfree(net->can.pkg_stats); + kfree(net->can.rcv_lists_stats); } -/* - * af_can module init/exit functions - */ +/* af_can module init/exit functions */ static struct packet_type can_packet __read_mostly = { .type = cpu_to_be16(ETH_P_CAN), diff --git a/net/can/af_can.h b/net/can/af_can.h index ef21f7c6bc80..7c2d9161e224 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -1,6 +1,5 @@ /* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) */ -/* - * Copyright (c) 2002-2007 Volkswagen Group Electronic Research +/* Copyright (c) 2002-2007 Volkswagen Group Electronic Research * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -54,32 +53,17 @@ struct receiver { canid_t can_id; canid_t mask; unsigned long matches; - void (*func)(struct sk_buff *, void *); + void (*func)(struct sk_buff *skb, void *data); void *data; char *ident; struct sock *sk; struct rcu_head rcu; }; -#define CAN_SFF_RCV_ARRAY_SZ (1 << CAN_SFF_ID_BITS) -#define CAN_EFF_RCV_HASH_BITS 10 -#define CAN_EFF_RCV_ARRAY_SZ (1 << CAN_EFF_RCV_HASH_BITS) - -enum { RX_ERR, RX_ALL, RX_FIL, RX_INV, RX_MAX }; - -/* per device receive filters linked at dev->ml_priv */ -struct can_dev_rcv_lists { - struct hlist_head rx[RX_MAX]; - struct hlist_head rx_sff[CAN_SFF_RCV_ARRAY_SZ]; - struct hlist_head rx_eff[CAN_EFF_RCV_ARRAY_SZ]; - int remove_on_zero_entries; - int entries; -}; - /* statistic structures */ /* can be reset e.g. by can_init_stats() */ -struct s_stats { +struct can_pkg_stats { unsigned long jiffies_init; unsigned long rx_frames; @@ -104,7 +88,7 @@ struct s_stats { }; /* persistent statistics */ -struct s_pstats { +struct can_rcv_lists_stats { unsigned long stats_reset; unsigned long user_reset; unsigned long rcv_entries; diff --git a/net/can/bcm.c b/net/can/bcm.c index bf1d0bbecec8..c96fa0f33db3 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -106,7 +106,6 @@ struct bcm_op { unsigned long frames_abs, frames_filtered; struct bcm_timeval ival1, ival2; struct hrtimer timer, thrtimer; - struct tasklet_struct tsklet, thrtsklet; ktime_t rx_stamp, kt_ival1, kt_ival2, kt_lastmsg; int rx_ifindex; int cfsiz; @@ -371,25 +370,34 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, } } -static void bcm_tx_start_timer(struct bcm_op *op) +static bool bcm_tx_set_expiry(struct bcm_op *op, struct hrtimer *hrt) { + ktime_t ival; + if (op->kt_ival1 && op->count) - hrtimer_start(&op->timer, - ktime_add(ktime_get(), op->kt_ival1), - HRTIMER_MODE_ABS); + ival = op->kt_ival1; else if (op->kt_ival2) - hrtimer_start(&op->timer, - ktime_add(ktime_get(), op->kt_ival2), - HRTIMER_MODE_ABS); + ival = op->kt_ival2; + else + return false; + + hrtimer_set_expires(hrt, ktime_add(ktime_get(), ival)); + return true; } -static void bcm_tx_timeout_tsklet(unsigned long data) +static void bcm_tx_start_timer(struct bcm_op *op) { - struct bcm_op *op = (struct bcm_op *)data; + if (bcm_tx_set_expiry(op, &op->timer)) + hrtimer_start_expires(&op->timer, HRTIMER_MODE_ABS_SOFT); +} + +/* bcm_tx_timeout_handler - performs cyclic CAN frame transmissions */ +static enum hrtimer_restart bcm_tx_timeout_handler(struct hrtimer *hrtimer) +{ + struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); struct bcm_msg_head msg_head; if (op->kt_ival1 && (op->count > 0)) { - op->count--; if (!op->count && (op->flags & TX_COUNTEVT)) { @@ -406,22 +414,12 @@ static void bcm_tx_timeout_tsklet(unsigned long data) } bcm_can_tx(op); - } else if (op->kt_ival2) + } else if (op->kt_ival2) { bcm_can_tx(op); + } - bcm_tx_start_timer(op); -} - -/* - * bcm_tx_timeout_handler - performs cyclic CAN frame transmissions - */ -static enum hrtimer_restart bcm_tx_timeout_handler(struct hrtimer *hrtimer) -{ - struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); - - tasklet_schedule(&op->tsklet); - - return HRTIMER_NORESTART; + return bcm_tx_set_expiry(op, &op->timer) ? + HRTIMER_RESTART : HRTIMER_NORESTART; } /* @@ -487,7 +485,7 @@ static void bcm_rx_update_and_send(struct bcm_op *op, /* do not send the saved data - only start throttle timer */ hrtimer_start(&op->thrtimer, ktime_add(op->kt_lastmsg, op->kt_ival2), - HRTIMER_MODE_ABS); + HRTIMER_MODE_ABS_SOFT); return; } @@ -546,14 +544,21 @@ static void bcm_rx_starttimer(struct bcm_op *op) return; if (op->kt_ival1) - hrtimer_start(&op->timer, op->kt_ival1, HRTIMER_MODE_REL); + hrtimer_start(&op->timer, op->kt_ival1, HRTIMER_MODE_REL_SOFT); } -static void bcm_rx_timeout_tsklet(unsigned long data) +/* bcm_rx_timeout_handler - when the (cyclic) CAN frame reception timed out */ +static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) { - struct bcm_op *op = (struct bcm_op *)data; + struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); struct bcm_msg_head msg_head; + /* if user wants to be informed, when cyclic CAN-Messages come back */ + if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) { + /* clear received CAN frames to indicate 'nothing received' */ + memset(op->last_frames, 0, op->nframes * op->cfsiz); + } + /* create notification to user */ msg_head.opcode = RX_TIMEOUT; msg_head.flags = op->flags; @@ -564,25 +569,6 @@ static void bcm_rx_timeout_tsklet(unsigned long data) msg_head.nframes = 0; bcm_send_to_user(op, &msg_head, NULL, 0); -} - -/* - * bcm_rx_timeout_handler - when the (cyclic) CAN frame reception timed out - */ -static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) -{ - struct bcm_op *op = container_of(hrtimer, struct bcm_op, timer); - - /* schedule before NET_RX_SOFTIRQ */ - tasklet_hi_schedule(&op->tsklet); - - /* no restart of the timer is done here! */ - - /* if user wants to be informed, when cyclic CAN-Messages come back */ - if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) { - /* clear received CAN frames to indicate 'nothing received' */ - memset(op->last_frames, 0, op->nframes * op->cfsiz); - } return HRTIMER_NORESTART; } @@ -590,14 +576,12 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) /* * bcm_rx_do_flush - helper for bcm_rx_thr_flush */ -static inline int bcm_rx_do_flush(struct bcm_op *op, int update, - unsigned int index) +static inline int bcm_rx_do_flush(struct bcm_op *op, unsigned int index) { struct canfd_frame *lcf = op->last_frames + op->cfsiz * index; if ((op->last_frames) && (lcf->flags & RX_THR)) { - if (update) - bcm_rx_changed(op, lcf); + bcm_rx_changed(op, lcf); return 1; } return 0; @@ -605,11 +589,8 @@ static inline int bcm_rx_do_flush(struct bcm_op *op, int update, /* * bcm_rx_thr_flush - Check for throttled data and send it to the userspace - * - * update == 0 : just check if throttled data is available (any irq context) - * update == 1 : check and send throttled data to userspace (soft_irq context) */ -static int bcm_rx_thr_flush(struct bcm_op *op, int update) +static int bcm_rx_thr_flush(struct bcm_op *op) { int updated = 0; @@ -618,24 +599,16 @@ static int bcm_rx_thr_flush(struct bcm_op *op, int update) /* for MUX filter we start at index 1 */ for (i = 1; i < op->nframes; i++) - updated += bcm_rx_do_flush(op, update, i); + updated += bcm_rx_do_flush(op, i); } else { /* for RX_FILTER_ID and simple filter */ - updated += bcm_rx_do_flush(op, update, 0); + updated += bcm_rx_do_flush(op, 0); } return updated; } -static void bcm_rx_thr_tsklet(unsigned long data) -{ - struct bcm_op *op = (struct bcm_op *)data; - - /* push the changed data to the userspace */ - bcm_rx_thr_flush(op, 1); -} - /* * bcm_rx_thr_handler - the time for blocked content updates is over now: * Check for throttled data and send it to the userspace @@ -644,9 +617,7 @@ static enum hrtimer_restart bcm_rx_thr_handler(struct hrtimer *hrtimer) { struct bcm_op *op = container_of(hrtimer, struct bcm_op, thrtimer); - tasklet_schedule(&op->thrtsklet); - - if (bcm_rx_thr_flush(op, 0)) { + if (bcm_rx_thr_flush(op)) { hrtimer_forward(hrtimer, ktime_get(), op->kt_ival2); return HRTIMER_RESTART; } else { @@ -742,23 +713,8 @@ static struct bcm_op *bcm_find_op(struct list_head *ops, static void bcm_remove_op(struct bcm_op *op) { - if (op->tsklet.func) { - while (test_bit(TASKLET_STATE_SCHED, &op->tsklet.state) || - test_bit(TASKLET_STATE_RUN, &op->tsklet.state) || - hrtimer_active(&op->timer)) { - hrtimer_cancel(&op->timer); - tasklet_kill(&op->tsklet); - } - } - - if (op->thrtsklet.func) { - while (test_bit(TASKLET_STATE_SCHED, &op->thrtsklet.state) || - test_bit(TASKLET_STATE_RUN, &op->thrtsklet.state) || - hrtimer_active(&op->thrtimer)) { - hrtimer_cancel(&op->thrtimer); - tasklet_kill(&op->thrtsklet); - } - } + hrtimer_cancel(&op->timer); + hrtimer_cancel(&op->thrtimer); if ((op->frames) && (op->frames != &op->sframe)) kfree(op->frames); @@ -991,15 +947,13 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->ifindex = ifindex; /* initialize uninitialized (kzalloc) structure */ - hrtimer_init(&op->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&op->timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); op->timer.function = bcm_tx_timeout_handler; - /* initialize tasklet for tx countevent notification */ - tasklet_init(&op->tsklet, bcm_tx_timeout_tsklet, - (unsigned long) op); - /* currently unused in tx_ops */ - hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); /* add this bcm_op to the list of the tx_ops */ list_add(&op->list, &bo->tx_ops); @@ -1168,20 +1122,14 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->rx_ifindex = ifindex; /* initialize uninitialized (kzalloc) structure */ - hrtimer_init(&op->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&op->timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); op->timer.function = bcm_rx_timeout_handler; - /* initialize tasklet for rx timeout notification */ - tasklet_init(&op->tsklet, bcm_rx_timeout_tsklet, - (unsigned long) op); - - hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&op->thrtimer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); op->thrtimer.function = bcm_rx_thr_handler; - /* initialize tasklet for rx throttle handling */ - tasklet_init(&op->thrtsklet, bcm_rx_thr_tsklet, - (unsigned long) op); - /* add this bcm_op to the list of the rx_ops */ list_add(&op->list, &bo->rx_ops); @@ -1227,12 +1175,12 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, */ op->kt_lastmsg = 0; hrtimer_cancel(&op->thrtimer); - bcm_rx_thr_flush(op, 1); + bcm_rx_thr_flush(op); } if ((op->flags & STARTTIMER) && op->kt_ival1) hrtimer_start(&op->timer, op->kt_ival1, - HRTIMER_MODE_REL); + HRTIMER_MODE_REL_SOFT); } /* now we can register for can_ids, if we added a new bcm_op */ @@ -1346,7 +1294,7 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) /* no bound device as default => check msg_name */ DECLARE_SOCKADDR(struct sockaddr_can *, addr, msg->msg_name); - if (msg->msg_namelen < sizeof(*addr)) + if (msg->msg_namelen < CAN_REQUIRED_SIZE(*addr, can_ifindex)) return -EINVAL; if (addr->can_family != AF_CAN) @@ -1588,7 +1536,7 @@ static int bcm_connect(struct socket *sock, struct sockaddr *uaddr, int len, struct net *net = sock_net(sk); int ret = 0; - if (len < sizeof(*addr)) + if (len < CAN_REQUIRED_SIZE(*addr, can_ifindex)) return -EINVAL; lock_sock(sk); @@ -1680,8 +1628,8 @@ static int bcm_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, return size; } -int bcm_sock_no_ioctlcmd(struct socket *sock, unsigned int cmd, - unsigned long arg) +static int bcm_sock_no_ioctlcmd(struct socket *sock, unsigned int cmd, + unsigned long arg) { /* no ioctls for socket layer -> hand it down to NIC layer */ return -ENOIOCTLCMD; diff --git a/net/can/gw.c b/net/can/gw.c index ce17f836262b..65d60c93af29 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -1,8 +1,7 @@ // SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) -/* - * gw.c - CAN frame Gateway/Router/Bridge with netlink interface +/* gw.c - CAN frame Gateway/Router/Bridge with netlink interface * - * Copyright (c) 2017 Volkswagen Group Electronic Research + * Copyright (c) 2019 Volkswagen Group Electronic Research * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -60,7 +59,7 @@ #include <net/net_namespace.h> #include <net/sock.h> -#define CAN_GW_VERSION "20170425" +#define CAN_GW_VERSION "20190810" #define CAN_GW_NAME "can-gw" MODULE_DESCRIPTION("PF_CAN netlink gateway"); @@ -86,10 +85,10 @@ static struct kmem_cache *cgw_cache __read_mostly; /* structure that contains the (on-the-fly) CAN frame modifications */ struct cf_mod { struct { - struct can_frame and; - struct can_frame or; - struct can_frame xor; - struct can_frame set; + struct canfd_frame and; + struct canfd_frame or; + struct canfd_frame xor; + struct canfd_frame set; } modframe; struct { u8 and; @@ -97,7 +96,7 @@ struct cf_mod { u8 xor; u8 set; } modtype; - void (*modfunc[MAX_MODFUNCTIONS])(struct can_frame *cf, + void (*modfunc[MAX_MODFUNCTIONS])(struct canfd_frame *cf, struct cf_mod *mod); /* CAN frame checksum calculation after CAN frame modifications */ @@ -106,15 +105,15 @@ struct cf_mod { struct cgw_csum_crc8 crc8; } csum; struct { - void (*xor)(struct can_frame *cf, struct cgw_csum_xor *xor); - void (*crc8)(struct can_frame *cf, struct cgw_csum_crc8 *crc8); + void (*xor)(struct canfd_frame *cf, + struct cgw_csum_xor *xor); + void (*crc8)(struct canfd_frame *cf, + struct cgw_csum_crc8 *crc8); } csumfunc; u32 uid; }; - -/* - * So far we just support CAN -> CAN routing and frame modifications. +/* So far we just support CAN -> CAN routing and frame modifications. * * The internal can_can_gw structure contains data and attributes for * a CAN -> CAN gateway job. @@ -152,39 +151,88 @@ struct cgw_job { /* modification functions that are invoked in the hot path in can_can_gw_rcv */ -#define MODFUNC(func, op) static void func(struct can_frame *cf, \ +#define MODFUNC(func, op) static void func(struct canfd_frame *cf, \ struct cf_mod *mod) { op ; } MODFUNC(mod_and_id, cf->can_id &= mod->modframe.and.can_id) -MODFUNC(mod_and_dlc, cf->can_dlc &= mod->modframe.and.can_dlc) +MODFUNC(mod_and_len, cf->len &= mod->modframe.and.len) +MODFUNC(mod_and_flags, cf->flags &= mod->modframe.and.flags) MODFUNC(mod_and_data, *(u64 *)cf->data &= *(u64 *)mod->modframe.and.data) MODFUNC(mod_or_id, cf->can_id |= mod->modframe.or.can_id) -MODFUNC(mod_or_dlc, cf->can_dlc |= mod->modframe.or.can_dlc) +MODFUNC(mod_or_len, cf->len |= mod->modframe.or.len) +MODFUNC(mod_or_flags, cf->flags |= mod->modframe.or.flags) MODFUNC(mod_or_data, *(u64 *)cf->data |= *(u64 *)mod->modframe.or.data) MODFUNC(mod_xor_id, cf->can_id ^= mod->modframe.xor.can_id) -MODFUNC(mod_xor_dlc, cf->can_dlc ^= mod->modframe.xor.can_dlc) +MODFUNC(mod_xor_len, cf->len ^= mod->modframe.xor.len) +MODFUNC(mod_xor_flags, cf->flags ^= mod->modframe.xor.flags) MODFUNC(mod_xor_data, *(u64 *)cf->data ^= *(u64 *)mod->modframe.xor.data) MODFUNC(mod_set_id, cf->can_id = mod->modframe.set.can_id) -MODFUNC(mod_set_dlc, cf->can_dlc = mod->modframe.set.can_dlc) +MODFUNC(mod_set_len, cf->len = mod->modframe.set.len) +MODFUNC(mod_set_flags, cf->flags = mod->modframe.set.flags) MODFUNC(mod_set_data, *(u64 *)cf->data = *(u64 *)mod->modframe.set.data) -static inline void canframecpy(struct can_frame *dst, struct can_frame *src) +static void mod_and_fddata(struct canfd_frame *cf, struct cf_mod *mod) +{ + int i; + + for (i = 0; i < CANFD_MAX_DLEN; i += 8) + *(u64 *)(cf->data + i) &= *(u64 *)(mod->modframe.and.data + i); +} + +static void mod_or_fddata(struct canfd_frame *cf, struct cf_mod *mod) +{ + int i; + + for (i = 0; i < CANFD_MAX_DLEN; i += 8) + *(u64 *)(cf->data + i) |= *(u64 *)(mod->modframe.or.data + i); +} + +static void mod_xor_fddata(struct canfd_frame *cf, struct cf_mod *mod) +{ + int i; + + for (i = 0; i < CANFD_MAX_DLEN; i += 8) + *(u64 *)(cf->data + i) ^= *(u64 *)(mod->modframe.xor.data + i); +} + +static void mod_set_fddata(struct canfd_frame *cf, struct cf_mod *mod) +{ + memcpy(cf->data, mod->modframe.set.data, CANFD_MAX_DLEN); +} + +static void canframecpy(struct canfd_frame *dst, struct can_frame *src) { - /* - * Copy the struct members separately to ensure that no uninitialized + /* Copy the struct members separately to ensure that no uninitialized * data are copied in the 3 bytes hole of the struct. This is needed * to make easy compares of the data in the struct cf_mod. */ dst->can_id = src->can_id; - dst->can_dlc = src->can_dlc; + dst->len = src->can_dlc; *(u64 *)dst->data = *(u64 *)src->data; } -static int cgw_chk_csum_parms(s8 fr, s8 to, s8 re) +static void canfdframecpy(struct canfd_frame *dst, struct canfd_frame *src) { - /* - * absolute dlc values 0 .. 7 => 0 .. 7, e.g. data [0] + /* Copy the struct members separately to ensure that no uninitialized + * data are copied in the 2 bytes hole of the struct. This is needed + * to make easy compares of the data in the struct cf_mod. + */ + + dst->can_id = src->can_id; + dst->flags = src->flags; + dst->len = src->len; + memcpy(dst->data, src->data, CANFD_MAX_DLEN); +} + +static int cgw_chk_csum_parms(s8 fr, s8 to, s8 re, struct rtcanmsg *r) +{ + s8 dlen = CAN_MAX_DLEN; + + if (r->flags & CGW_FLAGS_CAN_FD) + dlen = CANFD_MAX_DLEN; + + /* absolute dlc values 0 .. 7 => 0 .. 7, e.g. data [0] * relative to received dlc -1 .. -8 : * e.g. for received dlc = 8 * -1 => index = 7 (data[7]) @@ -192,27 +240,27 @@ static int cgw_chk_csum_parms(s8 fr, s8 to, s8 re) * -8 => index = 0 (data[0]) */ - if (fr > -9 && fr < 8 && - to > -9 && to < 8 && - re > -9 && re < 8) + if (fr >= -dlen && fr < dlen && + to >= -dlen && to < dlen && + re >= -dlen && re < dlen) return 0; else return -EINVAL; } -static inline int calc_idx(int idx, int rx_dlc) +static inline int calc_idx(int idx, int rx_len) { if (idx < 0) - return rx_dlc + idx; + return rx_len + idx; else return idx; } -static void cgw_csum_xor_rel(struct can_frame *cf, struct cgw_csum_xor *xor) +static void cgw_csum_xor_rel(struct canfd_frame *cf, struct cgw_csum_xor *xor) { - int from = calc_idx(xor->from_idx, cf->can_dlc); - int to = calc_idx(xor->to_idx, cf->can_dlc); - int res = calc_idx(xor->result_idx, cf->can_dlc); + int from = calc_idx(xor->from_idx, cf->len); + int to = calc_idx(xor->to_idx, cf->len); + int res = calc_idx(xor->result_idx, cf->len); u8 val = xor->init_xor_val; int i; @@ -230,7 +278,7 @@ static void cgw_csum_xor_rel(struct can_frame *cf, struct cgw_csum_xor *xor) cf->data[res] = val; } -static void cgw_csum_xor_pos(struct can_frame *cf, struct cgw_csum_xor *xor) +static void cgw_csum_xor_pos(struct canfd_frame *cf, struct cgw_csum_xor *xor) { u8 val = xor->init_xor_val; int i; @@ -241,7 +289,7 @@ static void cgw_csum_xor_pos(struct can_frame *cf, struct cgw_csum_xor *xor) cf->data[xor->result_idx] = val; } -static void cgw_csum_xor_neg(struct can_frame *cf, struct cgw_csum_xor *xor) +static void cgw_csum_xor_neg(struct canfd_frame *cf, struct cgw_csum_xor *xor) { u8 val = xor->init_xor_val; int i; @@ -252,11 +300,12 @@ static void cgw_csum_xor_neg(struct can_frame *cf, struct cgw_csum_xor *xor) cf->data[xor->result_idx] = val; } -static void cgw_csum_crc8_rel(struct can_frame *cf, struct cgw_csum_crc8 *crc8) +static void cgw_csum_crc8_rel(struct canfd_frame *cf, + struct cgw_csum_crc8 *crc8) { - int from = calc_idx(crc8->from_idx, cf->can_dlc); - int to = calc_idx(crc8->to_idx, cf->can_dlc); - int res = calc_idx(crc8->result_idx, cf->can_dlc); + int from = calc_idx(crc8->from_idx, cf->len); + int to = calc_idx(crc8->to_idx, cf->len); + int res = calc_idx(crc8->result_idx, cf->len); u8 crc = crc8->init_crc_val; int i; @@ -265,96 +314,102 @@ static void cgw_csum_crc8_rel(struct can_frame *cf, struct cgw_csum_crc8 *crc8) if (from <= to) { for (i = crc8->from_idx; i <= crc8->to_idx; i++) - crc = crc8->crctab[crc^cf->data[i]]; + crc = crc8->crctab[crc ^ cf->data[i]]; } else { for (i = crc8->from_idx; i >= crc8->to_idx; i--) - crc = crc8->crctab[crc^cf->data[i]]; + crc = crc8->crctab[crc ^ cf->data[i]]; } switch (crc8->profile) { - case CGW_CRC8PRF_1U8: - crc = crc8->crctab[crc^crc8->profile_data[0]]; + crc = crc8->crctab[crc ^ crc8->profile_data[0]]; break; case CGW_CRC8PRF_16U8: - crc = crc8->crctab[crc^crc8->profile_data[cf->data[1] & 0xF]]; + crc = crc8->crctab[crc ^ crc8->profile_data[cf->data[1] & 0xF]]; break; case CGW_CRC8PRF_SFFID_XOR: - crc = crc8->crctab[crc^(cf->can_id & 0xFF)^ + crc = crc8->crctab[crc ^ (cf->can_id & 0xFF) ^ (cf->can_id >> 8 & 0xFF)]; break; - } - cf->data[crc8->result_idx] = crc^crc8->final_xor_val; + cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val; } -static void cgw_csum_crc8_pos(struct can_frame *cf, struct cgw_csum_crc8 *crc8) +static void cgw_csum_crc8_pos(struct canfd_frame *cf, + struct cgw_csum_crc8 *crc8) { u8 crc = crc8->init_crc_val; int i; for (i = crc8->from_idx; i <= crc8->to_idx; i++) - crc = crc8->crctab[crc^cf->data[i]]; + crc = crc8->crctab[crc ^ cf->data[i]]; switch (crc8->profile) { - case CGW_CRC8PRF_1U8: - crc = crc8->crctab[crc^crc8->profile_data[0]]; + crc = crc8->crctab[crc ^ crc8->profile_data[0]]; break; case CGW_CRC8PRF_16U8: - crc = crc8->crctab[crc^crc8->profile_data[cf->data[1] & 0xF]]; + crc = crc8->crctab[crc ^ crc8->profile_data[cf->data[1] & 0xF]]; break; case CGW_CRC8PRF_SFFID_XOR: - crc = crc8->crctab[crc^(cf->can_id & 0xFF)^ + crc = crc8->crctab[crc ^ (cf->can_id & 0xFF) ^ (cf->can_id >> 8 & 0xFF)]; break; } - cf->data[crc8->result_idx] = crc^crc8->final_xor_val; + cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val; } -static void cgw_csum_crc8_neg(struct can_frame *cf, struct cgw_csum_crc8 *crc8) +static void cgw_csum_crc8_neg(struct canfd_frame *cf, + struct cgw_csum_crc8 *crc8) { u8 crc = crc8->init_crc_val; int i; for (i = crc8->from_idx; i >= crc8->to_idx; i--) - crc = crc8->crctab[crc^cf->data[i]]; + crc = crc8->crctab[crc ^ cf->data[i]]; switch (crc8->profile) { - case CGW_CRC8PRF_1U8: - crc = crc8->crctab[crc^crc8->profile_data[0]]; + crc = crc8->crctab[crc ^ crc8->profile_data[0]]; break; case CGW_CRC8PRF_16U8: - crc = crc8->crctab[crc^crc8->profile_data[cf->data[1] & 0xF]]; + crc = crc8->crctab[crc ^ crc8->profile_data[cf->data[1] & 0xF]]; break; case CGW_CRC8PRF_SFFID_XOR: - crc = crc8->crctab[crc^(cf->can_id & 0xFF)^ + crc = crc8->crctab[crc ^ (cf->can_id & 0xFF) ^ (cf->can_id >> 8 & 0xFF)]; break; } - cf->data[crc8->result_idx] = crc^crc8->final_xor_val; + cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val; } /* the receive & process & send function */ static void can_can_gw_rcv(struct sk_buff *skb, void *data) { struct cgw_job *gwj = (struct cgw_job *)data; - struct can_frame *cf; + struct canfd_frame *cf; struct sk_buff *nskb; int modidx = 0; - /* - * Do not handle CAN frames routed more than 'max_hops' times. + /* process strictly Classic CAN or CAN FD frames */ + if (gwj->flags & CGW_FLAGS_CAN_FD) { + if (skb->len != CANFD_MTU) + return; + } else { + if (skb->len != CAN_MTU) + return; + } + + /* Do not handle CAN frames routed more than 'max_hops' times. * In general we should never catch this delimiter which is intended * to cover a misconfiguration protection (e.g. circular CAN routes). * @@ -385,8 +440,7 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) can_skb_prv(skb)->ifindex == gwj->dst.dev->ifindex) return; - /* - * clone the given skb, which has not been done in can_rcv() + /* clone the given skb, which has not been done in can_rcv() * * When there is at least one modification function activated, * we need to copy the skb as we want to modify skb->data. @@ -411,7 +465,7 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) nskb->dev = gwj->dst.dev; /* pointer to modifiable CAN frame */ - cf = (struct can_frame *)nskb->data; + cf = (struct canfd_frame *)nskb->data; /* perform preprocessed modification functions if there are any */ while (modidx < MAX_MODFUNCTIONS && gwj->mod.modfunc[modidx]) @@ -420,26 +474,22 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) /* Has the CAN frame been modified? */ if (modidx) { /* get available space for the processed CAN frame type */ - int max_len = nskb->len - offsetof(struct can_frame, data); + int max_len = nskb->len - offsetof(struct canfd_frame, data); /* dlc may have changed, make sure it fits to the CAN frame */ - if (cf->can_dlc > max_len) - goto out_delete; - - /* check for checksum updates in classic CAN length only */ - if (gwj->mod.csumfunc.crc8) { - if (cf->can_dlc > 8) - goto out_delete; - - (*gwj->mod.csumfunc.crc8)(cf, &gwj->mod.csum.crc8); + if (cf->len > max_len) { + /* delete frame due to misconfiguration */ + gwj->deleted_frames++; + kfree_skb(nskb); + return; } - if (gwj->mod.csumfunc.xor) { - if (cf->can_dlc > 8) - goto out_delete; + /* check for checksum updates */ + if (gwj->mod.csumfunc.crc8) + (*gwj->mod.csumfunc.crc8)(cf, &gwj->mod.csum.crc8); + if (gwj->mod.csumfunc.xor) (*gwj->mod.csumfunc.xor)(cf, &gwj->mod.csum.xor); - } } /* clear the skb timestamp if not configured the other way */ @@ -451,14 +501,6 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) gwj->dropped_frames++; else gwj->handled_frames++; - - return; - - out_delete: - /* delete frame due to misconfiguration */ - gwj->deleted_frames++; - kfree_skb(nskb); - return; } static inline int cgw_register_filter(struct net *net, struct cgw_job *gwj) @@ -484,14 +526,12 @@ static int cgw_notifier(struct notifier_block *nb, return NOTIFY_DONE; if (msg == NETDEV_UNREGISTER) { - struct cgw_job *gwj = NULL; struct hlist_node *nx; ASSERT_RTNL(); hlist_for_each_entry_safe(gwj, nx, &net->can.cgw_list, list) { - if (gwj->src.dev == dev || gwj->dst.dev == dev) { hlist_del(&gwj->list); cgw_unregister_filter(net, gwj); @@ -506,7 +546,6 @@ static int cgw_notifier(struct notifier_block *nb, static int cgw_put_job(struct sk_buff *skb, struct cgw_job *gwj, int type, u32 pid, u32 seq, int flags) { - struct cgw_frame_mod mb; struct rtcanmsg *rtcan; struct nlmsghdr *nlh; @@ -543,32 +582,66 @@ static int cgw_put_job(struct sk_buff *skb, struct cgw_job *gwj, int type, goto cancel; } - if (gwj->mod.modtype.and) { - memcpy(&mb.cf, &gwj->mod.modframe.and, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.and; - if (nla_put(skb, CGW_MOD_AND, sizeof(mb), &mb) < 0) - goto cancel; - } + if (gwj->flags & CGW_FLAGS_CAN_FD) { + struct cgw_fdframe_mod mb; - if (gwj->mod.modtype.or) { - memcpy(&mb.cf, &gwj->mod.modframe.or, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.or; - if (nla_put(skb, CGW_MOD_OR, sizeof(mb), &mb) < 0) - goto cancel; - } + if (gwj->mod.modtype.and) { + memcpy(&mb.cf, &gwj->mod.modframe.and, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.and; + if (nla_put(skb, CGW_FDMOD_AND, sizeof(mb), &mb) < 0) + goto cancel; + } - if (gwj->mod.modtype.xor) { - memcpy(&mb.cf, &gwj->mod.modframe.xor, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.xor; - if (nla_put(skb, CGW_MOD_XOR, sizeof(mb), &mb) < 0) - goto cancel; - } + if (gwj->mod.modtype.or) { + memcpy(&mb.cf, &gwj->mod.modframe.or, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.or; + if (nla_put(skb, CGW_FDMOD_OR, sizeof(mb), &mb) < 0) + goto cancel; + } - if (gwj->mod.modtype.set) { - memcpy(&mb.cf, &gwj->mod.modframe.set, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.set; - if (nla_put(skb, CGW_MOD_SET, sizeof(mb), &mb) < 0) - goto cancel; + if (gwj->mod.modtype.xor) { + memcpy(&mb.cf, &gwj->mod.modframe.xor, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.xor; + if (nla_put(skb, CGW_FDMOD_XOR, sizeof(mb), &mb) < 0) + goto cancel; + } + + if (gwj->mod.modtype.set) { + memcpy(&mb.cf, &gwj->mod.modframe.set, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.set; + if (nla_put(skb, CGW_FDMOD_SET, sizeof(mb), &mb) < 0) + goto cancel; + } + } else { + struct cgw_frame_mod mb; + + if (gwj->mod.modtype.and) { + memcpy(&mb.cf, &gwj->mod.modframe.and, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.and; + if (nla_put(skb, CGW_MOD_AND, sizeof(mb), &mb) < 0) + goto cancel; + } + + if (gwj->mod.modtype.or) { + memcpy(&mb.cf, &gwj->mod.modframe.or, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.or; + if (nla_put(skb, CGW_MOD_OR, sizeof(mb), &mb) < 0) + goto cancel; + } + + if (gwj->mod.modtype.xor) { + memcpy(&mb.cf, &gwj->mod.modframe.xor, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.xor; + if (nla_put(skb, CGW_MOD_XOR, sizeof(mb), &mb) < 0) + goto cancel; + } + + if (gwj->mod.modtype.set) { + memcpy(&mb.cf, &gwj->mod.modframe.set, sizeof(mb.cf)); + mb.modtype = gwj->mod.modtype.set; + if (nla_put(skb, CGW_MOD_SET, sizeof(mb), &mb) < 0) + goto cancel; + } } if (gwj->mod.uid) { @@ -589,7 +662,6 @@ static int cgw_put_job(struct sk_buff *skb, struct cgw_job *gwj, int type, } if (gwj->gwtype == CGW_TYPE_CAN_CAN) { - if (gwj->ccgw.filter.can_id || gwj->ccgw.filter.can_mask) { if (nla_put(skb, CGW_FILTER, sizeof(struct can_filter), &gwj->ccgw.filter) < 0) @@ -624,8 +696,9 @@ static int cgw_dump_jobs(struct sk_buff *skb, struct netlink_callback *cb) if (idx < s_idx) goto cont; - if (cgw_put_job(skb, gwj, RTM_NEWROUTE, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI) < 0) + if (cgw_put_job(skb, gwj, RTM_NEWROUTE, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI) < 0) break; cont: idx++; @@ -637,7 +710,7 @@ cont: return skb->len; } -static const struct nla_policy cgw_policy[CGW_MAX+1] = { +static const struct nla_policy cgw_policy[CGW_MAX + 1] = { [CGW_MOD_AND] = { .len = sizeof(struct cgw_frame_mod) }, [CGW_MOD_OR] = { .len = sizeof(struct cgw_frame_mod) }, [CGW_MOD_XOR] = { .len = sizeof(struct cgw_frame_mod) }, @@ -649,14 +722,18 @@ static const struct nla_policy cgw_policy[CGW_MAX+1] = { [CGW_FILTER] = { .len = sizeof(struct can_filter) }, [CGW_LIM_HOPS] = { .type = NLA_U8 }, [CGW_MOD_UID] = { .type = NLA_U32 }, + [CGW_FDMOD_AND] = { .len = sizeof(struct cgw_fdframe_mod) }, + [CGW_FDMOD_OR] = { .len = sizeof(struct cgw_fdframe_mod) }, + [CGW_FDMOD_XOR] = { .len = sizeof(struct cgw_fdframe_mod) }, + [CGW_FDMOD_SET] = { .len = sizeof(struct cgw_fdframe_mod) }, }; /* check for common and gwtype specific attributes */ static int cgw_parse_attr(struct nlmsghdr *nlh, struct cf_mod *mod, u8 gwtype, void *gwtypeattr, u8 *limhops) { - struct nlattr *tb[CGW_MAX+1]; - struct cgw_frame_mod mb; + struct nlattr *tb[CGW_MAX + 1]; + struct rtcanmsg *r = nlmsg_data(nlh); int modidx = 0; int err = 0; @@ -676,87 +753,166 @@ static int cgw_parse_attr(struct nlmsghdr *nlh, struct cf_mod *mod, } /* check for AND/OR/XOR/SET modifications */ + if (r->flags & CGW_FLAGS_CAN_FD) { + struct cgw_fdframe_mod mb; - if (tb[CGW_MOD_AND]) { - nla_memcpy(&mb, tb[CGW_MOD_AND], CGW_MODATTR_LEN); + if (tb[CGW_FDMOD_AND]) { + nla_memcpy(&mb, tb[CGW_FDMOD_AND], CGW_FDMODATTR_LEN); - canframecpy(&mod->modframe.and, &mb.cf); - mod->modtype.and = mb.modtype; + canfdframecpy(&mod->modframe.and, &mb.cf); + mod->modtype.and = mb.modtype; - if (mb.modtype & CGW_MOD_ID) - mod->modfunc[modidx++] = mod_and_id; + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_and_id; - if (mb.modtype & CGW_MOD_DLC) - mod->modfunc[modidx++] = mod_and_dlc; + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_and_len; - if (mb.modtype & CGW_MOD_DATA) - mod->modfunc[modidx++] = mod_and_data; - } + if (mb.modtype & CGW_MOD_FLAGS) + mod->modfunc[modidx++] = mod_and_flags; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_and_fddata; + } - if (tb[CGW_MOD_OR]) { - nla_memcpy(&mb, tb[CGW_MOD_OR], CGW_MODATTR_LEN); + if (tb[CGW_FDMOD_OR]) { + nla_memcpy(&mb, tb[CGW_FDMOD_OR], CGW_FDMODATTR_LEN); - canframecpy(&mod->modframe.or, &mb.cf); - mod->modtype.or = mb.modtype; + canfdframecpy(&mod->modframe.or, &mb.cf); + mod->modtype.or = mb.modtype; - if (mb.modtype & CGW_MOD_ID) - mod->modfunc[modidx++] = mod_or_id; + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_or_id; - if (mb.modtype & CGW_MOD_DLC) - mod->modfunc[modidx++] = mod_or_dlc; + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_or_len; - if (mb.modtype & CGW_MOD_DATA) - mod->modfunc[modidx++] = mod_or_data; - } + if (mb.modtype & CGW_MOD_FLAGS) + mod->modfunc[modidx++] = mod_or_flags; - if (tb[CGW_MOD_XOR]) { - nla_memcpy(&mb, tb[CGW_MOD_XOR], CGW_MODATTR_LEN); + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_or_fddata; + } - canframecpy(&mod->modframe.xor, &mb.cf); - mod->modtype.xor = mb.modtype; + if (tb[CGW_FDMOD_XOR]) { + nla_memcpy(&mb, tb[CGW_FDMOD_XOR], CGW_FDMODATTR_LEN); - if (mb.modtype & CGW_MOD_ID) - mod->modfunc[modidx++] = mod_xor_id; + canfdframecpy(&mod->modframe.xor, &mb.cf); + mod->modtype.xor = mb.modtype; - if (mb.modtype & CGW_MOD_DLC) - mod->modfunc[modidx++] = mod_xor_dlc; + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_xor_id; - if (mb.modtype & CGW_MOD_DATA) - mod->modfunc[modidx++] = mod_xor_data; - } + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_xor_len; - if (tb[CGW_MOD_SET]) { - nla_memcpy(&mb, tb[CGW_MOD_SET], CGW_MODATTR_LEN); + if (mb.modtype & CGW_MOD_FLAGS) + mod->modfunc[modidx++] = mod_xor_flags; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_xor_fddata; + } - canframecpy(&mod->modframe.set, &mb.cf); - mod->modtype.set = mb.modtype; + if (tb[CGW_FDMOD_SET]) { + nla_memcpy(&mb, tb[CGW_FDMOD_SET], CGW_FDMODATTR_LEN); + + canfdframecpy(&mod->modframe.set, &mb.cf); + mod->modtype.set = mb.modtype; + + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_set_id; + + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_set_len; + + if (mb.modtype & CGW_MOD_FLAGS) + mod->modfunc[modidx++] = mod_set_flags; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_set_fddata; + } + } else { + struct cgw_frame_mod mb; - if (mb.modtype & CGW_MOD_ID) - mod->modfunc[modidx++] = mod_set_id; + if (tb[CGW_MOD_AND]) { + nla_memcpy(&mb, tb[CGW_MOD_AND], CGW_MODATTR_LEN); - if (mb.modtype & CGW_MOD_DLC) - mod->modfunc[modidx++] = mod_set_dlc; + canframecpy(&mod->modframe.and, &mb.cf); + mod->modtype.and = mb.modtype; - if (mb.modtype & CGW_MOD_DATA) - mod->modfunc[modidx++] = mod_set_data; + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_and_id; + + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_and_len; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_and_data; + } + + if (tb[CGW_MOD_OR]) { + nla_memcpy(&mb, tb[CGW_MOD_OR], CGW_MODATTR_LEN); + + canframecpy(&mod->modframe.or, &mb.cf); + mod->modtype.or = mb.modtype; + + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_or_id; + + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_or_len; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_or_data; + } + + if (tb[CGW_MOD_XOR]) { + nla_memcpy(&mb, tb[CGW_MOD_XOR], CGW_MODATTR_LEN); + + canframecpy(&mod->modframe.xor, &mb.cf); + mod->modtype.xor = mb.modtype; + + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_xor_id; + + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_xor_len; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_xor_data; + } + + if (tb[CGW_MOD_SET]) { + nla_memcpy(&mb, tb[CGW_MOD_SET], CGW_MODATTR_LEN); + + canframecpy(&mod->modframe.set, &mb.cf); + mod->modtype.set = mb.modtype; + + if (mb.modtype & CGW_MOD_ID) + mod->modfunc[modidx++] = mod_set_id; + + if (mb.modtype & CGW_MOD_LEN) + mod->modfunc[modidx++] = mod_set_len; + + if (mb.modtype & CGW_MOD_DATA) + mod->modfunc[modidx++] = mod_set_data; + } } /* check for checksum operations after CAN frame modifications */ if (modidx) { - if (tb[CGW_CS_CRC8]) { struct cgw_csum_crc8 *c = nla_data(tb[CGW_CS_CRC8]); err = cgw_chk_csum_parms(c->from_idx, c->to_idx, - c->result_idx); + c->result_idx, r); if (err) return err; nla_memcpy(&mod->csum.crc8, tb[CGW_CS_CRC8], CGW_CS_CRC8_LEN); - /* - * select dedicated processing function to reduce + /* select dedicated processing function to reduce * runtime operations in receive hot path. */ if (c->from_idx < 0 || c->to_idx < 0 || @@ -772,15 +928,14 @@ static int cgw_parse_attr(struct nlmsghdr *nlh, struct cf_mod *mod, struct cgw_csum_xor *c = nla_data(tb[CGW_CS_XOR]); err = cgw_chk_csum_parms(c->from_idx, c->to_idx, - c->result_idx); + c->result_idx, r); if (err) return err; nla_memcpy(&mod->csum.xor, tb[CGW_CS_XOR], CGW_CS_XOR_LEN); - /* - * select dedicated processing function to reduce + /* select dedicated processing function to reduce * runtime operations in receive hot path. */ if (c->from_idx < 0 || c->to_idx < 0 || @@ -792,16 +947,14 @@ static int cgw_parse_attr(struct nlmsghdr *nlh, struct cf_mod *mod, mod->csumfunc.xor = cgw_csum_xor_neg; } - if (tb[CGW_MOD_UID]) { + if (tb[CGW_MOD_UID]) nla_memcpy(&mod->uid, tb[CGW_MOD_UID], sizeof(u32)); - } } if (gwtype == CGW_TYPE_CAN_CAN) { - /* check CGW_TYPE_CAN_CAN specific attributes */ - struct can_can_gw *ccgw = (struct can_can_gw *)gwtypeattr; + memset(ccgw, 0, sizeof(*ccgw)); /* check for can_filter in attributes */ @@ -862,12 +1015,10 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, return err; if (mod.uid) { - ASSERT_RTNL(); /* check for updating an existing job with identical uid */ hlist_for_each_entry(gwj, &net->can.cgw_list, list) { - if (gwj->mod.uid != mod.uid) continue; @@ -988,7 +1139,6 @@ static int cgw_remove_job(struct sk_buff *skb, struct nlmsghdr *nlh, /* remove only the first matching entry */ hlist_for_each_entry_safe(gwj, nx, &net->can.cgw_list, list) { - if (gwj->flags != r->flags) continue; diff --git a/net/can/j1939/Kconfig b/net/can/j1939/Kconfig new file mode 100644 index 000000000000..2998298b71ec --- /dev/null +++ b/net/can/j1939/Kconfig @@ -0,0 +1,15 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# SAE J1939 network layer core configuration +# + +config CAN_J1939 + tristate "SAE J1939" + depends on CAN + help + SAE J1939 + Say Y to have in-kernel support for j1939 socket type. This + allows communication according to SAE j1939. + The relevant parts in kernel are + SAE j1939-21 (datalink & transport protocol) + & SAE j1939-81 (network management). diff --git a/net/can/j1939/Makefile b/net/can/j1939/Makefile new file mode 100644 index 000000000000..19181bdae173 --- /dev/null +++ b/net/can/j1939/Makefile @@ -0,0 +1,10 @@ +# SPDX-License-Identifier: GPL-2.0 + +obj-$(CONFIG_CAN_J1939) += can-j1939.o + +can-j1939-objs := \ + address-claim.o \ + bus.o \ + main.o \ + socket.o \ + transport.o diff --git a/net/can/j1939/address-claim.c b/net/can/j1939/address-claim.c new file mode 100644 index 000000000000..f33c47327927 --- /dev/null +++ b/net/can/j1939/address-claim.c @@ -0,0 +1,230 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2010-2011 EIA Electronics, +// Pieter Beyens <pieter.beyens@eia.be> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +/* J1939 Address Claiming. + * Address Claiming in the kernel + * - keeps track of the AC states of ECU's, + * - resolves NAME<=>SA taking into account the AC states of ECU's. + * + * All Address Claim msgs (including host-originated msg) are processed + * at the receive path (a sent msg is always received again via CAN echo). + * As such, the processing of AC msgs is done in the order on which msgs + * are sent on the bus. + * + * This module doesn't send msgs itself (e.g. replies on Address Claims), + * this is the responsibility of a user space application or daemon. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/netdevice.h> +#include <linux/skbuff.h> + +#include "j1939-priv.h" + +static inline name_t j1939_skb_to_name(const struct sk_buff *skb) +{ + return le64_to_cpup((__le64 *)skb->data); +} + +static inline bool j1939_ac_msg_is_request(struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + int req_pgn; + + if (skb->len < 3 || skcb->addr.pgn != J1939_PGN_REQUEST) + return false; + + req_pgn = skb->data[0] | (skb->data[1] << 8) | (skb->data[2] << 16); + + return req_pgn == J1939_PGN_ADDRESS_CLAIMED; +} + +static int j1939_ac_verify_outgoing(struct j1939_priv *priv, + struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + + if (skb->len != 8) { + netdev_notice(priv->ndev, "tx address claim with dlc %i\n", + skb->len); + return -EPROTO; + } + + if (skcb->addr.src_name != j1939_skb_to_name(skb)) { + netdev_notice(priv->ndev, "tx address claim with different name\n"); + return -EPROTO; + } + + if (skcb->addr.sa == J1939_NO_ADDR) { + netdev_notice(priv->ndev, "tx address claim with broadcast sa\n"); + return -EPROTO; + } + + /* ac must always be a broadcast */ + if (skcb->addr.dst_name || skcb->addr.da != J1939_NO_ADDR) { + netdev_notice(priv->ndev, "tx address claim with dest, not broadcast\n"); + return -EPROTO; + } + return 0; +} + +int j1939_ac_fixup(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + int ret; + u8 addr; + + /* network mgmt: address claiming msgs */ + if (skcb->addr.pgn == J1939_PGN_ADDRESS_CLAIMED) { + struct j1939_ecu *ecu; + + ret = j1939_ac_verify_outgoing(priv, skb); + /* return both when failure & when successful */ + if (ret < 0) + return ret; + ecu = j1939_ecu_get_by_name(priv, skcb->addr.src_name); + if (!ecu) + return -ENODEV; + + if (ecu->addr != skcb->addr.sa) + /* hold further traffic for ecu, remove from parent */ + j1939_ecu_unmap(ecu); + j1939_ecu_put(ecu); + } else if (skcb->addr.src_name) { + /* assign source address */ + addr = j1939_name_to_addr(priv, skcb->addr.src_name); + if (!j1939_address_is_unicast(addr) && + !j1939_ac_msg_is_request(skb)) { + netdev_notice(priv->ndev, "tx drop: invalid sa for name 0x%016llx\n", + skcb->addr.src_name); + return -EADDRNOTAVAIL; + } + skcb->addr.sa = addr; + } + + /* assign destination address */ + if (skcb->addr.dst_name) { + addr = j1939_name_to_addr(priv, skcb->addr.dst_name); + if (!j1939_address_is_unicast(addr)) { + netdev_notice(priv->ndev, "tx drop: invalid da for name 0x%016llx\n", + skcb->addr.dst_name); + return -EADDRNOTAVAIL; + } + skcb->addr.da = addr; + } + return 0; +} + +static void j1939_ac_process(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_ecu *ecu, *prev; + name_t name; + + if (skb->len != 8) { + netdev_notice(priv->ndev, "rx address claim with wrong dlc %i\n", + skb->len); + return; + } + + name = j1939_skb_to_name(skb); + skcb->addr.src_name = name; + if (!name) { + netdev_notice(priv->ndev, "rx address claim without name\n"); + return; + } + + if (!j1939_address_is_valid(skcb->addr.sa)) { + netdev_notice(priv->ndev, "rx address claim with broadcast sa\n"); + return; + } + + write_lock_bh(&priv->lock); + + /* Few words on the ECU ref counting: + * + * First we get an ECU handle, either with + * j1939_ecu_get_by_name_locked() (increments the ref counter) + * or j1939_ecu_create_locked() (initializes an ECU object + * with a ref counter of 1). + * + * j1939_ecu_unmap_locked() will decrement the ref counter, + * but only if the ECU was mapped before. So "ecu" still + * belongs to us. + * + * j1939_ecu_timer_start() will increment the ref counter + * before it starts the timer, so we can put the ecu when + * leaving this function. + */ + ecu = j1939_ecu_get_by_name_locked(priv, name); + if (!ecu && j1939_address_is_unicast(skcb->addr.sa)) + ecu = j1939_ecu_create_locked(priv, name); + + if (IS_ERR_OR_NULL(ecu)) + goto out_unlock_bh; + + /* cancel pending (previous) address claim */ + j1939_ecu_timer_cancel(ecu); + + if (j1939_address_is_idle(skcb->addr.sa)) { + j1939_ecu_unmap_locked(ecu); + goto out_ecu_put; + } + + /* save new addr */ + if (ecu->addr != skcb->addr.sa) + j1939_ecu_unmap_locked(ecu); + ecu->addr = skcb->addr.sa; + + prev = j1939_ecu_get_by_addr_locked(priv, skcb->addr.sa); + if (prev) { + if (ecu->name > prev->name) { + j1939_ecu_unmap_locked(ecu); + j1939_ecu_put(prev); + goto out_ecu_put; + } else { + /* kick prev if less or equal */ + j1939_ecu_unmap_locked(prev); + j1939_ecu_put(prev); + } + } + + j1939_ecu_timer_start(ecu); + out_ecu_put: + j1939_ecu_put(ecu); + out_unlock_bh: + write_unlock_bh(&priv->lock); +} + +void j1939_ac_recv(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_ecu *ecu; + + /* network mgmt */ + if (skcb->addr.pgn == J1939_PGN_ADDRESS_CLAIMED) { + j1939_ac_process(priv, skb); + } else if (j1939_address_is_unicast(skcb->addr.sa)) { + /* assign source name */ + ecu = j1939_ecu_get_by_addr(priv, skcb->addr.sa); + if (ecu) { + skcb->addr.src_name = ecu->name; + j1939_ecu_put(ecu); + } + } + + /* assign destination name */ + ecu = j1939_ecu_get_by_addr(priv, skcb->addr.da); + if (ecu) { + skcb->addr.dst_name = ecu->name; + j1939_ecu_put(ecu); + } +} diff --git a/net/can/j1939/bus.c b/net/can/j1939/bus.c new file mode 100644 index 000000000000..486687901602 --- /dev/null +++ b/net/can/j1939/bus.c @@ -0,0 +1,333 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +/* bus for j1939 remote devices + * Since rtnetlink, no real bus is used. + */ + +#include <net/sock.h> + +#include "j1939-priv.h" + +static void __j1939_ecu_release(struct kref *kref) +{ + struct j1939_ecu *ecu = container_of(kref, struct j1939_ecu, kref); + struct j1939_priv *priv = ecu->priv; + + list_del(&ecu->list); + kfree(ecu); + j1939_priv_put(priv); +} + +void j1939_ecu_put(struct j1939_ecu *ecu) +{ + kref_put(&ecu->kref, __j1939_ecu_release); +} + +static void j1939_ecu_get(struct j1939_ecu *ecu) +{ + kref_get(&ecu->kref); +} + +static bool j1939_ecu_is_mapped_locked(struct j1939_ecu *ecu) +{ + struct j1939_priv *priv = ecu->priv; + + lockdep_assert_held(&priv->lock); + + return j1939_ecu_find_by_addr_locked(priv, ecu->addr) == ecu; +} + +/* ECU device interface */ +/* map ECU to a bus address space */ +static void j1939_ecu_map_locked(struct j1939_ecu *ecu) +{ + struct j1939_priv *priv = ecu->priv; + struct j1939_addr_ent *ent; + + lockdep_assert_held(&priv->lock); + + if (!j1939_address_is_unicast(ecu->addr)) + return; + + ent = &priv->ents[ecu->addr]; + + if (ent->ecu) { + netdev_warn(priv->ndev, "Trying to map already mapped ECU, addr: 0x%02x, name: 0x%016llx. Skip it.\n", + ecu->addr, ecu->name); + return; + } + + j1939_ecu_get(ecu); + ent->ecu = ecu; + ent->nusers += ecu->nusers; +} + +/* unmap ECU from a bus address space */ +void j1939_ecu_unmap_locked(struct j1939_ecu *ecu) +{ + struct j1939_priv *priv = ecu->priv; + struct j1939_addr_ent *ent; + + lockdep_assert_held(&priv->lock); + + if (!j1939_address_is_unicast(ecu->addr)) + return; + + if (!j1939_ecu_is_mapped_locked(ecu)) + return; + + ent = &priv->ents[ecu->addr]; + ent->ecu = NULL; + ent->nusers -= ecu->nusers; + j1939_ecu_put(ecu); +} + +void j1939_ecu_unmap(struct j1939_ecu *ecu) +{ + write_lock_bh(&ecu->priv->lock); + j1939_ecu_unmap_locked(ecu); + write_unlock_bh(&ecu->priv->lock); +} + +void j1939_ecu_unmap_all(struct j1939_priv *priv) +{ + int i; + + write_lock_bh(&priv->lock); + for (i = 0; i < ARRAY_SIZE(priv->ents); i++) + if (priv->ents[i].ecu) + j1939_ecu_unmap_locked(priv->ents[i].ecu); + write_unlock_bh(&priv->lock); +} + +void j1939_ecu_timer_start(struct j1939_ecu *ecu) +{ + /* The ECU is held here and released in the + * j1939_ecu_timer_handler() or j1939_ecu_timer_cancel(). + */ + j1939_ecu_get(ecu); + + /* Schedule timer in 250 msec to commit address change. */ + hrtimer_start(&ecu->ac_timer, ms_to_ktime(250), + HRTIMER_MODE_REL_SOFT); +} + +void j1939_ecu_timer_cancel(struct j1939_ecu *ecu) +{ + if (hrtimer_cancel(&ecu->ac_timer)) + j1939_ecu_put(ecu); +} + +static enum hrtimer_restart j1939_ecu_timer_handler(struct hrtimer *hrtimer) +{ + struct j1939_ecu *ecu = + container_of(hrtimer, struct j1939_ecu, ac_timer); + struct j1939_priv *priv = ecu->priv; + + write_lock_bh(&priv->lock); + /* TODO: can we test if ecu->addr is unicast before starting + * the timer? + */ + j1939_ecu_map_locked(ecu); + + /* The corresponding j1939_ecu_get() is in + * j1939_ecu_timer_start(). + */ + j1939_ecu_put(ecu); + write_unlock_bh(&priv->lock); + + return HRTIMER_NORESTART; +} + +struct j1939_ecu *j1939_ecu_create_locked(struct j1939_priv *priv, name_t name) +{ + struct j1939_ecu *ecu; + + lockdep_assert_held(&priv->lock); + + ecu = kzalloc(sizeof(*ecu), gfp_any()); + if (!ecu) + return ERR_PTR(-ENOMEM); + kref_init(&ecu->kref); + ecu->addr = J1939_IDLE_ADDR; + ecu->name = name; + + hrtimer_init(&ecu->ac_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT); + ecu->ac_timer.function = j1939_ecu_timer_handler; + INIT_LIST_HEAD(&ecu->list); + + j1939_priv_get(priv); + ecu->priv = priv; + list_add_tail(&ecu->list, &priv->ecus); + + return ecu; +} + +struct j1939_ecu *j1939_ecu_find_by_addr_locked(struct j1939_priv *priv, + u8 addr) +{ + lockdep_assert_held(&priv->lock); + + return priv->ents[addr].ecu; +} + +struct j1939_ecu *j1939_ecu_get_by_addr_locked(struct j1939_priv *priv, u8 addr) +{ + struct j1939_ecu *ecu; + + lockdep_assert_held(&priv->lock); + + if (!j1939_address_is_unicast(addr)) + return NULL; + + ecu = j1939_ecu_find_by_addr_locked(priv, addr); + if (ecu) + j1939_ecu_get(ecu); + + return ecu; +} + +struct j1939_ecu *j1939_ecu_get_by_addr(struct j1939_priv *priv, u8 addr) +{ + struct j1939_ecu *ecu; + + read_lock_bh(&priv->lock); + ecu = j1939_ecu_get_by_addr_locked(priv, addr); + read_unlock_bh(&priv->lock); + + return ecu; +} + +/* get pointer to ecu without increasing ref counter */ +static struct j1939_ecu *j1939_ecu_find_by_name_locked(struct j1939_priv *priv, + name_t name) +{ + struct j1939_ecu *ecu; + + lockdep_assert_held(&priv->lock); + + list_for_each_entry(ecu, &priv->ecus, list) { + if (ecu->name == name) + return ecu; + } + + return NULL; +} + +struct j1939_ecu *j1939_ecu_get_by_name_locked(struct j1939_priv *priv, + name_t name) +{ + struct j1939_ecu *ecu; + + lockdep_assert_held(&priv->lock); + + if (!name) + return NULL; + + ecu = j1939_ecu_find_by_name_locked(priv, name); + if (ecu) + j1939_ecu_get(ecu); + + return ecu; +} + +struct j1939_ecu *j1939_ecu_get_by_name(struct j1939_priv *priv, name_t name) +{ + struct j1939_ecu *ecu; + + read_lock_bh(&priv->lock); + ecu = j1939_ecu_get_by_name_locked(priv, name); + read_unlock_bh(&priv->lock); + + return ecu; +} + +u8 j1939_name_to_addr(struct j1939_priv *priv, name_t name) +{ + struct j1939_ecu *ecu; + int addr = J1939_IDLE_ADDR; + + if (!name) + return J1939_NO_ADDR; + + read_lock_bh(&priv->lock); + ecu = j1939_ecu_find_by_name_locked(priv, name); + if (ecu && j1939_ecu_is_mapped_locked(ecu)) + /* ecu's SA is registered */ + addr = ecu->addr; + + read_unlock_bh(&priv->lock); + + return addr; +} + +/* TX addr/name accounting + * Transport protocol needs to know if a SA is local or not + * These functions originate from userspace manipulating sockets, + * so locking is straigforward + */ + +int j1939_local_ecu_get(struct j1939_priv *priv, name_t name, u8 sa) +{ + struct j1939_ecu *ecu; + int err = 0; + + write_lock_bh(&priv->lock); + + if (j1939_address_is_unicast(sa)) + priv->ents[sa].nusers++; + + if (!name) + goto done; + + ecu = j1939_ecu_get_by_name_locked(priv, name); + if (!ecu) + ecu = j1939_ecu_create_locked(priv, name); + err = PTR_ERR_OR_ZERO(ecu); + if (err) + goto done; + + ecu->nusers++; + /* TODO: do we care if ecu->addr != sa? */ + if (j1939_ecu_is_mapped_locked(ecu)) + /* ecu's sa is active already */ + priv->ents[ecu->addr].nusers++; + + done: + write_unlock_bh(&priv->lock); + + return err; +} + +void j1939_local_ecu_put(struct j1939_priv *priv, name_t name, u8 sa) +{ + struct j1939_ecu *ecu; + + write_lock_bh(&priv->lock); + + if (j1939_address_is_unicast(sa)) + priv->ents[sa].nusers--; + + if (!name) + goto done; + + ecu = j1939_ecu_find_by_name_locked(priv, name); + if (WARN_ON_ONCE(!ecu)) + goto done; + + ecu->nusers--; + /* TODO: do we care if ecu->addr != sa? */ + if (j1939_ecu_is_mapped_locked(ecu)) + /* ecu's sa is active already */ + priv->ents[ecu->addr].nusers--; + j1939_ecu_put(ecu); + + done: + write_unlock_bh(&priv->lock); +} diff --git a/net/can/j1939/j1939-priv.h b/net/can/j1939/j1939-priv.h new file mode 100644 index 000000000000..12369b604ce9 --- /dev/null +++ b/net/can/j1939/j1939-priv.h @@ -0,0 +1,338 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +#ifndef _J1939_PRIV_H_ +#define _J1939_PRIV_H_ + +#include <linux/can/j1939.h> +#include <net/sock.h> + +/* Timeout to receive the abort signal over loop back. In case CAN + * bus is open, the timeout should be triggered. + */ +#define J1939_XTP_ABORT_TIMEOUT_MS 500 +#define J1939_SIMPLE_ECHO_TIMEOUT_MS (10 * 1000) + +struct j1939_session; +enum j1939_sk_errqueue_type { + J1939_ERRQUEUE_ACK, + J1939_ERRQUEUE_SCHED, + J1939_ERRQUEUE_ABORT, +}; + +/* j1939 devices */ +struct j1939_ecu { + struct list_head list; + name_t name; + u8 addr; + + /* indicates that this ecu successfully claimed @sa as its address */ + struct hrtimer ac_timer; + struct kref kref; + struct j1939_priv *priv; + + /* count users, to help transport protocol decide for interaction */ + int nusers; +}; + +struct j1939_priv { + struct list_head ecus; + /* local list entry in priv + * These allow irq (& softirq) context lookups on j1939 devices + * This approach (separate lists) is done as the other 2 alternatives + * are not easier or even wrong + * 1) using the pure kobject methods involves mutexes, which are not + * allowed in irq context. + * 2) duplicating data structures would require a lot of synchronization + * code + * usage: + */ + + /* segments need a lock to protect the above list */ + rwlock_t lock; + + struct net_device *ndev; + + /* list of 256 ecu ptrs, that cache the claimed addresses. + * also protected by the above lock + */ + struct j1939_addr_ent { + struct j1939_ecu *ecu; + /* count users, to help transport protocol */ + int nusers; + } ents[256]; + + struct kref kref; + + /* List of active sessions to prevent start of conflicting + * one. + * + * Do not start two sessions of same type, addresses and + * direction. + */ + struct list_head active_session_list; + + /* protects active_session_list */ + spinlock_t active_session_list_lock; + + unsigned int tp_max_packet_size; + + /* lock for j1939_socks list */ + spinlock_t j1939_socks_lock; + struct list_head j1939_socks; + + struct kref rx_kref; +}; + +void j1939_ecu_put(struct j1939_ecu *ecu); + +/* keep the cache of what is local */ +int j1939_local_ecu_get(struct j1939_priv *priv, name_t name, u8 sa); +void j1939_local_ecu_put(struct j1939_priv *priv, name_t name, u8 sa); + +static inline bool j1939_address_is_unicast(u8 addr) +{ + return addr <= J1939_MAX_UNICAST_ADDR; +} + +static inline bool j1939_address_is_idle(u8 addr) +{ + return addr == J1939_IDLE_ADDR; +} + +static inline bool j1939_address_is_valid(u8 addr) +{ + return addr != J1939_NO_ADDR; +} + +static inline bool j1939_pgn_is_pdu1(pgn_t pgn) +{ + /* ignore dp & res bits for this */ + return (pgn & 0xff00) < 0xf000; +} + +/* utility to correctly unmap an ECU */ +void j1939_ecu_unmap_locked(struct j1939_ecu *ecu); +void j1939_ecu_unmap(struct j1939_ecu *ecu); + +u8 j1939_name_to_addr(struct j1939_priv *priv, name_t name); +struct j1939_ecu *j1939_ecu_find_by_addr_locked(struct j1939_priv *priv, + u8 addr); +struct j1939_ecu *j1939_ecu_get_by_addr(struct j1939_priv *priv, u8 addr); +struct j1939_ecu *j1939_ecu_get_by_addr_locked(struct j1939_priv *priv, + u8 addr); +struct j1939_ecu *j1939_ecu_get_by_name(struct j1939_priv *priv, name_t name); +struct j1939_ecu *j1939_ecu_get_by_name_locked(struct j1939_priv *priv, + name_t name); + +enum j1939_transfer_type { + J1939_TP, + J1939_ETP, + J1939_SIMPLE, +}; + +struct j1939_addr { + name_t src_name; + name_t dst_name; + pgn_t pgn; + + u8 sa; + u8 da; + + u8 type; +}; + +/* control buffer of the sk_buff */ +struct j1939_sk_buff_cb { + /* Offset in bytes within one ETP session */ + u32 offset; + + /* for tx, MSG_SYN will be used to sync on sockets */ + u32 msg_flags; + u32 tskey; + + struct j1939_addr addr; + + /* Flags for quick lookups during skb processing. + * These are set in the receive path only. + */ +#define J1939_ECU_LOCAL_SRC BIT(0) +#define J1939_ECU_LOCAL_DST BIT(1) + u8 flags; + + priority_t priority; +}; + +static inline +struct j1939_sk_buff_cb *j1939_skb_to_cb(const struct sk_buff *skb) +{ + BUILD_BUG_ON(sizeof(struct j1939_sk_buff_cb) > sizeof(skb->cb)); + + return (struct j1939_sk_buff_cb *)skb->cb; +} + +int j1939_send_one(struct j1939_priv *priv, struct sk_buff *skb); +void j1939_sk_recv(struct j1939_priv *priv, struct sk_buff *skb); +bool j1939_sk_recv_match(struct j1939_priv *priv, + struct j1939_sk_buff_cb *skcb); +void j1939_sk_send_loop_abort(struct sock *sk, int err); +void j1939_sk_errqueue(struct j1939_session *session, + enum j1939_sk_errqueue_type type); +void j1939_sk_queue_activate_next(struct j1939_session *session); + +/* stack entries */ +struct j1939_session *j1939_tp_send(struct j1939_priv *priv, + struct sk_buff *skb, size_t size); +int j1939_tp_recv(struct j1939_priv *priv, struct sk_buff *skb); +int j1939_ac_fixup(struct j1939_priv *priv, struct sk_buff *skb); +void j1939_ac_recv(struct j1939_priv *priv, struct sk_buff *skb); +void j1939_simple_recv(struct j1939_priv *priv, struct sk_buff *skb); + +/* network management */ +struct j1939_ecu *j1939_ecu_create_locked(struct j1939_priv *priv, name_t name); + +void j1939_ecu_timer_start(struct j1939_ecu *ecu); +void j1939_ecu_timer_cancel(struct j1939_ecu *ecu); +void j1939_ecu_unmap_all(struct j1939_priv *priv); + +struct j1939_priv *j1939_netdev_start(struct net_device *ndev); +void j1939_netdev_stop(struct j1939_priv *priv); + +void j1939_priv_put(struct j1939_priv *priv); +void j1939_priv_get(struct j1939_priv *priv); + +/* notify/alert all j1939 sockets bound to ifindex */ +void j1939_sk_netdev_event_netdown(struct j1939_priv *priv); +int j1939_cancel_active_session(struct j1939_priv *priv, struct sock *sk); +void j1939_tp_init(struct j1939_priv *priv); + +/* decrement pending skb for a j1939 socket */ +void j1939_sock_pending_del(struct sock *sk); + +enum j1939_session_state { + J1939_SESSION_NEW, + J1939_SESSION_ACTIVE, + /* waiting for abort signal on the bus */ + J1939_SESSION_WAITING_ABORT, + J1939_SESSION_ACTIVE_MAX, + J1939_SESSION_DONE, +}; + +struct j1939_session { + struct j1939_priv *priv; + struct list_head active_session_list_entry; + struct list_head sk_session_queue_entry; + struct kref kref; + struct sock *sk; + + /* ifindex, src, dst, pgn define the session block + * the are _never_ modified after insertion in the list + * this decreases locking problems a _lot_ + */ + struct j1939_sk_buff_cb skcb; + struct sk_buff_head skb_queue; + + /* all tx related stuff (last_txcmd, pkt.tx) + * is protected (modified only) with the txtimer hrtimer + * 'total' & 'block' are never changed, + * last_cmd, last & block are protected by ->lock + * this means that the tx may run after cts is received that should + * have stopped tx, but this time discrepancy is never avoided anyhow + */ + u8 last_cmd, last_txcmd; + bool transmission; + bool extd; + /* Total message size, number of bytes */ + unsigned int total_message_size; + /* Total number of bytes queue from socket to the session */ + unsigned int total_queued_size; + unsigned int tx_retry; + + int err; + u32 tskey; + enum j1939_session_state state; + + /* Packets counters for a (extended) transfer session. The packet is + * maximal of 7 bytes. + */ + struct { + /* total - total number of packets for this session */ + unsigned int total; + /* last - last packet of a transfer block after which + * responder should send ETP.CM_CTS and originator + * ETP.CM_DPO + */ + unsigned int last; + /* tx - number of packets send by originator node. + * this counter can be set back if responder node + * didn't received all packets send by originator. + */ + unsigned int tx; + unsigned int tx_acked; + /* rx - number of packets received */ + unsigned int rx; + /* block - amount of packets expected in one block */ + unsigned int block; + /* dpo - ETP.CM_DPO, Data Packet Offset */ + unsigned int dpo; + } pkt; + struct hrtimer txtimer, rxtimer; +}; + +struct j1939_sock { + struct sock sk; /* must be first to skip with memset */ + struct j1939_priv *priv; + struct list_head list; + +#define J1939_SOCK_BOUND BIT(0) +#define J1939_SOCK_CONNECTED BIT(1) +#define J1939_SOCK_PROMISC BIT(2) +#define J1939_SOCK_ERRQUEUE BIT(3) + int state; + + int ifindex; + struct j1939_addr addr; + struct j1939_filter *filters; + int nfilters; + pgn_t pgn_rx_filter; + + /* j1939 may emit equal PGN (!= equal CAN-id's) out of order + * when transport protocol comes in. + * To allow emitting in order, keep a 'pending' nr. of packets + */ + atomic_t skb_pending; + wait_queue_head_t waitq; + + /* lock for the sk_session_queue list */ + spinlock_t sk_session_queue_lock; + struct list_head sk_session_queue; +}; + +static inline struct j1939_sock *j1939_sk(const struct sock *sk) +{ + return container_of(sk, struct j1939_sock, sk); +} + +void j1939_session_get(struct j1939_session *session); +void j1939_session_put(struct j1939_session *session); +void j1939_session_skb_queue(struct j1939_session *session, + struct sk_buff *skb); +int j1939_session_activate(struct j1939_session *session); +void j1939_tp_schedule_txtimer(struct j1939_session *session, int msec); +void j1939_session_timers_cancel(struct j1939_session *session); + +#define J1939_MAX_TP_PACKET_SIZE (7 * 0xff) +#define J1939_MAX_ETP_PACKET_SIZE (7 * 0x00ffffff) + +#define J1939_REGULAR 0 +#define J1939_EXTENDED 1 + +/* CAN protocol */ +extern const struct can_proto j1939_can_proto; + +#endif /* _J1939_PRIV_H_ */ diff --git a/net/can/j1939/main.c b/net/can/j1939/main.c new file mode 100644 index 000000000000..def2f813ffce --- /dev/null +++ b/net/can/j1939/main.c @@ -0,0 +1,403 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2010-2011 EIA Electronics, +// Pieter Beyens <pieter.beyens@eia.be> +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2018 Protonic, +// Robin van der Gracht <robin@protonic.nl> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +/* Core of can-j1939 that links j1939 to CAN. */ + +#include <linux/can/can-ml.h> +#include <linux/can/core.h> +#include <linux/can/skb.h> +#include <linux/if_arp.h> +#include <linux/module.h> + +#include "j1939-priv.h" + +MODULE_DESCRIPTION("PF_CAN SAE J1939"); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("EIA Electronics (Kurt Van Dijck & Pieter Beyens)"); +MODULE_ALIAS("can-proto-" __stringify(CAN_J1939)); + +/* LOWLEVEL CAN interface */ + +/* CAN_HDR: #bytes before can_frame data part */ +#define J1939_CAN_HDR (offsetof(struct can_frame, data)) + +/* CAN_FTR: #bytes beyond data part */ +#define J1939_CAN_FTR (sizeof(struct can_frame) - J1939_CAN_HDR - \ + sizeof(((struct can_frame *)0)->data)) + +/* lowest layer */ +static void j1939_can_recv(struct sk_buff *iskb, void *data) +{ + struct j1939_priv *priv = data; + struct sk_buff *skb; + struct j1939_sk_buff_cb *skcb, *iskcb; + struct can_frame *cf; + + /* create a copy of the skb + * j1939 only delivers the real data bytes, + * the header goes into sockaddr. + * j1939 may not touch the incoming skb in such way + */ + skb = skb_clone(iskb, GFP_ATOMIC); + if (!skb) + return; + + can_skb_set_owner(skb, iskb->sk); + + /* get a pointer to the header of the skb + * the skb payload (pointer) is moved, so that the next skb_data + * returns the actual payload + */ + cf = (void *)skb->data; + skb_pull(skb, J1939_CAN_HDR); + + /* fix length, set to dlc, with 8 maximum */ + skb_trim(skb, min_t(uint8_t, cf->can_dlc, 8)); + + /* set addr */ + skcb = j1939_skb_to_cb(skb); + memset(skcb, 0, sizeof(*skcb)); + + iskcb = j1939_skb_to_cb(iskb); + skcb->tskey = iskcb->tskey; + skcb->priority = (cf->can_id >> 26) & 0x7; + skcb->addr.sa = cf->can_id; + skcb->addr.pgn = (cf->can_id >> 8) & J1939_PGN_MAX; + /* set default message type */ + skcb->addr.type = J1939_TP; + if (j1939_pgn_is_pdu1(skcb->addr.pgn)) { + /* Type 1: with destination address */ + skcb->addr.da = skcb->addr.pgn; + /* normalize pgn: strip dst address */ + skcb->addr.pgn &= 0x3ff00; + } else { + /* set broadcast address */ + skcb->addr.da = J1939_NO_ADDR; + } + + /* update localflags */ + read_lock_bh(&priv->lock); + if (j1939_address_is_unicast(skcb->addr.sa) && + priv->ents[skcb->addr.sa].nusers) + skcb->flags |= J1939_ECU_LOCAL_SRC; + if (j1939_address_is_unicast(skcb->addr.da) && + priv->ents[skcb->addr.da].nusers) + skcb->flags |= J1939_ECU_LOCAL_DST; + read_unlock_bh(&priv->lock); + + /* deliver into the j1939 stack ... */ + j1939_ac_recv(priv, skb); + + if (j1939_tp_recv(priv, skb)) + /* this means the transport layer processed the message */ + goto done; + + j1939_simple_recv(priv, skb); + j1939_sk_recv(priv, skb); + done: + kfree_skb(skb); +} + +/* NETDEV MANAGEMENT */ + +/* values for can_rx_(un)register */ +#define J1939_CAN_ID CAN_EFF_FLAG +#define J1939_CAN_MASK (CAN_EFF_FLAG | CAN_RTR_FLAG) + +static DEFINE_SPINLOCK(j1939_netdev_lock); + +static struct j1939_priv *j1939_priv_create(struct net_device *ndev) +{ + struct j1939_priv *priv; + + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (!priv) + return NULL; + + rwlock_init(&priv->lock); + INIT_LIST_HEAD(&priv->ecus); + priv->ndev = ndev; + kref_init(&priv->kref); + kref_init(&priv->rx_kref); + dev_hold(ndev); + + netdev_dbg(priv->ndev, "%s : 0x%p\n", __func__, priv); + + return priv; +} + +static inline void j1939_priv_set(struct net_device *ndev, + struct j1939_priv *priv) +{ + struct can_ml_priv *can_ml_priv = ndev->ml_priv; + + can_ml_priv->j1939_priv = priv; +} + +static void __j1939_priv_release(struct kref *kref) +{ + struct j1939_priv *priv = container_of(kref, struct j1939_priv, kref); + struct net_device *ndev = priv->ndev; + + netdev_dbg(priv->ndev, "%s: 0x%p\n", __func__, priv); + + dev_put(ndev); + kfree(priv); +} + +void j1939_priv_put(struct j1939_priv *priv) +{ + kref_put(&priv->kref, __j1939_priv_release); +} + +void j1939_priv_get(struct j1939_priv *priv) +{ + kref_get(&priv->kref); +} + +static int j1939_can_rx_register(struct j1939_priv *priv) +{ + struct net_device *ndev = priv->ndev; + int ret; + + j1939_priv_get(priv); + ret = can_rx_register(dev_net(ndev), ndev, J1939_CAN_ID, J1939_CAN_MASK, + j1939_can_recv, priv, "j1939", NULL); + if (ret < 0) { + j1939_priv_put(priv); + return ret; + } + + return 0; +} + +static void j1939_can_rx_unregister(struct j1939_priv *priv) +{ + struct net_device *ndev = priv->ndev; + + can_rx_unregister(dev_net(ndev), ndev, J1939_CAN_ID, J1939_CAN_MASK, + j1939_can_recv, priv); + + j1939_priv_put(priv); +} + +static void __j1939_rx_release(struct kref *kref) + __releases(&j1939_netdev_lock) +{ + struct j1939_priv *priv = container_of(kref, struct j1939_priv, + rx_kref); + + j1939_can_rx_unregister(priv); + j1939_ecu_unmap_all(priv); + j1939_priv_set(priv->ndev, NULL); + spin_unlock(&j1939_netdev_lock); +} + +/* get pointer to priv without increasing ref counter */ +static inline struct j1939_priv *j1939_ndev_to_priv(struct net_device *ndev) +{ + struct can_ml_priv *can_ml_priv = ndev->ml_priv; + + return can_ml_priv->j1939_priv; +} + +static struct j1939_priv *j1939_priv_get_by_ndev_locked(struct net_device *ndev) +{ + struct j1939_priv *priv; + + lockdep_assert_held(&j1939_netdev_lock); + + if (ndev->type != ARPHRD_CAN) + return NULL; + + priv = j1939_ndev_to_priv(ndev); + if (priv) + j1939_priv_get(priv); + + return priv; +} + +static struct j1939_priv *j1939_priv_get_by_ndev(struct net_device *ndev) +{ + struct j1939_priv *priv; + + spin_lock(&j1939_netdev_lock); + priv = j1939_priv_get_by_ndev_locked(ndev); + spin_unlock(&j1939_netdev_lock); + + return priv; +} + +struct j1939_priv *j1939_netdev_start(struct net_device *ndev) +{ + struct j1939_priv *priv, *priv_new; + int ret; + + priv = j1939_priv_get_by_ndev(ndev); + if (priv) { + kref_get(&priv->rx_kref); + return priv; + } + + priv = j1939_priv_create(ndev); + if (!priv) + return ERR_PTR(-ENOMEM); + + j1939_tp_init(priv); + spin_lock_init(&priv->j1939_socks_lock); + INIT_LIST_HEAD(&priv->j1939_socks); + + spin_lock(&j1939_netdev_lock); + priv_new = j1939_priv_get_by_ndev_locked(ndev); + if (priv_new) { + /* Someone was faster than us, use their priv and roll + * back our's. + */ + spin_unlock(&j1939_netdev_lock); + dev_put(ndev); + kfree(priv); + kref_get(&priv_new->rx_kref); + return priv_new; + } + j1939_priv_set(ndev, priv); + spin_unlock(&j1939_netdev_lock); + + ret = j1939_can_rx_register(priv); + if (ret < 0) + goto out_priv_put; + + return priv; + + out_priv_put: + j1939_priv_set(ndev, NULL); + dev_put(ndev); + kfree(priv); + + return ERR_PTR(ret); +} + +void j1939_netdev_stop(struct j1939_priv *priv) +{ + kref_put_lock(&priv->rx_kref, __j1939_rx_release, &j1939_netdev_lock); + j1939_priv_put(priv); +} + +int j1939_send_one(struct j1939_priv *priv, struct sk_buff *skb) +{ + int ret, dlc; + canid_t canid; + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct can_frame *cf; + + /* apply sanity checks */ + if (j1939_pgn_is_pdu1(skcb->addr.pgn)) + skcb->addr.pgn &= J1939_PGN_PDU1_MAX; + else + skcb->addr.pgn &= J1939_PGN_MAX; + + if (skcb->priority > 7) + skcb->priority = 6; + + ret = j1939_ac_fixup(priv, skb); + if (unlikely(ret)) + goto failed; + dlc = skb->len; + + /* re-claim the CAN_HDR from the SKB */ + cf = skb_push(skb, J1939_CAN_HDR); + + /* make it a full can frame again */ + skb_put(skb, J1939_CAN_FTR + (8 - dlc)); + + canid = CAN_EFF_FLAG | + (skcb->priority << 26) | + (skcb->addr.pgn << 8) | + skcb->addr.sa; + if (j1939_pgn_is_pdu1(skcb->addr.pgn)) + canid |= skcb->addr.da << 8; + + cf->can_id = canid; + cf->can_dlc = dlc; + + return can_send(skb, 1); + + failed: + kfree_skb(skb); + return ret; +} + +static int j1939_netdev_notify(struct notifier_block *nb, + unsigned long msg, void *data) +{ + struct net_device *ndev = netdev_notifier_info_to_dev(data); + struct j1939_priv *priv; + + priv = j1939_priv_get_by_ndev(ndev); + if (!priv) + goto notify_done; + + if (ndev->type != ARPHRD_CAN) + goto notify_put; + + switch (msg) { + case NETDEV_DOWN: + j1939_cancel_active_session(priv, NULL); + j1939_sk_netdev_event_netdown(priv); + j1939_ecu_unmap_all(priv); + break; + } + +notify_put: + j1939_priv_put(priv); + +notify_done: + return NOTIFY_DONE; +} + +static struct notifier_block j1939_netdev_notifier = { + .notifier_call = j1939_netdev_notify, +}; + +/* MODULE interface */ +static __init int j1939_module_init(void) +{ + int ret; + + pr_info("can: SAE J1939\n"); + + ret = register_netdevice_notifier(&j1939_netdev_notifier); + if (ret) + goto fail_notifier; + + ret = can_proto_register(&j1939_can_proto); + if (ret < 0) { + pr_err("can: registration of j1939 protocol failed\n"); + goto fail_sk; + } + + return 0; + + fail_sk: + unregister_netdevice_notifier(&j1939_netdev_notifier); + fail_notifier: + return ret; +} + +static __exit void j1939_module_exit(void) +{ + can_proto_unregister(&j1939_can_proto); + + unregister_netdevice_notifier(&j1939_netdev_notifier); +} + +module_init(j1939_module_init); +module_exit(j1939_module_exit); diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c new file mode 100644 index 000000000000..37c1040bcb9c --- /dev/null +++ b/net/can/j1939/socket.c @@ -0,0 +1,1160 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2010-2011 EIA Electronics, +// Pieter Beyens <pieter.beyens@eia.be> +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2018 Protonic, +// Robin van der Gracht <robin@protonic.nl> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/can/core.h> +#include <linux/can/skb.h> +#include <linux/errqueue.h> +#include <linux/if_arp.h> + +#include "j1939-priv.h" + +#define J1939_MIN_NAMELEN CAN_REQUIRED_SIZE(struct sockaddr_can, can_addr.j1939) + +/* conversion function between struct sock::sk_priority from linux and + * j1939 priority field + */ +static inline priority_t j1939_prio(u32 sk_priority) +{ + sk_priority = min(sk_priority, 7U); + + return 7 - sk_priority; +} + +static inline u32 j1939_to_sk_priority(priority_t prio) +{ + return 7 - prio; +} + +/* function to see if pgn is to be evaluated */ +static inline bool j1939_pgn_is_valid(pgn_t pgn) +{ + return pgn <= J1939_PGN_MAX; +} + +/* test function to avoid non-zero DA placeholder for pdu1 pgn's */ +static inline bool j1939_pgn_is_clean_pdu(pgn_t pgn) +{ + if (j1939_pgn_is_pdu1(pgn)) + return !(pgn & 0xff); + else + return true; +} + +static inline void j1939_sock_pending_add(struct sock *sk) +{ + struct j1939_sock *jsk = j1939_sk(sk); + + atomic_inc(&jsk->skb_pending); +} + +static int j1939_sock_pending_get(struct sock *sk) +{ + struct j1939_sock *jsk = j1939_sk(sk); + + return atomic_read(&jsk->skb_pending); +} + +void j1939_sock_pending_del(struct sock *sk) +{ + struct j1939_sock *jsk = j1939_sk(sk); + + /* atomic_dec_return returns the new value */ + if (!atomic_dec_return(&jsk->skb_pending)) + wake_up(&jsk->waitq); /* no pending SKB's */ +} + +static void j1939_jsk_add(struct j1939_priv *priv, struct j1939_sock *jsk) +{ + jsk->state |= J1939_SOCK_BOUND; + j1939_priv_get(priv); + jsk->priv = priv; + + spin_lock_bh(&priv->j1939_socks_lock); + list_add_tail(&jsk->list, &priv->j1939_socks); + spin_unlock_bh(&priv->j1939_socks_lock); +} + +static void j1939_jsk_del(struct j1939_priv *priv, struct j1939_sock *jsk) +{ + spin_lock_bh(&priv->j1939_socks_lock); + list_del_init(&jsk->list); + spin_unlock_bh(&priv->j1939_socks_lock); + + jsk->priv = NULL; + j1939_priv_put(priv); + jsk->state &= ~J1939_SOCK_BOUND; +} + +static bool j1939_sk_queue_session(struct j1939_session *session) +{ + struct j1939_sock *jsk = j1939_sk(session->sk); + bool empty; + + spin_lock_bh(&jsk->sk_session_queue_lock); + empty = list_empty(&jsk->sk_session_queue); + j1939_session_get(session); + list_add_tail(&session->sk_session_queue_entry, &jsk->sk_session_queue); + spin_unlock_bh(&jsk->sk_session_queue_lock); + j1939_sock_pending_add(&jsk->sk); + + return empty; +} + +static struct +j1939_session *j1939_sk_get_incomplete_session(struct j1939_sock *jsk) +{ + struct j1939_session *session = NULL; + + spin_lock_bh(&jsk->sk_session_queue_lock); + if (!list_empty(&jsk->sk_session_queue)) { + session = list_last_entry(&jsk->sk_session_queue, + struct j1939_session, + sk_session_queue_entry); + if (session->total_queued_size == session->total_message_size) + session = NULL; + else + j1939_session_get(session); + } + spin_unlock_bh(&jsk->sk_session_queue_lock); + + return session; +} + +static void j1939_sk_queue_drop_all(struct j1939_priv *priv, + struct j1939_sock *jsk, int err) +{ + struct j1939_session *session, *tmp; + + netdev_dbg(priv->ndev, "%s: err: %i\n", __func__, err); + spin_lock_bh(&jsk->sk_session_queue_lock); + list_for_each_entry_safe(session, tmp, &jsk->sk_session_queue, + sk_session_queue_entry) { + list_del_init(&session->sk_session_queue_entry); + session->err = err; + j1939_session_put(session); + } + spin_unlock_bh(&jsk->sk_session_queue_lock); +} + +static void j1939_sk_queue_activate_next_locked(struct j1939_session *session) +{ + struct j1939_sock *jsk; + struct j1939_session *first; + int err; + + /* RX-Session don't have a socket (yet) */ + if (!session->sk) + return; + + jsk = j1939_sk(session->sk); + lockdep_assert_held(&jsk->sk_session_queue_lock); + + err = session->err; + + first = list_first_entry_or_null(&jsk->sk_session_queue, + struct j1939_session, + sk_session_queue_entry); + + /* Some else has already activated the next session */ + if (first != session) + return; + +activate_next: + list_del_init(&first->sk_session_queue_entry); + j1939_session_put(first); + first = list_first_entry_or_null(&jsk->sk_session_queue, + struct j1939_session, + sk_session_queue_entry); + if (!first) + return; + + if (WARN_ON_ONCE(j1939_session_activate(first))) { + first->err = -EBUSY; + goto activate_next; + } else { + /* Give receiver some time (arbitrary chosen) to recover */ + int time_ms = 0; + + if (err) + time_ms = 10 + prandom_u32_max(16); + + j1939_tp_schedule_txtimer(first, time_ms); + } +} + +void j1939_sk_queue_activate_next(struct j1939_session *session) +{ + struct j1939_sock *jsk; + + if (!session->sk) + return; + + jsk = j1939_sk(session->sk); + + spin_lock_bh(&jsk->sk_session_queue_lock); + j1939_sk_queue_activate_next_locked(session); + spin_unlock_bh(&jsk->sk_session_queue_lock); +} + +static bool j1939_sk_match_dst(struct j1939_sock *jsk, + const struct j1939_sk_buff_cb *skcb) +{ + if ((jsk->state & J1939_SOCK_PROMISC)) + return true; + + /* Destination address filter */ + if (jsk->addr.src_name && skcb->addr.dst_name) { + if (jsk->addr.src_name != skcb->addr.dst_name) + return false; + } else { + /* receive (all sockets) if + * - all packages that match our bind() address + * - all broadcast on a socket if SO_BROADCAST + * is set + */ + if (j1939_address_is_unicast(skcb->addr.da)) { + if (jsk->addr.sa != skcb->addr.da) + return false; + } else if (!sock_flag(&jsk->sk, SOCK_BROADCAST)) { + /* receiving broadcast without SO_BROADCAST + * flag is not allowed + */ + return false; + } + } + + /* Source address filter */ + if (jsk->state & J1939_SOCK_CONNECTED) { + /* receive (all sockets) if + * - all packages that match our connect() name or address + */ + if (jsk->addr.dst_name && skcb->addr.src_name) { + if (jsk->addr.dst_name != skcb->addr.src_name) + return false; + } else { + if (jsk->addr.da != skcb->addr.sa) + return false; + } + } + + /* PGN filter */ + if (j1939_pgn_is_valid(jsk->pgn_rx_filter) && + jsk->pgn_rx_filter != skcb->addr.pgn) + return false; + + return true; +} + +/* matches skb control buffer (addr) with a j1939 filter */ +static bool j1939_sk_match_filter(struct j1939_sock *jsk, + const struct j1939_sk_buff_cb *skcb) +{ + const struct j1939_filter *f = jsk->filters; + int nfilter = jsk->nfilters; + + if (!nfilter) + /* receive all when no filters are assigned */ + return true; + + for (; nfilter; ++f, --nfilter) { + if ((skcb->addr.pgn & f->pgn_mask) != f->pgn) + continue; + if ((skcb->addr.sa & f->addr_mask) != f->addr) + continue; + if ((skcb->addr.src_name & f->name_mask) != f->name) + continue; + return true; + } + return false; +} + +static bool j1939_sk_recv_match_one(struct j1939_sock *jsk, + const struct j1939_sk_buff_cb *skcb) +{ + if (!(jsk->state & J1939_SOCK_BOUND)) + return false; + + if (!j1939_sk_match_dst(jsk, skcb)) + return false; + + if (!j1939_sk_match_filter(jsk, skcb)) + return false; + + return true; +} + +static void j1939_sk_recv_one(struct j1939_sock *jsk, struct sk_buff *oskb) +{ + const struct j1939_sk_buff_cb *oskcb = j1939_skb_to_cb(oskb); + struct j1939_sk_buff_cb *skcb; + struct sk_buff *skb; + + if (oskb->sk == &jsk->sk) + return; + + if (!j1939_sk_recv_match_one(jsk, oskcb)) + return; + + skb = skb_clone(oskb, GFP_ATOMIC); + if (!skb) { + pr_warn("skb clone failed\n"); + return; + } + can_skb_set_owner(skb, oskb->sk); + + skcb = j1939_skb_to_cb(skb); + skcb->msg_flags &= ~(MSG_DONTROUTE); + if (skb->sk) + skcb->msg_flags |= MSG_DONTROUTE; + + if (sock_queue_rcv_skb(&jsk->sk, skb) < 0) + kfree_skb(skb); +} + +bool j1939_sk_recv_match(struct j1939_priv *priv, struct j1939_sk_buff_cb *skcb) +{ + struct j1939_sock *jsk; + bool match = false; + + spin_lock_bh(&priv->j1939_socks_lock); + list_for_each_entry(jsk, &priv->j1939_socks, list) { + match = j1939_sk_recv_match_one(jsk, skcb); + if (match) + break; + } + spin_unlock_bh(&priv->j1939_socks_lock); + + return match; +} + +void j1939_sk_recv(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sock *jsk; + + spin_lock_bh(&priv->j1939_socks_lock); + list_for_each_entry(jsk, &priv->j1939_socks, list) { + j1939_sk_recv_one(jsk, skb); + } + spin_unlock_bh(&priv->j1939_socks_lock); +} + +static int j1939_sk_init(struct sock *sk) +{ + struct j1939_sock *jsk = j1939_sk(sk); + + /* Ensure that "sk" is first member in "struct j1939_sock", so that we + * can skip it during memset(). + */ + BUILD_BUG_ON(offsetof(struct j1939_sock, sk) != 0); + memset((void *)jsk + sizeof(jsk->sk), 0x0, + sizeof(*jsk) - sizeof(jsk->sk)); + + INIT_LIST_HEAD(&jsk->list); + init_waitqueue_head(&jsk->waitq); + jsk->sk.sk_priority = j1939_to_sk_priority(6); + jsk->sk.sk_reuse = 1; /* per default */ + jsk->addr.sa = J1939_NO_ADDR; + jsk->addr.da = J1939_NO_ADDR; + jsk->addr.pgn = J1939_NO_PGN; + jsk->pgn_rx_filter = J1939_NO_PGN; + atomic_set(&jsk->skb_pending, 0); + spin_lock_init(&jsk->sk_session_queue_lock); + INIT_LIST_HEAD(&jsk->sk_session_queue); + + return 0; +} + +static int j1939_sk_sanity_check(struct sockaddr_can *addr, int len) +{ + if (!addr) + return -EDESTADDRREQ; + if (len < J1939_MIN_NAMELEN) + return -EINVAL; + if (addr->can_family != AF_CAN) + return -EINVAL; + if (!addr->can_ifindex) + return -ENODEV; + if (j1939_pgn_is_valid(addr->can_addr.j1939.pgn) && + !j1939_pgn_is_clean_pdu(addr->can_addr.j1939.pgn)) + return -EINVAL; + + return 0; +} + +static int j1939_sk_bind(struct socket *sock, struct sockaddr *uaddr, int len) +{ + struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; + struct j1939_sock *jsk = j1939_sk(sock->sk); + struct j1939_priv *priv = jsk->priv; + struct sock *sk = sock->sk; + struct net *net = sock_net(sk); + int ret = 0; + + ret = j1939_sk_sanity_check(addr, len); + if (ret) + return ret; + + lock_sock(sock->sk); + + /* Already bound to an interface? */ + if (jsk->state & J1939_SOCK_BOUND) { + /* A re-bind() to a different interface is not + * supported. + */ + if (jsk->ifindex != addr->can_ifindex) { + ret = -EINVAL; + goto out_release_sock; + } + + /* drop old references */ + j1939_jsk_del(priv, jsk); + j1939_local_ecu_put(priv, jsk->addr.src_name, jsk->addr.sa); + } else { + struct net_device *ndev; + + ndev = dev_get_by_index(net, addr->can_ifindex); + if (!ndev) { + ret = -ENODEV; + goto out_release_sock; + } + + if (ndev->type != ARPHRD_CAN) { + dev_put(ndev); + ret = -ENODEV; + goto out_release_sock; + } + + priv = j1939_netdev_start(ndev); + dev_put(ndev); + if (IS_ERR(priv)) { + ret = PTR_ERR(priv); + goto out_release_sock; + } + + jsk->ifindex = addr->can_ifindex; + } + + /* set default transmit pgn */ + if (j1939_pgn_is_valid(addr->can_addr.j1939.pgn)) + jsk->pgn_rx_filter = addr->can_addr.j1939.pgn; + jsk->addr.src_name = addr->can_addr.j1939.name; + jsk->addr.sa = addr->can_addr.j1939.addr; + + /* get new references */ + ret = j1939_local_ecu_get(priv, jsk->addr.src_name, jsk->addr.sa); + if (ret) { + j1939_netdev_stop(priv); + goto out_release_sock; + } + + j1939_jsk_add(priv, jsk); + + out_release_sock: /* fall through */ + release_sock(sock->sk); + + return ret; +} + +static int j1939_sk_connect(struct socket *sock, struct sockaddr *uaddr, + int len, int flags) +{ + struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; + struct j1939_sock *jsk = j1939_sk(sock->sk); + int ret = 0; + + ret = j1939_sk_sanity_check(addr, len); + if (ret) + return ret; + + lock_sock(sock->sk); + + /* bind() before connect() is mandatory */ + if (!(jsk->state & J1939_SOCK_BOUND)) { + ret = -EINVAL; + goto out_release_sock; + } + + /* A connect() to a different interface is not supported. */ + if (jsk->ifindex != addr->can_ifindex) { + ret = -EINVAL; + goto out_release_sock; + } + + if (!addr->can_addr.j1939.name && + addr->can_addr.j1939.addr == J1939_NO_ADDR && + !sock_flag(&jsk->sk, SOCK_BROADCAST)) { + /* broadcast, but SO_BROADCAST not set */ + ret = -EACCES; + goto out_release_sock; + } + + jsk->addr.dst_name = addr->can_addr.j1939.name; + jsk->addr.da = addr->can_addr.j1939.addr; + + if (j1939_pgn_is_valid(addr->can_addr.j1939.pgn)) + jsk->addr.pgn = addr->can_addr.j1939.pgn; + + jsk->state |= J1939_SOCK_CONNECTED; + + out_release_sock: /* fall through */ + release_sock(sock->sk); + + return ret; +} + +static void j1939_sk_sock2sockaddr_can(struct sockaddr_can *addr, + const struct j1939_sock *jsk, int peer) +{ + addr->can_family = AF_CAN; + addr->can_ifindex = jsk->ifindex; + addr->can_addr.j1939.pgn = jsk->addr.pgn; + if (peer) { + addr->can_addr.j1939.name = jsk->addr.dst_name; + addr->can_addr.j1939.addr = jsk->addr.da; + } else { + addr->can_addr.j1939.name = jsk->addr.src_name; + addr->can_addr.j1939.addr = jsk->addr.sa; + } +} + +static int j1939_sk_getname(struct socket *sock, struct sockaddr *uaddr, + int peer) +{ + struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; + struct sock *sk = sock->sk; + struct j1939_sock *jsk = j1939_sk(sk); + int ret = 0; + + lock_sock(sk); + + if (peer && !(jsk->state & J1939_SOCK_CONNECTED)) { + ret = -EADDRNOTAVAIL; + goto failure; + } + + j1939_sk_sock2sockaddr_can(addr, jsk, peer); + ret = J1939_MIN_NAMELEN; + + failure: + release_sock(sk); + + return ret; +} + +static int j1939_sk_release(struct socket *sock) +{ + struct sock *sk = sock->sk; + struct j1939_sock *jsk; + + if (!sk) + return 0; + + jsk = j1939_sk(sk); + lock_sock(sk); + + if (jsk->state & J1939_SOCK_BOUND) { + struct j1939_priv *priv = jsk->priv; + + if (wait_event_interruptible(jsk->waitq, + !j1939_sock_pending_get(&jsk->sk))) { + j1939_cancel_active_session(priv, sk); + j1939_sk_queue_drop_all(priv, jsk, ESHUTDOWN); + } + + j1939_jsk_del(priv, jsk); + + j1939_local_ecu_put(priv, jsk->addr.src_name, + jsk->addr.sa); + + j1939_netdev_stop(priv); + } + + sock_orphan(sk); + sock->sk = NULL; + + release_sock(sk); + sock_put(sk); + + return 0; +} + +static int j1939_sk_setsockopt_flag(struct j1939_sock *jsk, char __user *optval, + unsigned int optlen, int flag) +{ + int tmp; + + if (optlen != sizeof(tmp)) + return -EINVAL; + if (copy_from_user(&tmp, optval, optlen)) + return -EFAULT; + lock_sock(&jsk->sk); + if (tmp) + jsk->state |= flag; + else + jsk->state &= ~flag; + release_sock(&jsk->sk); + return tmp; +} + +static int j1939_sk_setsockopt(struct socket *sock, int level, int optname, + char __user *optval, unsigned int optlen) +{ + struct sock *sk = sock->sk; + struct j1939_sock *jsk = j1939_sk(sk); + int tmp, count = 0, ret = 0; + struct j1939_filter *filters = NULL, *ofilters; + + if (level != SOL_CAN_J1939) + return -EINVAL; + + switch (optname) { + case SO_J1939_FILTER: + if (optval) { + struct j1939_filter *f; + int c; + + if (optlen % sizeof(*filters) != 0) + return -EINVAL; + + if (optlen > J1939_FILTER_MAX * + sizeof(struct j1939_filter)) + return -EINVAL; + + count = optlen / sizeof(*filters); + filters = memdup_user(optval, optlen); + if (IS_ERR(filters)) + return PTR_ERR(filters); + + for (f = filters, c = count; c; f++, c--) { + f->name &= f->name_mask; + f->pgn &= f->pgn_mask; + f->addr &= f->addr_mask; + } + } + + lock_sock(&jsk->sk); + ofilters = jsk->filters; + jsk->filters = filters; + jsk->nfilters = count; + release_sock(&jsk->sk); + kfree(ofilters); + return 0; + case SO_J1939_PROMISC: + return j1939_sk_setsockopt_flag(jsk, optval, optlen, + J1939_SOCK_PROMISC); + case SO_J1939_ERRQUEUE: + ret = j1939_sk_setsockopt_flag(jsk, optval, optlen, + J1939_SOCK_ERRQUEUE); + if (ret < 0) + return ret; + + if (!(jsk->state & J1939_SOCK_ERRQUEUE)) + skb_queue_purge(&sk->sk_error_queue); + return ret; + case SO_J1939_SEND_PRIO: + if (optlen != sizeof(tmp)) + return -EINVAL; + if (copy_from_user(&tmp, optval, optlen)) + return -EFAULT; + if (tmp < 0 || tmp > 7) + return -EDOM; + if (tmp < 2 && !capable(CAP_NET_ADMIN)) + return -EPERM; + lock_sock(&jsk->sk); + jsk->sk.sk_priority = j1939_to_sk_priority(tmp); + release_sock(&jsk->sk); + return 0; + default: + return -ENOPROTOOPT; + } +} + +static int j1939_sk_getsockopt(struct socket *sock, int level, int optname, + char __user *optval, int __user *optlen) +{ + struct sock *sk = sock->sk; + struct j1939_sock *jsk = j1939_sk(sk); + int ret, ulen; + /* set defaults for using 'int' properties */ + int tmp = 0; + int len = sizeof(tmp); + void *val = &tmp; + + if (level != SOL_CAN_J1939) + return -EINVAL; + if (get_user(ulen, optlen)) + return -EFAULT; + if (ulen < 0) + return -EINVAL; + + lock_sock(&jsk->sk); + switch (optname) { + case SO_J1939_PROMISC: + tmp = (jsk->state & J1939_SOCK_PROMISC) ? 1 : 0; + break; + case SO_J1939_ERRQUEUE: + tmp = (jsk->state & J1939_SOCK_ERRQUEUE) ? 1 : 0; + break; + case SO_J1939_SEND_PRIO: + tmp = j1939_prio(jsk->sk.sk_priority); + break; + default: + ret = -ENOPROTOOPT; + goto no_copy; + } + + /* copy to user, based on 'len' & 'val' + * but most sockopt's are 'int' properties, and have 'len' & 'val' + * left unchanged, but instead modified 'tmp' + */ + if (len > ulen) + ret = -EFAULT; + else if (put_user(len, optlen)) + ret = -EFAULT; + else if (copy_to_user(optval, val, len)) + ret = -EFAULT; + else + ret = 0; + no_copy: + release_sock(&jsk->sk); + return ret; +} + +static int j1939_sk_recvmsg(struct socket *sock, struct msghdr *msg, + size_t size, int flags) +{ + struct sock *sk = sock->sk; + struct sk_buff *skb; + struct j1939_sk_buff_cb *skcb; + int ret = 0; + + if (flags & ~(MSG_DONTWAIT | MSG_ERRQUEUE)) + return -EINVAL; + + if (flags & MSG_ERRQUEUE) + return sock_recv_errqueue(sock->sk, msg, size, SOL_CAN_J1939, + SCM_J1939_ERRQUEUE); + + skb = skb_recv_datagram(sk, flags, 0, &ret); + if (!skb) + return ret; + + if (size < skb->len) + msg->msg_flags |= MSG_TRUNC; + else + size = skb->len; + + ret = memcpy_to_msg(msg, skb->data, size); + if (ret < 0) { + skb_free_datagram(sk, skb); + return ret; + } + + skcb = j1939_skb_to_cb(skb); + if (j1939_address_is_valid(skcb->addr.da)) + put_cmsg(msg, SOL_CAN_J1939, SCM_J1939_DEST_ADDR, + sizeof(skcb->addr.da), &skcb->addr.da); + + if (skcb->addr.dst_name) + put_cmsg(msg, SOL_CAN_J1939, SCM_J1939_DEST_NAME, + sizeof(skcb->addr.dst_name), &skcb->addr.dst_name); + + put_cmsg(msg, SOL_CAN_J1939, SCM_J1939_PRIO, + sizeof(skcb->priority), &skcb->priority); + + if (msg->msg_name) { + struct sockaddr_can *paddr = msg->msg_name; + + msg->msg_namelen = J1939_MIN_NAMELEN; + memset(msg->msg_name, 0, msg->msg_namelen); + paddr->can_family = AF_CAN; + paddr->can_ifindex = skb->skb_iif; + paddr->can_addr.j1939.name = skcb->addr.src_name; + paddr->can_addr.j1939.addr = skcb->addr.sa; + paddr->can_addr.j1939.pgn = skcb->addr.pgn; + } + + sock_recv_ts_and_drops(msg, sk, skb); + msg->msg_flags |= skcb->msg_flags; + skb_free_datagram(sk, skb); + + return size; +} + +static struct sk_buff *j1939_sk_alloc_skb(struct net_device *ndev, + struct sock *sk, + struct msghdr *msg, size_t size, + int *errcode) +{ + struct j1939_sock *jsk = j1939_sk(sk); + struct j1939_sk_buff_cb *skcb; + struct sk_buff *skb; + int ret; + + skb = sock_alloc_send_skb(sk, + size + + sizeof(struct can_frame) - + sizeof(((struct can_frame *)NULL)->data) + + sizeof(struct can_skb_priv), + msg->msg_flags & MSG_DONTWAIT, &ret); + if (!skb) + goto failure; + + can_skb_reserve(skb); + can_skb_prv(skb)->ifindex = ndev->ifindex; + can_skb_prv(skb)->skbcnt = 0; + skb_reserve(skb, offsetof(struct can_frame, data)); + + ret = memcpy_from_msg(skb_put(skb, size), msg, size); + if (ret < 0) + goto free_skb; + + skb->dev = ndev; + + skcb = j1939_skb_to_cb(skb); + memset(skcb, 0, sizeof(*skcb)); + skcb->addr = jsk->addr; + skcb->priority = j1939_prio(sk->sk_priority); + + if (msg->msg_name) { + struct sockaddr_can *addr = msg->msg_name; + + if (addr->can_addr.j1939.name || + addr->can_addr.j1939.addr != J1939_NO_ADDR) { + skcb->addr.dst_name = addr->can_addr.j1939.name; + skcb->addr.da = addr->can_addr.j1939.addr; + } + if (j1939_pgn_is_valid(addr->can_addr.j1939.pgn)) + skcb->addr.pgn = addr->can_addr.j1939.pgn; + } + + *errcode = ret; + return skb; + +free_skb: + kfree_skb(skb); +failure: + *errcode = ret; + return NULL; +} + +static size_t j1939_sk_opt_stats_get_size(void) +{ + return + nla_total_size(sizeof(u32)) + /* J1939_NLA_BYTES_ACKED */ + 0; +} + +static struct sk_buff * +j1939_sk_get_timestamping_opt_stats(struct j1939_session *session) +{ + struct sk_buff *stats; + u32 size; + + stats = alloc_skb(j1939_sk_opt_stats_get_size(), GFP_ATOMIC); + if (!stats) + return NULL; + + if (session->skcb.addr.type == J1939_SIMPLE) + size = session->total_message_size; + else + size = min(session->pkt.tx_acked * 7, + session->total_message_size); + + nla_put_u32(stats, J1939_NLA_BYTES_ACKED, size); + + return stats; +} + +void j1939_sk_errqueue(struct j1939_session *session, + enum j1939_sk_errqueue_type type) +{ + struct j1939_priv *priv = session->priv; + struct sock *sk = session->sk; + struct j1939_sock *jsk; + struct sock_exterr_skb *serr; + struct sk_buff *skb; + char *state = "UNK"; + int err; + + /* currently we have no sk for the RX session */ + if (!sk) + return; + + jsk = j1939_sk(sk); + + if (!(jsk->state & J1939_SOCK_ERRQUEUE)) + return; + + skb = j1939_sk_get_timestamping_opt_stats(session); + if (!skb) + return; + + skb->tstamp = ktime_get_real(); + + BUILD_BUG_ON(sizeof(struct sock_exterr_skb) > sizeof(skb->cb)); + + serr = SKB_EXT_ERR(skb); + memset(serr, 0, sizeof(*serr)); + switch (type) { + case J1939_ERRQUEUE_ACK: + if (!(sk->sk_tsflags & SOF_TIMESTAMPING_TX_ACK)) + return; + + serr->ee.ee_errno = ENOMSG; + serr->ee.ee_origin = SO_EE_ORIGIN_TIMESTAMPING; + serr->ee.ee_info = SCM_TSTAMP_ACK; + state = "ACK"; + break; + case J1939_ERRQUEUE_SCHED: + if (!(sk->sk_tsflags & SOF_TIMESTAMPING_TX_SCHED)) + return; + + serr->ee.ee_errno = ENOMSG; + serr->ee.ee_origin = SO_EE_ORIGIN_TIMESTAMPING; + serr->ee.ee_info = SCM_TSTAMP_SCHED; + state = "SCH"; + break; + case J1939_ERRQUEUE_ABORT: + serr->ee.ee_errno = session->err; + serr->ee.ee_origin = SO_EE_ORIGIN_LOCAL; + serr->ee.ee_info = J1939_EE_INFO_TX_ABORT; + state = "ABT"; + break; + default: + netdev_err(priv->ndev, "Unknown errqueue type %i\n", type); + } + + serr->opt_stats = true; + if (sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID) + serr->ee.ee_data = session->tskey; + + netdev_dbg(session->priv->ndev, "%s: 0x%p tskey: %i, state: %s\n", + __func__, session, session->tskey, state); + err = sock_queue_err_skb(sk, skb); + + if (err) + kfree_skb(skb); +}; + +void j1939_sk_send_loop_abort(struct sock *sk, int err) +{ + sk->sk_err = err; + + sk->sk_error_report(sk); +} + +static int j1939_sk_send_loop(struct j1939_priv *priv, struct sock *sk, + struct msghdr *msg, size_t size) + +{ + struct j1939_sock *jsk = j1939_sk(sk); + struct j1939_session *session = j1939_sk_get_incomplete_session(jsk); + struct sk_buff *skb; + size_t segment_size, todo_size; + int ret = 0; + + if (session && + session->total_message_size != session->total_queued_size + size) { + j1939_session_put(session); + return -EIO; + } + + todo_size = size; + + while (todo_size) { + struct j1939_sk_buff_cb *skcb; + + segment_size = min_t(size_t, J1939_MAX_TP_PACKET_SIZE, + todo_size); + + /* Allocate skb for one segment */ + skb = j1939_sk_alloc_skb(priv->ndev, sk, msg, segment_size, + &ret); + if (ret) + break; + + skcb = j1939_skb_to_cb(skb); + + if (!session) { + /* at this point the size should be full size + * of the session + */ + skcb->offset = 0; + session = j1939_tp_send(priv, skb, size); + if (IS_ERR(session)) { + ret = PTR_ERR(session); + goto kfree_skb; + } + if (j1939_sk_queue_session(session)) { + /* try to activate session if we a + * fist in the queue + */ + if (!j1939_session_activate(session)) { + j1939_tp_schedule_txtimer(session, 0); + } else { + ret = -EBUSY; + session->err = ret; + j1939_sk_queue_drop_all(priv, jsk, + EBUSY); + break; + } + } + } else { + skcb->offset = session->total_queued_size; + j1939_session_skb_queue(session, skb); + } + + todo_size -= segment_size; + session->total_queued_size += segment_size; + } + + switch (ret) { + case 0: /* OK */ + if (todo_size) + netdev_warn(priv->ndev, + "no error found and not completely queued?! %zu\n", + todo_size); + ret = size; + break; + case -ERESTARTSYS: + ret = -EINTR; + /* fall through */ + case -EAGAIN: /* OK */ + if (todo_size != size) + ret = size - todo_size; + break; + default: /* ERROR */ + break; + } + + if (session) + j1939_session_put(session); + + return ret; + + kfree_skb: + kfree_skb(skb); + return ret; +} + +static int j1939_sk_sendmsg(struct socket *sock, struct msghdr *msg, + size_t size) +{ + struct sock *sk = sock->sk; + struct j1939_sock *jsk = j1939_sk(sk); + struct j1939_priv *priv = jsk->priv; + int ifindex; + int ret; + + /* various socket state tests */ + if (!(jsk->state & J1939_SOCK_BOUND)) + return -EBADFD; + + ifindex = jsk->ifindex; + + if (!jsk->addr.src_name && jsk->addr.sa == J1939_NO_ADDR) + /* no source address assigned yet */ + return -EBADFD; + + /* deal with provided destination address info */ + if (msg->msg_name) { + struct sockaddr_can *addr = msg->msg_name; + + if (msg->msg_namelen < J1939_MIN_NAMELEN) + return -EINVAL; + + if (addr->can_family != AF_CAN) + return -EINVAL; + + if (addr->can_ifindex && addr->can_ifindex != ifindex) + return -EBADFD; + + if (j1939_pgn_is_valid(addr->can_addr.j1939.pgn) && + !j1939_pgn_is_clean_pdu(addr->can_addr.j1939.pgn)) + return -EINVAL; + + if (!addr->can_addr.j1939.name && + addr->can_addr.j1939.addr == J1939_NO_ADDR && + !sock_flag(sk, SOCK_BROADCAST)) + /* broadcast, but SO_BROADCAST not set */ + return -EACCES; + } else { + if (!jsk->addr.dst_name && jsk->addr.da == J1939_NO_ADDR && + !sock_flag(sk, SOCK_BROADCAST)) + /* broadcast, but SO_BROADCAST not set */ + return -EACCES; + } + + ret = j1939_sk_send_loop(priv, sk, msg, size); + + return ret; +} + +void j1939_sk_netdev_event_netdown(struct j1939_priv *priv) +{ + struct j1939_sock *jsk; + int error_code = ENETDOWN; + + spin_lock_bh(&priv->j1939_socks_lock); + list_for_each_entry(jsk, &priv->j1939_socks, list) { + jsk->sk.sk_err = error_code; + if (!sock_flag(&jsk->sk, SOCK_DEAD)) + jsk->sk.sk_error_report(&jsk->sk); + + j1939_sk_queue_drop_all(priv, jsk, error_code); + } + spin_unlock_bh(&priv->j1939_socks_lock); +} + +static int j1939_sk_no_ioctlcmd(struct socket *sock, unsigned int cmd, + unsigned long arg) +{ + /* no ioctls for socket layer -> hand it down to NIC layer */ + return -ENOIOCTLCMD; +} + +static const struct proto_ops j1939_ops = { + .family = PF_CAN, + .release = j1939_sk_release, + .bind = j1939_sk_bind, + .connect = j1939_sk_connect, + .socketpair = sock_no_socketpair, + .accept = sock_no_accept, + .getname = j1939_sk_getname, + .poll = datagram_poll, + .ioctl = j1939_sk_no_ioctlcmd, + .listen = sock_no_listen, + .shutdown = sock_no_shutdown, + .setsockopt = j1939_sk_setsockopt, + .getsockopt = j1939_sk_getsockopt, + .sendmsg = j1939_sk_sendmsg, + .recvmsg = j1939_sk_recvmsg, + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage, +}; + +static struct proto j1939_proto __read_mostly = { + .name = "CAN_J1939", + .owner = THIS_MODULE, + .obj_size = sizeof(struct j1939_sock), + .init = j1939_sk_init, +}; + +const struct can_proto j1939_can_proto = { + .type = SOCK_DGRAM, + .protocol = CAN_J1939, + .ops = &j1939_ops, + .prot = &j1939_proto, +}; diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c new file mode 100644 index 000000000000..fe000ea757ea --- /dev/null +++ b/net/can/j1939/transport.c @@ -0,0 +1,2027 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2010-2011 EIA Electronics, +// Kurt Van Dijck <kurt.van.dijck@eia.be> +// Copyright (c) 2018 Protonic, +// Robin van der Gracht <robin@protonic.nl> +// Copyright (c) 2017-2019 Pengutronix, +// Marc Kleine-Budde <kernel@pengutronix.de> +// Copyright (c) 2017-2019 Pengutronix, +// Oleksij Rempel <kernel@pengutronix.de> + +#include <linux/can/skb.h> + +#include "j1939-priv.h" + +#define J1939_XTP_TX_RETRY_LIMIT 100 + +#define J1939_ETP_PGN_CTL 0xc800 +#define J1939_ETP_PGN_DAT 0xc700 +#define J1939_TP_PGN_CTL 0xec00 +#define J1939_TP_PGN_DAT 0xeb00 + +#define J1939_TP_CMD_RTS 0x10 +#define J1939_TP_CMD_CTS 0x11 +#define J1939_TP_CMD_EOMA 0x13 +#define J1939_TP_CMD_BAM 0x20 +#define J1939_TP_CMD_ABORT 0xff + +#define J1939_ETP_CMD_RTS 0x14 +#define J1939_ETP_CMD_CTS 0x15 +#define J1939_ETP_CMD_DPO 0x16 +#define J1939_ETP_CMD_EOMA 0x17 +#define J1939_ETP_CMD_ABORT 0xff + +enum j1939_xtp_abort { + J1939_XTP_NO_ABORT = 0, + J1939_XTP_ABORT_BUSY = 1, + /* Already in one or more connection managed sessions and + * cannot support another. + * + * EALREADY: + * Operation already in progress + */ + + J1939_XTP_ABORT_RESOURCE = 2, + /* System resources were needed for another task so this + * connection managed session was terminated. + * + * EMSGSIZE: + * The socket type requires that message be sent atomically, + * and the size of the message to be sent made this + * impossible. + */ + + J1939_XTP_ABORT_TIMEOUT = 3, + /* A timeout occurred and this is the connection abort to + * close the session. + * + * EHOSTUNREACH: + * The destination host cannot be reached (probably because + * the host is down or a remote router cannot reach it). + */ + + J1939_XTP_ABORT_GENERIC = 4, + /* CTS messages received when data transfer is in progress + * + * EBADMSG: + * Not a data message + */ + + J1939_XTP_ABORT_FAULT = 5, + /* Maximal retransmit request limit reached + * + * ENOTRECOVERABLE: + * State not recoverable + */ + + J1939_XTP_ABORT_UNEXPECTED_DATA = 6, + /* Unexpected data transfer packet + * + * ENOTCONN: + * Transport endpoint is not connected + */ + + J1939_XTP_ABORT_BAD_SEQ = 7, + /* Bad sequence number (and software is not able to recover) + * + * EILSEQ: + * Illegal byte sequence + */ + + J1939_XTP_ABORT_DUP_SEQ = 8, + /* Duplicate sequence number (and software is not able to + * recover) + */ + + J1939_XTP_ABORT_EDPO_UNEXPECTED = 9, + /* Unexpected EDPO packet (ETP) or Message size > 1785 bytes + * (TP) + */ + + J1939_XTP_ABORT_BAD_EDPO_PGN = 10, + /* Unexpected EDPO PGN (PGN in EDPO is bad) */ + + J1939_XTP_ABORT_EDPO_OUTOF_CTS = 11, + /* EDPO number of packets is greater than CTS */ + + J1939_XTP_ABORT_BAD_EDPO_OFFSET = 12, + /* Bad EDPO offset */ + + J1939_XTP_ABORT_OTHER_DEPRECATED = 13, + /* Deprecated. Use 250 instead (Any other reason) */ + + J1939_XTP_ABORT_ECTS_UNXPECTED_PGN = 14, + /* Unexpected ECTS PGN (PGN in ECTS is bad) */ + + J1939_XTP_ABORT_ECTS_TOO_BIG = 15, + /* ECTS requested packets exceeds message size */ + + J1939_XTP_ABORT_OTHER = 250, + /* Any other reason (if a Connection Abort reason is + * identified that is not listed in the table use code 250) + */ +}; + +static unsigned int j1939_tp_block = 255; +static unsigned int j1939_tp_packet_delay; +static unsigned int j1939_tp_padding = 1; + +/* helpers */ +static const char *j1939_xtp_abort_to_str(enum j1939_xtp_abort abort) +{ + switch (abort) { + case J1939_XTP_ABORT_BUSY: + return "Already in one or more connection managed sessions and cannot support another."; + case J1939_XTP_ABORT_RESOURCE: + return "System resources were needed for another task so this connection managed session was terminated."; + case J1939_XTP_ABORT_TIMEOUT: + return "A timeout occurred and this is the connection abort to close the session."; + case J1939_XTP_ABORT_GENERIC: + return "CTS messages received when data transfer is in progress"; + case J1939_XTP_ABORT_FAULT: + return "Maximal retransmit request limit reached"; + case J1939_XTP_ABORT_UNEXPECTED_DATA: + return "Unexpected data transfer packet"; + case J1939_XTP_ABORT_BAD_SEQ: + return "Bad sequence number (and software is not able to recover)"; + case J1939_XTP_ABORT_DUP_SEQ: + return "Duplicate sequence number (and software is not able to recover)"; + case J1939_XTP_ABORT_EDPO_UNEXPECTED: + return "Unexpected EDPO packet (ETP) or Message size > 1785 bytes (TP)"; + case J1939_XTP_ABORT_BAD_EDPO_PGN: + return "Unexpected EDPO PGN (PGN in EDPO is bad)"; + case J1939_XTP_ABORT_EDPO_OUTOF_CTS: + return "EDPO number of packets is greater than CTS"; + case J1939_XTP_ABORT_BAD_EDPO_OFFSET: + return "Bad EDPO offset"; + case J1939_XTP_ABORT_OTHER_DEPRECATED: + return "Deprecated. Use 250 instead (Any other reason)"; + case J1939_XTP_ABORT_ECTS_UNXPECTED_PGN: + return "Unexpected ECTS PGN (PGN in ECTS is bad)"; + case J1939_XTP_ABORT_ECTS_TOO_BIG: + return "ECTS requested packets exceeds message size"; + case J1939_XTP_ABORT_OTHER: + return "Any other reason (if a Connection Abort reason is identified that is not listed in the table use code 250)"; + default: + return "<unknown>"; + } +} + +static int j1939_xtp_abort_to_errno(struct j1939_priv *priv, + enum j1939_xtp_abort abort) +{ + int err; + + switch (abort) { + case J1939_XTP_NO_ABORT: + WARN_ON_ONCE(abort == J1939_XTP_NO_ABORT); + err = 0; + break; + case J1939_XTP_ABORT_BUSY: + err = EALREADY; + break; + case J1939_XTP_ABORT_RESOURCE: + err = EMSGSIZE; + break; + case J1939_XTP_ABORT_TIMEOUT: + err = EHOSTUNREACH; + break; + case J1939_XTP_ABORT_GENERIC: + err = EBADMSG; + break; + case J1939_XTP_ABORT_FAULT: + err = ENOTRECOVERABLE; + break; + case J1939_XTP_ABORT_UNEXPECTED_DATA: + err = ENOTCONN; + break; + case J1939_XTP_ABORT_BAD_SEQ: + err = EILSEQ; + break; + case J1939_XTP_ABORT_DUP_SEQ: + err = EPROTO; + break; + case J1939_XTP_ABORT_EDPO_UNEXPECTED: + err = EPROTO; + break; + case J1939_XTP_ABORT_BAD_EDPO_PGN: + err = EPROTO; + break; + case J1939_XTP_ABORT_EDPO_OUTOF_CTS: + err = EPROTO; + break; + case J1939_XTP_ABORT_BAD_EDPO_OFFSET: + err = EPROTO; + break; + case J1939_XTP_ABORT_OTHER_DEPRECATED: + err = EPROTO; + break; + case J1939_XTP_ABORT_ECTS_UNXPECTED_PGN: + err = EPROTO; + break; + case J1939_XTP_ABORT_ECTS_TOO_BIG: + err = EPROTO; + break; + case J1939_XTP_ABORT_OTHER: + err = EPROTO; + break; + default: + netdev_warn(priv->ndev, "Unknown abort code %i", abort); + err = EPROTO; + } + + return err; +} + +static inline void j1939_session_list_lock(struct j1939_priv *priv) +{ + spin_lock_bh(&priv->active_session_list_lock); +} + +static inline void j1939_session_list_unlock(struct j1939_priv *priv) +{ + spin_unlock_bh(&priv->active_session_list_lock); +} + +void j1939_session_get(struct j1939_session *session) +{ + kref_get(&session->kref); +} + +/* session completion functions */ +static void __j1939_session_drop(struct j1939_session *session) +{ + if (!session->transmission) + return; + + j1939_sock_pending_del(session->sk); +} + +static void j1939_session_destroy(struct j1939_session *session) +{ + if (session->err) + j1939_sk_errqueue(session, J1939_ERRQUEUE_ABORT); + else + j1939_sk_errqueue(session, J1939_ERRQUEUE_ACK); + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + skb_queue_purge(&session->skb_queue); + __j1939_session_drop(session); + j1939_priv_put(session->priv); + kfree(session); +} + +static void __j1939_session_release(struct kref *kref) +{ + struct j1939_session *session = container_of(kref, struct j1939_session, + kref); + + j1939_session_destroy(session); +} + +void j1939_session_put(struct j1939_session *session) +{ + kref_put(&session->kref, __j1939_session_release); +} + +static void j1939_session_txtimer_cancel(struct j1939_session *session) +{ + if (hrtimer_cancel(&session->txtimer)) + j1939_session_put(session); +} + +static void j1939_session_rxtimer_cancel(struct j1939_session *session) +{ + if (hrtimer_cancel(&session->rxtimer)) + j1939_session_put(session); +} + +void j1939_session_timers_cancel(struct j1939_session *session) +{ + j1939_session_txtimer_cancel(session); + j1939_session_rxtimer_cancel(session); +} + +static inline bool j1939_cb_is_broadcast(const struct j1939_sk_buff_cb *skcb) +{ + return (!skcb->addr.dst_name && (skcb->addr.da == 0xff)); +} + +static void j1939_session_skb_drop_old(struct j1939_session *session) +{ + struct sk_buff *do_skb; + struct j1939_sk_buff_cb *do_skcb; + unsigned int offset_start; + unsigned long flags; + + if (skb_queue_len(&session->skb_queue) < 2) + return; + + offset_start = session->pkt.tx_acked * 7; + + spin_lock_irqsave(&session->skb_queue.lock, flags); + do_skb = skb_peek(&session->skb_queue); + do_skcb = j1939_skb_to_cb(do_skb); + + if ((do_skcb->offset + do_skb->len) < offset_start) { + __skb_unlink(do_skb, &session->skb_queue); + kfree_skb(do_skb); + } + spin_unlock_irqrestore(&session->skb_queue.lock, flags); +} + +void j1939_session_skb_queue(struct j1939_session *session, + struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_priv *priv = session->priv; + + j1939_ac_fixup(priv, skb); + + if (j1939_address_is_unicast(skcb->addr.da) && + priv->ents[skcb->addr.da].nusers) + skcb->flags |= J1939_ECU_LOCAL_DST; + + skcb->flags |= J1939_ECU_LOCAL_SRC; + + skb_queue_tail(&session->skb_queue, skb); +} + +static struct sk_buff *j1939_session_skb_find(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + struct sk_buff *skb = NULL; + struct sk_buff *do_skb; + struct j1939_sk_buff_cb *do_skcb; + unsigned int offset_start; + unsigned long flags; + + offset_start = session->pkt.dpo * 7; + + spin_lock_irqsave(&session->skb_queue.lock, flags); + skb_queue_walk(&session->skb_queue, do_skb) { + do_skcb = j1939_skb_to_cb(do_skb); + + if (offset_start >= do_skcb->offset && + offset_start < (do_skcb->offset + do_skb->len)) { + skb = do_skb; + } + } + spin_unlock_irqrestore(&session->skb_queue.lock, flags); + + if (!skb) + netdev_dbg(priv->ndev, "%s: 0x%p: no skb found for start: %i, queue size: %i\n", + __func__, session, offset_start, + skb_queue_len(&session->skb_queue)); + + return skb; +} + +/* see if we are receiver + * returns 0 for broadcasts, although we will receive them + */ +static inline int j1939_tp_im_receiver(const struct j1939_sk_buff_cb *skcb) +{ + return skcb->flags & J1939_ECU_LOCAL_DST; +} + +/* see if we are sender */ +static inline int j1939_tp_im_transmitter(const struct j1939_sk_buff_cb *skcb) +{ + return skcb->flags & J1939_ECU_LOCAL_SRC; +} + +/* see if we are involved as either receiver or transmitter */ +static int j1939_tp_im_involved(const struct j1939_sk_buff_cb *skcb, bool swap) +{ + if (swap) + return j1939_tp_im_receiver(skcb); + else + return j1939_tp_im_transmitter(skcb); +} + +static int j1939_tp_im_involved_anydir(struct j1939_sk_buff_cb *skcb) +{ + return skcb->flags & (J1939_ECU_LOCAL_SRC | J1939_ECU_LOCAL_DST); +} + +/* extract pgn from flow-ctl message */ +static inline pgn_t j1939_xtp_ctl_to_pgn(const u8 *dat) +{ + pgn_t pgn; + + pgn = (dat[7] << 16) | (dat[6] << 8) | (dat[5] << 0); + if (j1939_pgn_is_pdu1(pgn)) + pgn &= 0xffff00; + return pgn; +} + +static inline unsigned int j1939_tp_ctl_to_size(const u8 *dat) +{ + return (dat[2] << 8) + (dat[1] << 0); +} + +static inline unsigned int j1939_etp_ctl_to_packet(const u8 *dat) +{ + return (dat[4] << 16) | (dat[3] << 8) | (dat[2] << 0); +} + +static inline unsigned int j1939_etp_ctl_to_size(const u8 *dat) +{ + return (dat[4] << 24) | (dat[3] << 16) | + (dat[2] << 8) | (dat[1] << 0); +} + +/* find existing session: + * reverse: swap cb's src & dst + * there is no problem with matching broadcasts, since + * broadcasts (no dst, no da) would never call this + * with reverse == true + */ +static bool j1939_session_match(struct j1939_addr *se_addr, + struct j1939_addr *sk_addr, bool reverse) +{ + if (se_addr->type != sk_addr->type) + return false; + + if (reverse) { + if (se_addr->src_name) { + if (se_addr->src_name != sk_addr->dst_name) + return false; + } else if (se_addr->sa != sk_addr->da) { + return false; + } + + if (se_addr->dst_name) { + if (se_addr->dst_name != sk_addr->src_name) + return false; + } else if (se_addr->da != sk_addr->sa) { + return false; + } + } else { + if (se_addr->src_name) { + if (se_addr->src_name != sk_addr->src_name) + return false; + } else if (se_addr->sa != sk_addr->sa) { + return false; + } + + if (se_addr->dst_name) { + if (se_addr->dst_name != sk_addr->dst_name) + return false; + } else if (se_addr->da != sk_addr->da) { + return false; + } + } + + return true; +} + +static struct +j1939_session *j1939_session_get_by_addr_locked(struct j1939_priv *priv, + struct list_head *root, + struct j1939_addr *addr, + bool reverse, bool transmitter) +{ + struct j1939_session *session; + + lockdep_assert_held(&priv->active_session_list_lock); + + list_for_each_entry(session, root, active_session_list_entry) { + j1939_session_get(session); + if (j1939_session_match(&session->skcb.addr, addr, reverse) && + session->transmission == transmitter) + return session; + j1939_session_put(session); + } + + return NULL; +} + +static struct +j1939_session *j1939_session_get_simple(struct j1939_priv *priv, + struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + + lockdep_assert_held(&priv->active_session_list_lock); + + list_for_each_entry(session, &priv->active_session_list, + active_session_list_entry) { + j1939_session_get(session); + if (session->skcb.addr.type == J1939_SIMPLE && + session->tskey == skcb->tskey && session->sk == skb->sk) + return session; + j1939_session_put(session); + } + + return NULL; +} + +static struct +j1939_session *j1939_session_get_by_addr(struct j1939_priv *priv, + struct j1939_addr *addr, + bool reverse, bool transmitter) +{ + struct j1939_session *session; + + j1939_session_list_lock(priv); + session = j1939_session_get_by_addr_locked(priv, + &priv->active_session_list, + addr, reverse, transmitter); + j1939_session_list_unlock(priv); + + return session; +} + +static void j1939_skbcb_swap(struct j1939_sk_buff_cb *skcb) +{ + u8 tmp = 0; + + swap(skcb->addr.dst_name, skcb->addr.src_name); + swap(skcb->addr.da, skcb->addr.sa); + + /* swap SRC and DST flags, leave other untouched */ + if (skcb->flags & J1939_ECU_LOCAL_SRC) + tmp |= J1939_ECU_LOCAL_DST; + if (skcb->flags & J1939_ECU_LOCAL_DST) + tmp |= J1939_ECU_LOCAL_SRC; + skcb->flags &= ~(J1939_ECU_LOCAL_SRC | J1939_ECU_LOCAL_DST); + skcb->flags |= tmp; +} + +static struct +sk_buff *j1939_tp_tx_dat_new(struct j1939_priv *priv, + const struct j1939_sk_buff_cb *re_skcb, + bool ctl, + bool swap_src_dst) +{ + struct sk_buff *skb; + struct j1939_sk_buff_cb *skcb; + + skb = alloc_skb(sizeof(struct can_frame) + sizeof(struct can_skb_priv), + GFP_ATOMIC); + if (unlikely(!skb)) + return ERR_PTR(-ENOMEM); + + skb->dev = priv->ndev; + can_skb_reserve(skb); + can_skb_prv(skb)->ifindex = priv->ndev->ifindex; + /* reserve CAN header */ + skb_reserve(skb, offsetof(struct can_frame, data)); + + memcpy(skb->cb, re_skcb, sizeof(skb->cb)); + skcb = j1939_skb_to_cb(skb); + if (swap_src_dst) + j1939_skbcb_swap(skcb); + + if (ctl) { + if (skcb->addr.type == J1939_ETP) + skcb->addr.pgn = J1939_ETP_PGN_CTL; + else + skcb->addr.pgn = J1939_TP_PGN_CTL; + } else { + if (skcb->addr.type == J1939_ETP) + skcb->addr.pgn = J1939_ETP_PGN_DAT; + else + skcb->addr.pgn = J1939_TP_PGN_DAT; + } + + return skb; +} + +/* TP transmit packet functions */ +static int j1939_tp_tx_dat(struct j1939_session *session, + const u8 *dat, int len) +{ + struct j1939_priv *priv = session->priv; + struct sk_buff *skb; + + skb = j1939_tp_tx_dat_new(priv, &session->skcb, + false, false); + if (IS_ERR(skb)) + return PTR_ERR(skb); + + skb_put_data(skb, dat, len); + if (j1939_tp_padding && len < 8) + memset(skb_put(skb, 8 - len), 0xff, 8 - len); + + return j1939_send_one(priv, skb); +} + +static int j1939_xtp_do_tx_ctl(struct j1939_priv *priv, + const struct j1939_sk_buff_cb *re_skcb, + bool swap_src_dst, pgn_t pgn, const u8 *dat) +{ + struct sk_buff *skb; + u8 *skdat; + + if (!j1939_tp_im_involved(re_skcb, swap_src_dst)) + return 0; + + skb = j1939_tp_tx_dat_new(priv, re_skcb, true, swap_src_dst); + if (IS_ERR(skb)) + return PTR_ERR(skb); + + skdat = skb_put(skb, 8); + memcpy(skdat, dat, 5); + skdat[5] = (pgn >> 0); + skdat[6] = (pgn >> 8); + skdat[7] = (pgn >> 16); + + return j1939_send_one(priv, skb); +} + +static inline int j1939_tp_tx_ctl(struct j1939_session *session, + bool swap_src_dst, const u8 *dat) +{ + struct j1939_priv *priv = session->priv; + + return j1939_xtp_do_tx_ctl(priv, &session->skcb, + swap_src_dst, + session->skcb.addr.pgn, dat); +} + +static int j1939_xtp_tx_abort(struct j1939_priv *priv, + const struct j1939_sk_buff_cb *re_skcb, + bool swap_src_dst, + enum j1939_xtp_abort err, + pgn_t pgn) +{ + u8 dat[5]; + + if (!j1939_tp_im_involved(re_skcb, swap_src_dst)) + return 0; + + memset(dat, 0xff, sizeof(dat)); + dat[0] = J1939_TP_CMD_ABORT; + dat[1] = err; + return j1939_xtp_do_tx_ctl(priv, re_skcb, swap_src_dst, pgn, dat); +} + +void j1939_tp_schedule_txtimer(struct j1939_session *session, int msec) +{ + j1939_session_get(session); + hrtimer_start(&session->txtimer, ms_to_ktime(msec), + HRTIMER_MODE_REL_SOFT); +} + +static inline void j1939_tp_set_rxtimeout(struct j1939_session *session, + int msec) +{ + j1939_session_rxtimer_cancel(session); + j1939_session_get(session); + hrtimer_start(&session->rxtimer, ms_to_ktime(msec), + HRTIMER_MODE_REL_SOFT); +} + +static int j1939_session_tx_rts(struct j1939_session *session) +{ + u8 dat[8]; + int ret; + + memset(dat, 0xff, sizeof(dat)); + + dat[1] = (session->total_message_size >> 0); + dat[2] = (session->total_message_size >> 8); + dat[3] = session->pkt.total; + + if (session->skcb.addr.type == J1939_ETP) { + dat[0] = J1939_ETP_CMD_RTS; + dat[1] = (session->total_message_size >> 0); + dat[2] = (session->total_message_size >> 8); + dat[3] = (session->total_message_size >> 16); + dat[4] = (session->total_message_size >> 24); + } else if (j1939_cb_is_broadcast(&session->skcb)) { + dat[0] = J1939_TP_CMD_BAM; + /* fake cts for broadcast */ + session->pkt.tx = 0; + } else { + dat[0] = J1939_TP_CMD_RTS; + dat[4] = dat[3]; + } + + if (dat[0] == session->last_txcmd) + /* done already */ + return 0; + + ret = j1939_tp_tx_ctl(session, false, dat); + if (ret < 0) + return ret; + + session->last_txcmd = dat[0]; + if (dat[0] == J1939_TP_CMD_BAM) + j1939_tp_schedule_txtimer(session, 50); + + j1939_tp_set_rxtimeout(session, 1250); + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + return 0; +} + +static int j1939_session_tx_dpo(struct j1939_session *session) +{ + unsigned int pkt; + u8 dat[8]; + int ret; + + memset(dat, 0xff, sizeof(dat)); + + dat[0] = J1939_ETP_CMD_DPO; + session->pkt.dpo = session->pkt.tx_acked; + pkt = session->pkt.dpo; + dat[1] = session->pkt.last - session->pkt.tx_acked; + dat[2] = (pkt >> 0); + dat[3] = (pkt >> 8); + dat[4] = (pkt >> 16); + + ret = j1939_tp_tx_ctl(session, false, dat); + if (ret < 0) + return ret; + + session->last_txcmd = dat[0]; + j1939_tp_set_rxtimeout(session, 1250); + session->pkt.tx = session->pkt.tx_acked; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + return 0; +} + +static int j1939_session_tx_dat(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + struct j1939_sk_buff_cb *skcb; + int offset, pkt_done, pkt_end; + unsigned int len, pdelay; + struct sk_buff *se_skb; + const u8 *tpdat; + int ret = 0; + u8 dat[8]; + + se_skb = j1939_session_skb_find(session); + if (!se_skb) + return -ENOBUFS; + + skcb = j1939_skb_to_cb(se_skb); + tpdat = se_skb->data; + ret = 0; + pkt_done = 0; + if (session->skcb.addr.type != J1939_ETP && + j1939_cb_is_broadcast(&session->skcb)) + pkt_end = session->pkt.total; + else + pkt_end = session->pkt.last; + + while (session->pkt.tx < pkt_end) { + dat[0] = session->pkt.tx - session->pkt.dpo + 1; + offset = (session->pkt.tx * 7) - skcb->offset; + len = se_skb->len - offset; + if (len > 7) + len = 7; + + memcpy(&dat[1], &tpdat[offset], len); + ret = j1939_tp_tx_dat(session, dat, len + 1); + if (ret < 0) { + /* ENOBUS == CAN interface TX queue is full */ + if (ret != -ENOBUFS) + netdev_alert(priv->ndev, + "%s: 0x%p: queue data error: %i\n", + __func__, session, ret); + break; + } + + session->last_txcmd = 0xff; + pkt_done++; + session->pkt.tx++; + pdelay = j1939_cb_is_broadcast(&session->skcb) ? 50 : + j1939_tp_packet_delay; + + if (session->pkt.tx < session->pkt.total && pdelay) { + j1939_tp_schedule_txtimer(session, pdelay); + break; + } + } + + if (pkt_done) + j1939_tp_set_rxtimeout(session, 250); + + return ret; +} + +static int j1939_xtp_txnext_transmiter(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + int ret = 0; + + if (!j1939_tp_im_transmitter(&session->skcb)) { + netdev_alert(priv->ndev, "%s: 0x%p: called by not transmitter!\n", + __func__, session); + return -EINVAL; + } + + switch (session->last_cmd) { + case 0: + ret = j1939_session_tx_rts(session); + break; + + case J1939_ETP_CMD_CTS: + if (session->last_txcmd != J1939_ETP_CMD_DPO) { + ret = j1939_session_tx_dpo(session); + if (ret) + return ret; + } + + /* fall through */ + case J1939_TP_CMD_CTS: + case 0xff: /* did some data */ + case J1939_ETP_CMD_DPO: + case J1939_TP_CMD_BAM: + ret = j1939_session_tx_dat(session); + + break; + default: + netdev_alert(priv->ndev, "%s: 0x%p: unexpected last_cmd: %x\n", + __func__, session, session->last_cmd); + } + + return ret; +} + +static int j1939_session_tx_cts(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + unsigned int pkt, len; + int ret; + u8 dat[8]; + + if (!j1939_sk_recv_match(priv, &session->skcb)) + return -ENOENT; + + len = session->pkt.total - session->pkt.rx; + len = min3(len, session->pkt.block, j1939_tp_block ?: 255); + memset(dat, 0xff, sizeof(dat)); + + if (session->skcb.addr.type == J1939_ETP) { + pkt = session->pkt.rx + 1; + dat[0] = J1939_ETP_CMD_CTS; + dat[1] = len; + dat[2] = (pkt >> 0); + dat[3] = (pkt >> 8); + dat[4] = (pkt >> 16); + } else { + dat[0] = J1939_TP_CMD_CTS; + dat[1] = len; + dat[2] = session->pkt.rx + 1; + } + + if (dat[0] == session->last_txcmd) + /* done already */ + return 0; + + ret = j1939_tp_tx_ctl(session, true, dat); + if (ret < 0) + return ret; + + if (len) + /* only mark cts done when len is set */ + session->last_txcmd = dat[0]; + j1939_tp_set_rxtimeout(session, 1250); + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + return 0; +} + +static int j1939_session_tx_eoma(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + u8 dat[8]; + int ret; + + if (!j1939_sk_recv_match(priv, &session->skcb)) + return -ENOENT; + + memset(dat, 0xff, sizeof(dat)); + + if (session->skcb.addr.type == J1939_ETP) { + dat[0] = J1939_ETP_CMD_EOMA; + dat[1] = session->total_message_size >> 0; + dat[2] = session->total_message_size >> 8; + dat[3] = session->total_message_size >> 16; + dat[4] = session->total_message_size >> 24; + } else { + dat[0] = J1939_TP_CMD_EOMA; + dat[1] = session->total_message_size; + dat[2] = session->total_message_size >> 8; + dat[3] = session->pkt.total; + } + + if (dat[0] == session->last_txcmd) + /* done already */ + return 0; + + ret = j1939_tp_tx_ctl(session, true, dat); + if (ret < 0) + return ret; + + session->last_txcmd = dat[0]; + + /* wait for the EOMA packet to come in */ + j1939_tp_set_rxtimeout(session, 1250); + + netdev_dbg(session->priv->ndev, "%p: 0x%p\n", __func__, session); + + return 0; +} + +static int j1939_xtp_txnext_receiver(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + int ret = 0; + + if (!j1939_tp_im_receiver(&session->skcb)) { + netdev_alert(priv->ndev, "%s: 0x%p: called by not receiver!\n", + __func__, session); + return -EINVAL; + } + + switch (session->last_cmd) { + case J1939_TP_CMD_RTS: + case J1939_ETP_CMD_RTS: + ret = j1939_session_tx_cts(session); + break; + + case J1939_ETP_CMD_CTS: + case J1939_TP_CMD_CTS: + case 0xff: /* did some data */ + case J1939_ETP_CMD_DPO: + if ((session->skcb.addr.type == J1939_TP && + j1939_cb_is_broadcast(&session->skcb))) + break; + + if (session->pkt.rx >= session->pkt.total) { + ret = j1939_session_tx_eoma(session); + } else if (session->pkt.rx >= session->pkt.last) { + session->last_txcmd = 0; + ret = j1939_session_tx_cts(session); + } + break; + default: + netdev_alert(priv->ndev, "%s: 0x%p: unexpected last_cmd: %x\n", + __func__, session, session->last_cmd); + } + + return ret; +} + +static int j1939_simple_txnext(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + struct sk_buff *se_skb = j1939_session_skb_find(session); + struct sk_buff *skb; + int ret; + + if (!se_skb) + return 0; + + skb = skb_clone(se_skb, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + can_skb_set_owner(skb, se_skb->sk); + + j1939_tp_set_rxtimeout(session, J1939_SIMPLE_ECHO_TIMEOUT_MS); + + ret = j1939_send_one(priv, skb); + if (ret) + return ret; + + j1939_sk_errqueue(session, J1939_ERRQUEUE_SCHED); + j1939_sk_queue_activate_next(session); + + return 0; +} + +static bool j1939_session_deactivate_locked(struct j1939_session *session) +{ + bool active = false; + + lockdep_assert_held(&session->priv->active_session_list_lock); + + if (session->state >= J1939_SESSION_ACTIVE && + session->state < J1939_SESSION_ACTIVE_MAX) { + active = true; + + list_del_init(&session->active_session_list_entry); + session->state = J1939_SESSION_DONE; + j1939_session_put(session); + } + + return active; +} + +static bool j1939_session_deactivate(struct j1939_session *session) +{ + bool active; + + j1939_session_list_lock(session->priv); + active = j1939_session_deactivate_locked(session); + j1939_session_list_unlock(session->priv); + + return active; +} + +static void +j1939_session_deactivate_activate_next(struct j1939_session *session) +{ + if (j1939_session_deactivate(session)) + j1939_sk_queue_activate_next(session); +} + +static void j1939_session_cancel(struct j1939_session *session, + enum j1939_xtp_abort err) +{ + struct j1939_priv *priv = session->priv; + + WARN_ON_ONCE(!err); + + session->err = j1939_xtp_abort_to_errno(priv, err); + /* do not send aborts on incoming broadcasts */ + if (!j1939_cb_is_broadcast(&session->skcb)) { + session->state = J1939_SESSION_WAITING_ABORT; + j1939_xtp_tx_abort(priv, &session->skcb, + !session->transmission, + err, session->skcb.addr.pgn); + } + + if (session->sk) + j1939_sk_send_loop_abort(session->sk, session->err); +} + +static enum hrtimer_restart j1939_tp_txtimer(struct hrtimer *hrtimer) +{ + struct j1939_session *session = + container_of(hrtimer, struct j1939_session, txtimer); + struct j1939_priv *priv = session->priv; + int ret = 0; + + if (session->skcb.addr.type == J1939_SIMPLE) { + ret = j1939_simple_txnext(session); + } else { + if (session->transmission) + ret = j1939_xtp_txnext_transmiter(session); + else + ret = j1939_xtp_txnext_receiver(session); + } + + switch (ret) { + case -ENOBUFS: + /* Retry limit is currently arbitrary chosen */ + if (session->tx_retry < J1939_XTP_TX_RETRY_LIMIT) { + session->tx_retry++; + j1939_tp_schedule_txtimer(session, + 10 + prandom_u32_max(16)); + } else { + netdev_alert(priv->ndev, "%s: 0x%p: tx retry count reached\n", + __func__, session); + session->err = -ENETUNREACH; + j1939_session_rxtimer_cancel(session); + j1939_session_deactivate_activate_next(session); + } + break; + case -ENETDOWN: + /* In this case we should get a netdev_event(), all active + * sessions will be cleared by + * j1939_cancel_all_active_sessions(). So handle this as an + * error, but let j1939_cancel_all_active_sessions() do the + * cleanup including propagation of the error to user space. + */ + break; + case 0: + session->tx_retry = 0; + break; + default: + netdev_alert(priv->ndev, "%s: 0x%p: tx aborted with unknown reason: %i\n", + __func__, session, ret); + if (session->skcb.addr.type != J1939_SIMPLE) { + j1939_tp_set_rxtimeout(session, + J1939_XTP_ABORT_TIMEOUT_MS); + j1939_session_cancel(session, J1939_XTP_ABORT_OTHER); + } else { + session->err = ret; + j1939_session_rxtimer_cancel(session); + j1939_session_deactivate_activate_next(session); + } + } + + j1939_session_put(session); + + return HRTIMER_NORESTART; +} + +static void j1939_session_completed(struct j1939_session *session) +{ + struct sk_buff *skb; + + if (!session->transmission) { + skb = j1939_session_skb_find(session); + /* distribute among j1939 receivers */ + j1939_sk_recv(session->priv, skb); + } + + j1939_session_deactivate_activate_next(session); +} + +static enum hrtimer_restart j1939_tp_rxtimer(struct hrtimer *hrtimer) +{ + struct j1939_session *session = container_of(hrtimer, + struct j1939_session, + rxtimer); + struct j1939_priv *priv = session->priv; + + if (session->state == J1939_SESSION_WAITING_ABORT) { + netdev_alert(priv->ndev, "%s: 0x%p: abort rx timeout. Force session deactivation\n", + __func__, session); + + j1939_session_deactivate_activate_next(session); + + } else if (session->skcb.addr.type == J1939_SIMPLE) { + netdev_alert(priv->ndev, "%s: 0x%p: Timeout. Failed to send simple message.\n", + __func__, session); + + /* The message is probably stuck in the CAN controller and can + * be send as soon as CAN bus is in working state again. + */ + session->err = -ETIME; + j1939_session_deactivate(session); + } else { + netdev_alert(priv->ndev, "%s: 0x%p: rx timeout, send abort\n", + __func__, session); + + j1939_session_list_lock(session->priv); + if (session->state >= J1939_SESSION_ACTIVE && + session->state < J1939_SESSION_ACTIVE_MAX) { + j1939_session_get(session); + hrtimer_start(&session->rxtimer, + ms_to_ktime(J1939_XTP_ABORT_TIMEOUT_MS), + HRTIMER_MODE_REL_SOFT); + j1939_session_cancel(session, J1939_XTP_ABORT_TIMEOUT); + } + j1939_session_list_unlock(session->priv); + } + + j1939_session_put(session); + + return HRTIMER_NORESTART; +} + +static bool j1939_xtp_rx_cmd_bad_pgn(struct j1939_session *session, + const struct sk_buff *skb) +{ + const struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + pgn_t pgn = j1939_xtp_ctl_to_pgn(skb->data); + struct j1939_priv *priv = session->priv; + enum j1939_xtp_abort abort = J1939_XTP_NO_ABORT; + u8 cmd = skb->data[0]; + + if (session->skcb.addr.pgn == pgn) + return false; + + switch (cmd) { + case J1939_TP_CMD_BAM: + abort = J1939_XTP_NO_ABORT; + break; + + case J1939_ETP_CMD_RTS: + case J1939_TP_CMD_RTS: /* fall through */ + abort = J1939_XTP_ABORT_BUSY; + break; + + case J1939_ETP_CMD_CTS: + case J1939_TP_CMD_CTS: /* fall through */ + abort = J1939_XTP_ABORT_ECTS_UNXPECTED_PGN; + break; + + case J1939_ETP_CMD_DPO: + abort = J1939_XTP_ABORT_BAD_EDPO_PGN; + break; + + case J1939_ETP_CMD_EOMA: + case J1939_TP_CMD_EOMA: /* fall through */ + abort = J1939_XTP_ABORT_OTHER; + break; + + case J1939_ETP_CMD_ABORT: /* && J1939_TP_CMD_ABORT */ + abort = J1939_XTP_NO_ABORT; + break; + + default: + WARN_ON_ONCE(1); + break; + } + + netdev_warn(priv->ndev, "%s: 0x%p: CMD 0x%02x with PGN 0x%05x for running session with different PGN 0x%05x.\n", + __func__, session, cmd, pgn, session->skcb.addr.pgn); + if (abort != J1939_XTP_NO_ABORT) + j1939_xtp_tx_abort(priv, skcb, true, abort, pgn); + + return true; +} + +static void j1939_xtp_rx_abort_one(struct j1939_priv *priv, struct sk_buff *skb, + bool reverse, bool transmitter) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + u8 abort = skb->data[1]; + + session = j1939_session_get_by_addr(priv, &skcb->addr, reverse, + transmitter); + if (!session) + return; + + if (j1939_xtp_rx_cmd_bad_pgn(session, skb)) + goto abort_put; + + netdev_info(priv->ndev, "%s: 0x%p: 0x%05x: (%u) %s\n", __func__, + session, j1939_xtp_ctl_to_pgn(skb->data), abort, + j1939_xtp_abort_to_str(abort)); + + j1939_session_timers_cancel(session); + session->err = j1939_xtp_abort_to_errno(priv, abort); + if (session->sk) + j1939_sk_send_loop_abort(session->sk, session->err); + j1939_session_deactivate_activate_next(session); + +abort_put: + j1939_session_put(session); +} + +/* abort packets may come in 2 directions */ +static void +j1939_xtp_rx_abort(struct j1939_priv *priv, struct sk_buff *skb, + bool transmitter) +{ + j1939_xtp_rx_abort_one(priv, skb, false, transmitter); + j1939_xtp_rx_abort_one(priv, skb, true, transmitter); +} + +static void +j1939_xtp_rx_eoma_one(struct j1939_session *session, struct sk_buff *skb) +{ + if (j1939_xtp_rx_cmd_bad_pgn(session, skb)) + return; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + session->pkt.tx_acked = session->pkt.total; + j1939_session_timers_cancel(session); + /* transmitted without problems */ + j1939_session_completed(session); +} + +static void +j1939_xtp_rx_eoma(struct j1939_priv *priv, struct sk_buff *skb, + bool transmitter) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + + session = j1939_session_get_by_addr(priv, &skcb->addr, true, + transmitter); + if (!session) + return; + + j1939_xtp_rx_eoma_one(session, skb); + j1939_session_put(session); +} + +static void +j1939_xtp_rx_cts_one(struct j1939_session *session, struct sk_buff *skb) +{ + enum j1939_xtp_abort err = J1939_XTP_ABORT_FAULT; + unsigned int pkt; + const u8 *dat; + + dat = skb->data; + + if (j1939_xtp_rx_cmd_bad_pgn(session, skb)) + return; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + if (session->last_cmd == dat[0]) { + err = J1939_XTP_ABORT_DUP_SEQ; + goto out_session_cancel; + } + + if (session->skcb.addr.type == J1939_ETP) + pkt = j1939_etp_ctl_to_packet(dat); + else + pkt = dat[2]; + + if (!pkt) + goto out_session_cancel; + else if (dat[1] > session->pkt.block /* 0xff for etp */) + goto out_session_cancel; + + /* set packet counters only when not CTS(0) */ + session->pkt.tx_acked = pkt - 1; + j1939_session_skb_drop_old(session); + session->pkt.last = session->pkt.tx_acked + dat[1]; + if (session->pkt.last > session->pkt.total) + /* safety measure */ + session->pkt.last = session->pkt.total; + /* TODO: do not set tx here, do it in txtimer */ + session->pkt.tx = session->pkt.tx_acked; + + session->last_cmd = dat[0]; + if (dat[1]) { + j1939_tp_set_rxtimeout(session, 1250); + if (session->transmission) { + if (session->pkt.tx_acked) + j1939_sk_errqueue(session, + J1939_ERRQUEUE_SCHED); + j1939_session_txtimer_cancel(session); + j1939_tp_schedule_txtimer(session, 0); + } + } else { + /* CTS(0) */ + j1939_tp_set_rxtimeout(session, 550); + } + return; + + out_session_cancel: + j1939_session_timers_cancel(session); + j1939_tp_set_rxtimeout(session, J1939_XTP_ABORT_TIMEOUT_MS); + j1939_session_cancel(session, err); +} + +static void +j1939_xtp_rx_cts(struct j1939_priv *priv, struct sk_buff *skb, bool transmitter) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + + session = j1939_session_get_by_addr(priv, &skcb->addr, true, + transmitter); + if (!session) + return; + j1939_xtp_rx_cts_one(session, skb); + j1939_session_put(session); +} + +static struct j1939_session *j1939_session_new(struct j1939_priv *priv, + struct sk_buff *skb, size_t size) +{ + struct j1939_session *session; + struct j1939_sk_buff_cb *skcb; + + session = kzalloc(sizeof(*session), gfp_any()); + if (!session) + return NULL; + + INIT_LIST_HEAD(&session->active_session_list_entry); + INIT_LIST_HEAD(&session->sk_session_queue_entry); + kref_init(&session->kref); + + j1939_priv_get(priv); + session->priv = priv; + session->total_message_size = size; + session->state = J1939_SESSION_NEW; + + skb_queue_head_init(&session->skb_queue); + skb_queue_tail(&session->skb_queue, skb); + + skcb = j1939_skb_to_cb(skb); + memcpy(&session->skcb, skcb, sizeof(session->skcb)); + + hrtimer_init(&session->txtimer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); + session->txtimer.function = j1939_tp_txtimer; + hrtimer_init(&session->rxtimer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_SOFT); + session->rxtimer.function = j1939_tp_rxtimer; + + netdev_dbg(priv->ndev, "%s: 0x%p: sa: %02x, da: %02x\n", + __func__, session, skcb->addr.sa, skcb->addr.da); + + return session; +} + +static struct +j1939_session *j1939_session_fresh_new(struct j1939_priv *priv, + int size, + const struct j1939_sk_buff_cb *rel_skcb) +{ + struct sk_buff *skb; + struct j1939_sk_buff_cb *skcb; + struct j1939_session *session; + + skb = alloc_skb(size + sizeof(struct can_skb_priv), GFP_ATOMIC); + if (unlikely(!skb)) + return NULL; + + skb->dev = priv->ndev; + can_skb_reserve(skb); + can_skb_prv(skb)->ifindex = priv->ndev->ifindex; + skcb = j1939_skb_to_cb(skb); + memcpy(skcb, rel_skcb, sizeof(*skcb)); + + session = j1939_session_new(priv, skb, skb->len); + if (!session) { + kfree_skb(skb); + return NULL; + } + + /* alloc data area */ + skb_put(skb, size); + /* skb is recounted in j1939_session_new() */ + return session; +} + +int j1939_session_activate(struct j1939_session *session) +{ + struct j1939_priv *priv = session->priv; + struct j1939_session *active = NULL; + int ret = 0; + + j1939_session_list_lock(priv); + if (session->skcb.addr.type != J1939_SIMPLE) + active = j1939_session_get_by_addr_locked(priv, + &priv->active_session_list, + &session->skcb.addr, false, + session->transmission); + if (active) { + j1939_session_put(active); + ret = -EAGAIN; + } else { + WARN_ON_ONCE(session->state != J1939_SESSION_NEW); + list_add_tail(&session->active_session_list_entry, + &priv->active_session_list); + j1939_session_get(session); + session->state = J1939_SESSION_ACTIVE; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", + __func__, session); + } + j1939_session_list_unlock(priv); + + return ret; +} + +static struct +j1939_session *j1939_xtp_rx_rts_session_new(struct j1939_priv *priv, + struct sk_buff *skb) +{ + enum j1939_xtp_abort abort = J1939_XTP_NO_ABORT; + struct j1939_sk_buff_cb skcb = *j1939_skb_to_cb(skb); + struct j1939_session *session; + const u8 *dat; + pgn_t pgn; + int len; + + netdev_dbg(priv->ndev, "%s\n", __func__); + + dat = skb->data; + pgn = j1939_xtp_ctl_to_pgn(dat); + skcb.addr.pgn = pgn; + + if (!j1939_sk_recv_match(priv, &skcb)) + return NULL; + + if (skcb.addr.type == J1939_ETP) { + len = j1939_etp_ctl_to_size(dat); + if (len > J1939_MAX_ETP_PACKET_SIZE) + abort = J1939_XTP_ABORT_FAULT; + else if (len > priv->tp_max_packet_size) + abort = J1939_XTP_ABORT_RESOURCE; + else if (len <= J1939_MAX_TP_PACKET_SIZE) + abort = J1939_XTP_ABORT_FAULT; + } else { + len = j1939_tp_ctl_to_size(dat); + if (len > J1939_MAX_TP_PACKET_SIZE) + abort = J1939_XTP_ABORT_FAULT; + else if (len > priv->tp_max_packet_size) + abort = J1939_XTP_ABORT_RESOURCE; + } + + if (abort != J1939_XTP_NO_ABORT) { + j1939_xtp_tx_abort(priv, &skcb, true, abort, pgn); + return NULL; + } + + session = j1939_session_fresh_new(priv, len, &skcb); + if (!session) { + j1939_xtp_tx_abort(priv, &skcb, true, + J1939_XTP_ABORT_RESOURCE, pgn); + return NULL; + } + + /* initialize the control buffer: plain copy */ + session->pkt.total = (len + 6) / 7; + session->pkt.block = 0xff; + if (skcb.addr.type != J1939_ETP) { + if (dat[3] != session->pkt.total) + netdev_alert(priv->ndev, "%s: 0x%p: strange total, %u != %u\n", + __func__, session, session->pkt.total, + dat[3]); + session->pkt.total = dat[3]; + session->pkt.block = min(dat[3], dat[4]); + } + + session->pkt.rx = 0; + session->pkt.tx = 0; + + WARN_ON_ONCE(j1939_session_activate(session)); + + return session; +} + +static int j1939_xtp_rx_rts_session_active(struct j1939_session *session, + struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_priv *priv = session->priv; + + if (!session->transmission) { + if (j1939_xtp_rx_cmd_bad_pgn(session, skb)) + return -EBUSY; + + /* RTS on active session */ + j1939_session_timers_cancel(session); + j1939_tp_set_rxtimeout(session, J1939_XTP_ABORT_TIMEOUT_MS); + j1939_session_cancel(session, J1939_XTP_ABORT_BUSY); + } + + if (session->last_cmd != 0) { + /* we received a second rts on the same connection */ + netdev_alert(priv->ndev, "%s: 0x%p: connection exists (%02x %02x). last cmd: %x\n", + __func__, session, skcb->addr.sa, skcb->addr.da, + session->last_cmd); + + j1939_session_timers_cancel(session); + j1939_tp_set_rxtimeout(session, J1939_XTP_ABORT_TIMEOUT_MS); + j1939_session_cancel(session, J1939_XTP_ABORT_BUSY); + + return -EBUSY; + } + + if (session->skcb.addr.sa != skcb->addr.sa || + session->skcb.addr.da != skcb->addr.da) + netdev_warn(priv->ndev, "%s: 0x%p: session->skcb.addr.sa=0x%02x skcb->addr.sa=0x%02x session->skcb.addr.da=0x%02x skcb->addr.da=0x%02x\n", + __func__, session, + session->skcb.addr.sa, skcb->addr.sa, + session->skcb.addr.da, skcb->addr.da); + /* make sure 'sa' & 'da' are correct ! + * They may be 'not filled in yet' for sending + * skb's, since they did not pass the Address Claim ever. + */ + session->skcb.addr.sa = skcb->addr.sa; + session->skcb.addr.da = skcb->addr.da; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + return 0; +} + +static void j1939_xtp_rx_rts(struct j1939_priv *priv, struct sk_buff *skb, + bool transmitter) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + u8 cmd = skb->data[0]; + + session = j1939_session_get_by_addr(priv, &skcb->addr, false, + transmitter); + + if (!session) { + if (transmitter) { + /* If we're the transmitter and this function is called, + * we received our own RTS. A session has already been + * created. + * + * For some reasons however it might have been destroyed + * already. So don't create a new one here (using + * "j1939_xtp_rx_rts_session_new()") as this will be a + * receiver session. + * + * The reasons the session is already destroyed might + * be: + * - user space closed socket was and the session was + * aborted + * - session was aborted due to external abort message + */ + return; + } + session = j1939_xtp_rx_rts_session_new(priv, skb); + if (!session) + return; + } else { + if (j1939_xtp_rx_rts_session_active(session, skb)) { + j1939_session_put(session); + return; + } + } + session->last_cmd = cmd; + + j1939_tp_set_rxtimeout(session, 1250); + + if (cmd != J1939_TP_CMD_BAM && !session->transmission) { + j1939_session_txtimer_cancel(session); + j1939_tp_schedule_txtimer(session, 0); + } + + j1939_session_put(session); +} + +static void j1939_xtp_rx_dpo_one(struct j1939_session *session, + struct sk_buff *skb) +{ + const u8 *dat = skb->data; + + if (j1939_xtp_rx_cmd_bad_pgn(session, skb)) + return; + + netdev_dbg(session->priv->ndev, "%s: 0x%p\n", __func__, session); + + /* transmitted without problems */ + session->pkt.dpo = j1939_etp_ctl_to_packet(skb->data); + session->last_cmd = dat[0]; + j1939_tp_set_rxtimeout(session, 750); +} + +static void j1939_xtp_rx_dpo(struct j1939_priv *priv, struct sk_buff *skb, + bool transmitter) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + + session = j1939_session_get_by_addr(priv, &skcb->addr, false, + transmitter); + if (!session) { + netdev_info(priv->ndev, + "%s: no connection found\n", __func__); + return; + } + + j1939_xtp_rx_dpo_one(session, skb); + j1939_session_put(session); +} + +static void j1939_xtp_rx_dat_one(struct j1939_session *session, + struct sk_buff *skb) +{ + struct j1939_priv *priv = session->priv; + struct j1939_sk_buff_cb *skcb; + struct sk_buff *se_skb; + const u8 *dat; + u8 *tpdat; + int offset; + int nbytes; + bool final = false; + bool do_cts_eoma = false; + int packet; + + skcb = j1939_skb_to_cb(skb); + dat = skb->data; + if (skb->len <= 1) + /* makes no sense */ + goto out_session_cancel; + + switch (session->last_cmd) { + case 0xff: + break; + case J1939_ETP_CMD_DPO: + if (skcb->addr.type == J1939_ETP) + break; + /* fall through */ + case J1939_TP_CMD_BAM: /* fall through */ + case J1939_TP_CMD_CTS: /* fall through */ + if (skcb->addr.type != J1939_ETP) + break; + /* fall through */ + default: + netdev_info(priv->ndev, "%s: 0x%p: last %02x\n", __func__, + session, session->last_cmd); + goto out_session_cancel; + } + + packet = (dat[0] - 1 + session->pkt.dpo); + if (packet > session->pkt.total || + (session->pkt.rx + 1) > session->pkt.total) { + netdev_info(priv->ndev, "%s: 0x%p: should have been completed\n", + __func__, session); + goto out_session_cancel; + } + se_skb = j1939_session_skb_find(session); + if (!se_skb) { + netdev_warn(priv->ndev, "%s: 0x%p: no skb found\n", __func__, + session); + goto out_session_cancel; + } + + skcb = j1939_skb_to_cb(se_skb); + offset = packet * 7 - skcb->offset; + nbytes = se_skb->len - offset; + if (nbytes > 7) + nbytes = 7; + if (nbytes <= 0 || (nbytes + 1) > skb->len) { + netdev_info(priv->ndev, "%s: 0x%p: nbytes %i, len %i\n", + __func__, session, nbytes, skb->len); + goto out_session_cancel; + } + + tpdat = se_skb->data; + memcpy(&tpdat[offset], &dat[1], nbytes); + if (packet == session->pkt.rx) + session->pkt.rx++; + + if (skcb->addr.type != J1939_ETP && + j1939_cb_is_broadcast(&session->skcb)) { + if (session->pkt.rx >= session->pkt.total) + final = true; + } else { + /* never final, an EOMA must follow */ + if (session->pkt.rx >= session->pkt.last) + do_cts_eoma = true; + } + + if (final) { + j1939_session_completed(session); + } else if (do_cts_eoma) { + j1939_tp_set_rxtimeout(session, 1250); + if (!session->transmission) + j1939_tp_schedule_txtimer(session, 0); + } else { + j1939_tp_set_rxtimeout(session, 250); + } + session->last_cmd = 0xff; + j1939_session_put(session); + + return; + + out_session_cancel: + j1939_session_timers_cancel(session); + j1939_tp_set_rxtimeout(session, J1939_XTP_ABORT_TIMEOUT_MS); + j1939_session_cancel(session, J1939_XTP_ABORT_FAULT); + j1939_session_put(session); +} + +static void j1939_xtp_rx_dat(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb; + struct j1939_session *session; + + skcb = j1939_skb_to_cb(skb); + + if (j1939_tp_im_transmitter(skcb)) { + session = j1939_session_get_by_addr(priv, &skcb->addr, false, + true); + if (!session) + netdev_info(priv->ndev, "%s: no tx connection found\n", + __func__); + else + j1939_xtp_rx_dat_one(session, skb); + } + + if (j1939_tp_im_receiver(skcb)) { + session = j1939_session_get_by_addr(priv, &skcb->addr, false, + false); + if (!session) + netdev_info(priv->ndev, "%s: no rx connection found\n", + __func__); + else + j1939_xtp_rx_dat_one(session, skb); + } +} + +/* j1939 main intf */ +struct j1939_session *j1939_tp_send(struct j1939_priv *priv, + struct sk_buff *skb, size_t size) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + struct j1939_session *session; + int ret; + + if (skcb->addr.pgn == J1939_TP_PGN_DAT || + skcb->addr.pgn == J1939_TP_PGN_CTL || + skcb->addr.pgn == J1939_ETP_PGN_DAT || + skcb->addr.pgn == J1939_ETP_PGN_CTL) + /* avoid conflict */ + return ERR_PTR(-EDOM); + + if (size > priv->tp_max_packet_size) + return ERR_PTR(-EMSGSIZE); + + if (size <= 8) + skcb->addr.type = J1939_SIMPLE; + else if (size > J1939_MAX_TP_PACKET_SIZE) + skcb->addr.type = J1939_ETP; + else + skcb->addr.type = J1939_TP; + + if (skcb->addr.type == J1939_ETP && + j1939_cb_is_broadcast(skcb)) + return ERR_PTR(-EDESTADDRREQ); + + /* fill in addresses from names */ + ret = j1939_ac_fixup(priv, skb); + if (unlikely(ret)) + return ERR_PTR(ret); + + /* fix DST flags, it may be used there soon */ + if (j1939_address_is_unicast(skcb->addr.da) && + priv->ents[skcb->addr.da].nusers) + skcb->flags |= J1939_ECU_LOCAL_DST; + + /* src is always local, I'm sending ... */ + skcb->flags |= J1939_ECU_LOCAL_SRC; + + /* prepare new session */ + session = j1939_session_new(priv, skb, size); + if (!session) + return ERR_PTR(-ENOMEM); + + /* skb is recounted in j1939_session_new() */ + session->sk = skb->sk; + session->transmission = true; + session->pkt.total = (size + 6) / 7; + session->pkt.block = skcb->addr.type == J1939_ETP ? 255 : + min(j1939_tp_block ?: 255, session->pkt.total); + + if (j1939_cb_is_broadcast(&session->skcb)) + /* set the end-packet for broadcast */ + session->pkt.last = session->pkt.total; + + skcb->tskey = session->sk->sk_tskey++; + session->tskey = skcb->tskey; + + return session; +} + +static void j1939_tp_cmd_recv(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + int extd = J1939_TP; + u8 cmd = skb->data[0]; + + switch (cmd) { + case J1939_ETP_CMD_RTS: + extd = J1939_ETP; + /* fall through */ + case J1939_TP_CMD_BAM: /* fall through */ + case J1939_TP_CMD_RTS: /* fall through */ + if (skcb->addr.type != extd) + return; + + if (cmd == J1939_TP_CMD_RTS && j1939_cb_is_broadcast(skcb)) { + netdev_alert(priv->ndev, "%s: rts without destination (%02x)\n", + __func__, skcb->addr.sa); + return; + } + + if (j1939_tp_im_transmitter(skcb)) + j1939_xtp_rx_rts(priv, skb, true); + + if (j1939_tp_im_receiver(skcb)) + j1939_xtp_rx_rts(priv, skb, false); + + break; + + case J1939_ETP_CMD_CTS: + extd = J1939_ETP; + /* fall through */ + case J1939_TP_CMD_CTS: + if (skcb->addr.type != extd) + return; + + if (j1939_tp_im_transmitter(skcb)) + j1939_xtp_rx_cts(priv, skb, false); + + if (j1939_tp_im_receiver(skcb)) + j1939_xtp_rx_cts(priv, skb, true); + + break; + + case J1939_ETP_CMD_DPO: + if (skcb->addr.type != J1939_ETP) + return; + + if (j1939_tp_im_transmitter(skcb)) + j1939_xtp_rx_dpo(priv, skb, true); + + if (j1939_tp_im_receiver(skcb)) + j1939_xtp_rx_dpo(priv, skb, false); + + break; + + case J1939_ETP_CMD_EOMA: + extd = J1939_ETP; + /* fall through */ + case J1939_TP_CMD_EOMA: + if (skcb->addr.type != extd) + return; + + if (j1939_tp_im_transmitter(skcb)) + j1939_xtp_rx_eoma(priv, skb, false); + + if (j1939_tp_im_receiver(skcb)) + j1939_xtp_rx_eoma(priv, skb, true); + + break; + + case J1939_ETP_CMD_ABORT: /* && J1939_TP_CMD_ABORT */ + if (j1939_tp_im_transmitter(skcb)) + j1939_xtp_rx_abort(priv, skb, true); + + if (j1939_tp_im_receiver(skcb)) + j1939_xtp_rx_abort(priv, skb, false); + + break; + default: + return; + } +} + +int j1939_tp_recv(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_sk_buff_cb *skcb = j1939_skb_to_cb(skb); + + if (!j1939_tp_im_involved_anydir(skcb)) + return 0; + + switch (skcb->addr.pgn) { + case J1939_ETP_PGN_DAT: + skcb->addr.type = J1939_ETP; + /* fall through */ + case J1939_TP_PGN_DAT: + j1939_xtp_rx_dat(priv, skb); + break; + + case J1939_ETP_PGN_CTL: + skcb->addr.type = J1939_ETP; + /* fall through */ + case J1939_TP_PGN_CTL: + if (skb->len < 8) + return 0; /* Don't care. Nothing to extract here */ + + j1939_tp_cmd_recv(priv, skb); + break; + default: + return 0; /* no problem */ + } + return 1; /* "I processed the message" */ +} + +void j1939_simple_recv(struct j1939_priv *priv, struct sk_buff *skb) +{ + struct j1939_session *session; + + if (!skb->sk) + return; + + j1939_session_list_lock(priv); + session = j1939_session_get_simple(priv, skb); + j1939_session_list_unlock(priv); + if (!session) { + netdev_warn(priv->ndev, + "%s: Received already invalidated message\n", + __func__); + return; + } + + j1939_session_timers_cancel(session); + j1939_session_deactivate(session); + j1939_session_put(session); +} + +int j1939_cancel_active_session(struct j1939_priv *priv, struct sock *sk) +{ + struct j1939_session *session, *saved; + + netdev_dbg(priv->ndev, "%s, sk: %p\n", __func__, sk); + j1939_session_list_lock(priv); + list_for_each_entry_safe(session, saved, + &priv->active_session_list, + active_session_list_entry) { + if (!sk || sk == session->sk) { + j1939_session_timers_cancel(session); + session->err = ESHUTDOWN; + j1939_session_deactivate_locked(session); + } + } + j1939_session_list_unlock(priv); + return NOTIFY_DONE; +} + +void j1939_tp_init(struct j1939_priv *priv) +{ + spin_lock_init(&priv->active_session_list_lock); + INIT_LIST_HEAD(&priv->active_session_list); + priv->tp_max_packet_size = J1939_MAX_ETP_PACKET_SIZE; +} diff --git a/net/can/proc.c b/net/can/proc.c index edb822c31902..e6881bfc3ed1 100644 --- a/net/can/proc.c +++ b/net/can/proc.c @@ -45,6 +45,7 @@ #include <linux/list.h> #include <linux/rcupdate.h> #include <linux/if_arp.h> +#include <linux/can/can-ml.h> #include <linux/can/core.h> #include "af_can.h" @@ -78,21 +79,21 @@ static const char rx_list_name[][8] = { static void can_init_stats(struct net *net) { - struct s_stats *can_stats = net->can.can_stats; - struct s_pstats *can_pstats = net->can.can_pstats; + struct can_pkg_stats *pkg_stats = net->can.pkg_stats; + struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats; /* * This memset function is called from a timer context (when * can_stattimer is active which is the default) OR in a process * context (reading the proc_fs when can_stattimer is disabled). */ - memset(can_stats, 0, sizeof(struct s_stats)); - can_stats->jiffies_init = jiffies; + memset(pkg_stats, 0, sizeof(struct can_pkg_stats)); + pkg_stats->jiffies_init = jiffies; - can_pstats->stats_reset++; + rcv_lists_stats->stats_reset++; if (user_reset) { user_reset = 0; - can_pstats->user_reset++; + rcv_lists_stats->user_reset++; } } @@ -118,8 +119,8 @@ static unsigned long calc_rate(unsigned long oldjif, unsigned long newjif, void can_stat_update(struct timer_list *t) { - struct net *net = from_timer(net, t, can.can_stattimer); - struct s_stats *can_stats = net->can.can_stats; + struct net *net = from_timer(net, t, can.stattimer); + struct can_pkg_stats *pkg_stats = net->can.pkg_stats; unsigned long j = jiffies; /* snapshot */ /* restart counting in timer context on user request */ @@ -127,57 +128,57 @@ void can_stat_update(struct timer_list *t) can_init_stats(net); /* restart counting on jiffies overflow */ - if (j < can_stats->jiffies_init) + if (j < pkg_stats->jiffies_init) can_init_stats(net); /* prevent overflow in calc_rate() */ - if (can_stats->rx_frames > (ULONG_MAX / HZ)) + if (pkg_stats->rx_frames > (ULONG_MAX / HZ)) can_init_stats(net); /* prevent overflow in calc_rate() */ - if (can_stats->tx_frames > (ULONG_MAX / HZ)) + if (pkg_stats->tx_frames > (ULONG_MAX / HZ)) can_init_stats(net); /* matches overflow - very improbable */ - if (can_stats->matches > (ULONG_MAX / 100)) + if (pkg_stats->matches > (ULONG_MAX / 100)) can_init_stats(net); /* calc total values */ - if (can_stats->rx_frames) - can_stats->total_rx_match_ratio = (can_stats->matches * 100) / - can_stats->rx_frames; + if (pkg_stats->rx_frames) + pkg_stats->total_rx_match_ratio = (pkg_stats->matches * 100) / + pkg_stats->rx_frames; - can_stats->total_tx_rate = calc_rate(can_stats->jiffies_init, j, - can_stats->tx_frames); - can_stats->total_rx_rate = calc_rate(can_stats->jiffies_init, j, - can_stats->rx_frames); + pkg_stats->total_tx_rate = calc_rate(pkg_stats->jiffies_init, j, + pkg_stats->tx_frames); + pkg_stats->total_rx_rate = calc_rate(pkg_stats->jiffies_init, j, + pkg_stats->rx_frames); /* calc current values */ - if (can_stats->rx_frames_delta) - can_stats->current_rx_match_ratio = - (can_stats->matches_delta * 100) / - can_stats->rx_frames_delta; + if (pkg_stats->rx_frames_delta) + pkg_stats->current_rx_match_ratio = + (pkg_stats->matches_delta * 100) / + pkg_stats->rx_frames_delta; - can_stats->current_tx_rate = calc_rate(0, HZ, can_stats->tx_frames_delta); - can_stats->current_rx_rate = calc_rate(0, HZ, can_stats->rx_frames_delta); + pkg_stats->current_tx_rate = calc_rate(0, HZ, pkg_stats->tx_frames_delta); + pkg_stats->current_rx_rate = calc_rate(0, HZ, pkg_stats->rx_frames_delta); /* check / update maximum values */ - if (can_stats->max_tx_rate < can_stats->current_tx_rate) - can_stats->max_tx_rate = can_stats->current_tx_rate; + if (pkg_stats->max_tx_rate < pkg_stats->current_tx_rate) + pkg_stats->max_tx_rate = pkg_stats->current_tx_rate; - if (can_stats->max_rx_rate < can_stats->current_rx_rate) - can_stats->max_rx_rate = can_stats->current_rx_rate; + if (pkg_stats->max_rx_rate < pkg_stats->current_rx_rate) + pkg_stats->max_rx_rate = pkg_stats->current_rx_rate; - if (can_stats->max_rx_match_ratio < can_stats->current_rx_match_ratio) - can_stats->max_rx_match_ratio = can_stats->current_rx_match_ratio; + if (pkg_stats->max_rx_match_ratio < pkg_stats->current_rx_match_ratio) + pkg_stats->max_rx_match_ratio = pkg_stats->current_rx_match_ratio; /* clear values for 'current rate' calculation */ - can_stats->tx_frames_delta = 0; - can_stats->rx_frames_delta = 0; - can_stats->matches_delta = 0; + pkg_stats->tx_frames_delta = 0; + pkg_stats->rx_frames_delta = 0; + pkg_stats->matches_delta = 0; /* restart timer (one second) */ - mod_timer(&net->can.can_stattimer, round_jiffies(jiffies + HZ)); + mod_timer(&net->can.stattimer, round_jiffies(jiffies + HZ)); } /* @@ -212,60 +213,60 @@ static void can_print_recv_banner(struct seq_file *m) static int can_stats_proc_show(struct seq_file *m, void *v) { struct net *net = m->private; - struct s_stats *can_stats = net->can.can_stats; - struct s_pstats *can_pstats = net->can.can_pstats; + struct can_pkg_stats *pkg_stats = net->can.pkg_stats; + struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats; seq_putc(m, '\n'); - seq_printf(m, " %8ld transmitted frames (TXF)\n", can_stats->tx_frames); - seq_printf(m, " %8ld received frames (RXF)\n", can_stats->rx_frames); - seq_printf(m, " %8ld matched frames (RXMF)\n", can_stats->matches); + seq_printf(m, " %8ld transmitted frames (TXF)\n", pkg_stats->tx_frames); + seq_printf(m, " %8ld received frames (RXF)\n", pkg_stats->rx_frames); + seq_printf(m, " %8ld matched frames (RXMF)\n", pkg_stats->matches); seq_putc(m, '\n'); - if (net->can.can_stattimer.function == can_stat_update) { + if (net->can.stattimer.function == can_stat_update) { seq_printf(m, " %8ld %% total match ratio (RXMR)\n", - can_stats->total_rx_match_ratio); + pkg_stats->total_rx_match_ratio); seq_printf(m, " %8ld frames/s total tx rate (TXR)\n", - can_stats->total_tx_rate); + pkg_stats->total_tx_rate); seq_printf(m, " %8ld frames/s total rx rate (RXR)\n", - can_stats->total_rx_rate); + pkg_stats->total_rx_rate); seq_putc(m, '\n'); seq_printf(m, " %8ld %% current match ratio (CRXMR)\n", - can_stats->current_rx_match_ratio); + pkg_stats->current_rx_match_ratio); seq_printf(m, " %8ld frames/s current tx rate (CTXR)\n", - can_stats->current_tx_rate); + pkg_stats->current_tx_rate); seq_printf(m, " %8ld frames/s current rx rate (CRXR)\n", - can_stats->current_rx_rate); + pkg_stats->current_rx_rate); seq_putc(m, '\n'); seq_printf(m, " %8ld %% max match ratio (MRXMR)\n", - can_stats->max_rx_match_ratio); + pkg_stats->max_rx_match_ratio); seq_printf(m, " %8ld frames/s max tx rate (MTXR)\n", - can_stats->max_tx_rate); + pkg_stats->max_tx_rate); seq_printf(m, " %8ld frames/s max rx rate (MRXR)\n", - can_stats->max_rx_rate); + pkg_stats->max_rx_rate); seq_putc(m, '\n'); } seq_printf(m, " %8ld current receive list entries (CRCV)\n", - can_pstats->rcv_entries); + rcv_lists_stats->rcv_entries); seq_printf(m, " %8ld maximum receive list entries (MRCV)\n", - can_pstats->rcv_entries_max); + rcv_lists_stats->rcv_entries_max); - if (can_pstats->stats_reset) + if (rcv_lists_stats->stats_reset) seq_printf(m, "\n %8ld statistic resets (STR)\n", - can_pstats->stats_reset); + rcv_lists_stats->stats_reset); - if (can_pstats->user_reset) + if (rcv_lists_stats->user_reset) seq_printf(m, " %8ld user statistic resets (USTR)\n", - can_pstats->user_reset); + rcv_lists_stats->user_reset); seq_putc(m, '\n'); return 0; @@ -274,20 +275,20 @@ static int can_stats_proc_show(struct seq_file *m, void *v) static int can_reset_stats_proc_show(struct seq_file *m, void *v) { struct net *net = m->private; - struct s_pstats *can_pstats = net->can.can_pstats; - struct s_stats *can_stats = net->can.can_stats; + struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats; + struct can_pkg_stats *pkg_stats = net->can.pkg_stats; user_reset = 1; - if (net->can.can_stattimer.function == can_stat_update) { + if (net->can.stattimer.function == can_stat_update) { seq_printf(m, "Scheduled statistic reset #%ld.\n", - can_pstats->stats_reset + 1); + rcv_lists_stats->stats_reset + 1); } else { - if (can_stats->jiffies_init != jiffies) + if (pkg_stats->jiffies_init != jiffies) can_init_stats(net); seq_printf(m, "Performed statistic reset #%ld.\n", - can_pstats->stats_reset); + rcv_lists_stats->stats_reset); } return 0; } @@ -300,11 +301,11 @@ static int can_version_proc_show(struct seq_file *m, void *v) static inline void can_rcvlist_proc_show_one(struct seq_file *m, int idx, struct net_device *dev, - struct can_dev_rcv_lists *d) + struct can_dev_rcv_lists *dev_rcv_lists) { - if (!hlist_empty(&d->rx[idx])) { + if (!hlist_empty(&dev_rcv_lists->rx[idx])) { can_print_recv_banner(m); - can_print_rcvlist(m, &d->rx[idx], dev); + can_print_rcvlist(m, &dev_rcv_lists->rx[idx], dev); } else seq_printf(m, " (%s: no entry)\n", DNAME(dev)); @@ -315,7 +316,7 @@ static int can_rcvlist_proc_show(struct seq_file *m, void *v) /* double cast to prevent GCC warning */ int idx = (int)(long)PDE_DATA(m->file->f_inode); struct net_device *dev; - struct can_dev_rcv_lists *d; + struct can_dev_rcv_lists *dev_rcv_lists; struct net *net = m->private; seq_printf(m, "\nreceive list '%s':\n", rx_list_name[idx]); @@ -323,8 +324,8 @@ static int can_rcvlist_proc_show(struct seq_file *m, void *v) rcu_read_lock(); /* receive list for 'all' CAN devices (dev == NULL) */ - d = net->can.can_rx_alldev_list; - can_rcvlist_proc_show_one(m, idx, NULL, d); + dev_rcv_lists = net->can.rx_alldev_list; + can_rcvlist_proc_show_one(m, idx, NULL, dev_rcv_lists); /* receive list for registered CAN devices */ for_each_netdev_rcu(net, dev) { @@ -366,7 +367,7 @@ static inline void can_rcvlist_proc_show_array(struct seq_file *m, static int can_rcvlist_sff_proc_show(struct seq_file *m, void *v) { struct net_device *dev; - struct can_dev_rcv_lists *d; + struct can_dev_rcv_lists *dev_rcv_lists; struct net *net = m->private; /* RX_SFF */ @@ -375,15 +376,16 @@ static int can_rcvlist_sff_proc_show(struct seq_file *m, void *v) rcu_read_lock(); /* sff receive list for 'all' CAN devices (dev == NULL) */ - d = net->can.can_rx_alldev_list; - can_rcvlist_proc_show_array(m, NULL, d->rx_sff, ARRAY_SIZE(d->rx_sff)); + dev_rcv_lists = net->can.rx_alldev_list; + can_rcvlist_proc_show_array(m, NULL, dev_rcv_lists->rx_sff, + ARRAY_SIZE(dev_rcv_lists->rx_sff)); /* sff receive list for registered CAN devices */ for_each_netdev_rcu(net, dev) { if (dev->type == ARPHRD_CAN && dev->ml_priv) { - d = dev->ml_priv; - can_rcvlist_proc_show_array(m, dev, d->rx_sff, - ARRAY_SIZE(d->rx_sff)); + dev_rcv_lists = dev->ml_priv; + can_rcvlist_proc_show_array(m, dev, dev_rcv_lists->rx_sff, + ARRAY_SIZE(dev_rcv_lists->rx_sff)); } } @@ -396,7 +398,7 @@ static int can_rcvlist_sff_proc_show(struct seq_file *m, void *v) static int can_rcvlist_eff_proc_show(struct seq_file *m, void *v) { struct net_device *dev; - struct can_dev_rcv_lists *d; + struct can_dev_rcv_lists *dev_rcv_lists; struct net *net = m->private; /* RX_EFF */ @@ -405,15 +407,16 @@ static int can_rcvlist_eff_proc_show(struct seq_file *m, void *v) rcu_read_lock(); /* eff receive list for 'all' CAN devices (dev == NULL) */ - d = net->can.can_rx_alldev_list; - can_rcvlist_proc_show_array(m, NULL, d->rx_eff, ARRAY_SIZE(d->rx_eff)); + dev_rcv_lists = net->can.rx_alldev_list; + can_rcvlist_proc_show_array(m, NULL, dev_rcv_lists->rx_eff, + ARRAY_SIZE(dev_rcv_lists->rx_eff)); /* eff receive list for registered CAN devices */ for_each_netdev_rcu(net, dev) { if (dev->type == ARPHRD_CAN && dev->ml_priv) { - d = dev->ml_priv; - can_rcvlist_proc_show_array(m, dev, d->rx_eff, - ARRAY_SIZE(d->rx_eff)); + dev_rcv_lists = dev->ml_priv; + can_rcvlist_proc_show_array(m, dev, dev_rcv_lists->rx_eff, + ARRAY_SIZE(dev_rcv_lists->rx_eff)); } } diff --git a/net/can/raw.c b/net/can/raw.c index da386f1fa815..59c039d73c6d 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -1,6 +1,5 @@ // SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) -/* - * raw.c - Raw sockets for protocol family CAN +/* raw.c - Raw sockets for protocol family CAN * * Copyright (c) 2002-2007 Volkswagen Group Electronic Research * All rights reserved. @@ -65,8 +64,7 @@ MODULE_ALIAS("can-proto-1"); #define MASK_ALL 0 -/* - * A raw socket has a list of can_filters attached to it, each receiving +/* A raw socket has a list of can_filters attached to it, each receiving * the CAN frames matching that filter. If the filter list is empty, * no CAN frames will be received by the socket. The default after * opening the socket, is to have one filter which receives all frames. @@ -97,8 +95,7 @@ struct raw_sock { struct uniqframe __percpu *uniq; }; -/* - * Return pointer to store the extra msg flags for raw_recvmsg(). +/* Return pointer to store the extra msg flags for raw_recvmsg(). * We use the space of one unsigned int beyond the 'struct sockaddr_can' * in skb->cb. */ @@ -157,8 +154,7 @@ static void raw_rcv(struct sk_buff *oskb, void *data) if (!skb) return; - /* - * Put the datagram to the queue so that raw_recvmsg() can + /* Put the datagram to the queue so that raw_recvmsg() can * get it from there. We need to pass the interface index to * raw_recvmsg(). We pass a whole struct sockaddr_can in skb->cb * containing the interface index. @@ -284,7 +280,6 @@ static int raw_notifier(struct notifier_block *nb, return NOTIFY_DONE; switch (msg) { - case NETDEV_UNREGISTER: lock_sock(sk); /* remove current filters & unregister */ @@ -370,8 +365,9 @@ static int raw_release(struct socket *sock) raw_disable_allfilters(dev_net(dev), dev, sk); dev_put(dev); } - } else + } else { raw_disable_allfilters(sock_net(sk), NULL, sk); + } } if (ro->count > 1) @@ -400,7 +396,7 @@ static int raw_bind(struct socket *sock, struct sockaddr *uaddr, int len) int err = 0; int notify_enetdown = 0; - if (len < sizeof(*addr)) + if (len < CAN_REQUIRED_SIZE(*addr, can_ifindex)) return -EINVAL; if (addr->can_family != AF_CAN) return -EINVAL; @@ -451,8 +447,9 @@ static int raw_bind(struct socket *sock, struct sockaddr *uaddr, int len) dev, sk); dev_put(dev); } - } else + } else { raw_disable_allfilters(sock_net(sk), NULL, sk); + } } ro->ifindex = ifindex; ro->bound = 1; @@ -503,7 +500,6 @@ static int raw_setsockopt(struct socket *sock, int level, int optname, return -EINVAL; switch (optname) { - case CAN_RAW_FILTER: if (optlen % sizeof(struct can_filter) != 0) return -EINVAL; @@ -666,17 +662,18 @@ static int raw_getsockopt(struct socket *sock, int level, int optname, return -EINVAL; switch (optname) { - case CAN_RAW_FILTER: lock_sock(sk); if (ro->count > 0) { int fsize = ro->count * sizeof(struct can_filter); + if (len > fsize) len = fsize; if (copy_to_user(optval, ro->filter, len)) err = -EFAULT; - } else + } else { len = 0; + } release_sock(sk); if (!err) @@ -736,15 +733,16 @@ static int raw_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if (msg->msg_name) { DECLARE_SOCKADDR(struct sockaddr_can *, addr, msg->msg_name); - if (msg->msg_namelen < sizeof(*addr)) + if (msg->msg_namelen < CAN_REQUIRED_SIZE(*addr, can_ifindex)) return -EINVAL; if (addr->can_family != AF_CAN) return -EINVAL; ifindex = addr->can_ifindex; - } else + } else { ifindex = ro->ifindex; + } dev = dev_get_by_index(sock_net(sk), ifindex); if (!dev) @@ -837,8 +835,8 @@ static int raw_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, return size; } -int raw_sock_no_ioctlcmd(struct socket *sock, unsigned int cmd, - unsigned long arg) +static int raw_sock_no_ioctlcmd(struct socket *sock, unsigned int cmd, + unsigned long arg) { /* no ioctls for socket layer -> hand it down to NIC layer */ return -ENOIOCTLCMD; @@ -887,7 +885,7 @@ static __init int raw_module_init(void) err = can_proto_register(&raw_can_proto); if (err < 0) - printk(KERN_ERR "can: registration of raw protocol failed\n"); + pr_err("can: registration of raw protocol failed\n"); return err; } diff --git a/net/ceph/crypto.c b/net/ceph/crypto.c index 5d6724cee38f..4f75df40fb12 100644 --- a/net/ceph/crypto.c +++ b/net/ceph/crypto.c @@ -136,8 +136,10 @@ void ceph_crypto_key_destroy(struct ceph_crypto_key *key) if (key) { kfree(key->key); key->key = NULL; - crypto_free_sync_skcipher(key->tfm); - key->tfm = NULL; + if (key->tfm) { + crypto_free_sync_skcipher(key->tfm); + key->tfm = NULL; + } } } diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 0b2df09b2554..78ae6e8c953d 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -1496,7 +1496,7 @@ static enum calc_target_result calc_target(struct ceph_osd_client *osdc, struct ceph_osds up, acting; bool force_resend = false; bool unpaused = false; - bool legacy_change; + bool legacy_change = false; bool split = false; bool sort_bitwise = ceph_osdmap_flag(osdc, CEPH_OSDMAP_SORTBITWISE); bool recovery_deletes = ceph_osdmap_flag(osdc, @@ -1584,15 +1584,14 @@ static enum calc_target_result calc_target(struct ceph_osd_client *osdc, t->osd = acting.primary; } - if (unpaused || legacy_change || force_resend || - (split && con && CEPH_HAVE_FEATURE(con->peer_features, - RESEND_ON_SPLIT))) + if (unpaused || legacy_change || force_resend || split) ct_res = CALC_TARGET_NEED_RESEND; else ct_res = CALC_TARGET_NO_ACTION; out: - dout("%s t %p -> ct_res %d osd %d\n", __func__, t, ct_res, t->osd); + dout("%s t %p -> %d%d%d%d ct_res %d osd%d\n", __func__, t, unpaused, + legacy_change, force_resend, split, ct_res, t->osd); return ct_res; } diff --git a/net/core/devlink.c b/net/core/devlink.c index d3dbb904bf3b..6e52d639dac6 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -18,6 +18,8 @@ #include <linux/spinlock.h> #include <linux/refcount.h> #include <linux/workqueue.h> +#include <linux/u64_stats_sync.h> +#include <linux/timekeeping.h> #include <rdma/ib_verbs.h> #include <net/netlink.h> #include <net/genetlink.h> @@ -25,6 +27,7 @@ #include <net/net_namespace.h> #include <net/sock.h> #include <net/devlink.h> +#include <net/drop_monitor.h> #define CREATE_TRACE_POINTS #include <trace/events/devlink.h> @@ -133,7 +136,7 @@ static struct devlink *devlink_get_from_info(struct genl_info *info) } static struct devlink_port *devlink_port_get_by_index(struct devlink *devlink, - int port_index) + unsigned int port_index) { struct devlink_port *devlink_port; @@ -144,7 +147,8 @@ static struct devlink_port *devlink_port_get_by_index(struct devlink *devlink, return NULL; } -static bool devlink_port_index_exists(struct devlink *devlink, int port_index) +static bool devlink_port_index_exists(struct devlink *devlink, + unsigned int port_index) { return devlink_port_get_by_index(devlink, port_index); } @@ -506,32 +510,37 @@ static int devlink_nl_port_attrs_put(struct sk_buff *msg, return 0; if (nla_put_u16(msg, DEVLINK_ATTR_PORT_FLAVOUR, attrs->flavour)) return -EMSGSIZE; - if (devlink_port->attrs.flavour == DEVLINK_PORT_FLAVOUR_PCI_PF) { + switch (devlink_port->attrs.flavour) { + case DEVLINK_PORT_FLAVOUR_PCI_PF: if (nla_put_u16(msg, DEVLINK_ATTR_PORT_PCI_PF_NUMBER, attrs->pci_pf.pf)) return -EMSGSIZE; - } else if (devlink_port->attrs.flavour == DEVLINK_PORT_FLAVOUR_PCI_VF) { + break; + case DEVLINK_PORT_FLAVOUR_PCI_VF: if (nla_put_u16(msg, DEVLINK_ATTR_PORT_PCI_PF_NUMBER, attrs->pci_vf.pf) || nla_put_u16(msg, DEVLINK_ATTR_PORT_PCI_VF_NUMBER, attrs->pci_vf.vf)) return -EMSGSIZE; + break; + case DEVLINK_PORT_FLAVOUR_PHYSICAL: + case DEVLINK_PORT_FLAVOUR_CPU: + case DEVLINK_PORT_FLAVOUR_DSA: + if (nla_put_u32(msg, DEVLINK_ATTR_PORT_NUMBER, + attrs->phys.port_number)) + return -EMSGSIZE; + if (!attrs->split) + return 0; + if (nla_put_u32(msg, DEVLINK_ATTR_PORT_SPLIT_GROUP, + attrs->phys.port_number)) + return -EMSGSIZE; + if (nla_put_u32(msg, DEVLINK_ATTR_PORT_SPLIT_SUBPORT_NUMBER, + attrs->phys.split_subport_number)) + return -EMSGSIZE; + break; + default: + break; } - if (devlink_port->attrs.flavour != DEVLINK_PORT_FLAVOUR_PHYSICAL && - devlink_port->attrs.flavour != DEVLINK_PORT_FLAVOUR_CPU && - devlink_port->attrs.flavour != DEVLINK_PORT_FLAVOUR_DSA) - return 0; - if (nla_put_u32(msg, DEVLINK_ATTR_PORT_NUMBER, - attrs->phys.port_number)) - return -EMSGSIZE; - if (!attrs->split) - return 0; - if (nla_put_u32(msg, DEVLINK_ATTR_PORT_SPLIT_GROUP, - attrs->phys.port_number)) - return -EMSGSIZE; - if (nla_put_u32(msg, DEVLINK_ATTR_PORT_SPLIT_SUBPORT_NUMBER, - attrs->phys.split_subport_number)) - return -EMSGSIZE; return 0; } @@ -551,7 +560,7 @@ static int devlink_nl_port_fill(struct sk_buff *msg, struct devlink *devlink, if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, devlink_port->index)) goto nla_put_failure; - spin_lock(&devlink_port->type_lock); + spin_lock_bh(&devlink_port->type_lock); if (nla_put_u16(msg, DEVLINK_ATTR_PORT_TYPE, devlink_port->type)) goto nla_put_failure_type_locked; if (devlink_port->desired_type != DEVLINK_PORT_TYPE_NOTSET && @@ -576,7 +585,7 @@ static int devlink_nl_port_fill(struct sk_buff *msg, struct devlink *devlink, ibdev->name)) goto nla_put_failure_type_locked; } - spin_unlock(&devlink_port->type_lock); + spin_unlock_bh(&devlink_port->type_lock); if (devlink_nl_port_attrs_put(msg, devlink_port)) goto nla_put_failure; @@ -584,7 +593,7 @@ static int devlink_nl_port_fill(struct sk_buff *msg, struct devlink *devlink, return 0; nla_put_failure_type_locked: - spin_unlock(&devlink_port->type_lock); + spin_unlock_bh(&devlink_port->type_lock); nla_put_failure: genlmsg_cancel(msg, hdr); return -EMSGSIZE; @@ -5154,6 +5163,571 @@ devlink_nl_cmd_health_reporter_dump_clear_doit(struct sk_buff *skb, return 0; } +struct devlink_stats { + u64 rx_bytes; + u64 rx_packets; + struct u64_stats_sync syncp; +}; + +/** + * struct devlink_trap_group_item - Packet trap group attributes. + * @group: Immutable packet trap group attributes. + * @refcount: Number of trap items using the group. + * @list: trap_group_list member. + * @stats: Trap group statistics. + * + * Describes packet trap group attributes. Created by devlink during trap + * registration. + */ +struct devlink_trap_group_item { + const struct devlink_trap_group *group; + refcount_t refcount; + struct list_head list; + struct devlink_stats __percpu *stats; +}; + +/** + * struct devlink_trap_item - Packet trap attributes. + * @trap: Immutable packet trap attributes. + * @group_item: Associated group item. + * @list: trap_list member. + * @action: Trap action. + * @stats: Trap statistics. + * @priv: Driver private information. + * + * Describes both mutable and immutable packet trap attributes. Created by + * devlink during trap registration and used for all trap related operations. + */ +struct devlink_trap_item { + const struct devlink_trap *trap; + struct devlink_trap_group_item *group_item; + struct list_head list; + enum devlink_trap_action action; + struct devlink_stats __percpu *stats; + void *priv; +}; + +static struct devlink_trap_item * +devlink_trap_item_lookup(struct devlink *devlink, const char *name) +{ + struct devlink_trap_item *trap_item; + + list_for_each_entry(trap_item, &devlink->trap_list, list) { + if (!strcmp(trap_item->trap->name, name)) + return trap_item; + } + + return NULL; +} + +static struct devlink_trap_item * +devlink_trap_item_get_from_info(struct devlink *devlink, + struct genl_info *info) +{ + struct nlattr *attr; + + if (!info->attrs[DEVLINK_ATTR_TRAP_NAME]) + return NULL; + attr = info->attrs[DEVLINK_ATTR_TRAP_NAME]; + + return devlink_trap_item_lookup(devlink, nla_data(attr)); +} + +static int +devlink_trap_action_get_from_info(struct genl_info *info, + enum devlink_trap_action *p_trap_action) +{ + u8 val; + + val = nla_get_u8(info->attrs[DEVLINK_ATTR_TRAP_ACTION]); + switch (val) { + case DEVLINK_TRAP_ACTION_DROP: /* fall-through */ + case DEVLINK_TRAP_ACTION_TRAP: + *p_trap_action = val; + break; + default: + return -EINVAL; + } + + return 0; +} + +static int devlink_trap_metadata_put(struct sk_buff *msg, + const struct devlink_trap *trap) +{ + struct nlattr *attr; + + attr = nla_nest_start(msg, DEVLINK_ATTR_TRAP_METADATA); + if (!attr) + return -EMSGSIZE; + + if ((trap->metadata_cap & DEVLINK_TRAP_METADATA_TYPE_F_IN_PORT) && + nla_put_flag(msg, DEVLINK_ATTR_TRAP_METADATA_TYPE_IN_PORT)) + goto nla_put_failure; + + nla_nest_end(msg, attr); + + return 0; + +nla_put_failure: + nla_nest_cancel(msg, attr); + return -EMSGSIZE; +} + +static void devlink_trap_stats_read(struct devlink_stats __percpu *trap_stats, + struct devlink_stats *stats) +{ + int i; + + memset(stats, 0, sizeof(*stats)); + for_each_possible_cpu(i) { + struct devlink_stats *cpu_stats; + u64 rx_packets, rx_bytes; + unsigned int start; + + cpu_stats = per_cpu_ptr(trap_stats, i); + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + rx_packets = cpu_stats->rx_packets; + rx_bytes = cpu_stats->rx_bytes; + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + stats->rx_packets += rx_packets; + stats->rx_bytes += rx_bytes; + } +} + +static int devlink_trap_stats_put(struct sk_buff *msg, + struct devlink_stats __percpu *trap_stats) +{ + struct devlink_stats stats; + struct nlattr *attr; + + devlink_trap_stats_read(trap_stats, &stats); + + attr = nla_nest_start(msg, DEVLINK_ATTR_STATS); + if (!attr) + return -EMSGSIZE; + + if (nla_put_u64_64bit(msg, DEVLINK_ATTR_STATS_RX_PACKETS, + stats.rx_packets, DEVLINK_ATTR_PAD)) + goto nla_put_failure; + + if (nla_put_u64_64bit(msg, DEVLINK_ATTR_STATS_RX_BYTES, + stats.rx_bytes, DEVLINK_ATTR_PAD)) + goto nla_put_failure; + + nla_nest_end(msg, attr); + + return 0; + +nla_put_failure: + nla_nest_cancel(msg, attr); + return -EMSGSIZE; +} + +static int devlink_nl_trap_fill(struct sk_buff *msg, struct devlink *devlink, + const struct devlink_trap_item *trap_item, + enum devlink_command cmd, u32 portid, u32 seq, + int flags) +{ + struct devlink_trap_group_item *group_item = trap_item->group_item; + void *hdr; + int err; + + hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd); + if (!hdr) + return -EMSGSIZE; + + if (devlink_nl_put_handle(msg, devlink)) + goto nla_put_failure; + + if (nla_put_string(msg, DEVLINK_ATTR_TRAP_GROUP_NAME, + group_item->group->name)) + goto nla_put_failure; + + if (nla_put_string(msg, DEVLINK_ATTR_TRAP_NAME, trap_item->trap->name)) + goto nla_put_failure; + + if (nla_put_u8(msg, DEVLINK_ATTR_TRAP_TYPE, trap_item->trap->type)) + goto nla_put_failure; + + if (trap_item->trap->generic && + nla_put_flag(msg, DEVLINK_ATTR_TRAP_GENERIC)) + goto nla_put_failure; + + if (nla_put_u8(msg, DEVLINK_ATTR_TRAP_ACTION, trap_item->action)) + goto nla_put_failure; + + err = devlink_trap_metadata_put(msg, trap_item->trap); + if (err) + goto nla_put_failure; + + err = devlink_trap_stats_put(msg, trap_item->stats); + if (err) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static int devlink_nl_cmd_trap_get_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct netlink_ext_ack *extack = info->extack; + struct devlink *devlink = info->user_ptr[0]; + struct devlink_trap_item *trap_item; + struct sk_buff *msg; + int err; + + if (list_empty(&devlink->trap_list)) + return -EOPNOTSUPP; + + trap_item = devlink_trap_item_get_from_info(devlink, info); + if (!trap_item) { + NL_SET_ERR_MSG_MOD(extack, "Device did not register this trap"); + return -ENOENT; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + err = devlink_nl_trap_fill(msg, devlink, trap_item, + DEVLINK_CMD_TRAP_NEW, info->snd_portid, + info->snd_seq, 0); + if (err) + goto err_trap_fill; + + return genlmsg_reply(msg, info); + +err_trap_fill: + nlmsg_free(msg); + return err; +} + +static int devlink_nl_cmd_trap_get_dumpit(struct sk_buff *msg, + struct netlink_callback *cb) +{ + struct devlink_trap_item *trap_item; + struct devlink *devlink; + int start = cb->args[0]; + int idx = 0; + int err; + + mutex_lock(&devlink_mutex); + list_for_each_entry(devlink, &devlink_list, list) { + if (!net_eq(devlink_net(devlink), sock_net(msg->sk))) + continue; + mutex_lock(&devlink->lock); + list_for_each_entry(trap_item, &devlink->trap_list, list) { + if (idx < start) { + idx++; + continue; + } + err = devlink_nl_trap_fill(msg, devlink, trap_item, + DEVLINK_CMD_TRAP_NEW, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, + NLM_F_MULTI); + if (err) { + mutex_unlock(&devlink->lock); + goto out; + } + idx++; + } + mutex_unlock(&devlink->lock); + } +out: + mutex_unlock(&devlink_mutex); + + cb->args[0] = idx; + return msg->len; +} + +static int __devlink_trap_action_set(struct devlink *devlink, + struct devlink_trap_item *trap_item, + enum devlink_trap_action trap_action, + struct netlink_ext_ack *extack) +{ + int err; + + if (trap_item->action != trap_action && + trap_item->trap->type != DEVLINK_TRAP_TYPE_DROP) { + NL_SET_ERR_MSG_MOD(extack, "Cannot change action of non-drop traps. Skipping"); + return 0; + } + + err = devlink->ops->trap_action_set(devlink, trap_item->trap, + trap_action); + if (err) + return err; + + trap_item->action = trap_action; + + return 0; +} + +static int devlink_trap_action_set(struct devlink *devlink, + struct devlink_trap_item *trap_item, + struct genl_info *info) +{ + enum devlink_trap_action trap_action; + int err; + + if (!info->attrs[DEVLINK_ATTR_TRAP_ACTION]) + return 0; + + err = devlink_trap_action_get_from_info(info, &trap_action); + if (err) { + NL_SET_ERR_MSG_MOD(info->extack, "Invalid trap action"); + return -EINVAL; + } + + return __devlink_trap_action_set(devlink, trap_item, trap_action, + info->extack); +} + +static int devlink_nl_cmd_trap_set_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct netlink_ext_ack *extack = info->extack; + struct devlink *devlink = info->user_ptr[0]; + struct devlink_trap_item *trap_item; + int err; + + if (list_empty(&devlink->trap_list)) + return -EOPNOTSUPP; + + trap_item = devlink_trap_item_get_from_info(devlink, info); + if (!trap_item) { + NL_SET_ERR_MSG_MOD(extack, "Device did not register this trap"); + return -ENOENT; + } + + err = devlink_trap_action_set(devlink, trap_item, info); + if (err) + return err; + + return 0; +} + +static struct devlink_trap_group_item * +devlink_trap_group_item_lookup(struct devlink *devlink, const char *name) +{ + struct devlink_trap_group_item *group_item; + + list_for_each_entry(group_item, &devlink->trap_group_list, list) { + if (!strcmp(group_item->group->name, name)) + return group_item; + } + + return NULL; +} + +static struct devlink_trap_group_item * +devlink_trap_group_item_get_from_info(struct devlink *devlink, + struct genl_info *info) +{ + char *name; + + if (!info->attrs[DEVLINK_ATTR_TRAP_GROUP_NAME]) + return NULL; + name = nla_data(info->attrs[DEVLINK_ATTR_TRAP_GROUP_NAME]); + + return devlink_trap_group_item_lookup(devlink, name); +} + +static int +devlink_nl_trap_group_fill(struct sk_buff *msg, struct devlink *devlink, + const struct devlink_trap_group_item *group_item, + enum devlink_command cmd, u32 portid, u32 seq, + int flags) +{ + void *hdr; + int err; + + hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd); + if (!hdr) + return -EMSGSIZE; + + if (devlink_nl_put_handle(msg, devlink)) + goto nla_put_failure; + + if (nla_put_string(msg, DEVLINK_ATTR_TRAP_GROUP_NAME, + group_item->group->name)) + goto nla_put_failure; + + if (group_item->group->generic && + nla_put_flag(msg, DEVLINK_ATTR_TRAP_GENERIC)) + goto nla_put_failure; + + err = devlink_trap_stats_put(msg, group_item->stats); + if (err) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static int devlink_nl_cmd_trap_group_get_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct netlink_ext_ack *extack = info->extack; + struct devlink *devlink = info->user_ptr[0]; + struct devlink_trap_group_item *group_item; + struct sk_buff *msg; + int err; + + if (list_empty(&devlink->trap_group_list)) + return -EOPNOTSUPP; + + group_item = devlink_trap_group_item_get_from_info(devlink, info); + if (!group_item) { + NL_SET_ERR_MSG_MOD(extack, "Device did not register this trap group"); + return -ENOENT; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + err = devlink_nl_trap_group_fill(msg, devlink, group_item, + DEVLINK_CMD_TRAP_GROUP_NEW, + info->snd_portid, info->snd_seq, 0); + if (err) + goto err_trap_group_fill; + + return genlmsg_reply(msg, info); + +err_trap_group_fill: + nlmsg_free(msg); + return err; +} + +static int devlink_nl_cmd_trap_group_get_dumpit(struct sk_buff *msg, + struct netlink_callback *cb) +{ + enum devlink_command cmd = DEVLINK_CMD_TRAP_GROUP_NEW; + struct devlink_trap_group_item *group_item; + u32 portid = NETLINK_CB(cb->skb).portid; + struct devlink *devlink; + int start = cb->args[0]; + int idx = 0; + int err; + + mutex_lock(&devlink_mutex); + list_for_each_entry(devlink, &devlink_list, list) { + if (!net_eq(devlink_net(devlink), sock_net(msg->sk))) + continue; + mutex_lock(&devlink->lock); + list_for_each_entry(group_item, &devlink->trap_group_list, + list) { + if (idx < start) { + idx++; + continue; + } + err = devlink_nl_trap_group_fill(msg, devlink, + group_item, cmd, + portid, + cb->nlh->nlmsg_seq, + NLM_F_MULTI); + if (err) { + mutex_unlock(&devlink->lock); + goto out; + } + idx++; + } + mutex_unlock(&devlink->lock); + } +out: + mutex_unlock(&devlink_mutex); + + cb->args[0] = idx; + return msg->len; +} + +static int +__devlink_trap_group_action_set(struct devlink *devlink, + struct devlink_trap_group_item *group_item, + enum devlink_trap_action trap_action, + struct netlink_ext_ack *extack) +{ + const char *group_name = group_item->group->name; + struct devlink_trap_item *trap_item; + int err; + + list_for_each_entry(trap_item, &devlink->trap_list, list) { + if (strcmp(trap_item->trap->group.name, group_name)) + continue; + err = __devlink_trap_action_set(devlink, trap_item, + trap_action, extack); + if (err) + return err; + } + + return 0; +} + +static int +devlink_trap_group_action_set(struct devlink *devlink, + struct devlink_trap_group_item *group_item, + struct genl_info *info) +{ + enum devlink_trap_action trap_action; + int err; + + if (!info->attrs[DEVLINK_ATTR_TRAP_ACTION]) + return 0; + + err = devlink_trap_action_get_from_info(info, &trap_action); + if (err) { + NL_SET_ERR_MSG_MOD(info->extack, "Invalid trap action"); + return -EINVAL; + } + + err = __devlink_trap_group_action_set(devlink, group_item, trap_action, + info->extack); + if (err) + return err; + + return 0; +} + +static int devlink_nl_cmd_trap_group_set_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct netlink_ext_ack *extack = info->extack; + struct devlink *devlink = info->user_ptr[0]; + struct devlink_trap_group_item *group_item; + int err; + + if (list_empty(&devlink->trap_group_list)) + return -EOPNOTSUPP; + + group_item = devlink_trap_group_item_get_from_info(devlink, info); + if (!group_item) { + NL_SET_ERR_MSG_MOD(extack, "Device did not register this trap group"); + return -ENOENT; + } + + err = devlink_trap_group_action_set(devlink, group_item, info); + if (err) + return err; + + return 0; +} + static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING }, @@ -5184,6 +5758,9 @@ static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_HEALTH_REPORTER_AUTO_RECOVER] = { .type = NLA_U8 }, [DEVLINK_ATTR_FLASH_UPDATE_FILE_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_FLASH_UPDATE_COMPONENT] = { .type = NLA_NUL_STRING }, + [DEVLINK_ATTR_TRAP_NAME] = { .type = NLA_NUL_STRING }, + [DEVLINK_ATTR_TRAP_ACTION] = { .type = NLA_U8 }, + [DEVLINK_ATTR_TRAP_GROUP_NAME] = { .type = NLA_NUL_STRING }, }; static const struct genl_ops devlink_nl_ops[] = { @@ -5483,6 +6060,32 @@ static const struct genl_ops devlink_nl_ops[] = { .flags = GENL_ADMIN_PERM, .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, }, + { + .cmd = DEVLINK_CMD_TRAP_GET, + .doit = devlink_nl_cmd_trap_get_doit, + .dumpit = devlink_nl_cmd_trap_get_dumpit, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + /* can be retrieved by unprivileged users */ + }, + { + .cmd = DEVLINK_CMD_TRAP_SET, + .doit = devlink_nl_cmd_trap_set_doit, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, + { + .cmd = DEVLINK_CMD_TRAP_GROUP_GET, + .doit = devlink_nl_cmd_trap_group_get_doit, + .dumpit = devlink_nl_cmd_trap_group_get_dumpit, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + /* can be retrieved by unprivileged users */ + }, + { + .cmd = DEVLINK_CMD_TRAP_GROUP_SET, + .doit = devlink_nl_cmd_trap_group_set_doit, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, }; static struct genl_family devlink_nl_family __ro_after_init = { @@ -5528,6 +6131,8 @@ struct devlink *devlink_alloc(const struct devlink_ops *ops, size_t priv_size) INIT_LIST_HEAD(&devlink->param_list); INIT_LIST_HEAD(&devlink->region_list); INIT_LIST_HEAD(&devlink->reporter_list); + INIT_LIST_HEAD(&devlink->trap_list); + INIT_LIST_HEAD(&devlink->trap_group_list); mutex_init(&devlink->lock); mutex_init(&devlink->reporters_lock); return devlink; @@ -5574,6 +6179,8 @@ void devlink_free(struct devlink *devlink) { mutex_destroy(&devlink->reporters_lock); mutex_destroy(&devlink->lock); + WARN_ON(!list_empty(&devlink->trap_group_list)); + WARN_ON(!list_empty(&devlink->trap_list)); WARN_ON(!list_empty(&devlink->reporter_list)); WARN_ON(!list_empty(&devlink->region_list)); WARN_ON(!list_empty(&devlink->param_list)); @@ -5678,10 +6285,10 @@ static void __devlink_port_type_set(struct devlink_port *devlink_port, if (WARN_ON(!devlink_port->registered)) return; devlink_port_type_warn_cancel(devlink_port); - spin_lock(&devlink_port->type_lock); + spin_lock_bh(&devlink_port->type_lock); devlink_port->type = type; devlink_port->type_dev = type_dev; - spin_unlock(&devlink_port->type_lock); + spin_unlock_bh(&devlink_port->type_lock); devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_NEW); } @@ -6834,6 +7441,475 @@ unlock: } EXPORT_SYMBOL_GPL(devlink_region_snapshot_create); +#define DEVLINK_TRAP(_id, _type) \ + { \ + .type = DEVLINK_TRAP_TYPE_##_type, \ + .id = DEVLINK_TRAP_GENERIC_ID_##_id, \ + .name = DEVLINK_TRAP_GENERIC_NAME_##_id, \ + } + +static const struct devlink_trap devlink_trap_generic[] = { + DEVLINK_TRAP(SMAC_MC, DROP), + DEVLINK_TRAP(VLAN_TAG_MISMATCH, DROP), + DEVLINK_TRAP(INGRESS_VLAN_FILTER, DROP), + DEVLINK_TRAP(INGRESS_STP_FILTER, DROP), + DEVLINK_TRAP(EMPTY_TX_LIST, DROP), + DEVLINK_TRAP(PORT_LOOPBACK_FILTER, DROP), + DEVLINK_TRAP(BLACKHOLE_ROUTE, DROP), + DEVLINK_TRAP(TTL_ERROR, EXCEPTION), + DEVLINK_TRAP(TAIL_DROP, DROP), +}; + +#define DEVLINK_TRAP_GROUP(_id) \ + { \ + .id = DEVLINK_TRAP_GROUP_GENERIC_ID_##_id, \ + .name = DEVLINK_TRAP_GROUP_GENERIC_NAME_##_id, \ + } + +static const struct devlink_trap_group devlink_trap_group_generic[] = { + DEVLINK_TRAP_GROUP(L2_DROPS), + DEVLINK_TRAP_GROUP(L3_DROPS), + DEVLINK_TRAP_GROUP(BUFFER_DROPS), +}; + +static int devlink_trap_generic_verify(const struct devlink_trap *trap) +{ + if (trap->id > DEVLINK_TRAP_GENERIC_ID_MAX) + return -EINVAL; + + if (strcmp(trap->name, devlink_trap_generic[trap->id].name)) + return -EINVAL; + + if (trap->type != devlink_trap_generic[trap->id].type) + return -EINVAL; + + return 0; +} + +static int devlink_trap_driver_verify(const struct devlink_trap *trap) +{ + int i; + + if (trap->id <= DEVLINK_TRAP_GENERIC_ID_MAX) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(devlink_trap_generic); i++) { + if (!strcmp(trap->name, devlink_trap_generic[i].name)) + return -EEXIST; + } + + return 0; +} + +static int devlink_trap_verify(const struct devlink_trap *trap) +{ + if (!trap || !trap->name || !trap->group.name) + return -EINVAL; + + if (trap->generic) + return devlink_trap_generic_verify(trap); + else + return devlink_trap_driver_verify(trap); +} + +static int +devlink_trap_group_generic_verify(const struct devlink_trap_group *group) +{ + if (group->id > DEVLINK_TRAP_GROUP_GENERIC_ID_MAX) + return -EINVAL; + + if (strcmp(group->name, devlink_trap_group_generic[group->id].name)) + return -EINVAL; + + return 0; +} + +static int +devlink_trap_group_driver_verify(const struct devlink_trap_group *group) +{ + int i; + + if (group->id <= DEVLINK_TRAP_GROUP_GENERIC_ID_MAX) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(devlink_trap_group_generic); i++) { + if (!strcmp(group->name, devlink_trap_group_generic[i].name)) + return -EEXIST; + } + + return 0; +} + +static int devlink_trap_group_verify(const struct devlink_trap_group *group) +{ + if (group->generic) + return devlink_trap_group_generic_verify(group); + else + return devlink_trap_group_driver_verify(group); +} + +static void +devlink_trap_group_notify(struct devlink *devlink, + const struct devlink_trap_group_item *group_item, + enum devlink_command cmd) +{ + struct sk_buff *msg; + int err; + + WARN_ON_ONCE(cmd != DEVLINK_CMD_TRAP_GROUP_NEW && + cmd != DEVLINK_CMD_TRAP_GROUP_DEL); + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return; + + err = devlink_nl_trap_group_fill(msg, devlink, group_item, cmd, 0, 0, + 0); + if (err) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast_netns(&devlink_nl_family, devlink_net(devlink), + msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL); +} + +static struct devlink_trap_group_item * +devlink_trap_group_item_create(struct devlink *devlink, + const struct devlink_trap_group *group) +{ + struct devlink_trap_group_item *group_item; + int err; + + err = devlink_trap_group_verify(group); + if (err) + return ERR_PTR(err); + + group_item = kzalloc(sizeof(*group_item), GFP_KERNEL); + if (!group_item) + return ERR_PTR(-ENOMEM); + + group_item->stats = netdev_alloc_pcpu_stats(struct devlink_stats); + if (!group_item->stats) { + err = -ENOMEM; + goto err_stats_alloc; + } + + group_item->group = group; + refcount_set(&group_item->refcount, 1); + + if (devlink->ops->trap_group_init) { + err = devlink->ops->trap_group_init(devlink, group); + if (err) + goto err_group_init; + } + + list_add_tail(&group_item->list, &devlink->trap_group_list); + devlink_trap_group_notify(devlink, group_item, + DEVLINK_CMD_TRAP_GROUP_NEW); + + return group_item; + +err_group_init: + free_percpu(group_item->stats); +err_stats_alloc: + kfree(group_item); + return ERR_PTR(err); +} + +static void +devlink_trap_group_item_destroy(struct devlink *devlink, + struct devlink_trap_group_item *group_item) +{ + devlink_trap_group_notify(devlink, group_item, + DEVLINK_CMD_TRAP_GROUP_DEL); + list_del(&group_item->list); + free_percpu(group_item->stats); + kfree(group_item); +} + +static struct devlink_trap_group_item * +devlink_trap_group_item_get(struct devlink *devlink, + const struct devlink_trap_group *group) +{ + struct devlink_trap_group_item *group_item; + + group_item = devlink_trap_group_item_lookup(devlink, group->name); + if (group_item) { + refcount_inc(&group_item->refcount); + return group_item; + } + + return devlink_trap_group_item_create(devlink, group); +} + +static void +devlink_trap_group_item_put(struct devlink *devlink, + struct devlink_trap_group_item *group_item) +{ + if (!refcount_dec_and_test(&group_item->refcount)) + return; + + devlink_trap_group_item_destroy(devlink, group_item); +} + +static int +devlink_trap_item_group_link(struct devlink *devlink, + struct devlink_trap_item *trap_item) +{ + struct devlink_trap_group_item *group_item; + + group_item = devlink_trap_group_item_get(devlink, + &trap_item->trap->group); + if (IS_ERR(group_item)) + return PTR_ERR(group_item); + + trap_item->group_item = group_item; + + return 0; +} + +static void +devlink_trap_item_group_unlink(struct devlink *devlink, + struct devlink_trap_item *trap_item) +{ + devlink_trap_group_item_put(devlink, trap_item->group_item); +} + +static void devlink_trap_notify(struct devlink *devlink, + const struct devlink_trap_item *trap_item, + enum devlink_command cmd) +{ + struct sk_buff *msg; + int err; + + WARN_ON_ONCE(cmd != DEVLINK_CMD_TRAP_NEW && + cmd != DEVLINK_CMD_TRAP_DEL); + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return; + + err = devlink_nl_trap_fill(msg, devlink, trap_item, cmd, 0, 0, 0); + if (err) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast_netns(&devlink_nl_family, devlink_net(devlink), + msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL); +} + +static int +devlink_trap_register(struct devlink *devlink, + const struct devlink_trap *trap, void *priv) +{ + struct devlink_trap_item *trap_item; + int err; + + if (devlink_trap_item_lookup(devlink, trap->name)) + return -EEXIST; + + trap_item = kzalloc(sizeof(*trap_item), GFP_KERNEL); + if (!trap_item) + return -ENOMEM; + + trap_item->stats = netdev_alloc_pcpu_stats(struct devlink_stats); + if (!trap_item->stats) { + err = -ENOMEM; + goto err_stats_alloc; + } + + trap_item->trap = trap; + trap_item->action = trap->init_action; + trap_item->priv = priv; + + err = devlink_trap_item_group_link(devlink, trap_item); + if (err) + goto err_group_link; + + err = devlink->ops->trap_init(devlink, trap, trap_item); + if (err) + goto err_trap_init; + + list_add_tail(&trap_item->list, &devlink->trap_list); + devlink_trap_notify(devlink, trap_item, DEVLINK_CMD_TRAP_NEW); + + return 0; + +err_trap_init: + devlink_trap_item_group_unlink(devlink, trap_item); +err_group_link: + free_percpu(trap_item->stats); +err_stats_alloc: + kfree(trap_item); + return err; +} + +static void devlink_trap_unregister(struct devlink *devlink, + const struct devlink_trap *trap) +{ + struct devlink_trap_item *trap_item; + + trap_item = devlink_trap_item_lookup(devlink, trap->name); + if (WARN_ON_ONCE(!trap_item)) + return; + + devlink_trap_notify(devlink, trap_item, DEVLINK_CMD_TRAP_DEL); + list_del(&trap_item->list); + if (devlink->ops->trap_fini) + devlink->ops->trap_fini(devlink, trap, trap_item); + devlink_trap_item_group_unlink(devlink, trap_item); + free_percpu(trap_item->stats); + kfree(trap_item); +} + +static void devlink_trap_disable(struct devlink *devlink, + const struct devlink_trap *trap) +{ + struct devlink_trap_item *trap_item; + + trap_item = devlink_trap_item_lookup(devlink, trap->name); + if (WARN_ON_ONCE(!trap_item)) + return; + + devlink->ops->trap_action_set(devlink, trap, DEVLINK_TRAP_ACTION_DROP); + trap_item->action = DEVLINK_TRAP_ACTION_DROP; +} + +/** + * devlink_traps_register - Register packet traps with devlink. + * @devlink: devlink. + * @traps: Packet traps. + * @traps_count: Count of provided packet traps. + * @priv: Driver private information. + * + * Return: Non-zero value on failure. + */ +int devlink_traps_register(struct devlink *devlink, + const struct devlink_trap *traps, + size_t traps_count, void *priv) +{ + int i, err; + + if (!devlink->ops->trap_init || !devlink->ops->trap_action_set) + return -EINVAL; + + mutex_lock(&devlink->lock); + for (i = 0; i < traps_count; i++) { + const struct devlink_trap *trap = &traps[i]; + + err = devlink_trap_verify(trap); + if (err) + goto err_trap_verify; + + err = devlink_trap_register(devlink, trap, priv); + if (err) + goto err_trap_register; + } + mutex_unlock(&devlink->lock); + + return 0; + +err_trap_register: +err_trap_verify: + for (i--; i >= 0; i--) + devlink_trap_unregister(devlink, &traps[i]); + mutex_unlock(&devlink->lock); + return err; +} +EXPORT_SYMBOL_GPL(devlink_traps_register); + +/** + * devlink_traps_unregister - Unregister packet traps from devlink. + * @devlink: devlink. + * @traps: Packet traps. + * @traps_count: Count of provided packet traps. + */ +void devlink_traps_unregister(struct devlink *devlink, + const struct devlink_trap *traps, + size_t traps_count) +{ + int i; + + mutex_lock(&devlink->lock); + /* Make sure we do not have any packets in-flight while unregistering + * traps by disabling all of them and waiting for a grace period. + */ + for (i = traps_count - 1; i >= 0; i--) + devlink_trap_disable(devlink, &traps[i]); + synchronize_rcu(); + for (i = traps_count - 1; i >= 0; i--) + devlink_trap_unregister(devlink, &traps[i]); + mutex_unlock(&devlink->lock); +} +EXPORT_SYMBOL_GPL(devlink_traps_unregister); + +static void +devlink_trap_stats_update(struct devlink_stats __percpu *trap_stats, + size_t skb_len) +{ + struct devlink_stats *stats; + + stats = this_cpu_ptr(trap_stats); + u64_stats_update_begin(&stats->syncp); + stats->rx_bytes += skb_len; + stats->rx_packets++; + u64_stats_update_end(&stats->syncp); +} + +static void +devlink_trap_report_metadata_fill(struct net_dm_hw_metadata *hw_metadata, + const struct devlink_trap_item *trap_item, + struct devlink_port *in_devlink_port) +{ + struct devlink_trap_group_item *group_item = trap_item->group_item; + + hw_metadata->trap_group_name = group_item->group->name; + hw_metadata->trap_name = trap_item->trap->name; + + spin_lock(&in_devlink_port->type_lock); + if (in_devlink_port->type == DEVLINK_PORT_TYPE_ETH) + hw_metadata->input_dev = in_devlink_port->type_dev; + spin_unlock(&in_devlink_port->type_lock); +} + +/** + * devlink_trap_report - Report trapped packet to drop monitor. + * @devlink: devlink. + * @skb: Trapped packet. + * @trap_ctx: Trap context. + * @in_devlink_port: Input devlink port. + */ +void devlink_trap_report(struct devlink *devlink, struct sk_buff *skb, + void *trap_ctx, struct devlink_port *in_devlink_port) +{ + struct devlink_trap_item *trap_item = trap_ctx; + struct net_dm_hw_metadata hw_metadata = {}; + + devlink_trap_stats_update(trap_item->stats, skb->len); + devlink_trap_stats_update(trap_item->group_item->stats, skb->len); + + devlink_trap_report_metadata_fill(&hw_metadata, trap_item, + in_devlink_port); + net_dm_hw_report(skb, &hw_metadata); +} +EXPORT_SYMBOL_GPL(devlink_trap_report); + +/** + * devlink_trap_ctx_priv - Trap context to driver private information. + * @trap_ctx: Trap context. + * + * Return: Driver private information passed during registration. + */ +void *devlink_trap_ctx_priv(void *trap_ctx) +{ + struct devlink_trap_item *trap_item = trap_ctx; + + return trap_item->priv; +} +EXPORT_SYMBOL_GPL(devlink_trap_ctx_priv); + static void __devlink_compat_running_version(struct devlink *devlink, char *buf, size_t len) { diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index 39e094907391..cc60cc22e2db 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -26,6 +26,7 @@ #include <linux/bitops.h> #include <linux/slab.h> #include <linux/module.h> +#include <net/drop_monitor.h> #include <net/genetlink.h> #include <net/netevent.h> @@ -43,6 +44,7 @@ * netlink alerts */ static int trace_state = TRACE_OFF; +static bool monitor_hw; /* net_dm_mutex * @@ -56,9 +58,26 @@ struct net_dm_stats { struct u64_stats_sync syncp; }; +#define NET_DM_MAX_HW_TRAP_NAME_LEN 40 + +struct net_dm_hw_entry { + char trap_name[NET_DM_MAX_HW_TRAP_NAME_LEN]; + u32 count; +}; + +struct net_dm_hw_entries { + u32 num_entries; + struct net_dm_hw_entry entries[0]; +}; + struct per_cpu_dm_data { - spinlock_t lock; /* Protects 'skb' and 'send_timer' */ - struct sk_buff *skb; + spinlock_t lock; /* Protects 'skb', 'hw_entries' and + * 'send_timer' + */ + union { + struct sk_buff *skb; + struct net_dm_hw_entries *hw_entries; + }; struct sk_buff_head drop_queue; struct work_struct dm_alert_work; struct timer_list send_timer; @@ -76,6 +95,7 @@ struct dm_hw_stat_delta { static struct genl_family net_drop_monitor_family; static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_cpu_data); +static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_hw_cpu_data); static int dm_hit_limit = 64; static int dm_delay = 1; @@ -92,10 +112,16 @@ struct net_dm_alert_ops { void (*napi_poll_probe)(void *ignore, struct napi_struct *napi, int work, int budget); void (*work_item_func)(struct work_struct *work); + void (*hw_work_item_func)(struct work_struct *work); + void (*hw_probe)(struct sk_buff *skb, + const struct net_dm_hw_metadata *hw_metadata); }; struct net_dm_skb_cb { - void *pc; + union { + struct net_dm_hw_metadata *hw_metadata; + void *pc; + }; }; #define NET_DM_SKB_CB(__skb) ((struct net_dm_skb_cb *)&((__skb)->cb[0])) @@ -266,10 +292,190 @@ static void trace_napi_poll_hit(void *ignore, struct napi_struct *napi, rcu_read_unlock(); } +static struct net_dm_hw_entries * +net_dm_hw_reset_per_cpu_data(struct per_cpu_dm_data *hw_data) +{ + struct net_dm_hw_entries *hw_entries; + unsigned long flags; + + hw_entries = kzalloc(struct_size(hw_entries, entries, dm_hit_limit), + GFP_KERNEL); + if (!hw_entries) { + /* If the memory allocation failed, we try to perform another + * allocation in 1/10 second. Otherwise, the probe function + * will constantly bail out. + */ + mod_timer(&hw_data->send_timer, jiffies + HZ / 10); + } + + spin_lock_irqsave(&hw_data->lock, flags); + swap(hw_data->hw_entries, hw_entries); + spin_unlock_irqrestore(&hw_data->lock, flags); + + return hw_entries; +} + +static int net_dm_hw_entry_put(struct sk_buff *msg, + const struct net_dm_hw_entry *hw_entry) +{ + struct nlattr *attr; + + attr = nla_nest_start(msg, NET_DM_ATTR_HW_ENTRY); + if (!attr) + return -EMSGSIZE; + + if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_NAME, hw_entry->trap_name)) + goto nla_put_failure; + + if (nla_put_u32(msg, NET_DM_ATTR_HW_TRAP_COUNT, hw_entry->count)) + goto nla_put_failure; + + nla_nest_end(msg, attr); + + return 0; + +nla_put_failure: + nla_nest_cancel(msg, attr); + return -EMSGSIZE; +} + +static int net_dm_hw_entries_put(struct sk_buff *msg, + const struct net_dm_hw_entries *hw_entries) +{ + struct nlattr *attr; + int i; + + attr = nla_nest_start(msg, NET_DM_ATTR_HW_ENTRIES); + if (!attr) + return -EMSGSIZE; + + for (i = 0; i < hw_entries->num_entries; i++) { + int rc; + + rc = net_dm_hw_entry_put(msg, &hw_entries->entries[i]); + if (rc) + goto nla_put_failure; + } + + nla_nest_end(msg, attr); + + return 0; + +nla_put_failure: + nla_nest_cancel(msg, attr); + return -EMSGSIZE; +} + +static int +net_dm_hw_summary_report_fill(struct sk_buff *msg, + const struct net_dm_hw_entries *hw_entries) +{ + struct net_dm_alert_msg anc_hdr = { 0 }; + void *hdr; + int rc; + + hdr = genlmsg_put(msg, 0, 0, &net_drop_monitor_family, 0, + NET_DM_CMD_ALERT); + if (!hdr) + return -EMSGSIZE; + + /* We need to put the ancillary header in order not to break user + * space. + */ + if (nla_put(msg, NLA_UNSPEC, sizeof(anc_hdr), &anc_hdr)) + goto nla_put_failure; + + rc = net_dm_hw_entries_put(msg, hw_entries); + if (rc) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static void net_dm_hw_summary_work(struct work_struct *work) +{ + struct net_dm_hw_entries *hw_entries; + struct per_cpu_dm_data *hw_data; + struct sk_buff *msg; + int rc; + + hw_data = container_of(work, struct per_cpu_dm_data, dm_alert_work); + + hw_entries = net_dm_hw_reset_per_cpu_data(hw_data); + if (!hw_entries) + return; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + goto out; + + rc = net_dm_hw_summary_report_fill(msg, hw_entries); + if (rc) { + nlmsg_free(msg); + goto out; + } + + genlmsg_multicast(&net_drop_monitor_family, msg, 0, 0, GFP_KERNEL); + +out: + kfree(hw_entries); +} + +static void +net_dm_hw_summary_probe(struct sk_buff *skb, + const struct net_dm_hw_metadata *hw_metadata) +{ + struct net_dm_hw_entries *hw_entries; + struct net_dm_hw_entry *hw_entry; + struct per_cpu_dm_data *hw_data; + unsigned long flags; + int i; + + hw_data = this_cpu_ptr(&dm_hw_cpu_data); + spin_lock_irqsave(&hw_data->lock, flags); + hw_entries = hw_data->hw_entries; + + if (!hw_entries) + goto out; + + for (i = 0; i < hw_entries->num_entries; i++) { + hw_entry = &hw_entries->entries[i]; + if (!strncmp(hw_entry->trap_name, hw_metadata->trap_name, + NET_DM_MAX_HW_TRAP_NAME_LEN - 1)) { + hw_entry->count++; + goto out; + } + } + if (WARN_ON_ONCE(hw_entries->num_entries == dm_hit_limit)) + goto out; + + hw_entry = &hw_entries->entries[hw_entries->num_entries]; + strlcpy(hw_entry->trap_name, hw_metadata->trap_name, + NET_DM_MAX_HW_TRAP_NAME_LEN - 1); + hw_entry->count = 1; + hw_entries->num_entries++; + + if (!timer_pending(&hw_data->send_timer)) { + hw_data->send_timer.expires = jiffies + dm_delay * HZ; + add_timer(&hw_data->send_timer); + } + +out: + spin_unlock_irqrestore(&hw_data->lock, flags); +} + static const struct net_dm_alert_ops net_dm_alert_summary_ops = { .kfree_skb_probe = trace_kfree_skb_hit, .napi_poll_probe = trace_napi_poll_hit, .work_item_func = send_dm_alert, + .hw_work_item_func = net_dm_hw_summary_work, + .hw_probe = net_dm_hw_summary_probe, }; static void net_dm_packet_trace_kfree_skb_hit(void *ignore, @@ -323,7 +529,9 @@ static size_t net_dm_in_port_size(void) /* NET_DM_ATTR_IN_PORT nest */ return nla_total_size(0) + /* NET_DM_ATTR_PORT_NETDEV_IFINDEX */ - nla_total_size(sizeof(u32)); + nla_total_size(sizeof(u32)) + + /* NET_DM_ATTR_PORT_NETDEV_NAME */ + nla_total_size(IFNAMSIZ + 1); } #define NET_DM_MAX_SYMBOL_LEN 40 @@ -335,6 +543,8 @@ static size_t net_dm_packet_report_size(size_t payload_len) size = nlmsg_msg_size(GENL_HDRLEN + net_drop_monitor_family.hdrsize); return NLMSG_ALIGN(size) + + /* NET_DM_ATTR_ORIGIN */ + nla_total_size(sizeof(u16)) + /* NET_DM_ATTR_PC */ nla_total_size(sizeof(u64)) + /* NET_DM_ATTR_SYMBOL */ @@ -342,7 +552,7 @@ static size_t net_dm_packet_report_size(size_t payload_len) /* NET_DM_ATTR_IN_PORT */ net_dm_in_port_size() + /* NET_DM_ATTR_TIMESTAMP */ - nla_total_size(sizeof(struct timespec)) + + nla_total_size(sizeof(u64)) + /* NET_DM_ATTR_ORIG_LEN */ nla_total_size(sizeof(u32)) + /* NET_DM_ATTR_PROTO */ @@ -351,7 +561,8 @@ static size_t net_dm_packet_report_size(size_t payload_len) nla_total_size(payload_len); } -static int net_dm_packet_report_in_port_put(struct sk_buff *msg, int ifindex) +static int net_dm_packet_report_in_port_put(struct sk_buff *msg, int ifindex, + const char *name) { struct nlattr *attr; @@ -363,6 +574,9 @@ static int net_dm_packet_report_in_port_put(struct sk_buff *msg, int ifindex) nla_put_u32(msg, NET_DM_ATTR_PORT_NETDEV_IFINDEX, ifindex)) goto nla_put_failure; + if (name && nla_put_string(msg, NET_DM_ATTR_PORT_NETDEV_NAME, name)) + goto nla_put_failure; + nla_nest_end(msg, attr); return 0; @@ -378,7 +592,6 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb, u64 pc = (u64)(uintptr_t) NET_DM_SKB_CB(skb)->pc; char buf[NET_DM_MAX_SYMBOL_LEN]; struct nlattr *attr; - struct timespec ts; void *hdr; int rc; @@ -387,6 +600,9 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb, if (!hdr) return -EMSGSIZE; + if (nla_put_u16(msg, NET_DM_ATTR_ORIGIN, NET_DM_ORIGIN_SW)) + goto nla_put_failure; + if (nla_put_u64_64bit(msg, NET_DM_ATTR_PC, pc, NET_DM_ATTR_PAD)) goto nla_put_failure; @@ -394,12 +610,12 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb, if (nla_put_string(msg, NET_DM_ATTR_SYMBOL, buf)) goto nla_put_failure; - rc = net_dm_packet_report_in_port_put(msg, skb->skb_iif); + rc = net_dm_packet_report_in_port_put(msg, skb->skb_iif, NULL); if (rc) goto nla_put_failure; - if (ktime_to_timespec_cond(skb->tstamp, &ts) && - nla_put(msg, NET_DM_ATTR_TIMESTAMP, sizeof(ts), &ts)) + if (nla_put_u64_64bit(msg, NET_DM_ATTR_TIMESTAMP, + ktime_to_ns(skb->tstamp), NET_DM_ATTR_PAD)) goto nla_put_failure; if (nla_put_u32(msg, NET_DM_ATTR_ORIG_LEN, skb->len)) @@ -481,10 +697,249 @@ static void net_dm_packet_work(struct work_struct *work) net_dm_packet_report(skb); } +static size_t +net_dm_hw_packet_report_size(size_t payload_len, + const struct net_dm_hw_metadata *hw_metadata) +{ + size_t size; + + size = nlmsg_msg_size(GENL_HDRLEN + net_drop_monitor_family.hdrsize); + + return NLMSG_ALIGN(size) + + /* NET_DM_ATTR_ORIGIN */ + nla_total_size(sizeof(u16)) + + /* NET_DM_ATTR_HW_TRAP_GROUP_NAME */ + nla_total_size(strlen(hw_metadata->trap_group_name) + 1) + + /* NET_DM_ATTR_HW_TRAP_NAME */ + nla_total_size(strlen(hw_metadata->trap_name) + 1) + + /* NET_DM_ATTR_IN_PORT */ + net_dm_in_port_size() + + /* NET_DM_ATTR_TIMESTAMP */ + nla_total_size(sizeof(u64)) + + /* NET_DM_ATTR_ORIG_LEN */ + nla_total_size(sizeof(u32)) + + /* NET_DM_ATTR_PROTO */ + nla_total_size(sizeof(u16)) + + /* NET_DM_ATTR_PAYLOAD */ + nla_total_size(payload_len); +} + +static int net_dm_hw_packet_report_fill(struct sk_buff *msg, + struct sk_buff *skb, size_t payload_len) +{ + struct net_dm_hw_metadata *hw_metadata; + struct nlattr *attr; + void *hdr; + + hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata; + + hdr = genlmsg_put(msg, 0, 0, &net_drop_monitor_family, 0, + NET_DM_CMD_PACKET_ALERT); + if (!hdr) + return -EMSGSIZE; + + if (nla_put_u16(msg, NET_DM_ATTR_ORIGIN, NET_DM_ORIGIN_HW)) + goto nla_put_failure; + + if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_GROUP_NAME, + hw_metadata->trap_group_name)) + goto nla_put_failure; + + if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_NAME, + hw_metadata->trap_name)) + goto nla_put_failure; + + if (hw_metadata->input_dev) { + struct net_device *dev = hw_metadata->input_dev; + int rc; + + rc = net_dm_packet_report_in_port_put(msg, dev->ifindex, + dev->name); + if (rc) + goto nla_put_failure; + } + + if (nla_put_u64_64bit(msg, NET_DM_ATTR_TIMESTAMP, + ktime_to_ns(skb->tstamp), NET_DM_ATTR_PAD)) + goto nla_put_failure; + + if (nla_put_u32(msg, NET_DM_ATTR_ORIG_LEN, skb->len)) + goto nla_put_failure; + + if (!payload_len) + goto out; + + if (nla_put_u16(msg, NET_DM_ATTR_PROTO, be16_to_cpu(skb->protocol))) + goto nla_put_failure; + + attr = skb_put(msg, nla_total_size(payload_len)); + attr->nla_type = NET_DM_ATTR_PAYLOAD; + attr->nla_len = nla_attr_size(payload_len); + if (skb_copy_bits(skb, 0, nla_data(attr), payload_len)) + goto nla_put_failure; + +out: + genlmsg_end(msg, hdr); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static struct net_dm_hw_metadata * +net_dm_hw_metadata_clone(const struct net_dm_hw_metadata *hw_metadata) +{ + struct net_dm_hw_metadata *n_hw_metadata; + const char *trap_group_name; + const char *trap_name; + + n_hw_metadata = kmalloc(sizeof(*hw_metadata), GFP_ATOMIC); + if (!n_hw_metadata) + return NULL; + + trap_group_name = kmemdup(hw_metadata->trap_group_name, + strlen(hw_metadata->trap_group_name) + 1, + GFP_ATOMIC | __GFP_ZERO); + if (!trap_group_name) + goto free_hw_metadata; + n_hw_metadata->trap_group_name = trap_group_name; + + trap_name = kmemdup(hw_metadata->trap_name, + strlen(hw_metadata->trap_name) + 1, + GFP_ATOMIC | __GFP_ZERO); + if (!trap_name) + goto free_trap_group; + n_hw_metadata->trap_name = trap_name; + + n_hw_metadata->input_dev = hw_metadata->input_dev; + if (n_hw_metadata->input_dev) + dev_hold(n_hw_metadata->input_dev); + + return n_hw_metadata; + +free_trap_group: + kfree(trap_group_name); +free_hw_metadata: + kfree(n_hw_metadata); + return NULL; +} + +static void +net_dm_hw_metadata_free(const struct net_dm_hw_metadata *hw_metadata) +{ + if (hw_metadata->input_dev) + dev_put(hw_metadata->input_dev); + kfree(hw_metadata->trap_name); + kfree(hw_metadata->trap_group_name); + kfree(hw_metadata); +} + +static void net_dm_hw_packet_report(struct sk_buff *skb) +{ + struct net_dm_hw_metadata *hw_metadata; + struct sk_buff *msg; + size_t payload_len; + int rc; + + if (skb->data > skb_mac_header(skb)) + skb_push(skb, skb->data - skb_mac_header(skb)); + else + skb_pull(skb, skb_mac_header(skb) - skb->data); + + payload_len = min_t(size_t, skb->len, NET_DM_MAX_PACKET_SIZE); + if (net_dm_trunc_len) + payload_len = min_t(size_t, net_dm_trunc_len, payload_len); + + hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata; + msg = nlmsg_new(net_dm_hw_packet_report_size(payload_len, hw_metadata), + GFP_KERNEL); + if (!msg) + goto out; + + rc = net_dm_hw_packet_report_fill(msg, skb, payload_len); + if (rc) { + nlmsg_free(msg); + goto out; + } + + genlmsg_multicast(&net_drop_monitor_family, msg, 0, 0, GFP_KERNEL); + +out: + net_dm_hw_metadata_free(NET_DM_SKB_CB(skb)->hw_metadata); + consume_skb(skb); +} + +static void net_dm_hw_packet_work(struct work_struct *work) +{ + struct per_cpu_dm_data *hw_data; + struct sk_buff_head list; + struct sk_buff *skb; + unsigned long flags; + + hw_data = container_of(work, struct per_cpu_dm_data, dm_alert_work); + + __skb_queue_head_init(&list); + + spin_lock_irqsave(&hw_data->drop_queue.lock, flags); + skb_queue_splice_tail_init(&hw_data->drop_queue, &list); + spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags); + + while ((skb = __skb_dequeue(&list))) + net_dm_hw_packet_report(skb); +} + +static void +net_dm_hw_packet_probe(struct sk_buff *skb, + const struct net_dm_hw_metadata *hw_metadata) +{ + struct net_dm_hw_metadata *n_hw_metadata; + ktime_t tstamp = ktime_get_real(); + struct per_cpu_dm_data *hw_data; + struct sk_buff *nskb; + unsigned long flags; + + nskb = skb_clone(skb, GFP_ATOMIC); + if (!nskb) + return; + + n_hw_metadata = net_dm_hw_metadata_clone(hw_metadata); + if (!n_hw_metadata) + goto free; + + NET_DM_SKB_CB(nskb)->hw_metadata = n_hw_metadata; + nskb->tstamp = tstamp; + + hw_data = this_cpu_ptr(&dm_hw_cpu_data); + + spin_lock_irqsave(&hw_data->drop_queue.lock, flags); + if (skb_queue_len(&hw_data->drop_queue) < net_dm_queue_len) + __skb_queue_tail(&hw_data->drop_queue, nskb); + else + goto unlock_free; + spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags); + + schedule_work(&hw_data->dm_alert_work); + + return; + +unlock_free: + spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags); + u64_stats_update_begin(&hw_data->stats.syncp); + hw_data->stats.dropped++; + u64_stats_update_end(&hw_data->stats.syncp); + net_dm_hw_metadata_free(n_hw_metadata); +free: + consume_skb(nskb); +} + static const struct net_dm_alert_ops net_dm_alert_packet_ops = { .kfree_skb_probe = net_dm_packet_trace_kfree_skb_hit, .napi_poll_probe = net_dm_packet_trace_napi_poll_hit, .work_item_func = net_dm_packet_work, + .hw_work_item_func = net_dm_hw_packet_work, + .hw_probe = net_dm_hw_packet_probe, }; static const struct net_dm_alert_ops *net_dm_alert_ops_arr[] = { @@ -492,6 +947,85 @@ static const struct net_dm_alert_ops *net_dm_alert_ops_arr[] = { [NET_DM_ALERT_MODE_PACKET] = &net_dm_alert_packet_ops, }; +void net_dm_hw_report(struct sk_buff *skb, + const struct net_dm_hw_metadata *hw_metadata) +{ + rcu_read_lock(); + + if (!monitor_hw) + goto out; + + net_dm_alert_ops_arr[net_dm_alert_mode]->hw_probe(skb, hw_metadata); + +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(net_dm_hw_report); + +static int net_dm_hw_monitor_start(struct netlink_ext_ack *extack) +{ + const struct net_dm_alert_ops *ops; + int cpu; + + if (monitor_hw) { + NL_SET_ERR_MSG_MOD(extack, "Hardware monitoring already enabled"); + return -EAGAIN; + } + + ops = net_dm_alert_ops_arr[net_dm_alert_mode]; + + if (!try_module_get(THIS_MODULE)) { + NL_SET_ERR_MSG_MOD(extack, "Failed to take reference on module"); + return -ENODEV; + } + + for_each_possible_cpu(cpu) { + struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu); + struct net_dm_hw_entries *hw_entries; + + INIT_WORK(&hw_data->dm_alert_work, ops->hw_work_item_func); + timer_setup(&hw_data->send_timer, sched_send_work, 0); + hw_entries = net_dm_hw_reset_per_cpu_data(hw_data); + kfree(hw_entries); + } + + monitor_hw = true; + + return 0; +} + +static void net_dm_hw_monitor_stop(struct netlink_ext_ack *extack) +{ + int cpu; + + if (!monitor_hw) + NL_SET_ERR_MSG_MOD(extack, "Hardware monitoring already disabled"); + + monitor_hw = false; + + /* After this call returns we are guaranteed that no CPU is processing + * any hardware drops. + */ + synchronize_rcu(); + + for_each_possible_cpu(cpu) { + struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu); + struct sk_buff *skb; + + del_timer_sync(&hw_data->send_timer); + cancel_work_sync(&hw_data->dm_alert_work); + while ((skb = __skb_dequeue(&hw_data->drop_queue))) { + struct net_dm_hw_metadata *hw_metadata; + + hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata; + net_dm_hw_metadata_free(hw_metadata); + consume_skb(skb); + } + } + + module_put(THIS_MODULE); +} + static int net_dm_trace_on_set(struct netlink_ext_ack *extack) { const struct net_dm_alert_ops *ops; @@ -604,6 +1138,11 @@ static int set_all_monitor_traces(int state, struct netlink_ext_ack *extack) return rc; } +static bool net_dm_is_monitoring(void) +{ + return trace_state == TRACE_ON || monitor_hw; +} + static int net_dm_alert_mode_get_from_info(struct genl_info *info, enum net_dm_alert_mode *p_alert_mode) { @@ -665,8 +1204,8 @@ static int net_dm_cmd_config(struct sk_buff *skb, struct netlink_ext_ack *extack = info->extack; int rc; - if (trace_state == TRACE_ON) { - NL_SET_ERR_MSG_MOD(extack, "Cannot configure drop monitor while tracing is on"); + if (net_dm_is_monitoring()) { + NL_SET_ERR_MSG_MOD(extack, "Cannot configure drop monitor during monitoring"); return -EBUSY; } @@ -681,14 +1220,61 @@ static int net_dm_cmd_config(struct sk_buff *skb, return 0; } +static int net_dm_monitor_start(bool set_sw, bool set_hw, + struct netlink_ext_ack *extack) +{ + bool sw_set = false; + int rc; + + if (set_sw) { + rc = set_all_monitor_traces(TRACE_ON, extack); + if (rc) + return rc; + sw_set = true; + } + + if (set_hw) { + rc = net_dm_hw_monitor_start(extack); + if (rc) + goto err_monitor_hw; + } + + return 0; + +err_monitor_hw: + if (sw_set) + set_all_monitor_traces(TRACE_OFF, extack); + return rc; +} + +static void net_dm_monitor_stop(bool set_sw, bool set_hw, + struct netlink_ext_ack *extack) +{ + if (set_hw) + net_dm_hw_monitor_stop(extack); + if (set_sw) + set_all_monitor_traces(TRACE_OFF, extack); +} + static int net_dm_cmd_trace(struct sk_buff *skb, struct genl_info *info) { + bool set_sw = !!info->attrs[NET_DM_ATTR_SW_DROPS]; + bool set_hw = !!info->attrs[NET_DM_ATTR_HW_DROPS]; + struct netlink_ext_ack *extack = info->extack; + + /* To maintain backward compatibility, we start / stop monitoring of + * software drops if no flag is specified. + */ + if (!set_sw && !set_hw) + set_sw = true; + switch (info->genlhdr->cmd) { case NET_DM_CMD_START: - return set_all_monitor_traces(TRACE_ON, info->extack); + return net_dm_monitor_start(set_sw, set_hw, extack); case NET_DM_CMD_STOP: - return set_all_monitor_traces(TRACE_OFF, info->extack); + net_dm_monitor_stop(set_sw, set_hw, extack); + return 0; } return -EOPNOTSUPP; @@ -785,6 +1371,50 @@ nla_put_failure: return -EMSGSIZE; } +static void net_dm_hw_stats_read(struct net_dm_stats *stats) +{ + int cpu; + + memset(stats, 0, sizeof(*stats)); + for_each_possible_cpu(cpu) { + struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu); + struct net_dm_stats *cpu_stats = &hw_data->stats; + unsigned int start; + u64 dropped; + + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + dropped = cpu_stats->dropped; + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + stats->dropped += dropped; + } +} + +static int net_dm_hw_stats_put(struct sk_buff *msg) +{ + struct net_dm_stats stats; + struct nlattr *attr; + + net_dm_hw_stats_read(&stats); + + attr = nla_nest_start(msg, NET_DM_ATTR_HW_STATS); + if (!attr) + return -EMSGSIZE; + + if (nla_put_u64_64bit(msg, NET_DM_ATTR_STATS_DROPPED, + stats.dropped, NET_DM_ATTR_PAD)) + goto nla_put_failure; + + nla_nest_end(msg, attr); + + return 0; + +nla_put_failure: + nla_nest_cancel(msg, attr); + return -EMSGSIZE; +} + static int net_dm_stats_fill(struct sk_buff *msg, struct genl_info *info) { void *hdr; @@ -799,6 +1429,10 @@ static int net_dm_stats_fill(struct sk_buff *msg, struct genl_info *info) if (rc) goto nla_put_failure; + rc = net_dm_hw_stats_put(msg); + if (rc) + goto nla_put_failure; + genlmsg_end(msg, hdr); return 0; @@ -872,6 +1506,8 @@ static const struct nla_policy net_dm_nl_policy[NET_DM_ATTR_MAX + 1] = { [NET_DM_ATTR_ALERT_MODE] = { .type = NLA_U8 }, [NET_DM_ATTR_TRUNC_LEN] = { .type = NLA_U32 }, [NET_DM_ATTR_QUEUE_LEN] = { .type = NLA_U32 }, + [NET_DM_ATTR_SW_DROPS] = {. type = NLA_FLAG }, + [NET_DM_ATTR_HW_DROPS] = {. type = NLA_FLAG }, }; static const struct genl_ops dropmon_ops[] = { @@ -934,9 +1570,57 @@ static struct notifier_block dropmon_net_notifier = { .notifier_call = dropmon_net_event }; -static int __init init_net_drop_monitor(void) +static void __net_dm_cpu_data_init(struct per_cpu_dm_data *data) +{ + spin_lock_init(&data->lock); + skb_queue_head_init(&data->drop_queue); + u64_stats_init(&data->stats.syncp); +} + +static void __net_dm_cpu_data_fini(struct per_cpu_dm_data *data) +{ + WARN_ON(!skb_queue_empty(&data->drop_queue)); +} + +static void net_dm_cpu_data_init(int cpu) +{ + struct per_cpu_dm_data *data; + + data = &per_cpu(dm_cpu_data, cpu); + __net_dm_cpu_data_init(data); +} + +static void net_dm_cpu_data_fini(int cpu) { struct per_cpu_dm_data *data; + + data = &per_cpu(dm_cpu_data, cpu); + /* At this point, we should have exclusive access + * to this struct and can free the skb inside it. + */ + consume_skb(data->skb); + __net_dm_cpu_data_fini(data); +} + +static void net_dm_hw_cpu_data_init(int cpu) +{ + struct per_cpu_dm_data *hw_data; + + hw_data = &per_cpu(dm_hw_cpu_data, cpu); + __net_dm_cpu_data_init(hw_data); +} + +static void net_dm_hw_cpu_data_fini(int cpu) +{ + struct per_cpu_dm_data *hw_data; + + hw_data = &per_cpu(dm_hw_cpu_data, cpu); + kfree(hw_data->hw_entries); + __net_dm_cpu_data_fini(hw_data); +} + +static int __init init_net_drop_monitor(void) +{ int cpu, rc; pr_info("Initializing network drop monitor service\n"); @@ -962,10 +1646,8 @@ static int __init init_net_drop_monitor(void) rc = 0; for_each_possible_cpu(cpu) { - data = &per_cpu(dm_cpu_data, cpu); - spin_lock_init(&data->lock); - skb_queue_head_init(&data->drop_queue); - u64_stats_init(&data->stats.syncp); + net_dm_cpu_data_init(cpu); + net_dm_hw_cpu_data_init(cpu); } goto out; @@ -978,7 +1660,6 @@ out: static void exit_net_drop_monitor(void) { - struct per_cpu_dm_data *data; int cpu; BUG_ON(unregister_netdevice_notifier(&dropmon_net_notifier)); @@ -989,13 +1670,8 @@ static void exit_net_drop_monitor(void) */ for_each_possible_cpu(cpu) { - data = &per_cpu(dm_cpu_data, cpu); - /* - * At this point, we should have exclusive access - * to this struct and can free the skb inside it - */ - kfree_skb(data->skb); - WARN_ON(!skb_queue_empty(&data->drop_queue)); + net_dm_hw_cpu_data_fini(cpu); + net_dm_cpu_data_fini(cpu); } BUG_ON(genl_unregister_family(&net_drop_monitor_family)); diff --git a/net/core/filter.c b/net/core/filter.c index 17bc9af8f156..ed6563622ce3 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -8841,13 +8841,13 @@ sk_reuseport_is_valid_access(int off, int size, return size == size_default; /* Fields that allow narrowing */ - case offsetof(struct sk_reuseport_md, eth_protocol): + case bpf_ctx_range(struct sk_reuseport_md, eth_protocol): if (size < FIELD_SIZEOF(struct sk_buff, protocol)) return false; /* fall through */ - case offsetof(struct sk_reuseport_md, ip_protocol): - case offsetof(struct sk_reuseport_md, bind_inany): - case offsetof(struct sk_reuseport_md, len): + case bpf_ctx_range(struct sk_reuseport_md, ip_protocol): + case bpf_ctx_range(struct sk_reuseport_md, bind_inany): + case bpf_ctx_range(struct sk_reuseport_md, len): bpf_ctx_record_field_size(info, size_default); return bpf_ctx_narrow_access_ok(off, size, size_default); diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 9741b593ea53..7c09d87d3269 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -142,8 +142,8 @@ int skb_flow_dissector_bpf_prog_detach(const union bpf_attr *attr) mutex_unlock(&flow_dissector_mutex); return -ENOENT; } - bpf_prog_put(attached); RCU_INIT_POINTER(net->flow_dissector_prog, NULL); + bpf_prog_put(attached); mutex_unlock(&flow_dissector_mutex); return 0; } diff --git a/net/core/flow_offload.c b/net/core/flow_offload.c index 64c3d4d72b9c..cf52d9c422fa 100644 --- a/net/core/flow_offload.c +++ b/net/core/flow_offload.c @@ -391,6 +391,8 @@ static void flow_indr_block_cb_del(struct flow_indr_block_cb *indr_block_cb) kfree(indr_block_cb); } +static DEFINE_MUTEX(flow_indr_block_ing_cb_lock); + static void flow_block_ing_cmd(struct net_device *dev, flow_indr_block_bind_cb_t *cb, void *cb_priv, @@ -398,11 +400,11 @@ static void flow_block_ing_cmd(struct net_device *dev, { struct flow_indr_block_ing_entry *entry; - rcu_read_lock(); - list_for_each_entry_rcu(entry, &block_ing_cb_list, list) { + mutex_lock(&flow_indr_block_ing_cb_lock); + list_for_each_entry(entry, &block_ing_cb_list, list) { entry->cb(dev, cb, cb_priv, command); } - rcu_read_unlock(); + mutex_unlock(&flow_indr_block_ing_cb_lock); } int __flow_indr_block_cb_register(struct net_device *dev, void *cb_priv, @@ -497,11 +499,10 @@ void flow_indr_block_call(struct net_device *dev, } EXPORT_SYMBOL_GPL(flow_indr_block_call); -static DEFINE_MUTEX(flow_indr_block_ing_cb_lock); void flow_indr_add_block_ing_cb(struct flow_indr_block_ing_entry *entry) { mutex_lock(&flow_indr_block_ing_cb_lock); - list_add_tail_rcu(&entry->list, &block_ing_cb_list); + list_add_tail(&entry->list, &block_ing_cb_list); mutex_unlock(&flow_indr_block_ing_cb_lock); } EXPORT_SYMBOL_GPL(flow_indr_add_block_ing_cb); @@ -509,7 +510,7 @@ EXPORT_SYMBOL_GPL(flow_indr_add_block_ing_cb); void flow_indr_del_block_ing_cb(struct flow_indr_block_ing_entry *entry) { mutex_lock(&flow_indr_block_ing_cb_lock); - list_del_rcu(&entry->list); + list_del(&entry->list); mutex_unlock(&flow_indr_block_ing_cb_lock); } EXPORT_SYMBOL_GPL(flow_indr_del_block_ing_cb); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 2cf27da1baeb..849380a622ef 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -122,7 +122,7 @@ static void queue_process(struct work_struct *work) txq = netdev_get_tx_queue(dev, q_index); HARD_TX_LOCK(dev, txq, smp_processor_id()); if (netif_xmit_frozen_or_stopped(txq) || - netpoll_start_xmit(skb, dev, txq) != NETDEV_TX_OK) { + !dev_xmit_complete(netpoll_start_xmit(skb, dev, txq))) { skb_queue_head(&npinfo->txq, skb); HARD_TX_UNLOCK(dev, txq); local_irq_restore(flags); @@ -335,7 +335,7 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, HARD_TX_UNLOCK(dev, txq); - if (status == NETDEV_TX_OK) + if (dev_xmit_complete(status)) break; } @@ -352,7 +352,7 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, } - if (status != NETDEV_TX_OK) { + if (!dev_xmit_complete(status)) { skb_queue_tail(&npinfo->txq, skb); schedule_delayed_work(&npinfo->tx_work,0); } diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 3272dc7a8c81..5bc65587f1c4 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -61,7 +61,7 @@ static int page_pool_init(struct page_pool *pool, struct page_pool *page_pool_create(const struct page_pool_params *params) { struct page_pool *pool; - int err = 0; + int err; pool = kzalloc_node(sizeof(*pool), GFP_KERNEL, params->nid); if (!pool) @@ -82,12 +82,9 @@ EXPORT_SYMBOL(page_pool_create); static struct page *__page_pool_get_cached(struct page_pool *pool) { struct ptr_ring *r = &pool->ring; + bool refill = false; struct page *page; - /* Quicker fallback, avoid locks when ring is empty */ - if (__ptr_ring_empty(r)) - return NULL; - /* Test for safe-context, caller should provide this guarantee */ if (likely(in_serving_softirq())) { if (likely(pool->alloc.count)) { @@ -95,27 +92,23 @@ static struct page *__page_pool_get_cached(struct page_pool *pool) page = pool->alloc.cache[--pool->alloc.count]; return page; } - /* Slower-path: Alloc array empty, time to refill - * - * Open-coded bulk ptr_ring consumer. - * - * Discussion: the ring consumer lock is not really - * needed due to the softirq/NAPI protection, but - * later need the ability to reclaim pages on the - * ring. Thus, keeping the locks. - */ - spin_lock(&r->consumer_lock); - while ((page = __ptr_ring_consume(r))) { - if (pool->alloc.count == PP_ALLOC_CACHE_REFILL) - break; - pool->alloc.cache[pool->alloc.count++] = page; - } - spin_unlock(&r->consumer_lock); - return page; + refill = true; } - /* Slow-path: Get page from locked ring queue */ - page = ptr_ring_consume(&pool->ring); + /* Quicker fallback, avoid locks when ring is empty */ + if (__ptr_ring_empty(r)) + return NULL; + + /* Slow-path: Get page from locked ring queue, + * refill alloc array if requested. + */ + spin_lock(&r->consumer_lock); + page = __ptr_ring_consume(r); + if (refill) + pool->alloc.count = __ptr_ring_consume_batched(r, + pool->alloc.cache, + PP_ALLOC_CACHE_REFILL); + spin_unlock(&r->consumer_lock); return page; } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index ea8e8d332d85..2b40b5a9425b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4087,6 +4087,9 @@ static const u8 skb_ext_type_len[] = { #ifdef CONFIG_XFRM [SKB_EXT_SEC_PATH] = SKB_EXT_CHUNKSIZEOF(struct sec_path), #endif +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + [TC_SKB_EXT] = SKB_EXT_CHUNKSIZEOF(struct tc_skb_ext), +#endif }; static __always_inline unsigned int skb_ext_total_length(void) @@ -4098,6 +4101,9 @@ static __always_inline unsigned int skb_ext_total_length(void) #ifdef CONFIG_XFRM skb_ext_type_len[SKB_EXT_SEC_PATH] + #endif +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + skb_ext_type_len[TC_SKB_EXT] + +#endif 0; } diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 6832eeb4b785..cf390e0aa73d 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -190,8 +190,7 @@ static int __sk_msg_free(struct sock *sk, struct sk_msg *msg, u32 i, sk_msg_check_to_free(msg, i, msg->sg.size); sge = sk_msg_elem(msg, i); } - if (msg->skb) - consume_skb(msg->skb); + consume_skb(msg->skb); sk_msg_init(msg); return freed; } diff --git a/net/core/sock.c b/net/core/sock.c index f5e801a9cea4..07863edbe6fc 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1995,6 +1995,19 @@ void skb_set_owner_w(struct sk_buff *skb, struct sock *sk) } EXPORT_SYMBOL(skb_set_owner_w); +static bool can_skb_orphan_partial(const struct sk_buff *skb) +{ +#ifdef CONFIG_TLS_DEVICE + /* Drivers depend on in-order delivery for crypto offload, + * partial orphan breaks out-of-order-OK logic. + */ + if (skb->decrypted) + return false; +#endif + return (skb->destructor == sock_wfree || + (IS_ENABLED(CONFIG_INET) && skb->destructor == tcp_wfree)); +} + /* This helper is used by netem, as it can hold packets in its * delay queue. We want to allow the owner socket to send more * packets, as if they were already TX completed by a typical driver. @@ -2006,11 +2019,7 @@ void skb_orphan_partial(struct sk_buff *skb) if (skb_is_tcp_pure_ack(skb)) return; - if (skb->destructor == sock_wfree -#ifdef CONFIG_INET - || skb->destructor == tcp_wfree -#endif - ) { + if (can_skb_orphan_partial(skb)) { struct sock *sk = skb->sk; if (refcount_inc_not_zero(&sk->sk_refcnt)) { @@ -3281,16 +3290,17 @@ static __init int net_inuse_init(void) core_initcall(net_inuse_init); -static void assign_proto_idx(struct proto *prot) +static int assign_proto_idx(struct proto *prot) { prot->inuse_idx = find_first_zero_bit(proto_inuse_idx, PROTO_INUSE_NR); if (unlikely(prot->inuse_idx == PROTO_INUSE_NR - 1)) { pr_err("PROTO_INUSE_NR exhausted\n"); - return; + return -ENOSPC; } set_bit(prot->inuse_idx, proto_inuse_idx); + return 0; } static void release_proto_idx(struct proto *prot) @@ -3299,8 +3309,9 @@ static void release_proto_idx(struct proto *prot) clear_bit(prot->inuse_idx, proto_inuse_idx); } #else -static inline void assign_proto_idx(struct proto *prot) +static inline int assign_proto_idx(struct proto *prot) { + return 0; } static inline void release_proto_idx(struct proto *prot) @@ -3349,6 +3360,8 @@ static int req_prot_init(const struct proto *prot) int proto_register(struct proto *prot, int alloc_slab) { + int ret = -ENOBUFS; + if (alloc_slab) { prot->slab = kmem_cache_create_usercopy(prot->name, prot->obj_size, 0, @@ -3385,20 +3398,27 @@ int proto_register(struct proto *prot, int alloc_slab) } mutex_lock(&proto_list_mutex); + ret = assign_proto_idx(prot); + if (ret) { + mutex_unlock(&proto_list_mutex); + goto out_free_timewait_sock_slab_name; + } list_add(&prot->node, &proto_list); - assign_proto_idx(prot); mutex_unlock(&proto_list_mutex); - return 0; + return ret; out_free_timewait_sock_slab_name: - kfree(prot->twsk_prot->twsk_slab_name); + if (alloc_slab && prot->twsk_prot) + kfree(prot->twsk_prot->twsk_slab_name); out_free_request_sock_slab: - req_prot_cleanup(prot->rsk_prot); + if (alloc_slab) { + req_prot_cleanup(prot->rsk_prot); - kmem_cache_destroy(prot->slab); - prot->slab = NULL; + kmem_cache_destroy(prot->slab); + prot->slab = NULL; + } out: - return -ENOBUFS; + return ret; } EXPORT_SYMBOL(proto_register); diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index 3312a5849a97..c13ffbd33d8d 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -19,6 +19,7 @@ static const struct sock_diag_handler *sock_diag_handlers[AF_MAX]; static int (*inet_rcv_compat)(struct sk_buff *skb, struct nlmsghdr *nlh); static DEFINE_MUTEX(sock_diag_table_mutex); static struct workqueue_struct *broadcast_wq; +static atomic64_t cookie_gen; u64 sock_gen_cookie(struct sock *sk) { @@ -27,7 +28,7 @@ u64 sock_gen_cookie(struct sock *sk) if (res) return res; - res = atomic64_inc_return(&sock_net(sk)->cookie_gen); + res = atomic64_inc_return(&cookie_gen); atomic64_cmpxchg(&sk->sk_cookie, 0, res); } } diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 1330a7442e5b..01998860afaa 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -345,7 +345,7 @@ static int sock_map_update_common(struct bpf_map *map, u32 idx, return -EINVAL; if (unlikely(idx >= map->max_entries)) return -E2BIG; - if (unlikely(icsk->icsk_ulp_data)) + if (unlikely(rcu_access_pointer(icsk->icsk_ulp_data))) return -EINVAL; link = sk_psock_init_link(); diff --git a/net/core/stream.c b/net/core/stream.c index e94bb02a5629..4f1d4aa5fb38 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -120,7 +120,6 @@ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) int err = 0; long vm_wait = 0; long current_timeo = *timeo_p; - bool noblock = (*timeo_p ? false : true); DEFINE_WAIT_FUNC(wait, woken_wake_function); if (sk_stream_memory_free(sk)) @@ -133,11 +132,8 @@ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) goto do_error; - if (!*timeo_p) { - if (noblock) - set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); - goto do_nonblock; - } + if (!*timeo_p) + goto do_eagain; if (signal_pending(current)) goto do_interrupted; sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); @@ -169,7 +165,13 @@ out: do_error: err = -EPIPE; goto out; -do_nonblock: +do_eagain: + /* Make sure that whenever EAGAIN is returned, EPOLLOUT event can + * be generated later. + * When TCP receives ACK packets that make room, tcp_check_space() + * only calls tcp_new_space() if SOCK_NOSPACE is set. + */ + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); err = -EAGAIN; goto out; do_interrupted: diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 3abd173ebacb..b501c90aabe4 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -254,88 +254,109 @@ static void dsa_tree_teardown_default_cpu(struct dsa_switch_tree *dst) static int dsa_port_setup(struct dsa_port *dp) { - enum devlink_port_flavour flavour; struct dsa_switch *ds = dp->ds; struct dsa_switch_tree *dst = ds->dst; + const unsigned char *id = (const unsigned char *)&dst->index; + const unsigned char len = sizeof(dst->index); + struct devlink_port *dlp = &dp->devlink_port; + bool dsa_port_link_registered = false; + bool devlink_port_registered = false; + struct devlink *dl = ds->devlink; + bool dsa_port_enabled = false; int err = 0; - if (dp->type == DSA_PORT_TYPE_UNUSED) - return 0; - - memset(&dp->devlink_port, 0, sizeof(dp->devlink_port)); - dp->mac = of_get_mac_address(dp->dn); - - switch (dp->type) { - case DSA_PORT_TYPE_CPU: - flavour = DEVLINK_PORT_FLAVOUR_CPU; - break; - case DSA_PORT_TYPE_DSA: - flavour = DEVLINK_PORT_FLAVOUR_DSA; - break; - case DSA_PORT_TYPE_USER: /* fall-through */ - default: - flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; - break; - } - - /* dp->index is used now as port_number. However - * CPU and DSA ports should have separate numbering - * independent from front panel port numbers. - */ - devlink_port_attrs_set(&dp->devlink_port, flavour, - dp->index, false, 0, - (const char *) &dst->index, sizeof(dst->index)); - err = devlink_port_register(ds->devlink, &dp->devlink_port, - dp->index); - if (err) - return err; - switch (dp->type) { case DSA_PORT_TYPE_UNUSED: + dsa_port_disable(dp); break; case DSA_PORT_TYPE_CPU: + memset(dlp, 0, sizeof(*dlp)); + devlink_port_attrs_set(dlp, DEVLINK_PORT_FLAVOUR_CPU, + dp->index, false, 0, id, len); + err = devlink_port_register(dl, dlp, dp->index); + if (err) + break; + devlink_port_registered = true; + err = dsa_port_link_register_of(dp); if (err) - dev_err(ds->dev, "failed to setup link for port %d.%d\n", - ds->index, dp->index); + break; + dsa_port_link_registered = true; + + err = dsa_port_enable(dp, NULL); + if (err) + break; + dsa_port_enabled = true; + break; case DSA_PORT_TYPE_DSA: + memset(dlp, 0, sizeof(*dlp)); + devlink_port_attrs_set(dlp, DEVLINK_PORT_FLAVOUR_DSA, + dp->index, false, 0, id, len); + err = devlink_port_register(dl, dlp, dp->index); + if (err) + break; + devlink_port_registered = true; + err = dsa_port_link_register_of(dp); if (err) - dev_err(ds->dev, "failed to setup link for port %d.%d\n", - ds->index, dp->index); + break; + dsa_port_link_registered = true; + + err = dsa_port_enable(dp, NULL); + if (err) + break; + dsa_port_enabled = true; + break; case DSA_PORT_TYPE_USER: + memset(dlp, 0, sizeof(*dlp)); + devlink_port_attrs_set(dlp, DEVLINK_PORT_FLAVOUR_PHYSICAL, + dp->index, false, 0, id, len); + err = devlink_port_register(dl, dlp, dp->index); + if (err) + break; + devlink_port_registered = true; + + dp->mac = of_get_mac_address(dp->dn); err = dsa_slave_create(dp); if (err) - dev_err(ds->dev, "failed to create slave for port %d.%d\n", - ds->index, dp->index); - else - devlink_port_type_eth_set(&dp->devlink_port, dp->slave); + break; + + devlink_port_type_eth_set(dlp, dp->slave); break; } - if (err) - devlink_port_unregister(&dp->devlink_port); + if (err && dsa_port_enabled) + dsa_port_disable(dp); + if (err && dsa_port_link_registered) + dsa_port_link_unregister_of(dp); + if (err && devlink_port_registered) + devlink_port_unregister(dlp); return err; } static void dsa_port_teardown(struct dsa_port *dp) { - if (dp->type != DSA_PORT_TYPE_UNUSED) - devlink_port_unregister(&dp->devlink_port); + struct devlink_port *dlp = &dp->devlink_port; switch (dp->type) { case DSA_PORT_TYPE_UNUSED: break; case DSA_PORT_TYPE_CPU: + dsa_port_disable(dp); dsa_tag_driver_put(dp->tag_ops); - /* fall-through */ + devlink_port_unregister(dlp); + dsa_port_link_unregister_of(dp); + break; case DSA_PORT_TYPE_DSA: + dsa_port_disable(dp); + devlink_port_unregister(dlp); dsa_port_link_unregister_of(dp); break; case DSA_PORT_TYPE_USER: + devlink_port_unregister(dlp); if (dp->slave) { dsa_slave_destroy(dp->slave); dp->slave = NULL; @@ -832,20 +853,6 @@ struct dsa_switch *dsa_switch_alloc(struct device *dev, size_t n) if (!ds) return NULL; - /* We avoid allocating memory outside dsa_switch - * if it is not needed. - */ - if (n <= sizeof(ds->_bitmap) * 8) { - ds->bitmap = &ds->_bitmap; - } else { - ds->bitmap = devm_kcalloc(dev, - BITS_TO_LONGS(n), - sizeof(unsigned long), - GFP_KERNEL); - if (unlikely(!ds->bitmap)) - return NULL; - } - ds->dev = dev; ds->num_ports = n; diff --git a/net/dsa/port.c b/net/dsa/port.c index f071acf2842b..9b54e5a76297 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -348,10 +348,7 @@ int dsa_port_vlan_add(struct dsa_port *dp, .vlan = vlan, }; - if (!dp->bridge_dev || br_vlan_enabled(dp->bridge_dev)) - return dsa_port_notify(dp, DSA_NOTIFIER_VLAN_ADD, &info); - - return 0; + return dsa_port_notify(dp, DSA_NOTIFIER_VLAN_ADD, &info); } int dsa_port_vlan_del(struct dsa_port *dp, @@ -363,10 +360,7 @@ int dsa_port_vlan_del(struct dsa_port *dp, .vlan = vlan, }; - if (!dp->bridge_dev || br_vlan_enabled(dp->bridge_dev)) - return dsa_port_notify(dp, DSA_NOTIFIER_VLAN_DEL, &info); - - return 0; + return dsa_port_notify(dp, DSA_NOTIFIER_VLAN_DEL, &info); } int dsa_port_vid_add(struct dsa_port *dp, u16 vid, u16 flags) @@ -382,8 +376,8 @@ int dsa_port_vid_add(struct dsa_port *dp, u16 vid, u16 flags) trans.ph_prepare = true; err = dsa_port_vlan_add(dp, &vlan, &trans); - if (err == -EOPNOTSUPP) - return 0; + if (err) + return err; trans.ph_prepare = false; return dsa_port_vlan_add(dp, &vlan, &trans); @@ -538,10 +532,6 @@ static int dsa_port_setup_phy_of(struct dsa_port *dp, bool enable) return PTR_ERR(phydev); if (enable) { - err = genphy_config_init(phydev); - if (err < 0) - goto err_put_dev; - err = genphy_resume(phydev); if (err < 0) goto err_put_dev; @@ -589,7 +579,6 @@ static int dsa_port_fixed_link_register_of(struct dsa_port *dp) mode = PHY_INTERFACE_MODE_NA; phydev->interface = mode; - genphy_config_init(phydev); genphy_read_status(phydev); if (ds->ops->adjust_link) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 33f41178afcc..9a88035517a6 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -312,6 +312,39 @@ static int dsa_slave_port_attr_set(struct net_device *dev, return ret; } +static int dsa_slave_vlan_add(struct net_device *dev, + const struct switchdev_obj *obj, + struct switchdev_trans *trans) +{ + struct dsa_port *dp = dsa_slave_to_port(dev); + struct switchdev_obj_port_vlan vlan; + int err; + + if (obj->orig_dev != dev) + return -EOPNOTSUPP; + + if (dp->bridge_dev && !br_vlan_enabled(dp->bridge_dev)) + return 0; + + vlan = *SWITCHDEV_OBJ_PORT_VLAN(obj); + + err = dsa_port_vlan_add(dp, &vlan, trans); + if (err) + return err; + + /* We need the dedicated CPU port to be a member of the VLAN as well. + * Even though drivers often handle CPU membership in special ways, + * it doesn't make sense to program a PVID, so clear this flag. + */ + vlan.flags &= ~BRIDGE_VLAN_INFO_PVID; + + err = dsa_port_vlan_add(dp->cpu_dp, &vlan, trans); + if (err) + return err; + + return 0; +} + static int dsa_slave_port_obj_add(struct net_device *dev, const struct switchdev_obj *obj, struct switchdev_trans *trans, @@ -339,10 +372,7 @@ static int dsa_slave_port_obj_add(struct net_device *dev, trans); break; case SWITCHDEV_OBJ_ID_PORT_VLAN: - if (obj->orig_dev != dev) - return -EOPNOTSUPP; - err = dsa_port_vlan_add(dp, SWITCHDEV_OBJ_PORT_VLAN(obj), - trans); + err = dsa_slave_vlan_add(dev, obj, trans); break; default: err = -EOPNOTSUPP; @@ -352,6 +382,23 @@ static int dsa_slave_port_obj_add(struct net_device *dev, return err; } +static int dsa_slave_vlan_del(struct net_device *dev, + const struct switchdev_obj *obj) +{ + struct dsa_port *dp = dsa_slave_to_port(dev); + + if (obj->orig_dev != dev) + return -EOPNOTSUPP; + + if (dp->bridge_dev && !br_vlan_enabled(dp->bridge_dev)) + return 0; + + /* Do not deprogram the CPU port as it may be shared with other user + * ports which can be members of this VLAN as well. + */ + return dsa_port_vlan_del(dp, SWITCHDEV_OBJ_PORT_VLAN(obj)); +} + static int dsa_slave_port_obj_del(struct net_device *dev, const struct switchdev_obj *obj) { @@ -371,9 +418,7 @@ static int dsa_slave_port_obj_del(struct net_device *dev, err = dsa_port_mdb_del(dp->cpu_dp, SWITCHDEV_OBJ_PORT_MDB(obj)); break; case SWITCHDEV_OBJ_ID_PORT_VLAN: - if (obj->orig_dev != dev) - return -EOPNOTSUPP; - err = dsa_port_vlan_del(dp, SWITCHDEV_OBJ_PORT_VLAN(obj)); + err = dsa_slave_vlan_del(dev, obj); break; default: err = -EOPNOTSUPP; @@ -1073,6 +1118,9 @@ static int dsa_slave_vlan_rx_add_vid(struct net_device *dev, __be16 proto, * need to emulate the switchdev prepare + commit phase. */ if (dp->bridge_dev) { + if (!br_vlan_enabled(dp->bridge_dev)) + return 0; + /* br_vlan_get_info() returns -EINVAL or -ENOENT if the * device, respectively the VID is not found, returning * 0 means success, which is a failure for us here. @@ -1082,8 +1130,15 @@ static int dsa_slave_vlan_rx_add_vid(struct net_device *dev, __be16 proto, return -EBUSY; } - /* This API only allows programming tagged, non-PVID VIDs */ - return dsa_port_vid_add(dp, vid, 0); + ret = dsa_port_vid_add(dp, vid, 0); + if (ret) + return ret; + + ret = dsa_port_vid_add(dp->cpu_dp, vid, 0); + if (ret) + return ret; + + return 0; } static int dsa_slave_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, @@ -1097,6 +1152,9 @@ static int dsa_slave_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, * need to emulate the switchdev prepare + commit phase. */ if (dp->bridge_dev) { + if (!br_vlan_enabled(dp->bridge_dev)) + return 0; + /* br_vlan_get_info() returns -EINVAL or -ENOENT if the * device, respectively the VID is not found, returning * 0 means success, which is a failure for us here. @@ -1106,11 +1164,10 @@ static int dsa_slave_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, return -EBUSY; } - ret = dsa_port_vid_del(dp, vid); - if (ret == -EOPNOTSUPP) - ret = 0; - - return ret; + /* Do not deprogram the CPU port as it may be shared with other user + * ports which can be members of this VLAN as well. + */ + return dsa_port_vid_del(dp, vid); } static const struct ethtool_ops dsa_slave_ethtool_ops = { @@ -1357,8 +1414,9 @@ int dsa_slave_create(struct dsa_port *port) if (slave_dev == NULL) return -ENOMEM; - slave_dev->features = master->vlan_features | NETIF_F_HW_TC | - NETIF_F_HW_VLAN_CTAG_FILTER; + slave_dev->features = master->vlan_features | NETIF_F_HW_TC; + if (ds->ops->port_vlan_add && ds->ops->port_vlan_del) + slave_dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; slave_dev->hw_features |= NETIF_F_HW_TC; slave_dev->ethtool_ops = &dsa_slave_ethtool_ops; if (!IS_ERR_OR_NULL(port->mac)) diff --git a/net/dsa/switch.c b/net/dsa/switch.c index 4ec5b7f85d51..6a9607518823 100644 --- a/net/dsa/switch.c +++ b/net/dsa/switch.c @@ -128,54 +128,51 @@ static int dsa_switch_fdb_del(struct dsa_switch *ds, return ds->ops->port_fdb_del(ds, port, info->addr, info->vid); } -static int -dsa_switch_mdb_prepare_bitmap(struct dsa_switch *ds, - const struct switchdev_obj_port_mdb *mdb, - const unsigned long *bitmap) +static bool dsa_switch_mdb_match(struct dsa_switch *ds, int port, + struct dsa_notifier_mdb_info *info) +{ + if (ds->index == info->sw_index && port == info->port) + return true; + + if (dsa_is_dsa_port(ds, port)) + return true; + + return false; +} + +static int dsa_switch_mdb_prepare(struct dsa_switch *ds, + struct dsa_notifier_mdb_info *info) { int port, err; if (!ds->ops->port_mdb_prepare || !ds->ops->port_mdb_add) return -EOPNOTSUPP; - for_each_set_bit(port, bitmap, ds->num_ports) { - err = ds->ops->port_mdb_prepare(ds, port, mdb); - if (err) - return err; + for (port = 0; port < ds->num_ports; port++) { + if (dsa_switch_mdb_match(ds, port, info)) { + err = ds->ops->port_mdb_prepare(ds, port, info->mdb); + if (err) + return err; + } } return 0; } -static void dsa_switch_mdb_add_bitmap(struct dsa_switch *ds, - const struct switchdev_obj_port_mdb *mdb, - const unsigned long *bitmap) -{ - int port; - - for_each_set_bit(port, bitmap, ds->num_ports) - ds->ops->port_mdb_add(ds, port, mdb); -} - static int dsa_switch_mdb_add(struct dsa_switch *ds, struct dsa_notifier_mdb_info *info) { - const struct switchdev_obj_port_mdb *mdb = info->mdb; - struct switchdev_trans *trans = info->trans; int port; - /* Build a mask of Multicast group members */ - bitmap_zero(ds->bitmap, ds->num_ports); - if (ds->index == info->sw_index) - set_bit(info->port, ds->bitmap); - for (port = 0; port < ds->num_ports; port++) - if (dsa_is_dsa_port(ds, port)) - set_bit(port, ds->bitmap); + if (switchdev_trans_ph_prepare(info->trans)) + return dsa_switch_mdb_prepare(ds, info); - if (switchdev_trans_ph_prepare(trans)) - return dsa_switch_mdb_prepare_bitmap(ds, mdb, ds->bitmap); + if (!ds->ops->port_mdb_add) + return 0; - dsa_switch_mdb_add_bitmap(ds, mdb, ds->bitmap); + for (port = 0; port < ds->num_ports; port++) + if (dsa_switch_mdb_match(ds, port, info)) + ds->ops->port_mdb_add(ds, port, info->mdb); return 0; } @@ -183,13 +180,11 @@ static int dsa_switch_mdb_add(struct dsa_switch *ds, static int dsa_switch_mdb_del(struct dsa_switch *ds, struct dsa_notifier_mdb_info *info) { - const struct switchdev_obj_port_mdb *mdb = info->mdb; - if (!ds->ops->port_mdb_del) return -EOPNOTSUPP; if (ds->index == info->sw_index) - return ds->ops->port_mdb_del(ds, info->port, mdb); + return ds->ops->port_mdb_del(ds, info->port, info->mdb); return 0; } @@ -231,59 +226,55 @@ static int dsa_port_vlan_check(struct dsa_switch *ds, int port, (void *)vlan); } -static int -dsa_switch_vlan_prepare_bitmap(struct dsa_switch *ds, - const struct switchdev_obj_port_vlan *vlan, - const unsigned long *bitmap) +static bool dsa_switch_vlan_match(struct dsa_switch *ds, int port, + struct dsa_notifier_vlan_info *info) +{ + if (ds->index == info->sw_index && port == info->port) + return true; + + if (dsa_is_dsa_port(ds, port)) + return true; + + return false; +} + +static int dsa_switch_vlan_prepare(struct dsa_switch *ds, + struct dsa_notifier_vlan_info *info) { int port, err; if (!ds->ops->port_vlan_prepare || !ds->ops->port_vlan_add) return -EOPNOTSUPP; - for_each_set_bit(port, bitmap, ds->num_ports) { - err = dsa_port_vlan_check(ds, port, vlan); - if (err) - return err; + for (port = 0; port < ds->num_ports; port++) { + if (dsa_switch_vlan_match(ds, port, info)) { + err = dsa_port_vlan_check(ds, port, info->vlan); + if (err) + return err; - err = ds->ops->port_vlan_prepare(ds, port, vlan); - if (err) - return err; + err = ds->ops->port_vlan_prepare(ds, port, info->vlan); + if (err) + return err; + } } return 0; } -static void -dsa_switch_vlan_add_bitmap(struct dsa_switch *ds, - const struct switchdev_obj_port_vlan *vlan, - const unsigned long *bitmap) -{ - int port; - - for_each_set_bit(port, bitmap, ds->num_ports) - ds->ops->port_vlan_add(ds, port, vlan); -} - static int dsa_switch_vlan_add(struct dsa_switch *ds, struct dsa_notifier_vlan_info *info) { - const struct switchdev_obj_port_vlan *vlan = info->vlan; - struct switchdev_trans *trans = info->trans; int port; - /* Build a mask of VLAN members */ - bitmap_zero(ds->bitmap, ds->num_ports); - if (ds->index == info->sw_index) - set_bit(info->port, ds->bitmap); - for (port = 0; port < ds->num_ports; port++) - if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) - set_bit(port, ds->bitmap); + if (switchdev_trans_ph_prepare(info->trans)) + return dsa_switch_vlan_prepare(ds, info); - if (switchdev_trans_ph_prepare(trans)) - return dsa_switch_vlan_prepare_bitmap(ds, vlan, ds->bitmap); + if (!ds->ops->port_vlan_add) + return 0; - dsa_switch_vlan_add_bitmap(ds, vlan, ds->bitmap); + for (port = 0; port < ds->num_ports; port++) + if (dsa_switch_vlan_match(ds, port, info)) + ds->ops->port_vlan_add(ds, port, info->vlan); return 0; } @@ -291,14 +282,15 @@ static int dsa_switch_vlan_add(struct dsa_switch *ds, static int dsa_switch_vlan_del(struct dsa_switch *ds, struct dsa_notifier_vlan_info *info) { - const struct switchdev_obj_port_vlan *vlan = info->vlan; - if (!ds->ops->port_vlan_del) return -EOPNOTSUPP; if (ds->index == info->sw_index) - return ds->ops->port_vlan_del(ds, info->port, vlan); + return ds->ops->port_vlan_del(ds, info->port, info->vlan); + /* Do not deprogram the DSA links as they may be used as conduit + * for other VLAN members in the fabric. + */ return 0; } diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c index 6ebbd799c4eb..9c1cc2482b68 100644 --- a/net/dsa/tag_8021q.c +++ b/net/dsa/tag_8021q.c @@ -28,6 +28,7 @@ * * RSV - VID[9]: * To be used for further expansion of SWITCH_ID or for other purposes. + * Must be transmitted as zero and ignored on receive. * * SWITCH_ID - VID[8:6]: * Index of switch within DSA tree. Must be between 0 and @@ -35,6 +36,7 @@ * * RSV - VID[5:4]: * To be used for further expansion of PORT or for other purposes. + * Must be transmitted as zero and ignored on receive. * * PORT - VID[3:0]: * Index of switch port. Must be between 0 and DSA_MAX_PORTS - 1. @@ -91,6 +93,79 @@ int dsa_8021q_rx_source_port(u16 vid) } EXPORT_SYMBOL_GPL(dsa_8021q_rx_source_port); +static int dsa_8021q_restore_pvid(struct dsa_switch *ds, int port) +{ + struct bridge_vlan_info vinfo; + struct net_device *slave; + u16 pvid; + int err; + + if (!dsa_is_user_port(ds, port)) + return 0; + + slave = ds->ports[port].slave; + + err = br_vlan_get_pvid(slave, &pvid); + if (err < 0) + /* There is no pvid on the bridge for this port, which is + * perfectly valid. Nothing to restore, bye-bye! + */ + return 0; + + err = br_vlan_get_info(slave, pvid, &vinfo); + if (err < 0) { + dev_err(ds->dev, "Couldn't determine PVID attributes\n"); + return err; + } + + return dsa_port_vid_add(&ds->ports[port], pvid, vinfo.flags); +} + +/* If @enabled is true, installs @vid with @flags into the switch port's HW + * filter. + * If @enabled is false, deletes @vid (ignores @flags) from the port. Had the + * user explicitly configured this @vid through the bridge core, then the @vid + * is installed again, but this time with the flags from the bridge layer. + */ +static int dsa_8021q_vid_apply(struct dsa_switch *ds, int port, u16 vid, + u16 flags, bool enabled) +{ + struct dsa_port *dp = &ds->ports[port]; + struct bridge_vlan_info vinfo; + int err; + + if (enabled) + return dsa_port_vid_add(dp, vid, flags); + + err = dsa_port_vid_del(dp, vid); + if (err < 0) + return err; + + /* Nothing to restore from the bridge for a non-user port. + * The CPU port VLANs are restored implicitly with the user ports, + * similar to how the bridge does in dsa_slave_vlan_add and + * dsa_slave_vlan_del. + */ + if (!dsa_is_user_port(ds, port)) + return 0; + + err = br_vlan_get_info(dp->slave, vid, &vinfo); + /* Couldn't determine bridge attributes for this vid, + * it means the bridge had not configured it. + */ + if (err < 0) + return 0; + + /* Restore the VID from the bridge */ + err = dsa_port_vid_add(dp, vid, vinfo.flags); + if (err < 0) + return err; + + vinfo.flags &= ~BRIDGE_VLAN_INFO_PVID; + + return dsa_port_vid_add(dp->cpu_dp, vid, vinfo.flags); +} + /* RX VLAN tagging (left) and TX VLAN tagging (right) setup shown for a single * front-panel switch port (here swp0). * @@ -146,8 +221,6 @@ EXPORT_SYMBOL_GPL(dsa_8021q_rx_source_port); int dsa_port_setup_8021q_tagging(struct dsa_switch *ds, int port, bool enabled) { int upstream = dsa_upstream_port(ds, port); - struct dsa_port *dp = &ds->ports[port]; - struct dsa_port *upstream_dp = &ds->ports[upstream]; u16 rx_vid = dsa_8021q_rx_vid(ds, port); u16 tx_vid = dsa_8021q_tx_vid(ds, port); int i, err; @@ -164,7 +237,6 @@ int dsa_port_setup_8021q_tagging(struct dsa_switch *ds, int port, bool enabled) * restrictions, so there are no concerns about leaking traffic. */ for (i = 0; i < ds->num_ports; i++) { - struct dsa_port *other_dp = &ds->ports[i]; u16 flags; if (i == upstream) @@ -177,10 +249,7 @@ int dsa_port_setup_8021q_tagging(struct dsa_switch *ds, int port, bool enabled) /* The RX VID is a regular VLAN on all others */ flags = BRIDGE_VLAN_INFO_UNTAGGED; - if (enabled) - err = dsa_port_vid_add(other_dp, rx_vid, flags); - else - err = dsa_port_vid_del(other_dp, rx_vid); + err = dsa_8021q_vid_apply(ds, i, rx_vid, flags, enabled); if (err) { dev_err(ds->dev, "Failed to apply RX VID %d to port %d: %d\n", rx_vid, port, err); @@ -191,10 +260,7 @@ int dsa_port_setup_8021q_tagging(struct dsa_switch *ds, int port, bool enabled) /* CPU port needs to see this port's RX VID * as tagged egress. */ - if (enabled) - err = dsa_port_vid_add(upstream_dp, rx_vid, 0); - else - err = dsa_port_vid_del(upstream_dp, rx_vid); + err = dsa_8021q_vid_apply(ds, upstream, rx_vid, 0, enabled); if (err) { dev_err(ds->dev, "Failed to apply RX VID %d to port %d: %d\n", rx_vid, port, err); @@ -202,26 +268,24 @@ int dsa_port_setup_8021q_tagging(struct dsa_switch *ds, int port, bool enabled) } /* Finally apply the TX VID on this port and on the CPU port */ - if (enabled) - err = dsa_port_vid_add(dp, tx_vid, BRIDGE_VLAN_INFO_UNTAGGED); - else - err = dsa_port_vid_del(dp, tx_vid); + err = dsa_8021q_vid_apply(ds, port, tx_vid, BRIDGE_VLAN_INFO_UNTAGGED, + enabled); if (err) { dev_err(ds->dev, "Failed to apply TX VID %d on port %d: %d\n", tx_vid, port, err); return err; } - if (enabled) - err = dsa_port_vid_add(upstream_dp, tx_vid, 0); - else - err = dsa_port_vid_del(upstream_dp, tx_vid); + err = dsa_8021q_vid_apply(ds, upstream, tx_vid, 0, enabled); if (err) { dev_err(ds->dev, "Failed to apply TX VID %d on port %d: %d\n", tx_vid, upstream, err); return err; } - return 0; + if (!enabled) + err = dsa_8021q_restore_pvid(ds, port); + + return err; } EXPORT_SYMBOL_GPL(dsa_port_setup_8021q_tagging); diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index e4aba5d485be..bbe9b3b2d395 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -170,7 +170,7 @@ static int lowpan_frag_reasm(struct lowpan_frag_queue *fq, struct sk_buff *skb, reasm_data = inet_frag_reasm_prepare(&fq->q, skb, prev_tail); if (!reasm_data) goto out_oom; - inet_frag_reasm_finish(&fq->q, skb, reasm_data); + inet_frag_reasm_finish(&fq->q, skb, reasm_data, false); skb->dev = ldev; skb->tstamp = fq->q.stamp; diff --git a/net/ieee802154/socket.c b/net/ieee802154/socket.c index dacbd58e1799..badc5cfe4dc6 100644 --- a/net/ieee802154/socket.c +++ b/net/ieee802154/socket.c @@ -1092,7 +1092,7 @@ static struct packet_type ieee802154_packet_type = { static int __init af_ieee802154_init(void) { - int rc = -EINVAL; + int rc; rc = proto_register(&ieee802154_raw_prot, 1); if (rc) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index ed2301ef872e..70f92aaca411 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1845,13 +1845,8 @@ static __net_init int inet_init_net(struct net *net) return 0; } -static __net_exit void inet_exit_net(struct net *net) -{ -} - static __net_initdata struct pernet_operations af_inet_ops = { .init = inet_init_net, - .exit = inet_exit_net, }; static int __init init_inet_pernet_ops(void) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 2b2b3d291ab0..1ab2fb6bb37d 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -2145,7 +2145,7 @@ static int fn_trie_dump_leaf(struct key_vector *l, struct fib_table *tb, if (filter->dump_exceptions) { err = fib_dump_info_fnhe(skb, cb, tb->tb_id, fi, - &i_fa, s_fa); + &i_fa, s_fa, flags); if (err < 0) goto stop; } diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 1510e951f451..4298aae74e0e 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -582,7 +582,13 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, if (!rt) goto out; - net = dev_net(rt->dst.dev); + + if (rt->dst.dev) + net = dev_net(rt->dst.dev); + else if (skb_in->dev) + net = dev_net(skb_in->dev); + else + goto out; /* * Find the original header. It is expected to be valid, of course. @@ -902,7 +908,7 @@ static bool icmp_redirect(struct sk_buff *skb) return false; } - icmp_socket_deliver(skb, icmp_hdr(skb)->un.gateway); + icmp_socket_deliver(skb, ntohl(icmp_hdr(skb)->un.gateway)); return true; } diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 180f6896b98b..480d0b22db1a 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1475,7 +1475,7 @@ EXPORT_SYMBOL(__ip_mc_inc_group); void ip_mc_inc_group(struct in_device *in_dev, __be32 addr) { - __ip_mc_inc_group(in_dev, addr, MCAST_EXCLUDE); + __ip_mc_inc_group(in_dev, addr, GFP_KERNEL); } EXPORT_SYMBOL(ip_mc_inc_group); @@ -2197,7 +2197,7 @@ static int __ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr, iml->sflist = NULL; iml->sfmode = mode; rcu_assign_pointer(inet->mc_list, iml); - __ip_mc_inc_group(in_dev, addr, mode); + ____ip_mc_inc_group(in_dev, addr, mode, GFP_KERNEL); err = 0; done: return err; diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index a999451345f9..10d31733297d 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -475,11 +475,12 @@ void *inet_frag_reasm_prepare(struct inet_frag_queue *q, struct sk_buff *skb, EXPORT_SYMBOL(inet_frag_reasm_prepare); void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, - void *reasm_data) + void *reasm_data, bool try_coalesce) { struct sk_buff **nextp = (struct sk_buff **)reasm_data; struct rb_node *rbn; struct sk_buff *fp; + int sum_truesize; skb_push(head, head->data - skb_network_header(head)); @@ -487,25 +488,41 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, fp = FRAG_CB(head)->next_frag; rbn = rb_next(&head->rbnode); rb_erase(&head->rbnode, &q->rb_fragments); + + sum_truesize = head->truesize; while (rbn || fp) { /* fp points to the next sk_buff in the current run; * rbn points to the next run. */ /* Go through the current run. */ while (fp) { - *nextp = fp; - nextp = &fp->next; - fp->prev = NULL; - memset(&fp->rbnode, 0, sizeof(fp->rbnode)); - fp->sk = NULL; - head->data_len += fp->len; - head->len += fp->len; + struct sk_buff *next_frag = FRAG_CB(fp)->next_frag; + bool stolen; + int delta; + + sum_truesize += fp->truesize; if (head->ip_summed != fp->ip_summed) head->ip_summed = CHECKSUM_NONE; else if (head->ip_summed == CHECKSUM_COMPLETE) head->csum = csum_add(head->csum, fp->csum); - head->truesize += fp->truesize; - fp = FRAG_CB(fp)->next_frag; + + if (try_coalesce && skb_try_coalesce(head, fp, &stolen, + &delta)) { + kfree_skb_partial(fp, stolen); + } else { + fp->prev = NULL; + memset(&fp->rbnode, 0, sizeof(fp->rbnode)); + fp->sk = NULL; + + head->data_len += fp->len; + head->len += fp->len; + head->truesize += fp->truesize; + + *nextp = fp; + nextp = &fp->next; + } + + fp = next_frag; } /* Move to the next run. */ if (rbn) { @@ -516,7 +533,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, rbn = rbnext; } } - sub_frag_mem_limit(q->fqdir, head->truesize); + sub_frag_mem_limit(q->fqdir, sum_truesize); *nextp = NULL; skb_mark_not_on_list(head); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 4385eb9e781f..cfeb8890f94e 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -393,6 +393,11 @@ err: return err; } +static bool ip_frag_coalesce_ok(const struct ipq *qp) +{ + return qp->q.key.v4.user == IP_DEFRAG_LOCAL_DELIVER; +} + /* Build a new IP datagram from all its fragments. */ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb, struct sk_buff *prev_tail, struct net_device *dev) @@ -421,7 +426,8 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb, if (len > 65535) goto out_oversize; - inet_frag_reasm_finish(&qp->q, skb, reasm_data); + inet_frag_reasm_finish(&qp->q, skb, reasm_data, + ip_frag_coalesce_ok(qp)); skb->dev = dev; IPCB(skb)->frag_max_size = max(qp->max_df_size, qp->q.max_size); diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index 5fe5a3981d43..fc34fd1668d6 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -1151,7 +1151,7 @@ static int nh_create_ipv4(struct net *net, struct nexthop *nh, .fc_encap_type = cfg->nh_encap_type, }; u32 tb_id = l3mdev_fib_table(cfg->dev); - int err = -EINVAL; + int err; err = fib_nh_init(net, fib_nh, &fib_cfg, 1, extack); if (err) { diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 517300d587a7..b6a6f18c3dd1 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2728,7 +2728,8 @@ EXPORT_SYMBOL_GPL(ip_route_output_flow); /* called with rcu_read_lock held */ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, struct rtable *rt, u32 table_id, struct flowi4 *fl4, - struct sk_buff *skb, u32 portid, u32 seq) + struct sk_buff *skb, u32 portid, u32 seq, + unsigned int flags) { struct rtmsg *r; struct nlmsghdr *nlh; @@ -2736,7 +2737,7 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, u32 error; u32 metrics[RTAX_MAX]; - nlh = nlmsg_put(skb, portid, seq, RTM_NEWROUTE, sizeof(*r), 0); + nlh = nlmsg_put(skb, portid, seq, RTM_NEWROUTE, sizeof(*r), flags); if (!nlh) return -EMSGSIZE; @@ -2860,7 +2861,7 @@ nla_put_failure: static int fnhe_dump_bucket(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, u32 table_id, struct fnhe_hash_bucket *bucket, int genid, - int *fa_index, int fa_start) + int *fa_index, int fa_start, unsigned int flags) { int i; @@ -2891,7 +2892,7 @@ static int fnhe_dump_bucket(struct net *net, struct sk_buff *skb, err = rt_fill_info(net, fnhe->fnhe_daddr, 0, rt, table_id, NULL, skb, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq); + cb->nlh->nlmsg_seq, flags); if (err) return err; next: @@ -2904,7 +2905,7 @@ next: int fib_dump_info_fnhe(struct sk_buff *skb, struct netlink_callback *cb, u32 table_id, struct fib_info *fi, - int *fa_index, int fa_start) + int *fa_index, int fa_start, unsigned int flags) { struct net *net = sock_net(cb->skb->sk); int nhsel, genid = fnhe_genid(net); @@ -2922,7 +2923,8 @@ int fib_dump_info_fnhe(struct sk_buff *skb, struct netlink_callback *cb, err = 0; if (bucket) err = fnhe_dump_bucket(net, skb, cb, table_id, bucket, - genid, fa_index, fa_start); + genid, fa_index, fa_start, + flags); rcu_read_unlock(); if (err) return err; @@ -3183,7 +3185,8 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, fl4.flowi4_tos, res.fi, 0); } else { err = rt_fill_info(net, dst, src, rt, table_id, &fl4, skb, - NETLINK_CB(in_skb).portid, nlh->nlmsg_seq); + NETLINK_CB(in_skb).portid, + nlh->nlmsg_seq, 0); } if (err < 0) goto errout_rcu; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f8fa1686f7f3..94df48bcecc2 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -935,6 +935,22 @@ static int tcp_send_mss(struct sock *sk, int *size_goal, int flags) return mss_now; } +/* In some cases, both sendpage() and sendmsg() could have added + * an skb to the write queue, but failed adding payload on it. + * We need to remove it to consume less memory, but more + * importantly be able to generate EPOLLOUT for Edge Trigger epoll() + * users. + */ +static void tcp_remove_empty_skb(struct sock *sk, struct sk_buff *skb) +{ + if (skb && !skb->len) { + tcp_unlink_write_queue(skb, sk); + if (tcp_write_queue_empty(sk)) + tcp_chrono_stop(sk, TCP_CHRONO_BUSY); + sk_wmem_free_skb(sk, skb); + } +} + ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset, size_t size, int flags) { @@ -984,6 +1000,9 @@ new_segment: if (!skb) goto wait_for_memory; +#ifdef CONFIG_TLS_DEVICE + skb->decrypted = !!(flags & MSG_SENDPAGE_DECRYPTED); +#endif skb_entail(sk, skb); copy = size_goal; } @@ -1061,6 +1080,7 @@ out: return copied; do_error: + tcp_remove_empty_skb(sk, tcp_write_queue_tail(sk)); if (copied) goto out; out_err: @@ -1386,18 +1406,11 @@ out_nopush: sock_zerocopy_put(uarg); return copied + copied_syn; +do_error: + skb = tcp_write_queue_tail(sk); do_fault: - if (!skb->len) { - tcp_unlink_write_queue(skb, sk); - /* It is the one place in all of TCP, except connection - * reset, where we can be unlinking the send_head. - */ - if (tcp_write_queue_empty(sk)) - tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - sk_wmem_free_skb(sk, skb); - } + tcp_remove_empty_skb(sk, skb); -do_error: if (copied + copied_syn) goto out; out_err: diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 56be7d27f208..95b59540eee1 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -346,7 +346,7 @@ static void bbr_cwnd_event(struct sock *sk, enum tcp_ca_event event) /* Calculate bdp based on min RTT and the estimated bottleneck bandwidth: * - * bdp = bw * min_rtt * gain + * bdp = ceil(bw * min_rtt * gain) * * The key factor, gain, controls the amount of queue. While a small gain * builds a smaller queue, it becomes more vulnerable to noise in RTT @@ -370,7 +370,9 @@ static u32 bbr_bdp(struct sock *sk, u32 bw, int gain) w = (u64)bw * bbr->min_rtt_us; - /* Apply a gain to the given value, then remove the BW_SCALE shift. */ + /* Apply a gain to the given value, remove the BW_SCALE shift, and + * round the value up to avoid a negative feedback loop. + */ bdp = (((w * gain) >> BBR_SCALE) + BW_UNIT - 1) / BW_UNIT; return bdp; diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 3d1e15401384..8a56e09cfb0e 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -398,10 +398,14 @@ more_data: static int tcp_bpf_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) { struct sk_msg tmp, *msg_tx = NULL; - int flags = msg->msg_flags | MSG_NO_SHARED_FRAGS; int copied = 0, err = 0; struct sk_psock *psock; long timeo; + int flags; + + /* Don't let internal do_tcp_sendpages() flags through */ + flags = (msg->msg_flags & ~MSG_SENDPAGE_DECRYPTED); + flags |= MSG_NO_SHARED_FRAGS; psock = sk_psock_get(sk); if (unlikely(!psock)) diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index a3a386236d93..babc156deabb 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -81,13 +81,42 @@ static int tcp_diag_put_md5sig(struct sk_buff *skb, } #endif +static int tcp_diag_put_ulp(struct sk_buff *skb, struct sock *sk, + const struct tcp_ulp_ops *ulp_ops) +{ + struct nlattr *nest; + int err; + + nest = nla_nest_start_noflag(skb, INET_DIAG_ULP_INFO); + if (!nest) + return -EMSGSIZE; + + err = nla_put_string(skb, INET_ULP_INFO_NAME, ulp_ops->name); + if (err) + goto nla_failure; + + if (ulp_ops->get_info) + err = ulp_ops->get_info(sk, skb); + if (err) + goto nla_failure; + + nla_nest_end(skb, nest); + return 0; + +nla_failure: + nla_nest_cancel(skb, nest); + return err; +} + static int tcp_diag_get_aux(struct sock *sk, bool net_admin, struct sk_buff *skb) { + struct inet_connection_sock *icsk = inet_csk(sk); + int err = 0; + #ifdef CONFIG_TCP_MD5SIG if (net_admin) { struct tcp_md5sig_info *md5sig; - int err = 0; rcu_read_lock(); md5sig = rcu_dereference(tcp_sk(sk)->md5sig_info); @@ -99,11 +128,21 @@ static int tcp_diag_get_aux(struct sock *sk, bool net_admin, } #endif + if (net_admin) { + const struct tcp_ulp_ops *ulp_ops; + + ulp_ops = icsk->icsk_ulp_ops; + if (ulp_ops) + err = tcp_diag_put_ulp(skb, sk, ulp_ops); + if (err) + return err; + } return 0; } static size_t tcp_diag_get_aux_size(struct sock *sk, bool net_admin) { + struct inet_connection_sock *icsk = inet_csk(sk); size_t size = 0; #ifdef CONFIG_TCP_MD5SIG @@ -124,6 +163,17 @@ static size_t tcp_diag_get_aux_size(struct sock *sk, bool net_admin) } #endif + if (net_admin) { + const struct tcp_ulp_ops *ulp_ops; + + ulp_ops = icsk->icsk_ulp_ops; + if (ulp_ops) { + size += nla_total_size(0) + + nla_total_size(TCP_ULP_NAME_MAX); + if (ulp_ops->get_info_size) + size += ulp_ops->get_info_size(sk); + } + } return size; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index e6d02e05bb1c..42abc9bd687a 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1320,6 +1320,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, buff = sk_stream_alloc_skb(sk, nsize, gfp, true); if (!buff) return -ENOMEM; /* We'll just try again later. */ + skb_copy_decrypted(buff, skb); sk->sk_wmem_queued += buff->truesize; sk_mem_charge(sk, buff->truesize); @@ -1874,6 +1875,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, buff = sk_stream_alloc_skb(sk, 0, gfp, true); if (unlikely(!buff)) return -ENOMEM; + skb_copy_decrypted(buff, skb); sk->sk_wmem_queued += buff->truesize; sk_mem_charge(sk, buff->truesize); @@ -2051,7 +2053,7 @@ static bool tcp_can_coalesce_send_queue_head(struct sock *sk, int len) if (len <= skb->len) break; - if (unlikely(TCP_SKB_CB(skb)->eor)) + if (unlikely(TCP_SKB_CB(skb)->eor) || tcp_has_tx_tstamp(skb)) return false; len -= skb->len; @@ -2143,6 +2145,7 @@ static int tcp_mtu_probe(struct sock *sk) sk_mem_charge(sk, nskb->truesize); skb = tcp_send_head(sk); + skb_copy_decrypted(nskb, skb); TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(skb)->seq; TCP_SKB_CB(nskb)->end_seq = TCP_SKB_CB(skb)->seq + probe_size; @@ -2167,6 +2170,7 @@ static int tcp_mtu_probe(struct sock *sk) * we need to propagate it to the new skb. */ TCP_SKB_CB(nskb)->eor = TCP_SKB_CB(skb)->eor; + tcp_skb_collapse_tstamp(nskb, skb); tcp_unlink_write_queue(skb, sk); sk_wmem_free_skb(sk, skb); } else { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index dc73888c7859..6a576ff92c39 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -478,7 +478,7 @@ static struct inet6_dev *ipv6_find_idev(struct net_device *dev) if (!idev) { idev = ipv6_add_dev(dev); if (IS_ERR(idev)) - return NULL; + return idev; } if (dev->flags&IFF_UP) @@ -1045,7 +1045,8 @@ ipv6_add_addr(struct inet6_dev *idev, struct ifa6_config *cfg, int err = 0; if (addr_type == IPV6_ADDR_ANY || - addr_type & IPV6_ADDR_MULTICAST || + (addr_type & IPV6_ADDR_MULTICAST && + !(cfg->ifa_flags & IFA_F_MCAUTOJOIN)) || (!(idev->dev->flags & IFF_LOOPBACK) && !netif_is_l3_master(idev->dev) && addr_type & IPV6_ADDR_LOOPBACK)) @@ -2465,8 +2466,8 @@ static struct inet6_dev *addrconf_add_dev(struct net_device *dev) ASSERT_RTNL(); idev = ipv6_find_idev(dev); - if (!idev) - return ERR_PTR(-ENOBUFS); + if (IS_ERR(idev)) + return idev; if (idev->cnf.disable_ipv6) return ERR_PTR(-EACCES); @@ -3158,7 +3159,7 @@ static void init_loopback(struct net_device *dev) ASSERT_RTNL(); idev = ipv6_find_idev(dev); - if (!idev) { + if (IS_ERR(idev)) { pr_debug("%s: add_dev failed\n", __func__); return; } @@ -3373,7 +3374,7 @@ static void addrconf_sit_config(struct net_device *dev) */ idev = ipv6_find_idev(dev); - if (!idev) { + if (IS_ERR(idev)) { pr_debug("%s: add_dev failed\n", __func__); return; } @@ -3398,7 +3399,7 @@ static void addrconf_gre_config(struct net_device *dev) ASSERT_RTNL(); idev = ipv6_find_idev(dev); - if (!idev) { + if (IS_ERR(idev)) { pr_debug("%s: add_dev failed\n", __func__); return; } @@ -4772,8 +4773,8 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, IFA_F_MCAUTOJOIN | IFA_F_OPTIMISTIC; idev = ipv6_find_idev(dev); - if (!idev) - return -ENOBUFS; + if (IS_ERR(idev)) + return PTR_ERR(idev); if (!ipv6_allow_optimistic_dad(net, idev)) cfg.ifa_flags &= ~IFA_F_OPTIMISTIC; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index fa014d5f1732..d432d0011c16 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -80,8 +80,10 @@ static void ip6_sublist_rcv_finish(struct list_head *head) { struct sk_buff *skb, *next; - list_for_each_entry_safe(skb, next, head, list) + list_for_each_entry_safe(skb, next, head, list) { + skb_list_del_init(skb); dst_input(skb); + } } static void ip6_list_rcv_finish(struct net *net, struct sock *sk, diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 7f3f13c37916..eaa4c2cc2fbb 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -787,14 +787,15 @@ static void mld_del_delrec(struct inet6_dev *idev, struct ifmcaddr6 *im) if (pmc) { im->idev = pmc->idev; if (im->mca_sfmode == MCAST_INCLUDE) { - im->mca_tomb = pmc->mca_tomb; - im->mca_sources = pmc->mca_sources; + swap(im->mca_tomb, pmc->mca_tomb); + swap(im->mca_sources, pmc->mca_sources); for (psf = im->mca_sources; psf; psf = psf->sf_next) psf->sf_crcount = idev->mc_qrv; } else { im->mca_crcount = idev->mc_qrv; } in6_dev_put(pmc->idev); + ip6_mc_clear_src(pmc); kfree(pmc); } spin_unlock_bh(&im->mca_lock); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 0f82c150543b..fed9666a2f7d 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -348,7 +348,7 @@ static int nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *skb, skb_reset_transport_header(skb); - inet_frag_reasm_finish(&fq->q, skb, reasm_data); + inet_frag_reasm_finish(&fq->q, skb, reasm_data, false); skb->ignore_df = 1; skb->dev = dev; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index ca05b16f1bb9..1f5d4d196dcc 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -282,7 +282,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *skb, skb_reset_transport_header(skb); - inet_frag_reasm_finish(&fq->q, skb, reasm_data); + inet_frag_reasm_finish(&fq->q, skb, reasm_data, true); skb->dev = dev; ipv6_hdr(skb)->payload_len = htons(payload_len); diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index ed56b0c6fe19..7c6edb7c5f10 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1549,6 +1549,11 @@ static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev, if (!is_valid_ether_addr(mac)) return -EINVAL; + if (params->sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER) && + sdata->vif.type == NL80211_IFTYPE_STATION && + !sdata->u.mgd.associated) + return -EINVAL; + sta = sta_info_alloc(sdata, mac, GFP_KERNEL); if (!sta) return -ENOMEM; @@ -1556,10 +1561,6 @@ static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev, if (params->sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)) sta->sta.tdls = true; - if (sta->sta.tdls && sdata->vif.type == NL80211_IFTYPE_STATION && - !sdata->u.mgd.associated) - return -EINVAL; - err = sta_apply_parameters(local, sta, params); if (err) { sta_info_free(local, sta); diff --git a/net/mac80211/he.c b/net/mac80211/he.c index a02abfc424aa..736da0035135 100644 --- a/net/mac80211/he.c +++ b/net/mac80211/he.c @@ -72,12 +72,13 @@ ieee80211_he_spr_ie_to_bss_conf(struct ieee80211_vif *vif, { struct ieee80211_he_obss_pd *he_obss_pd = &vif->bss_conf.he_obss_pd; - const u8 *data = he_spr_ie_elem->optional; + const u8 *data; memset(he_obss_pd, 0, sizeof(*he_obss_pd)); if (!he_spr_ie_elem) return; + data = he_spr_ie_elem->optional; if (he_spr_ie_elem->he_sr_control & IEEE80211_HE_SPR_NON_SRG_OFFSET_PRESENT) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 641876982ab9..6471f552a942 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -158,10 +158,10 @@ ieee80211_determine_chantype(struct ieee80211_sub_if_data *sdata, memcpy(&sta_ht_cap, &sband->ht_cap, sizeof(sta_ht_cap)); ieee80211_apply_htcap_overrides(sdata, &sta_ht_cap); + memset(chandef, 0, sizeof(struct cfg80211_chan_def)); chandef->chan = channel; chandef->width = NL80211_CHAN_WIDTH_20_NOHT; chandef->center_freq1 = channel->center_freq; - chandef->center_freq2 = 0; if (!ht_oper || !sta_ht_cap.ht_supported) { ret = IEEE80211_STA_DISABLE_HT | IEEE80211_STA_DISABLE_VHT; diff --git a/net/mac80211/rc80211_minstrel.h b/net/mac80211/rc80211_minstrel.h index 3c96a853adbd..51d8b2c846e7 100644 --- a/net/mac80211/rc80211_minstrel.h +++ b/net/mac80211/rc80211_minstrel.h @@ -95,6 +95,7 @@ struct minstrel_sta_info { struct minstrel_priv { struct ieee80211_hw *hw; bool has_mrr; + u32 sample_switch; unsigned int cw_min; unsigned int cw_max; unsigned int max_retry; diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c index 5a882da82f0e..a01168514840 100644 --- a/net/mac80211/rc80211_minstrel_ht.c +++ b/net/mac80211/rc80211_minstrel_ht.c @@ -18,6 +18,8 @@ #define AVG_AMPDU_SIZE 16 #define AVG_PKT_SIZE 1200 +#define SAMPLE_SWITCH_THR 100 + /* Number of bits for an average sized packet */ #define MCS_NBITS ((AVG_PKT_SIZE * AVG_AMPDU_SIZE) << 3) @@ -58,6 +60,7 @@ [GROUP_IDX(_streams, _sgi, _ht40)] = { \ .streams = _streams, \ .shift = _s, \ + .bw = _ht40, \ .flags = \ IEEE80211_TX_RC_MCS | \ (_sgi ? IEEE80211_TX_RC_SHORT_GI : 0) | \ @@ -94,6 +97,7 @@ [VHT_GROUP_IDX(_streams, _sgi, _bw)] = { \ .streams = _streams, \ .shift = _s, \ + .bw = _bw, \ .flags = \ IEEE80211_TX_RC_VHT_MCS | \ (_sgi ? IEEE80211_TX_RC_SHORT_GI : 0) | \ @@ -486,7 +490,7 @@ minstrel_ht_assign_best_tp_rates(struct minstrel_ht_sta *mi, tmp_prob = mi->groups[tmp_group].rates[tmp_idx].prob_ewma; tmp_mcs_tp = minstrel_ht_get_tp_avg(mi, tmp_group, tmp_idx, tmp_prob); - if (tmp_cck_tp > tmp_mcs_tp) { + if (tmp_cck_tp_rate && tmp_cck_tp > tmp_mcs_tp) { for(i = 0; i < MAX_THR_RATES; i++) { minstrel_ht_sort_best_tp_rates(mi, tmp_cck_tp_rate[i], tmp_mcs_tp_rate); @@ -526,6 +530,133 @@ minstrel_ht_prob_rate_reduce_streams(struct minstrel_ht_sta *mi) } } +static inline int +minstrel_get_duration(int index) +{ + const struct mcs_group *group = &minstrel_mcs_groups[index / MCS_GROUP_RATES]; + unsigned int duration = group->duration[index % MCS_GROUP_RATES]; + return duration << group->shift; +} + +static bool +minstrel_ht_probe_group(struct minstrel_ht_sta *mi, const struct mcs_group *tp_group, + int tp_idx, const struct mcs_group *group) +{ + if (group->bw < tp_group->bw) + return false; + + if (group->streams == tp_group->streams) + return true; + + if (tp_idx < 4 && group->streams == tp_group->streams - 1) + return true; + + return group->streams == tp_group->streams + 1; +} + +static void +minstrel_ht_find_probe_rates(struct minstrel_ht_sta *mi, u16 *rates, int *n_rates, + bool faster_rate) +{ + const struct mcs_group *group, *tp_group; + int i, g, max_dur; + int tp_idx; + + tp_group = &minstrel_mcs_groups[mi->max_tp_rate[0] / MCS_GROUP_RATES]; + tp_idx = mi->max_tp_rate[0] % MCS_GROUP_RATES; + + max_dur = minstrel_get_duration(mi->max_tp_rate[0]); + if (faster_rate) + max_dur -= max_dur / 16; + + for (g = 0; g < MINSTREL_GROUPS_NB; g++) { + u16 supported = mi->supported[g]; + + if (!supported) + continue; + + group = &minstrel_mcs_groups[g]; + if (!minstrel_ht_probe_group(mi, tp_group, tp_idx, group)) + continue; + + for (i = 0; supported; supported >>= 1, i++) { + int idx; + + if (!(supported & 1)) + continue; + + if ((group->duration[i] << group->shift) > max_dur) + continue; + + idx = g * MCS_GROUP_RATES + i; + if (idx == mi->max_tp_rate[0]) + continue; + + rates[(*n_rates)++] = idx; + break; + } + } +} + +static void +minstrel_ht_rate_sample_switch(struct minstrel_priv *mp, + struct minstrel_ht_sta *mi) +{ + struct minstrel_rate_stats *mrs; + u16 rates[MINSTREL_GROUPS_NB]; + int n_rates = 0; + int probe_rate = 0; + bool faster_rate; + int i; + u8 random; + + /* + * Use rate switching instead of probing packets for devices with + * little control over retry fallback behavior + */ + if (mp->hw->max_rates > 1) + return; + + /* + * If the current EWMA prob is >75%, look for a rate that's 6.25% + * faster than the max tp rate. + * If that fails, look again for a rate that is at least as fast + */ + mrs = minstrel_get_ratestats(mi, mi->max_tp_rate[0]); + faster_rate = mrs->prob_ewma > MINSTREL_FRAC(75, 100); + minstrel_ht_find_probe_rates(mi, rates, &n_rates, faster_rate); + if (!n_rates && faster_rate) + minstrel_ht_find_probe_rates(mi, rates, &n_rates, false); + + /* If no suitable rate was found, try to pick the next one in the group */ + if (!n_rates) { + int g_idx = mi->max_tp_rate[0] / MCS_GROUP_RATES; + u16 supported = mi->supported[g_idx]; + + supported >>= mi->max_tp_rate[0] % MCS_GROUP_RATES; + for (i = 0; supported; i++) { + if (!(supported & 1)) + continue; + + probe_rate = mi->max_tp_rate[0] + i; + goto out; + } + + return; + } + + i = 0; + if (n_rates > 1) { + random = prandom_u32(); + i = random % n_rates; + } + probe_rate = rates[i]; + +out: + mi->sample_rate = probe_rate; + mi->sample_mode = MINSTREL_SAMPLE_ACTIVE; +} + /* * Update rate statistics and select new primary rates * @@ -536,7 +667,8 @@ minstrel_ht_prob_rate_reduce_streams(struct minstrel_ht_sta *mi) * higher throughput rates, even if the probablity is a bit lower */ static void -minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) +minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, + bool sample) { struct minstrel_mcs_group_data *mg; struct minstrel_rate_stats *mrs; @@ -544,6 +676,18 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) u16 tmp_mcs_tp_rate[MAX_THR_RATES], tmp_group_tp_rate[MAX_THR_RATES]; u16 tmp_cck_tp_rate[MAX_THR_RATES], index; + mi->sample_mode = MINSTREL_SAMPLE_IDLE; + + if (sample) { + mi->total_packets_cur = mi->total_packets - + mi->total_packets_last; + mi->total_packets_last = mi->total_packets; + } + if (!mp->sample_switch) + sample = false; + if (mi->total_packets_cur < SAMPLE_SWITCH_THR && mp->sample_switch != 1) + sample = false; + if (mi->ampdu_packets > 0) { if (!ieee80211_hw_check(mp->hw, TX_STATUS_NO_AMPDU_LEN)) mi->avg_ampdu_len = minstrel_ewma(mi->avg_ampdu_len, @@ -558,11 +702,19 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) mi->sample_slow = 0; mi->sample_count = 0; - /* Initialize global rate indexes */ - for(j = 0; j < MAX_THR_RATES; j++){ - tmp_mcs_tp_rate[j] = 0; - tmp_cck_tp_rate[j] = 0; - } + memset(tmp_mcs_tp_rate, 0, sizeof(tmp_mcs_tp_rate)); + memset(tmp_cck_tp_rate, 0, sizeof(tmp_cck_tp_rate)); + if (mi->supported[MINSTREL_CCK_GROUP]) + for (j = 0; j < ARRAY_SIZE(tmp_cck_tp_rate); j++) + tmp_cck_tp_rate[j] = MINSTREL_CCK_GROUP * MCS_GROUP_RATES; + + if (mi->supported[MINSTREL_VHT_GROUP_0]) + index = MINSTREL_VHT_GROUP_0 * MCS_GROUP_RATES; + else + index = MINSTREL_HT_GROUP_0 * MCS_GROUP_RATES; + + for (j = 0; j < ARRAY_SIZE(tmp_mcs_tp_rate); j++) + tmp_mcs_tp_rate[j] = index; /* Find best rate sets within all MCS groups*/ for (group = 0; group < ARRAY_SIZE(minstrel_mcs_groups); group++) { @@ -575,7 +727,7 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) /* (re)Initialize group rate indexes */ for(j = 0; j < MAX_THR_RATES; j++) - tmp_group_tp_rate[j] = group; + tmp_group_tp_rate[j] = MCS_GROUP_RATES * group; for (i = 0; i < MCS_GROUP_RATES; i++) { if (!(mi->supported[group] & BIT(i))) @@ -622,12 +774,16 @@ minstrel_ht_update_stats(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) /* try to sample all available rates during each interval */ mi->sample_count *= 8; + if (sample) + minstrel_ht_rate_sample_switch(mp, mi); + #ifdef CONFIG_MAC80211_DEBUGFS /* use fixed index if set */ if (mp->fixed_rate_idx != -1) { for (i = 0; i < 4; i++) mi->max_tp_rate[i] = mp->fixed_rate_idx; mi->max_prob_rate = mp->fixed_rate_idx; + mi->sample_mode = MINSTREL_SAMPLE_IDLE; } #endif @@ -731,15 +887,17 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, struct minstrel_ht_sta_priv *msp = priv_sta; struct minstrel_ht_sta *mi = &msp->ht; struct ieee80211_tx_rate *ar = info->status.rates; - struct minstrel_rate_stats *rate, *rate2; + struct minstrel_rate_stats *rate, *rate2, *rate_sample = NULL; struct minstrel_priv *mp = priv; bool last, update = false; + bool sample_status = false; int i; if (!msp->is_ht) return mac80211_minstrel.tx_status_ext(priv, sband, &msp->legacy, st); + /* This packet was aggregated but doesn't carry status info */ if ((info->flags & IEEE80211_TX_CTL_AMPDU) && !(info->flags & IEEE80211_TX_STAT_AMPDU)) @@ -765,12 +923,17 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, if (info->flags & IEEE80211_TX_CTL_RATE_CTRL_PROBE) mi->sample_packets += info->status.ampdu_len; + if (mi->sample_mode != MINSTREL_SAMPLE_IDLE) + rate_sample = minstrel_get_ratestats(mi, mi->sample_rate); + last = !minstrel_ht_txstat_valid(mp, &ar[0]); for (i = 0; !last; i++) { last = (i == IEEE80211_TX_MAX_RATES - 1) || !minstrel_ht_txstat_valid(mp, &ar[i + 1]); rate = minstrel_ht_get_stats(mp, mi, &ar[i]); + if (rate == rate_sample) + sample_status = true; if (last) rate->success += info->status.ampdu_ack_len; @@ -778,44 +941,60 @@ minstrel_ht_tx_status(void *priv, struct ieee80211_supported_band *sband, rate->attempts += ar[i].count * info->status.ampdu_len; } - /* - * check for sudden death of spatial multiplexing, - * downgrade to a lower number of streams if necessary. - */ - rate = minstrel_get_ratestats(mi, mi->max_tp_rate[0]); - if (rate->attempts > 30 && - MINSTREL_FRAC(rate->success, rate->attempts) < - MINSTREL_FRAC(20, 100)) { - minstrel_downgrade_rate(mi, &mi->max_tp_rate[0], true); + switch (mi->sample_mode) { + case MINSTREL_SAMPLE_IDLE: + break; + + case MINSTREL_SAMPLE_ACTIVE: + if (!sample_status) + break; + + mi->sample_mode = MINSTREL_SAMPLE_PENDING; update = true; - } + break; + + case MINSTREL_SAMPLE_PENDING: + if (sample_status) + break; - rate2 = minstrel_get_ratestats(mi, mi->max_tp_rate[1]); - if (rate2->attempts > 30 && - MINSTREL_FRAC(rate2->success, rate2->attempts) < - MINSTREL_FRAC(20, 100)) { - minstrel_downgrade_rate(mi, &mi->max_tp_rate[1], false); update = true; + minstrel_ht_update_stats(mp, mi, false); + break; + } + + + if (mp->hw->max_rates > 1) { + /* + * check for sudden death of spatial multiplexing, + * downgrade to a lower number of streams if necessary. + */ + rate = minstrel_get_ratestats(mi, mi->max_tp_rate[0]); + if (rate->attempts > 30 && + MINSTREL_FRAC(rate->success, rate->attempts) < + MINSTREL_FRAC(20, 100)) { + minstrel_downgrade_rate(mi, &mi->max_tp_rate[0], true); + update = true; + } + + rate2 = minstrel_get_ratestats(mi, mi->max_tp_rate[1]); + if (rate2->attempts > 30 && + MINSTREL_FRAC(rate2->success, rate2->attempts) < + MINSTREL_FRAC(20, 100)) { + minstrel_downgrade_rate(mi, &mi->max_tp_rate[1], false); + update = true; + } } if (time_after(jiffies, mi->last_stats_update + (mp->update_interval / 2 * HZ) / 1000)) { update = true; - minstrel_ht_update_stats(mp, mi); + minstrel_ht_update_stats(mp, mi, true); } if (update) minstrel_ht_update_rates(mp, mi); } -static inline int -minstrel_get_duration(int index) -{ - const struct mcs_group *group = &minstrel_mcs_groups[index / MCS_GROUP_RATES]; - unsigned int duration = group->duration[index % MCS_GROUP_RATES]; - return duration << group->shift; -} - static void minstrel_calc_retransmit(struct minstrel_priv *mp, struct minstrel_ht_sta *mi, int index) @@ -980,14 +1159,18 @@ static void minstrel_ht_update_rates(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) { struct ieee80211_sta_rates *rates; + u16 first_rate = mi->max_tp_rate[0]; int i = 0; + if (mi->sample_mode == MINSTREL_SAMPLE_ACTIVE) + first_rate = mi->sample_rate; + rates = kzalloc(sizeof(*rates), GFP_ATOMIC); if (!rates) return; /* Start with max_tp_rate[0] */ - minstrel_ht_set_rate(mp, mi, rates, i++, mi->max_tp_rate[0]); + minstrel_ht_set_rate(mp, mi, rates, i++, first_rate); if (mp->hw->max_rates >= 3) { /* At least 3 tx rates supported, use max_tp_rate[1] next */ @@ -1012,6 +1195,11 @@ minstrel_get_sample_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) int tp_rate1, tp_rate2; int sample_idx = 0; + if (mp->hw->max_rates == 1 && mp->sample_switch && + (mi->total_packets_cur >= SAMPLE_SWITCH_THR || + mp->sample_switch == 1)) + return -1; + if (mi->sample_wait > 0) { mi->sample_wait--; return -1; @@ -1059,6 +1247,21 @@ minstrel_get_sample_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi) minstrel_get_duration(mi->max_prob_rate) * 3 < sample_dur) return -1; + + /* + * For devices with no configurable multi-rate retry, skip sampling + * below the per-group max throughput rate, and only use one sampling + * attempt per rate + */ + if (mp->hw->max_rates == 1 && + (minstrel_get_duration(mg->max_group_tp_rate[0]) < sample_dur || + mrs->attempts)) + return -1; + + /* Skip already sampled slow rates */ + if (sample_dur >= minstrel_get_duration(tp_rate1) && mrs->attempts) + return -1; + /* * Make sure that lower rates get sampled only occasionally, * if the link is working perfectly. @@ -1318,7 +1521,7 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, mi->supported[MINSTREL_CCK_GROUP] |= mi->cck_supported_short << 4; /* create an initial rate table with the lowest supported rates */ - minstrel_ht_update_stats(mp, mi); + minstrel_ht_update_stats(mp, mi, true); minstrel_ht_update_rates(mp, mi); return; @@ -1436,6 +1639,8 @@ minstrel_ht_alloc(struct ieee80211_hw *hw, struct dentry *debugfsdir) if (!mp) return NULL; + mp->sample_switch = -1; + /* contention window settings * Just an approximation. Using the per-queue values would complicate * the calculations and is probably unnecessary */ @@ -1467,6 +1672,8 @@ minstrel_ht_alloc(struct ieee80211_hw *hw, struct dentry *debugfsdir) mp->fixed_rate_idx = (u32) -1; debugfs_create_u32("fixed_rate_idx", S_IRUGO | S_IWUGO, debugfsdir, &mp->fixed_rate_idx); + debugfs_create_u32("sample_switch", S_IRUGO | S_IWUSR, debugfsdir, + &mp->sample_switch); #endif minstrel_ht_init_cck_rates(mp); diff --git a/net/mac80211/rc80211_minstrel_ht.h b/net/mac80211/rc80211_minstrel_ht.h index 80296268c778..f938701e7ab7 100644 --- a/net/mac80211/rc80211_minstrel_ht.h +++ b/net/mac80211/rc80211_minstrel_ht.h @@ -33,6 +33,7 @@ struct mcs_group { u16 flags; u8 streams; u8 shift; + u8 bw; u16 duration[MCS_GROUP_RATES]; }; @@ -50,6 +51,12 @@ struct minstrel_mcs_group_data { struct minstrel_rate_stats rates[MCS_GROUP_RATES]; }; +enum minstrel_sample_mode { + MINSTREL_SAMPLE_IDLE, + MINSTREL_SAMPLE_ACTIVE, + MINSTREL_SAMPLE_PENDING, +}; + struct minstrel_ht_sta { struct ieee80211_sta *sta; @@ -71,6 +78,8 @@ struct minstrel_ht_sta { unsigned int overhead; unsigned int overhead_rtscts; + unsigned int total_packets_last; + unsigned int total_packets_cur; unsigned int total_packets; unsigned int sample_packets; @@ -82,6 +91,9 @@ struct minstrel_ht_sta { u8 sample_count; u8 sample_slow; + enum minstrel_sample_mode sample_mode; + u16 sample_rate; + /* current MCS group to be sampled */ u8 sample_group; diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 3c1ab870fefe..768d14c9a716 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2447,11 +2447,13 @@ static void ieee80211_deliver_skb_to_local_stack(struct sk_buff *skb, skb->protocol == cpu_to_be16(ETH_P_PREAUTH)) && sdata->control_port_over_nl80211)) { struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); - bool noencrypt = status->flag & RX_FLAG_DECRYPTED; + bool noencrypt = !(status->flag & RX_FLAG_DECRYPTED); cfg80211_rx_control_port(dev, skb, noencrypt); dev_kfree_skb(skb); } else { + memset(skb->cb, 0, sizeof(skb->cb)); + /* deliver to local stack */ if (rx->napi) napi_gro_receive(rx->napi, skb); @@ -2546,8 +2548,6 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx) if (skb) { skb->protocol = eth_type_trans(skb, dev); - memset(skb->cb, 0, sizeof(skb->cb)); - ieee80211_deliver_skb_to_local_stack(skb, rx); } diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index fb6614f57cbc..df553070206c 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -1961,6 +1961,7 @@ int sta_info_move_state(struct sta_info *sta, case IEEE80211_STA_ASSOC: if (sta->sta_state == IEEE80211_STA_AUTH) { set_bit(WLAN_STA_ASSOC, &sta->_flags); + sta->assoc_at = ktime_get_boottime_ns(); ieee80211_recalc_min_chandef(sta->sdata); if (!sta->sta.support_p2p_ps) ieee80211_recalc_p2p_go_ps_allowed(sta->sdata); @@ -2190,6 +2191,7 @@ void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo, BIT_ULL(NL80211_STA_INFO_STA_FLAGS) | BIT_ULL(NL80211_STA_INFO_BSS_PARAM) | BIT_ULL(NL80211_STA_INFO_CONNECTED_TIME) | + BIT_ULL(NL80211_STA_INFO_ASSOC_AT_BOOTTIME) | BIT_ULL(NL80211_STA_INFO_RX_DROP_MISC); if (sdata->vif.type == NL80211_IFTYPE_STATION) { @@ -2198,6 +2200,7 @@ void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo, } sinfo->connected_time = ktime_get_seconds() - sta->last_connected; + sinfo->assoc_at = sta->assoc_at; sinfo->inactive_time = jiffies_to_msecs(jiffies - ieee80211_sta_last_active(sta)); diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 3260d4234920..369c2dddce52 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -466,6 +466,7 @@ struct ieee80211_sta_rx_stats { * the station when it leaves powersave or polls for frames * @driver_buffered_tids: bitmap of TIDs the driver has data buffered on * @txq_buffered_tids: bitmap of TIDs that mac80211 has txq data buffered on + * @assoc_at: clock boottime (in ns) of last association * @last_connected: time (in seconds) when a station got connected * @last_seq_ctrl: last received seq/frag number from this STA (per TID * plus one for non-QoS frames) @@ -562,6 +563,7 @@ struct sta_info { unsigned long driver_buffered_tids; unsigned long txq_buffered_tids; + u64 assoc_at; long last_connected; /* Updated from RX path only, no locking requirements */ diff --git a/net/mac80211/status.c b/net/mac80211/status.c index f03aa8924d23..ab8ba5835ca0 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -260,9 +260,16 @@ static int ieee80211_tx_radiotap_len(struct ieee80211_tx_info *info, int len = sizeof(struct ieee80211_radiotap_header); /* IEEE80211_RADIOTAP_RATE rate */ - if (info->status.rates[0].idx >= 0 && - !(info->status.rates[0].flags & (IEEE80211_TX_RC_MCS | - IEEE80211_TX_RC_VHT_MCS))) + if (status && status->rate && !(status->rate->flags & + (RATE_INFO_FLAGS_MCS | + RATE_INFO_FLAGS_DMG | + RATE_INFO_FLAGS_EDMG | + RATE_INFO_FLAGS_VHT_MCS | + RATE_INFO_FLAGS_HE_MCS))) + len += 2; + else if (info->status.rates[0].idx >= 0 && + !(info->status.rates[0].flags & + (IEEE80211_TX_RC_MCS | IEEE80211_TX_RC_VHT_MCS))) len += 2; /* IEEE80211_RADIOTAP_TX_FLAGS */ @@ -321,13 +328,14 @@ ieee80211_add_tx_radiotap_header(struct ieee80211_local *local, /* IEEE80211_RADIOTAP_RATE */ - if (status && status->rate && !(status->rate->flags & - (RATE_INFO_FLAGS_MCS | - RATE_INFO_FLAGS_60G | - RATE_INFO_FLAGS_VHT_MCS | - RATE_INFO_FLAGS_HE_MCS))) - legacy_rate = status->rate->legacy; - else if (info->status.rates[0].idx >= 0 && + if (status && status->rate) { + if (!(status->rate->flags & (RATE_INFO_FLAGS_MCS | + RATE_INFO_FLAGS_DMG | + RATE_INFO_FLAGS_EDMG | + RATE_INFO_FLAGS_VHT_MCS | + RATE_INFO_FLAGS_HE_MCS))) + legacy_rate = status->rate->legacy; + } else if (info->status.rates[0].idx >= 0 && !(info->status.rates[0].flags & (IEEE80211_TX_RC_MCS | IEEE80211_TX_RC_VHT_MCS))) legacy_rate = @@ -397,7 +405,7 @@ ieee80211_add_tx_radiotap_header(struct ieee80211_local *local, *pos = 11; break; case RATE_INFO_BW_80: - *pos = 2; + *pos = 4; break; case RATE_INFO_BW_40: *pos = 1; @@ -406,6 +414,7 @@ ieee80211_add_tx_radiotap_header(struct ieee80211_local *local, *pos = 0; break; } + pos++; /* u8 mcs_nss[4] */ *pos = (status->rate->mcs << 4) | status->rate->nss; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 235c6377a203..1fa422782905 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -162,6 +162,7 @@ static __le16 ieee80211_duration(struct ieee80211_tx_data *tx, break; } case NL80211_BAND_5GHZ: + case NL80211_BAND_6GHZ: if (r->flags & IEEE80211_RATE_MANDATORY_A) mrate = r->bitrate; break; diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index d25e91d7bdc1..44b675016393 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -133,12 +133,12 @@ static int mpls_xmit(struct sk_buff *skb) mpls_stats_inc_outucastpkts(out_dev, skb); if (rt) { - if (rt->rt_gw_family == AF_INET) - err = neigh_xmit(NEIGH_ARP_TABLE, out_dev, &rt->rt_gw4, - skb); - else if (rt->rt_gw_family == AF_INET6) + if (rt->rt_gw_family == AF_INET6) err = neigh_xmit(NEIGH_ND_TABLE, out_dev, &rt->rt_gw6, skb); + else + err = neigh_xmit(NEIGH_ARP_TABLE, out_dev, &rt->rt_gw4, + skb); } else if (rt6) { if (ipv6_addr_v4mapped(&rt6->rt6i_gateway)) { /* 6PE (RFC 4798) */ diff --git a/net/ncsi/ncsi-cmd.c b/net/ncsi/ncsi-cmd.c index 5c3fad8cba57..0187e65176c0 100644 --- a/net/ncsi/ncsi-cmd.c +++ b/net/ncsi/ncsi-cmd.c @@ -54,7 +54,7 @@ static void ncsi_cmd_build_header(struct ncsi_pkt_hdr *h, checksum = ncsi_calculate_checksum((unsigned char *)h, sizeof(*h) + nca->payload); pchecksum = (__be32 *)((void *)h + sizeof(struct ncsi_pkt_hdr) + - nca->payload); + ALIGN(nca->payload, 4)); *pchecksum = htonl(checksum); } @@ -309,14 +309,21 @@ static struct ncsi_request *ncsi_alloc_command(struct ncsi_cmd_arg *nca) int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca) { + struct ncsi_cmd_handler *nch = NULL; struct ncsi_request *nr; + unsigned char type; struct ethhdr *eh; - struct ncsi_cmd_handler *nch = NULL; int i, ret; + /* Use OEM generic handler for Netlink request */ + if (nca->req_flags == NCSI_REQ_FLAG_NETLINK_DRIVEN) + type = NCSI_PKT_CMD_OEM; + else + type = nca->type; + /* Search for the handler */ for (i = 0; i < ARRAY_SIZE(ncsi_cmd_handlers); i++) { - if (ncsi_cmd_handlers[i].type == nca->type) { + if (ncsi_cmd_handlers[i].type == type) { if (ncsi_cmd_handlers[i].handler) nch = &ncsi_cmd_handlers[i]; else diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h index a8e9def593f2..80938b338fee 100644 --- a/net/ncsi/ncsi-pkt.h +++ b/net/ncsi/ncsi-pkt.h @@ -387,6 +387,9 @@ struct ncsi_aen_hncdsc_pkt { #define NCSI_PKT_CMD_OEM 0x50 /* OEM */ #define NCSI_PKT_CMD_PLDM 0x51 /* PLDM request over NCSI over RBT */ #define NCSI_PKT_CMD_GPUUID 0x52 /* Get package UUID */ +#define NCSI_PKT_CMD_QPNPR 0x56 /* Query Pending NC PLDM request */ +#define NCSI_PKT_CMD_SNPR 0x57 /* Send NC PLDM Reply */ + /* NCSI packet responses */ #define NCSI_PKT_RSP_CIS (NCSI_PKT_CMD_CIS + 0x80) @@ -419,6 +422,8 @@ struct ncsi_aen_hncdsc_pkt { #define NCSI_PKT_RSP_OEM (NCSI_PKT_CMD_OEM + 0x80) #define NCSI_PKT_RSP_PLDM (NCSI_PKT_CMD_PLDM + 0x80) #define NCSI_PKT_RSP_GPUUID (NCSI_PKT_CMD_GPUUID + 0x80) +#define NCSI_PKT_RSP_QPNPR (NCSI_PKT_CMD_QPNPR + 0x80) +#define NCSI_PKT_RSP_SNPR (NCSI_PKT_CMD_SNPR + 0x80) /* NCSI response code/reason */ #define NCSI_PKT_RSP_C_COMPLETED 0x0000 /* Command Completed */ diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c index 7581bf919885..d5611f04926d 100644 --- a/net/ncsi/ncsi-rsp.c +++ b/net/ncsi/ncsi-rsp.c @@ -47,7 +47,8 @@ static int ncsi_validate_rsp_pkt(struct ncsi_request *nr, if (ntohs(h->code) != NCSI_PKT_RSP_C_COMPLETED || ntohs(h->reason) != NCSI_PKT_RSP_R_NO_ERROR) { netdev_dbg(nr->ndp->ndev.dev, - "NCSI: non zero response/reason code\n"); + "NCSI: non zero response/reason code %04xh, %04xh\n", + ntohs(h->code), ntohs(h->reason)); return -EPERM; } @@ -55,7 +56,7 @@ static int ncsi_validate_rsp_pkt(struct ncsi_request *nr, * sender doesn't support checksum according to NCSI * specification. */ - pchecksum = (__be32 *)((void *)(h + 1) + payload - 4); + pchecksum = (__be32 *)((void *)(h + 1) + ALIGN(payload, 4) - 4); if (ntohl(*pchecksum) == 0) return 0; @@ -63,7 +64,9 @@ static int ncsi_validate_rsp_pkt(struct ncsi_request *nr, sizeof(*h) + payload - 4); if (*pchecksum != htonl(checksum)) { - netdev_dbg(nr->ndp->ndev.dev, "NCSI: checksum mismatched\n"); + netdev_dbg(nr->ndp->ndev.dev, + "NCSI: checksum mismatched; recd: %08x calc: %08x\n", + *pchecksum, htonl(checksum)); return -EINVAL; } @@ -1035,6 +1038,11 @@ static int ncsi_rsp_handler_gpuuid(struct ncsi_request *nr) return 0; } +static int ncsi_rsp_handler_pldm(struct ncsi_request *nr) +{ + return 0; +} + static int ncsi_rsp_handler_netlink(struct ncsi_request *nr) { struct ncsi_dev_priv *ndp = nr->ndp; @@ -1083,13 +1091,15 @@ static struct ncsi_rsp_handler { { NCSI_PKT_RSP_GVI, 40, ncsi_rsp_handler_gvi }, { NCSI_PKT_RSP_GC, 32, ncsi_rsp_handler_gc }, { NCSI_PKT_RSP_GP, -1, ncsi_rsp_handler_gp }, - { NCSI_PKT_RSP_GCPS, 172, ncsi_rsp_handler_gcps }, - { NCSI_PKT_RSP_GNS, 172, ncsi_rsp_handler_gns }, - { NCSI_PKT_RSP_GNPTS, 172, ncsi_rsp_handler_gnpts }, + { NCSI_PKT_RSP_GCPS, 204, ncsi_rsp_handler_gcps }, + { NCSI_PKT_RSP_GNS, 32, ncsi_rsp_handler_gns }, + { NCSI_PKT_RSP_GNPTS, 48, ncsi_rsp_handler_gnpts }, { NCSI_PKT_RSP_GPS, 8, ncsi_rsp_handler_gps }, { NCSI_PKT_RSP_OEM, -1, ncsi_rsp_handler_oem }, - { NCSI_PKT_RSP_PLDM, 0, NULL }, - { NCSI_PKT_RSP_GPUUID, 20, ncsi_rsp_handler_gpuuid } + { NCSI_PKT_RSP_PLDM, -1, ncsi_rsp_handler_pldm }, + { NCSI_PKT_RSP_GPUUID, 20, ncsi_rsp_handler_gpuuid }, + { NCSI_PKT_RSP_QPNPR, -1, ncsi_rsp_handler_pldm }, + { NCSI_PKT_RSP_SNPR, -1, ncsi_rsp_handler_pldm } }; int ncsi_rcv_rsp(struct sk_buff *skb, struct net_device *dev, diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index a542761e90d1..81a8ef42b88d 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -453,13 +453,12 @@ EXPORT_SYMBOL_GPL(nf_ct_invert_tuple); * table location, we assume id gets exposed to userspace. * * Following nf_conn items do not change throughout lifetime - * of the nf_conn after it has been committed to main hash table: + * of the nf_conn: * * 1. nf_conn address - * 2. nf_conn->ext address - * 3. nf_conn->master address (normally NULL) - * 4. tuple - * 5. the associated net namespace + * 2. nf_conn->master address (normally NULL) + * 3. the associated net namespace + * 4. the original direction tuple */ u32 nf_ct_get_id(const struct nf_conn *ct) { @@ -469,9 +468,10 @@ u32 nf_ct_get_id(const struct nf_conn *ct) net_get_random_once(&ct_id_seed, sizeof(ct_id_seed)); a = (unsigned long)ct; - b = (unsigned long)ct->master ^ net_hash_mix(nf_ct_net(ct)); - c = (unsigned long)ct->ext; - d = (unsigned long)siphash(&ct->tuplehash, sizeof(ct->tuplehash), + b = (unsigned long)ct->master; + c = (unsigned long)nf_ct_net(ct); + d = (unsigned long)siphash(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + sizeof(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple), &ct_id_seed); #ifdef CONFIG_64BIT return siphash_4u64((u64)a, (u64)b, (u64)c, (u64)d, &ct_id_seed); diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index c57d2348c505..9eca90414bb7 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -322,7 +322,7 @@ static int find_pattern(const char *data, size_t dlen, i++; } - pr_debug("Skipped up to `%c'!\n", skip); + pr_debug("Skipped up to 0x%hhx delimiter!\n", skip); *numoff = i; *numlen = getnum(data + i, dlen - i, cmd, term, numoff); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index d97f4ea47cf3..88d4127df863 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -1035,9 +1035,14 @@ static int nf_conntrack_standalone_init_sysctl(struct net *net) table[NF_SYSCTL_CT_COUNT].data = &net->ct.count; table[NF_SYSCTL_CT_CHECKSUM].data = &net->ct.sysctl_checksum; table[NF_SYSCTL_CT_LOG_INVALID].data = &net->ct.sysctl_log_invalid; + table[NF_SYSCTL_CT_ACCT].data = &net->ct.sysctl_acct; + table[NF_SYSCTL_CT_HELPER].data = &net->ct.sysctl_auto_assign_helper; #ifdef CONFIG_NF_CONNTRACK_EVENTS table[NF_SYSCTL_CT_EVENTS].data = &net->ct.sysctl_events; #endif +#ifdef CONFIG_NF_CONNTRACK_TIMESTAMP + table[NF_SYSCTL_CT_TIMESTAMP].data = &net->ct.sysctl_tstamp; +#endif table[NF_SYSCTL_CT_PROTO_TIMEOUT_GENERIC].data = &nf_generic_pernet(net)->timeout; table[NF_SYSCTL_CT_PROTO_TIMEOUT_ICMP].data = &nf_icmp_pernet(net)->timeout; table[NF_SYSCTL_CT_PROTO_TIMEOUT_ICMPV6].data = &nf_icmpv6_pernet(net)->timeout; diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index e3d797252a98..80a8f9ae4c93 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -111,15 +111,16 @@ static void flow_offload_fixup_tcp(struct ip_ct_tcp *tcp) #define NF_FLOWTABLE_TCP_PICKUP_TIMEOUT (120 * HZ) #define NF_FLOWTABLE_UDP_PICKUP_TIMEOUT (30 * HZ) -static void flow_offload_fixup_ct_state(struct nf_conn *ct) +static inline __s32 nf_flow_timeout_delta(unsigned int timeout) +{ + return (__s32)(timeout - (u32)jiffies); +} + +static void flow_offload_fixup_ct_timeout(struct nf_conn *ct) { const struct nf_conntrack_l4proto *l4proto; + int l4num = nf_ct_protonum(ct); unsigned int timeout; - int l4num; - - l4num = nf_ct_protonum(ct); - if (l4num == IPPROTO_TCP) - flow_offload_fixup_tcp(&ct->proto.tcp); l4proto = nf_ct_l4proto_find(l4num); if (!l4proto) @@ -132,7 +133,20 @@ static void flow_offload_fixup_ct_state(struct nf_conn *ct) else return; - ct->timeout = nfct_time_stamp + timeout; + if (nf_flow_timeout_delta(ct->timeout) > (__s32)timeout) + ct->timeout = nfct_time_stamp + timeout; +} + +static void flow_offload_fixup_ct_state(struct nf_conn *ct) +{ + if (nf_ct_protonum(ct) == IPPROTO_TCP) + flow_offload_fixup_tcp(&ct->proto.tcp); +} + +static void flow_offload_fixup_ct(struct nf_conn *ct) +{ + flow_offload_fixup_ct_state(ct); + flow_offload_fixup_ct_timeout(ct); } void flow_offload_free(struct flow_offload *flow) @@ -208,6 +222,11 @@ int flow_offload_add(struct nf_flowtable *flow_table, struct flow_offload *flow) } EXPORT_SYMBOL_GPL(flow_offload_add); +static inline bool nf_flow_has_expired(const struct flow_offload *flow) +{ + return nf_flow_timeout_delta(flow->timeout) <= 0; +} + static void flow_offload_del(struct nf_flowtable *flow_table, struct flow_offload *flow) { @@ -223,6 +242,11 @@ static void flow_offload_del(struct nf_flowtable *flow_table, e = container_of(flow, struct flow_offload_entry, flow); clear_bit(IPS_OFFLOAD_BIT, &e->ct->status); + if (nf_flow_has_expired(flow)) + flow_offload_fixup_ct(e->ct); + else if (flow->flags & FLOW_OFFLOAD_TEARDOWN) + flow_offload_fixup_ct_timeout(e->ct); + flow_offload_free(flow); } @@ -298,11 +322,6 @@ nf_flow_table_iterate(struct nf_flowtable *flow_table, return err; } -static inline bool nf_flow_has_expired(const struct flow_offload *flow) -{ - return (__s32)(flow->timeout - (u32)jiffies) <= 0; -} - static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data) { struct nf_flowtable *flow_table = data; diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index cdfc33517e85..b9e7dd6e60ce 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -214,6 +214,24 @@ static bool nf_flow_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) return true; } +static int nf_flow_offload_dst_check(struct dst_entry *dst) +{ + if (unlikely(dst_xfrm(dst))) + return dst_check(dst, 0) ? 0 : -1; + + return 0; +} + +static unsigned int nf_flow_xmit_xfrm(struct sk_buff *skb, + const struct nf_hook_state *state, + struct dst_entry *dst) +{ + skb_orphan(skb); + skb_dst_set_noref(skb, dst); + dst_output(state->net, state->sk, skb); + return NF_STOLEN; +} + unsigned int nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) @@ -254,12 +272,25 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, if (nf_flow_state_check(flow, ip_hdr(skb)->protocol, skb, thoff)) return NF_ACCEPT; + if (nf_flow_offload_dst_check(&rt->dst)) { + flow_offload_teardown(flow); + return NF_ACCEPT; + } + if (nf_flow_nat_ip(flow, skb, thoff, dir) < 0) return NF_DROP; flow->timeout = (u32)jiffies + NF_FLOW_TIMEOUT; iph = ip_hdr(skb); ip_decrease_ttl(iph); + skb->tstamp = 0; + + if (unlikely(dst_xfrm(&rt->dst))) { + memset(skb->cb, 0, sizeof(struct inet_skb_parm)); + IPCB(skb)->iif = skb->dev->ifindex; + IPCB(skb)->flags = IPSKB_FORWARDED; + return nf_flow_xmit_xfrm(skb, state, &rt->dst); + } skb->dev = outdev; nexthop = rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr); @@ -467,6 +498,11 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, sizeof(*ip6h))) return NF_ACCEPT; + if (nf_flow_offload_dst_check(&rt->dst)) { + flow_offload_teardown(flow); + return NF_ACCEPT; + } + if (skb_try_make_writable(skb, sizeof(*ip6h))) return NF_DROP; @@ -476,6 +512,14 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, flow->timeout = (u32)jiffies + NF_FLOW_TIMEOUT; ip6h = ipv6_hdr(skb); ip6h->hop_limit--; + skb->tstamp = 0; + + if (unlikely(dst_xfrm(&rt->dst))) { + memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); + IP6CB(skb)->iif = skb->dev->ifindex; + IP6CB(skb)->flags = IP6SKB_FORWARDED; + return nf_flow_xmit_xfrm(skb, state, &rt->dst); + } skb->dev = outdev; nexthop = rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index fe3b7b0c6c66..6d00bef023c4 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -138,9 +138,14 @@ static void nft_set_trans_bind(const struct nft_ctx *ctx, struct nft_set *set) return; list_for_each_entry_reverse(trans, &net->nft.commit_list, list) { - if (trans->msg_type == NFT_MSG_NEWSET && - nft_trans_set(trans) == set) { - set->bound = true; + switch (trans->msg_type) { + case NFT_MSG_NEWSET: + if (nft_trans_set(trans) == set) + nft_trans_set_bound(trans) = true; + break; + case NFT_MSG_NEWSETELEM: + if (nft_trans_elem_set(trans) == set) + nft_trans_elem_set_bound(trans) = true; break; } } @@ -1662,6 +1667,10 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, chain->flags |= NFT_BASE_CHAIN | flags; basechain->policy = NF_ACCEPT; + if (chain->flags & NFT_CHAIN_HW_OFFLOAD && + nft_chain_offload_priority(basechain) < 0) + return -EOPNOTSUPP; + flow_block_init(&basechain->flow_block); } else { chain = kzalloc(sizeof(*chain), GFP_KERNEL); @@ -6906,7 +6915,7 @@ static int __nf_tables_abort(struct net *net) break; case NFT_MSG_NEWSET: trans->ctx.table->use--; - if (nft_trans_set(trans)->bound) { + if (nft_trans_set_bound(trans)) { nft_trans_destroy(trans); break; } @@ -6918,7 +6927,7 @@ static int __nf_tables_abort(struct net *net) nft_trans_destroy(trans); break; case NFT_MSG_NEWSETELEM: - if (nft_trans_elem_set(trans)->bound) { + if (nft_trans_elem_set_bound(trans)) { nft_trans_destroy(trans); break; } diff --git a/net/netfilter/nf_tables_offload.c b/net/netfilter/nf_tables_offload.c index d3c4c9c88bc8..3c2725ade61b 100644 --- a/net/netfilter/nf_tables_offload.c +++ b/net/netfilter/nf_tables_offload.c @@ -103,10 +103,11 @@ void nft_offload_update_dependency(struct nft_offload_ctx *ctx, } static void nft_flow_offload_common_init(struct flow_cls_common_offload *common, - __be16 proto, - struct netlink_ext_ack *extack) + __be16 proto, int priority, + struct netlink_ext_ack *extack) { common->protocol = proto; + common->prio = priority; common->extack = extack; } @@ -124,6 +125,15 @@ static int nft_setup_cb_call(struct nft_base_chain *basechain, return 0; } +int nft_chain_offload_priority(struct nft_base_chain *basechain) +{ + if (basechain->ops.priority <= 0 || + basechain->ops.priority > USHRT_MAX) + return -1; + + return 0; +} + static int nft_flow_offload_rule(struct nft_trans *trans, enum flow_cls_command command) { @@ -142,7 +152,8 @@ static int nft_flow_offload_rule(struct nft_trans *trans, if (flow) proto = flow->proto; - nft_flow_offload_common_init(&cls_flow.common, proto, &extack); + nft_flow_offload_common_init(&cls_flow.common, proto, + basechain->ops.priority, &extack); cls_flow.command = command; cls_flow.cookie = (unsigned long) rule; if (flow) diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c index 1f04ed5c518c..974300178fa9 100644 --- a/net/netfilter/nft_bitwise.c +++ b/net/netfilter/nft_bitwise.c @@ -135,8 +135,8 @@ static int nft_bitwise_offload(struct nft_offload_ctx *ctx, { const struct nft_bitwise *priv = nft_expr_priv(expr); - if (memcmp(&priv->xor, &zero, sizeof(priv->xor) || - priv->sreg != priv->dreg)) + if (memcmp(&priv->xor, &zero, sizeof(priv->xor)) || + priv->sreg != priv->dreg) return -EOPNOTSUPP; memcpy(&ctx->regs[priv->dreg].mask, &priv->mask, sizeof(priv->mask)); diff --git a/net/netfilter/nft_flow_offload.c b/net/netfilter/nft_flow_offload.c index aa5f571d4361..01705ad74a9a 100644 --- a/net/netfilter/nft_flow_offload.c +++ b/net/netfilter/nft_flow_offload.c @@ -72,11 +72,11 @@ static void nft_flow_offload_eval(const struct nft_expr *expr, { struct nft_flow_offload *priv = nft_expr_priv(expr); struct nf_flowtable *flowtable = &priv->flowtable->data; + struct tcphdr _tcph, *tcph = NULL; enum ip_conntrack_info ctinfo; struct nf_flow_route route; struct flow_offload *flow; enum ip_conntrack_dir dir; - bool is_tcp = false; struct nf_conn *ct; int ret; @@ -89,7 +89,10 @@ static void nft_flow_offload_eval(const struct nft_expr *expr, switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) { case IPPROTO_TCP: - is_tcp = true; + tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff, + sizeof(_tcph), &_tcph); + if (unlikely(!tcph || tcph->fin || tcph->rst)) + goto out; break; case IPPROTO_UDP: break; @@ -115,7 +118,7 @@ static void nft_flow_offload_eval(const struct nft_expr *expr, if (!flow) goto err_flow_alloc; - if (is_tcp) { + if (tcph) { ct->proto.tcp.seen[0].flags |= IP_CT_TCP_FLAG_BE_LIBERAL; ct->proto.tcp.seen[1].flags |= IP_CT_TCP_FLAG_BE_LIBERAL; } @@ -146,6 +149,11 @@ static int nft_flow_offload_validate(const struct nft_ctx *ctx, return nft_chain_validate_hooks(ctx->chain, hook_mask); } +static const struct nla_policy nft_flow_offload_policy[NFTA_FLOW_MAX + 1] = { + [NFTA_FLOW_TABLE_NAME] = { .type = NLA_STRING, + .len = NFT_NAME_MAXLEN - 1 }, +}; + static int nft_flow_offload_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]) @@ -204,6 +212,7 @@ static const struct nft_expr_ops nft_flow_offload_ops = { static struct nft_expr_type nft_flow_offload_type __read_mostly = { .name = "flow_offload", .ops = &nft_flow_offload_ops, + .policy = nft_flow_offload_policy, .maxattr = NFTA_FLOW_MAX, .owner = THIS_MODULE, }; diff --git a/net/netfilter/xt_nfacct.c b/net/netfilter/xt_nfacct.c index d0ab1adf5bff..5aab6df74e0f 100644 --- a/net/netfilter/xt_nfacct.c +++ b/net/netfilter/xt_nfacct.c @@ -54,25 +54,39 @@ nfacct_mt_destroy(const struct xt_mtdtor_param *par) nfnl_acct_put(info->nfacct); } -static struct xt_match nfacct_mt_reg __read_mostly = { - .name = "nfacct", - .family = NFPROTO_UNSPEC, - .checkentry = nfacct_mt_checkentry, - .match = nfacct_mt, - .destroy = nfacct_mt_destroy, - .matchsize = sizeof(struct xt_nfacct_match_info), - .usersize = offsetof(struct xt_nfacct_match_info, nfacct), - .me = THIS_MODULE, +static struct xt_match nfacct_mt_reg[] __read_mostly = { + { + .name = "nfacct", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = nfacct_mt_checkentry, + .match = nfacct_mt, + .destroy = nfacct_mt_destroy, + .matchsize = sizeof(struct xt_nfacct_match_info), + .usersize = offsetof(struct xt_nfacct_match_info, nfacct), + .me = THIS_MODULE, + }, + { + .name = "nfacct", + .revision = 1, + .family = NFPROTO_UNSPEC, + .checkentry = nfacct_mt_checkentry, + .match = nfacct_mt, + .destroy = nfacct_mt_destroy, + .matchsize = sizeof(struct xt_nfacct_match_info_v1), + .usersize = offsetof(struct xt_nfacct_match_info_v1, nfacct), + .me = THIS_MODULE, + }, }; static int __init nfacct_mt_init(void) { - return xt_register_match(&nfacct_mt_reg); + return xt_register_matches(nfacct_mt_reg, ARRAY_SIZE(nfacct_mt_reg)); } static void __exit nfacct_mt_exit(void) { - xt_unregister_match(&nfacct_mt_reg); + xt_unregister_matches(nfacct_mt_reg, ARRAY_SIZE(nfacct_mt_reg)); } module_init(nfacct_mt_init); diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c index ead7c6022208..b92b22ce8abd 100644 --- a/net/netfilter/xt_physdev.c +++ b/net/netfilter/xt_physdev.c @@ -101,11 +101,9 @@ static int physdev_mt_check(const struct xt_mtchk_param *par) if (info->bitmask & (XT_PHYSDEV_OP_OUT | XT_PHYSDEV_OP_ISOUT) && (!(info->bitmask & XT_PHYSDEV_OP_BRIDGED) || info->invert & XT_PHYSDEV_OP_BRIDGED) && - par->hook_mask & ((1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_FORWARD) | (1 << NF_INET_POST_ROUTING))) { + par->hook_mask & (1 << NF_INET_LOCAL_OUT)) { pr_info_ratelimited("--physdev-out and --physdev-is-out only supported in the FORWARD and POSTROUTING chains with bridged traffic\n"); - if (par->hook_mask & (1 << NF_INET_LOCAL_OUT)) - return -EINVAL; + return -EINVAL; } if (!brnf_probed) { diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 2b0ef55cf89e..409a3ae47ce2 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -607,7 +607,7 @@ catmap_getnode_alloc: */ int netlbl_catmap_walk(struct netlbl_lsm_catmap *catmap, u32 offset) { - struct netlbl_lsm_catmap *iter = catmap; + struct netlbl_lsm_catmap *iter; u32 idx; u32 bit; NETLBL_CATMAP_MAPTYPE bitmap; diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 848c6eb55064..05249eb45082 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -67,6 +67,7 @@ struct ovs_conntrack_info { struct md_mark mark; struct md_labels labels; char timeout[CTNL_TIMEOUT_NAME_MAX]; + struct nf_ct_timeout *nf_ct_timeout; #if IS_ENABLED(CONFIG_NF_NAT) struct nf_nat_range2 range; /* Only present for SRC NAT and DST NAT. */ #endif @@ -524,6 +525,11 @@ static int handle_fragments(struct net *net, struct sw_flow_key *key, return -EPFNOSUPPORT; } + /* The key extracted from the fragment that completed this datagram + * likely didn't have an L4 header, so regenerate it. + */ + ovs_flow_key_update_l3l4(skb, key); + key->ip.frag = OVS_FRAG_TYPE_NONE; skb_clear_hash(skb); skb->ignore_df = 1; @@ -697,6 +703,14 @@ static bool skb_nfct_cached(struct net *net, if (help && rcu_access_pointer(help->helper) != info->helper) return false; } + if (info->nf_ct_timeout) { + struct nf_conn_timeout *timeout_ext; + + timeout_ext = nf_ct_timeout_find(ct); + if (!timeout_ext || info->nf_ct_timeout != + rcu_dereference(timeout_ext->timeout)) + return false; + } /* Force conntrack entry direction to the current packet? */ if (info->force && CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) { /* Delete the conntrack entry if confirmed, else just release @@ -1565,7 +1579,7 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, case OVS_CT_ATTR_TIMEOUT: memcpy(info->timeout, nla_data(a), nla_len(a)); if (!memchr(info->timeout, '\0', nla_len(a))) { - OVS_NLERR(log, "Invalid conntrack helper"); + OVS_NLERR(log, "Invalid conntrack timeout"); return -EINVAL; } break; @@ -1657,6 +1671,10 @@ int ovs_ct_copy_action(struct net *net, const struct nlattr *attr, ct_info.timeout)) pr_info_ratelimited("Failed to associated timeout " "policy `%s'\n", ct_info.timeout); + else + ct_info.nf_ct_timeout = rcu_dereference( + nf_ct_timeout_find(ct_info.ct)->timeout); + } if (helper) { diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 65122bbccd27..dde9d762edee 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -1545,10 +1545,34 @@ static void ovs_dp_reset_user_features(struct sk_buff *skb, struct genl_info *in dp->user_features = 0; } -static void ovs_dp_change(struct datapath *dp, struct nlattr *a[]) +DEFINE_STATIC_KEY_FALSE(tc_recirc_sharing_support); + +static int ovs_dp_change(struct datapath *dp, struct nlattr *a[]) { - if (a[OVS_DP_ATTR_USER_FEATURES]) - dp->user_features = nla_get_u32(a[OVS_DP_ATTR_USER_FEATURES]); + u32 user_features = 0; + + if (a[OVS_DP_ATTR_USER_FEATURES]) { + user_features = nla_get_u32(a[OVS_DP_ATTR_USER_FEATURES]); + + if (user_features & ~(OVS_DP_F_VPORT_PIDS | + OVS_DP_F_UNALIGNED | + OVS_DP_F_TC_RECIRC_SHARING)) + return -EOPNOTSUPP; + +#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + if (user_features & OVS_DP_F_TC_RECIRC_SHARING) + return -EOPNOTSUPP; +#endif + } + + dp->user_features = user_features; + + if (dp->user_features & OVS_DP_F_TC_RECIRC_SHARING) + static_branch_enable(&tc_recirc_sharing_support); + else + static_branch_disable(&tc_recirc_sharing_support); + + return 0; } static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) @@ -1610,7 +1634,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) parms.port_no = OVSP_LOCAL; parms.upcall_portids = a[OVS_DP_ATTR_UPCALL_PID]; - ovs_dp_change(dp, a); + err = ovs_dp_change(dp, a); + if (err) + goto err_destroy_meters; /* So far only local changes have been made, now need the lock. */ ovs_lock(); @@ -1736,7 +1762,9 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(dp)) goto err_unlock_free; - ovs_dp_change(dp, info->attrs); + err = ovs_dp_change(dp, info->attrs); + if (err) + goto err_unlock_free; err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, info->snd_seq, 0, OVS_DP_CMD_SET); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 751d34accdf9..81e85dde8217 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -218,6 +218,8 @@ static inline struct datapath *get_dp(struct net *net, int dp_ifindex) extern struct notifier_block ovs_dp_device_notifier; extern struct genl_family dp_vport_genl_family; +DECLARE_STATIC_KEY_FALSE(tc_recirc_sharing_support); + void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key); void ovs_dp_detach_port(struct vport *); int ovs_dp_upcall(struct datapath *, struct sk_buff *, diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index bc89e16e0505..38147e6a20f5 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -523,78 +523,15 @@ static int parse_nsh(struct sk_buff *skb, struct sw_flow_key *key) } /** - * key_extract - extracts a flow key from an Ethernet frame. + * key_extract_l3l4 - extracts L3/L4 header information. * @skb: sk_buff that contains the frame, with skb->data pointing to the - * Ethernet header + * L3 header * @key: output flow key * - * The caller must ensure that skb->len >= ETH_HLEN. - * - * Returns 0 if successful, otherwise a negative errno value. - * - * Initializes @skb header fields as follows: - * - * - skb->mac_header: the L2 header. - * - * - skb->network_header: just past the L2 header, or just past the - * VLAN header, to the first byte of the L2 payload. - * - * - skb->transport_header: If key->eth.type is ETH_P_IP or ETH_P_IPV6 - * on output, then just past the IP header, if one is present and - * of a correct length, otherwise the same as skb->network_header. - * For other key->eth.type values it is left untouched. - * - * - skb->protocol: the type of the data starting at skb->network_header. - * Equals to key->eth.type. */ -static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) +static int key_extract_l3l4(struct sk_buff *skb, struct sw_flow_key *key) { int error; - struct ethhdr *eth; - - /* Flags are always used as part of stats */ - key->tp.flags = 0; - - skb_reset_mac_header(skb); - - /* Link layer. */ - clear_vlan(key); - if (ovs_key_mac_proto(key) == MAC_PROTO_NONE) { - if (unlikely(eth_type_vlan(skb->protocol))) - return -EINVAL; - - skb_reset_network_header(skb); - key->eth.type = skb->protocol; - } else { - eth = eth_hdr(skb); - ether_addr_copy(key->eth.src, eth->h_source); - ether_addr_copy(key->eth.dst, eth->h_dest); - - __skb_pull(skb, 2 * ETH_ALEN); - /* We are going to push all headers that we pull, so no need to - * update skb->csum here. - */ - - if (unlikely(parse_vlan(skb, key))) - return -ENOMEM; - - key->eth.type = parse_ethertype(skb); - if (unlikely(key->eth.type == htons(0))) - return -ENOMEM; - - /* Multiple tagged packets need to retain TPID to satisfy - * skb_vlan_pop(), which will later shift the ethertype into - * skb->protocol. - */ - if (key->eth.cvlan.tci & htons(VLAN_CFI_MASK)) - skb->protocol = key->eth.cvlan.tpid; - else - skb->protocol = key->eth.type; - - skb_reset_network_header(skb); - __skb_push(skb, skb->data - skb_mac_header(skb)); - } - skb_reset_mac_len(skb); /* Network layer. */ if (key->eth.type == htons(ETH_P_IP)) { @@ -623,6 +560,7 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) offset = nh->frag_off & htons(IP_OFFSET); if (offset) { key->ip.frag = OVS_FRAG_TYPE_LATER; + memset(&key->tp, 0, sizeof(key->tp)); return 0; } if (nh->frag_off & htons(IP_MF) || @@ -740,8 +678,10 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) return error; } - if (key->ip.frag == OVS_FRAG_TYPE_LATER) + if (key->ip.frag == OVS_FRAG_TYPE_LATER) { + memset(&key->tp, 0, sizeof(key->tp)); return 0; + } if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP) key->ip.frag = OVS_FRAG_TYPE_FIRST; @@ -788,6 +728,92 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) return 0; } +/** + * key_extract - extracts a flow key from an Ethernet frame. + * @skb: sk_buff that contains the frame, with skb->data pointing to the + * Ethernet header + * @key: output flow key + * + * The caller must ensure that skb->len >= ETH_HLEN. + * + * Returns 0 if successful, otherwise a negative errno value. + * + * Initializes @skb header fields as follows: + * + * - skb->mac_header: the L2 header. + * + * - skb->network_header: just past the L2 header, or just past the + * VLAN header, to the first byte of the L2 payload. + * + * - skb->transport_header: If key->eth.type is ETH_P_IP or ETH_P_IPV6 + * on output, then just past the IP header, if one is present and + * of a correct length, otherwise the same as skb->network_header. + * For other key->eth.type values it is left untouched. + * + * - skb->protocol: the type of the data starting at skb->network_header. + * Equals to key->eth.type. + */ +static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) +{ + struct ethhdr *eth; + + /* Flags are always used as part of stats */ + key->tp.flags = 0; + + skb_reset_mac_header(skb); + + /* Link layer. */ + clear_vlan(key); + if (ovs_key_mac_proto(key) == MAC_PROTO_NONE) { + if (unlikely(eth_type_vlan(skb->protocol))) + return -EINVAL; + + skb_reset_network_header(skb); + key->eth.type = skb->protocol; + } else { + eth = eth_hdr(skb); + ether_addr_copy(key->eth.src, eth->h_source); + ether_addr_copy(key->eth.dst, eth->h_dest); + + __skb_pull(skb, 2 * ETH_ALEN); + /* We are going to push all headers that we pull, so no need to + * update skb->csum here. + */ + + if (unlikely(parse_vlan(skb, key))) + return -ENOMEM; + + key->eth.type = parse_ethertype(skb); + if (unlikely(key->eth.type == htons(0))) + return -ENOMEM; + + /* Multiple tagged packets need to retain TPID to satisfy + * skb_vlan_pop(), which will later shift the ethertype into + * skb->protocol. + */ + if (key->eth.cvlan.tci & htons(VLAN_CFI_MASK)) + skb->protocol = key->eth.cvlan.tpid; + else + skb->protocol = key->eth.type; + + skb_reset_network_header(skb); + __skb_push(skb, skb->data - skb_mac_header(skb)); + } + + skb_reset_mac_len(skb); + + /* Fill out L3/L4 key info, if any */ + return key_extract_l3l4(skb, key); +} + +/* In the case of conntrack fragment handling it expects L3 headers, + * add a helper. + */ +int ovs_flow_key_update_l3l4(struct sk_buff *skb, struct sw_flow_key *key) +{ + return key_extract_l3l4(skb, key); +} + int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key) { int res; @@ -816,6 +842,9 @@ static int key_extract_mac_proto(struct sk_buff *skb) int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info, struct sk_buff *skb, struct sw_flow_key *key) { +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + struct tc_skb_ext *tc_ext; +#endif int res, err; /* Extract metadata from packet. */ @@ -848,7 +877,17 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info, if (res < 0) return res; key->mac_proto = res; + +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + if (static_branch_unlikely(&tc_recirc_sharing_support)) { + tc_ext = skb_ext_find(skb, TC_SKB_EXT); + key->recirc_id = tc_ext ? tc_ext->chain : 0; + } else { + key->recirc_id = 0; + } +#else key->recirc_id = 0; +#endif err = key_extract(skb, key); if (!err) diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index a5506e2d4b7a..b830d5ff7af4 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -270,6 +270,7 @@ void ovs_flow_stats_clear(struct sw_flow *); u64 ovs_flow_used_time(unsigned long flow_jiffies); int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key); +int ovs_flow_key_update_l3l4(struct sk_buff *skb, struct sw_flow_key *key); int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info, struct sk_buff *skb, struct sw_flow_key *key); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8d54f3047768..e2742b006d25 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2618,6 +2618,13 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) mutex_lock(&po->pg_vec_lock); + /* packet_sendmsg() check on tx_ring.pg_vec was lockless, + * we need to confirm it under protection of pg_vec_lock. + */ + if (unlikely(!po->tx_ring.pg_vec)) { + err = -EBUSY; + goto out; + } if (likely(saddr == NULL)) { dev = packet_cached_dev_get(po); proto = po->num; diff --git a/net/psample/psample.c b/net/psample/psample.c index 841f198ea1a8..66e4b61a350d 100644 --- a/net/psample/psample.c +++ b/net/psample/psample.c @@ -154,7 +154,7 @@ static void psample_group_destroy(struct psample_group *group) { psample_group_notify(group, PSAMPLE_CMD_DEL_GROUP); list_del(&group->list); - kfree(group); + kfree_rcu(group, rcu); } static struct psample_group * diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index 2b969f99ef13..1a5bf3fa4578 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2006, 2018 Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2006, 2019 Oracle and/or its affiliates. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -559,7 +559,7 @@ static int rds_connect(struct socket *sock, struct sockaddr *uaddr, ret = -EDESTADDRREQ; break; } - if (IN_MULTICAST(ntohl(sin->sin_addr.s_addr)) || + if (ipv4_is_multicast(sin->sin_addr.s_addr) || sin->sin_addr.s_addr == htonl(INADDR_BROADCAST)) { ret = -EINVAL; break; @@ -593,7 +593,7 @@ static int rds_connect(struct socket *sock, struct sockaddr *uaddr, addr4 = sin6->sin6_addr.s6_addr32[3]; if (addr4 == htonl(INADDR_ANY) || addr4 == htonl(INADDR_BROADCAST) || - IN_MULTICAST(ntohl(addr4))) { + ipv4_is_multicast(addr4)) { ret = -EPROTOTYPE; break; } @@ -705,7 +705,7 @@ static int rds_create(struct net *net, struct socket *sock, int protocol, if (sock->type != SOCK_SEQPACKET || protocol) return -ESOCKTNOSUPPORT; - sk = sk_alloc(net, AF_RDS, GFP_ATOMIC, &rds_proto, kern); + sk = sk_alloc(net, AF_RDS, GFP_KERNEL, &rds_proto, kern); if (!sk) return -ENOMEM; @@ -741,6 +741,10 @@ static void rds_sock_inc_info(struct socket *sock, unsigned int len, spin_lock_bh(&rds_sock_lock); list_for_each_entry(rs, &rds_sock_list, rs_item) { + /* This option only supports IPv4 sockets. */ + if (!ipv6_addr_v4mapped(&rs->rs_bound_addr)) + continue; + read_lock(&rs->rs_recv_lock); /* XXX too lazy to maintain counts.. */ @@ -762,21 +766,60 @@ static void rds_sock_inc_info(struct socket *sock, unsigned int len, lens->each = sizeof(struct rds_info_message); } +#if IS_ENABLED(CONFIG_IPV6) +static void rds6_sock_inc_info(struct socket *sock, unsigned int len, + struct rds_info_iterator *iter, + struct rds_info_lengths *lens) +{ + struct rds_incoming *inc; + unsigned int total = 0; + struct rds_sock *rs; + + len /= sizeof(struct rds6_info_message); + + spin_lock_bh(&rds_sock_lock); + + list_for_each_entry(rs, &rds_sock_list, rs_item) { + read_lock(&rs->rs_recv_lock); + + list_for_each_entry(inc, &rs->rs_recv_queue, i_item) { + total++; + if (total <= len) + rds6_inc_info_copy(inc, iter, &inc->i_saddr, + &rs->rs_bound_addr, 1); + } + + read_unlock(&rs->rs_recv_lock); + } + + spin_unlock_bh(&rds_sock_lock); + + lens->nr = total; + lens->each = sizeof(struct rds6_info_message); +} +#endif + static void rds_sock_info(struct socket *sock, unsigned int len, struct rds_info_iterator *iter, struct rds_info_lengths *lens) { struct rds_info_socket sinfo; + unsigned int cnt = 0; struct rds_sock *rs; len /= sizeof(struct rds_info_socket); spin_lock_bh(&rds_sock_lock); - if (len < rds_sock_count) + if (len < rds_sock_count) { + cnt = rds_sock_count; goto out; + } list_for_each_entry(rs, &rds_sock_list, rs_item) { + /* This option only supports IPv4 sockets. */ + if (!ipv6_addr_v4mapped(&rs->rs_bound_addr)) + continue; sinfo.sndbuf = rds_sk_sndbuf(rs); sinfo.rcvbuf = rds_sk_rcvbuf(rs); sinfo.bound_addr = rs->rs_bound_addr_v4; @@ -786,15 +829,51 @@ static void rds_sock_info(struct socket *sock, unsigned int len, sinfo.inum = sock_i_ino(rds_rs_to_sk(rs)); rds_info_copy(iter, &sinfo, sizeof(sinfo)); + cnt++; } out: - lens->nr = rds_sock_count; + lens->nr = cnt; lens->each = sizeof(struct rds_info_socket); spin_unlock_bh(&rds_sock_lock); } +#if IS_ENABLED(CONFIG_IPV6) +static void rds6_sock_info(struct socket *sock, unsigned int len, + struct rds_info_iterator *iter, + struct rds_info_lengths *lens) +{ + struct rds6_info_socket sinfo6; + struct rds_sock *rs; + + len /= sizeof(struct rds6_info_socket); + + spin_lock_bh(&rds_sock_lock); + + if (len < rds_sock_count) + goto out; + + list_for_each_entry(rs, &rds_sock_list, rs_item) { + sinfo6.sndbuf = rds_sk_sndbuf(rs); + sinfo6.rcvbuf = rds_sk_rcvbuf(rs); + sinfo6.bound_addr = rs->rs_bound_addr; + sinfo6.connected_addr = rs->rs_conn_addr; + sinfo6.bound_port = rs->rs_bound_port; + sinfo6.connected_port = rs->rs_conn_port; + sinfo6.inum = sock_i_ino(rds_rs_to_sk(rs)); + + rds_info_copy(iter, &sinfo6, sizeof(sinfo6)); + } + + out: + lens->nr = rds_sock_count; + lens->each = sizeof(struct rds6_info_socket); + + spin_unlock_bh(&rds_sock_lock); +} +#endif + static void rds_exit(void) { sock_unregister(rds_family_ops.family); @@ -808,6 +887,10 @@ static void rds_exit(void) rds_bind_lock_destroy(); rds_info_deregister_func(RDS_INFO_SOCKETS, rds_sock_info); rds_info_deregister_func(RDS_INFO_RECV_MESSAGES, rds_sock_inc_info); +#if IS_ENABLED(CONFIG_IPV6) + rds_info_deregister_func(RDS6_INFO_SOCKETS, rds6_sock_info); + rds_info_deregister_func(RDS6_INFO_RECV_MESSAGES, rds6_sock_inc_info); +#endif } module_exit(rds_exit); @@ -845,6 +928,10 @@ static int rds_init(void) rds_info_register_func(RDS_INFO_SOCKETS, rds_sock_info); rds_info_register_func(RDS_INFO_RECV_MESSAGES, rds_sock_inc_info); +#if IS_ENABLED(CONFIG_IPV6) + rds_info_register_func(RDS6_INFO_SOCKETS, rds6_sock_info); + rds_info_register_func(RDS6_INFO_RECV_MESSAGES, rds6_sock_inc_info); +#endif goto out; diff --git a/net/rds/bind.c b/net/rds/bind.c index 0f4398e7f2a7..6dbb763bc1fd 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -181,7 +181,7 @@ int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (addr_len < sizeof(struct sockaddr_in) || sin->sin_addr.s_addr == htonl(INADDR_ANY) || sin->sin_addr.s_addr == htonl(INADDR_BROADCAST) || - IN_MULTICAST(ntohl(sin->sin_addr.s_addr))) + ipv4_is_multicast(sin->sin_addr.s_addr)) return -EINVAL; ipv6_addr_set_v4mapped(sin->sin_addr.s_addr, &v6addr); binding_addr = &v6addr; @@ -206,7 +206,7 @@ int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) addr4 = sin6->sin6_addr.s6_addr32[3]; if (addr4 == htonl(INADDR_ANY) || addr4 == htonl(INADDR_BROADCAST) || - IN_MULTICAST(ntohl(addr4))) + ipv4_is_multicast(addr4)) return -EINVAL; } /* The scope ID must be specified for link local address. */ diff --git a/net/rds/ib.c b/net/rds/ib.c index ec05d91aa9a2..45acab2de0cf 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -291,7 +291,7 @@ static int rds_ib_conn_info_visitor(struct rds_connection *conn, void *buffer) { struct rds_info_rdma_connection *iinfo = buffer; - struct rds_ib_connection *ic; + struct rds_ib_connection *ic = conn->c_transport_data; /* We will only ever look at IB transports */ if (conn->c_trans != &rds_ib_transport) @@ -301,15 +301,16 @@ static int rds_ib_conn_info_visitor(struct rds_connection *conn, iinfo->src_addr = conn->c_laddr.s6_addr32[3]; iinfo->dst_addr = conn->c_faddr.s6_addr32[3]; - iinfo->tos = conn->c_tos; + if (ic) { + iinfo->tos = conn->c_tos; + iinfo->sl = ic->i_sl; + } memset(&iinfo->src_gid, 0, sizeof(iinfo->src_gid)); memset(&iinfo->dst_gid, 0, sizeof(iinfo->dst_gid)); if (rds_conn_state(conn) == RDS_CONN_UP) { struct rds_ib_device *rds_ibdev; - ic = conn->c_transport_data; - rdma_read_gids(ic->i_cm_id, (union ib_gid *)&iinfo->src_gid, (union ib_gid *)&iinfo->dst_gid); @@ -329,7 +330,7 @@ static int rds6_ib_conn_info_visitor(struct rds_connection *conn, void *buffer) { struct rds6_info_rdma_connection *iinfo6 = buffer; - struct rds_ib_connection *ic; + struct rds_ib_connection *ic = conn->c_transport_data; /* We will only ever look at IB transports */ if (conn->c_trans != &rds_ib_transport) @@ -337,6 +338,10 @@ static int rds6_ib_conn_info_visitor(struct rds_connection *conn, iinfo6->src_addr = conn->c_laddr; iinfo6->dst_addr = conn->c_faddr; + if (ic) { + iinfo6->tos = conn->c_tos; + iinfo6->sl = ic->i_sl; + } memset(&iinfo6->src_gid, 0, sizeof(iinfo6->src_gid)); memset(&iinfo6->dst_gid, 0, sizeof(iinfo6->dst_gid)); @@ -344,7 +349,6 @@ static int rds6_ib_conn_info_visitor(struct rds_connection *conn, if (rds_conn_state(conn) == RDS_CONN_UP) { struct rds_ib_device *rds_ibdev; - ic = conn->c_transport_data; rdma_read_gids(ic->i_cm_id, (union ib_gid *)&iinfo6->src_gid, (union ib_gid *)&iinfo6->dst_gid); rds_ibdev = ic->rds_ibdev; diff --git a/net/rds/ib.h b/net/rds/ib.h index 303c6ee8bdb7..f2b558e8b5ea 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -220,6 +220,7 @@ struct rds_ib_connection { /* Send/Recv vectors */ int i_scq_vector; int i_rcq_vector; + u8 i_sl; }; /* This assumes that atomic_t is at least 32 bits */ diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index fddaa09f7b0d..233f1368162b 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -152,6 +152,9 @@ void rds_ib_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_even RDS_PROTOCOL_MINOR(conn->c_version), ic->i_flowctl ? ", flow control" : ""); + /* receive sl from the peer */ + ic->i_sl = ic->i_cm_id->route.path_rec->sl; + atomic_set(&ic->i_cq_quiesce, 0); /* Init rings and fill recv. this needs to wait until protocol diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index 3cae88cbdaa0..a0f99bbf362c 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -385,6 +385,7 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp) unsigned int posted = 0; int ret = 0; bool can_wait = !!(gfp & __GFP_DIRECT_RECLAIM); + bool must_wake = false; u32 pos; /* the goal here is to just make sure that someone, somewhere @@ -405,6 +406,7 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp) recv = &ic->i_recvs[pos]; ret = rds_ib_recv_refill_one(conn, recv, gfp); if (ret) { + must_wake = true; break; } @@ -423,6 +425,11 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp) } posted++; + + if ((posted > 128 && need_resched()) || posted > 8192) { + must_wake = true; + break; + } } /* We're doing flow control - update the window. */ @@ -445,10 +452,13 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp) * if we should requeue. */ if (rds_conn_up(conn) && - ((can_wait && rds_ib_ring_low(&ic->i_recv_ring)) || + (must_wake || + (can_wait && rds_ib_ring_low(&ic->i_recv_ring)) || rds_ib_ring_empty(&ic->i_recv_ring))) { queue_delayed_work(rds_wq, &conn->c_recv_w, 1); } + if (can_wait) + cond_resched(); } /* @@ -1038,9 +1048,14 @@ int rds_ib_recv_init(void) si_meminfo(&si); rds_ib_sysctl_max_recv_allocation = si.totalram / 3 * PAGE_SIZE / RDS_FRAG_SIZE; - rds_ib_incoming_slab = kmem_cache_create("rds_ib_incoming", - sizeof(struct rds_ib_incoming), - 0, SLAB_HWCACHE_ALIGN, NULL); + rds_ib_incoming_slab = + kmem_cache_create_usercopy("rds_ib_incoming", + sizeof(struct rds_ib_incoming), + 0, SLAB_HWCACHE_ALIGN, + offsetof(struct rds_ib_incoming, + ii_inc.i_usercopy), + sizeof(struct rds_inc_usercopy), + NULL); if (!rds_ib_incoming_slab) goto out; diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c index 9986d6065c4d..5f741e51b4ba 100644 --- a/net/rds/rdma_transport.c +++ b/net/rds/rdma_transport.c @@ -43,6 +43,9 @@ static struct rdma_cm_id *rds_rdma_listen_id; static struct rdma_cm_id *rds6_rdma_listen_id; #endif +/* Per IB specification 7.7.3, service level is a 4-bit field. */ +#define TOS_TO_SL(tos) ((tos) & 0xF) + static int rds_rdma_cm_event_handler_cmn(struct rdma_cm_id *cm_id, struct rdma_cm_event *event, bool isv6) @@ -97,10 +100,13 @@ static int rds_rdma_cm_event_handler_cmn(struct rdma_cm_id *cm_id, struct rds_ib_connection *ibic; ibic = conn->c_transport_data; - if (ibic && ibic->i_cm_id == cm_id) + if (ibic && ibic->i_cm_id == cm_id) { + cm_id->route.path_rec[0].sl = + TOS_TO_SL(conn->c_tos); ret = trans->cm_initiate_connect(cm_id, isv6); - else + } else { rds_conn_drop(conn); + } } break; diff --git a/net/rds/rds.h b/net/rds/rds.h index f0066d168499..53e86911773a 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -271,6 +271,12 @@ struct rds_ext_header_rdma_dest { #define RDS_MSG_RX_END 2 #define RDS_MSG_RX_CMSG 3 +/* The following values are whitelisted for usercopy */ +struct rds_inc_usercopy { + rds_rdma_cookie_t rdma_cookie; + ktime_t rx_tstamp; +}; + struct rds_incoming { refcount_t i_refcount; struct list_head i_item; @@ -280,8 +286,7 @@ struct rds_incoming { unsigned long i_rx_jiffies; struct in6_addr i_saddr; - rds_rdma_cookie_t i_rdma_cookie; - ktime_t i_rx_tstamp; + struct rds_inc_usercopy i_usercopy; u64 i_rx_lat_trace[RDS_RX_MAX_TRACES]; }; @@ -717,7 +722,7 @@ struct rds_statistics { uint64_t s_cong_send_blocked; uint64_t s_recv_bytes_added_to_socket; uint64_t s_recv_bytes_removed_from_socket; - + uint64_t s_send_stuck_rm; }; /* af_rds.c */ diff --git a/net/rds/recv.c b/net/rds/recv.c index 853de4876088..c8404971d5ab 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2006, 2018 Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2006, 2019 Oracle and/or its affiliates. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -47,8 +47,8 @@ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, INIT_LIST_HEAD(&inc->i_item); inc->i_conn = conn; inc->i_saddr = *saddr; - inc->i_rdma_cookie = 0; - inc->i_rx_tstamp = ktime_set(0, 0); + inc->i_usercopy.rdma_cookie = 0; + inc->i_usercopy.rx_tstamp = ktime_set(0, 0); memset(inc->i_rx_lat_trace, 0, sizeof(inc->i_rx_lat_trace)); } @@ -62,8 +62,8 @@ void rds_inc_path_init(struct rds_incoming *inc, struct rds_conn_path *cp, inc->i_conn = cp->cp_conn; inc->i_conn_path = cp; inc->i_saddr = *saddr; - inc->i_rdma_cookie = 0; - inc->i_rx_tstamp = ktime_set(0, 0); + inc->i_usercopy.rdma_cookie = 0; + inc->i_usercopy.rx_tstamp = ktime_set(0, 0); } EXPORT_SYMBOL_GPL(rds_inc_path_init); @@ -186,7 +186,7 @@ static void rds_recv_incoming_exthdrs(struct rds_incoming *inc, struct rds_sock case RDS_EXTHDR_RDMA_DEST: /* We ignore the size for now. We could stash it * somewhere and use it for error checking. */ - inc->i_rdma_cookie = rds_rdma_make_cookie( + inc->i_usercopy.rdma_cookie = rds_rdma_make_cookie( be32_to_cpu(buffer.rdma_dest.h_rdma_rkey), be32_to_cpu(buffer.rdma_dest.h_rdma_offset)); @@ -380,7 +380,7 @@ void rds_recv_incoming(struct rds_connection *conn, struct in6_addr *saddr, be32_to_cpu(inc->i_hdr.h_len), inc->i_hdr.h_dport); if (sock_flag(sk, SOCK_RCVTSTAMP)) - inc->i_rx_tstamp = ktime_get_real(); + inc->i_usercopy.rx_tstamp = ktime_get_real(); rds_inc_addref(inc); inc->i_rx_lat_trace[RDS_MSG_RX_END] = local_clock(); list_add_tail(&inc->i_item, &rs->rs_recv_queue); @@ -540,16 +540,18 @@ static int rds_cmsg_recv(struct rds_incoming *inc, struct msghdr *msg, { int ret = 0; - if (inc->i_rdma_cookie) { + if (inc->i_usercopy.rdma_cookie) { ret = put_cmsg(msg, SOL_RDS, RDS_CMSG_RDMA_DEST, - sizeof(inc->i_rdma_cookie), &inc->i_rdma_cookie); + sizeof(inc->i_usercopy.rdma_cookie), + &inc->i_usercopy.rdma_cookie); if (ret) goto out; } - if ((inc->i_rx_tstamp != 0) && + if ((inc->i_usercopy.rx_tstamp != 0) && sock_flag(rds_rs_to_sk(rs), SOCK_RCVTSTAMP)) { - struct __kernel_old_timeval tv = ns_to_kernel_old_timeval(inc->i_rx_tstamp); + struct __kernel_old_timeval tv = + ns_to_kernel_old_timeval(inc->i_usercopy.rx_tstamp); if (!sock_flag(rds_rs_to_sk(rs), SOCK_TSTAMP_NEW)) { ret = put_cmsg(msg, SOL_SOCKET, SO_TIMESTAMP_OLD, @@ -811,6 +813,7 @@ void rds6_inc_info_copy(struct rds_incoming *inc, minfo6.seq = be64_to_cpu(inc->i_hdr.h_sequence); minfo6.len = be32_to_cpu(inc->i_hdr.h_len); + minfo6.tos = inc->i_conn->c_tos; if (flip) { minfo6.laddr = *daddr; @@ -824,6 +827,8 @@ void rds6_inc_info_copy(struct rds_incoming *inc, minfo6.fport = inc->i_hdr.h_dport; } + minfo6.flags = 0; + rds_info_copy(iter, &minfo6, sizeof(minfo6)); } #endif diff --git a/net/rds/send.c b/net/rds/send.c index 031b1e97a466..82dcd8b84fe7 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -145,6 +145,7 @@ int rds_send_xmit(struct rds_conn_path *cp) LIST_HEAD(to_be_dropped); int batch_count; unsigned long send_gen = 0; + int same_rm = 0; restart: batch_count = 0; @@ -200,6 +201,17 @@ restart: rm = cp->cp_xmit_rm; + if (!rm) { + same_rm = 0; + } else { + same_rm++; + if (same_rm >= 4096) { + rds_stats_inc(s_send_stuck_rm); + ret = -EAGAIN; + break; + } + } + /* * If between sending messages, we can send a pending congestion * map update. @@ -1132,7 +1144,7 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) case AF_INET: if (usin->sin_addr.s_addr == htonl(INADDR_ANY) || usin->sin_addr.s_addr == htonl(INADDR_BROADCAST) || - IN_MULTICAST(ntohl(usin->sin_addr.s_addr))) { + ipv4_is_multicast(usin->sin_addr.s_addr)) { ret = -EINVAL; goto out; } @@ -1163,7 +1175,7 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) addr4 = sin6->sin6_addr.s6_addr32[3]; if (addr4 == htonl(INADDR_ANY) || addr4 == htonl(INADDR_BROADCAST) || - IN_MULTICAST(ntohl(addr4))) { + ipv4_is_multicast(addr4)) { ret = -EINVAL; goto out; } diff --git a/net/rds/stats.c b/net/rds/stats.c index 73be187d389e..9e87da43c004 100644 --- a/net/rds/stats.c +++ b/net/rds/stats.c @@ -76,6 +76,9 @@ static const char *const rds_stat_names[] = { "cong_update_received", "cong_send_error", "cong_send_blocked", + "recv_bytes_added_to_sock", + "recv_bytes_freed_fromsock", + "send_stuck_rm", }; void rds_stats_info_copy(struct rds_info_iterator *iter, diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index d09eaf153544..d72ddb67bb74 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -193,7 +193,7 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) service_in_use: write_unlock(&local->services_lock); - rxrpc_put_local(local); + rxrpc_unuse_local(local); ret = -EADDRINUSE; error_unlock: release_sock(&rx->sk); @@ -402,7 +402,7 @@ EXPORT_SYMBOL(rxrpc_kernel_check_life); */ void rxrpc_kernel_probe_life(struct socket *sock, struct rxrpc_call *call) { - rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, 0, true, false, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, true, false, rxrpc_propose_ack_ping_for_check_life); rxrpc_send_ack_packet(call, true, NULL); } @@ -862,7 +862,6 @@ static void rxrpc_sock_destructor(struct sock *sk) static int rxrpc_release_sock(struct sock *sk) { struct rxrpc_sock *rx = rxrpc_sk(sk); - struct rxrpc_net *rxnet = rxrpc_net(sock_net(&rx->sk)); _enter("%p{%d,%d}", sk, sk->sk_state, refcount_read(&sk->sk_refcnt)); @@ -898,10 +897,8 @@ static int rxrpc_release_sock(struct sock *sk) rxrpc_release_calls_on_socket(rx); flush_workqueue(rxrpc_workqueue); rxrpc_purge_queue(&sk->sk_receive_queue); - rxrpc_queue_work(&rxnet->service_conn_reaper); - rxrpc_queue_work(&rxnet->client_conn_reaper); - rxrpc_put_local(rx->local); + rxrpc_unuse_local(rx->local); rx->local = NULL; key_put(rx->key); rx->key = NULL; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 63b26baa108a..1091bf35a199 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -185,11 +185,17 @@ struct rxrpc_host_header { * - max 48 bytes (struct sk_buff::cb) */ struct rxrpc_skb_priv { - union { - u8 nr_jumbo; /* Number of jumbo subpackets */ - }; + atomic_t nr_ring_pins; /* Number of rxtx ring pins */ + u8 nr_subpackets; /* Number of subpackets */ + u8 rx_flags; /* Received packet flags */ +#define RXRPC_SKB_INCL_LAST 0x01 /* - Includes last packet */ +#define RXRPC_SKB_TX_BUFFER 0x02 /* - Is transmit buffer */ union { int remain; /* amount of space remaining for next write */ + + /* List of requested ACKs on subpackets */ + unsigned long rx_req_ack[(RXRPC_MAX_NR_JUMBO + BITS_PER_LONG - 1) / + BITS_PER_LONG]; }; struct rxrpc_host_header hdr; /* RxRPC packet header from this packet */ @@ -257,7 +263,8 @@ struct rxrpc_security { */ struct rxrpc_local { struct rcu_head rcu; - atomic_t usage; + atomic_t active_users; /* Number of users of the local endpoint */ + atomic_t usage; /* Number of references to the structure */ struct rxrpc_net *rxnet; /* The network ns in which this resides */ struct list_head link; struct socket *socket; /* my UDP socket */ @@ -616,8 +623,7 @@ struct rxrpc_call { #define RXRPC_TX_ANNO_LAST 0x04 #define RXRPC_TX_ANNO_RESENT 0x08 -#define RXRPC_RX_ANNO_JUMBO 0x3f /* Jumbo subpacket number + 1 if not zero */ -#define RXRPC_RX_ANNO_JLAST 0x40 /* Set if last element of a jumbo packet */ +#define RXRPC_RX_ANNO_SUBPACKET 0x3f /* Subpacket number in jumbogram */ #define RXRPC_RX_ANNO_VERIFIED 0x80 /* Set if verified and decrypted */ rxrpc_seq_t tx_hard_ack; /* Dead slot in buffer; the first transmitted but * not hard-ACK'd packet follows this. @@ -653,7 +659,6 @@ struct rxrpc_call { /* receive-phase ACK management */ u8 ackr_reason; /* reason to ACK */ - u16 ackr_skew; /* skew on packet being ACK'd */ rxrpc_serial_t ackr_serial; /* serial of packet being ACK'd */ rxrpc_serial_t ackr_first_seq; /* first sequence number received */ rxrpc_seq_t ackr_prev_seq; /* previous sequence number received */ @@ -747,7 +752,7 @@ int rxrpc_reject_call(struct rxrpc_sock *); /* * call_event.c */ -void rxrpc_propose_ACK(struct rxrpc_call *, u8, u16, u32, bool, bool, +void rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool, bool, enum rxrpc_propose_ack_trace); void rxrpc_process_call(struct work_struct *); @@ -909,6 +914,7 @@ void rxrpc_disconnect_client_call(struct rxrpc_call *); void rxrpc_put_client_conn(struct rxrpc_connection *); void rxrpc_discard_expired_client_conns(struct work_struct *); void rxrpc_destroy_all_client_connections(struct rxrpc_net *); +void rxrpc_clean_up_local_conns(struct rxrpc_local *); /* * conn_event.c @@ -1006,6 +1012,8 @@ struct rxrpc_local *rxrpc_lookup_local(struct net *, const struct sockaddr_rxrpc struct rxrpc_local *rxrpc_get_local(struct rxrpc_local *); struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *); void rxrpc_put_local(struct rxrpc_local *); +struct rxrpc_local *rxrpc_use_local(struct rxrpc_local *); +void rxrpc_unuse_local(struct rxrpc_local *); void rxrpc_queue_local(struct rxrpc_local *); void rxrpc_destroy_all_locals(struct rxrpc_net *); @@ -1107,6 +1115,7 @@ void rxrpc_kernel_data_consumed(struct rxrpc_call *, struct sk_buff *); void rxrpc_packet_destructor(struct sk_buff *); void rxrpc_new_skb(struct sk_buff *, enum rxrpc_skb_trace); void rxrpc_see_skb(struct sk_buff *, enum rxrpc_skb_trace); +void rxrpc_eaten_skb(struct sk_buff *, enum rxrpc_skb_trace); void rxrpc_get_skb(struct sk_buff *, enum rxrpc_skb_trace); void rxrpc_free_skb(struct sk_buff *, enum rxrpc_skb_trace); void rxrpc_purge_queue(struct sk_buff_head *); diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index bc2adeb3acb9..cedbbb3a7c2e 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -43,8 +43,7 @@ static void rxrpc_propose_ping(struct rxrpc_call *call, * propose an ACK be sent */ static void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, - u16 skew, u32 serial, bool immediate, - bool background, + u32 serial, bool immediate, bool background, enum rxrpc_propose_ack_trace why) { enum rxrpc_propose_ack_outcome outcome = rxrpc_propose_ack_use; @@ -69,14 +68,12 @@ static void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, if (RXRPC_ACK_UPDATEABLE & (1 << ack_reason)) { outcome = rxrpc_propose_ack_update; call->ackr_serial = serial; - call->ackr_skew = skew; } if (!immediate) goto trace; } else if (prior > rxrpc_ack_priority[call->ackr_reason]) { call->ackr_reason = ack_reason; call->ackr_serial = serial; - call->ackr_skew = skew; } else { outcome = rxrpc_propose_ack_subsume; } @@ -137,11 +134,11 @@ trace: * propose an ACK be sent, locking the call structure */ void rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, - u16 skew, u32 serial, bool immediate, bool background, + u32 serial, bool immediate, bool background, enum rxrpc_propose_ack_trace why) { spin_lock_bh(&call->lock); - __rxrpc_propose_ACK(call, ack_reason, skew, serial, + __rxrpc_propose_ACK(call, ack_reason, serial, immediate, background, why); spin_unlock_bh(&call->lock); } @@ -202,7 +199,7 @@ static void rxrpc_resend(struct rxrpc_call *call, unsigned long now_j) continue; skb = call->rxtx_buffer[ix]; - rxrpc_see_skb(skb, rxrpc_skb_tx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); if (anno_type == RXRPC_TX_ANNO_UNACK) { if (ktime_after(skb->tstamp, max_age)) { @@ -239,7 +236,7 @@ static void rxrpc_resend(struct rxrpc_call *call, unsigned long now_j) ack_ts = ktime_sub(now, call->acks_latest_ts); if (ktime_to_ns(ack_ts) < call->peer->rtt) goto out; - rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, 0, true, false, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, true, false, rxrpc_propose_ack_ping_for_lost_ack); rxrpc_send_ack_packet(call, true, NULL); goto out; @@ -258,18 +255,18 @@ static void rxrpc_resend(struct rxrpc_call *call, unsigned long now_j) continue; skb = call->rxtx_buffer[ix]; - rxrpc_get_skb(skb, rxrpc_skb_tx_got); + rxrpc_get_skb(skb, rxrpc_skb_got); spin_unlock_bh(&call->lock); if (rxrpc_send_data_packet(call, skb, true) < 0) { - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); return; } if (rxrpc_is_client_call(call)) rxrpc_expose_client_call(call); - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); spin_lock_bh(&call->lock); /* We need to clear the retransmit state, but there are two @@ -372,7 +369,7 @@ recheck_state: if (time_after_eq(now, t)) { trace_rxrpc_timer(call, rxrpc_timer_exp_keepalive, now); cmpxchg(&call->keepalive_at, t, now + MAX_JIFFY_OFFSET); - rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, 0, true, true, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, true, true, rxrpc_propose_ack_ping_for_keepalive); set_bit(RXRPC_CALL_EV_PING, &call->events); } @@ -407,7 +404,7 @@ recheck_state: send_ack = NULL; if (test_and_clear_bit(RXRPC_CALL_EV_ACK_LOST, &call->events)) { call->acks_lost_top = call->tx_top; - rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, 0, true, false, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, 0, true, false, rxrpc_propose_ack_ping_for_lost_ack); send_ack = &call->acks_lost_ping; } diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 60cbc81dc461..32d8dc677142 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -422,6 +422,19 @@ void rxrpc_get_call(struct rxrpc_call *call, enum rxrpc_call_trace op) } /* + * Clean up the RxTx skb ring. + */ +static void rxrpc_cleanup_ring(struct rxrpc_call *call) +{ + int i; + + for (i = 0; i < RXRPC_RXTX_BUFF_SIZE; i++) { + rxrpc_free_skb(call->rxtx_buffer[i], rxrpc_skb_cleaned); + call->rxtx_buffer[i] = NULL; + } +} + +/* * Detach a call from its owning socket. */ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) @@ -429,7 +442,6 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) const void *here = __builtin_return_address(0); struct rxrpc_connection *conn = call->conn; bool put = false; - int i; _enter("{%d,%d}", call->debug_id, atomic_read(&call->usage)); @@ -481,13 +493,7 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) conn->security->free_call_crypto(call); } - for (i = 0; i < RXRPC_RXTX_BUFF_SIZE; i++) { - rxrpc_free_skb(call->rxtx_buffer[i], - (call->tx_phase ? rxrpc_skb_tx_cleaned : - rxrpc_skb_rx_cleaned)); - call->rxtx_buffer[i] = NULL; - } - + rxrpc_cleanup_ring(call); _leave(""); } @@ -570,8 +576,6 @@ static void rxrpc_rcu_destroy_call(struct rcu_head *rcu) */ void rxrpc_cleanup_call(struct rxrpc_call *call) { - int i; - _net("DESTROY CALL %d", call->debug_id); memset(&call->sock_node, 0xcd, sizeof(call->sock_node)); @@ -582,13 +586,8 @@ void rxrpc_cleanup_call(struct rxrpc_call *call) ASSERT(test_bit(RXRPC_CALL_RELEASED, &call->flags)); ASSERTCMP(call->conn, ==, NULL); - /* Clean up the Rx/Tx buffer */ - for (i = 0; i < RXRPC_RXTX_BUFF_SIZE; i++) - rxrpc_free_skb(call->rxtx_buffer[i], - (call->tx_phase ? rxrpc_skb_tx_cleaned : - rxrpc_skb_rx_cleaned)); - - rxrpc_free_skb(call->tx_pending, rxrpc_skb_tx_cleaned); + rxrpc_cleanup_ring(call); + rxrpc_free_skb(call->tx_pending, rxrpc_skb_cleaned); call_rcu(&call->rcu, rxrpc_rcu_destroy_call); } diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index aea82f909c60..3f1da1b49f69 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -1162,3 +1162,47 @@ void rxrpc_destroy_all_client_connections(struct rxrpc_net *rxnet) _leave(""); } + +/* + * Clean up the client connections on a local endpoint. + */ +void rxrpc_clean_up_local_conns(struct rxrpc_local *local) +{ + struct rxrpc_connection *conn, *tmp; + struct rxrpc_net *rxnet = local->rxnet; + unsigned int nr_active; + LIST_HEAD(graveyard); + + _enter(""); + + spin_lock(&rxnet->client_conn_cache_lock); + nr_active = rxnet->nr_active_client_conns; + + list_for_each_entry_safe(conn, tmp, &rxnet->idle_client_conns, + cache_link) { + if (conn->params.local == local) { + ASSERTCMP(conn->cache_state, ==, RXRPC_CONN_CLIENT_IDLE); + + trace_rxrpc_client(conn, -1, rxrpc_client_discard); + if (!test_and_clear_bit(RXRPC_CONN_EXPOSED, &conn->flags)) + BUG(); + conn->cache_state = RXRPC_CONN_CLIENT_INACTIVE; + list_move(&conn->cache_link, &graveyard); + nr_active--; + } + } + + rxnet->nr_active_client_conns = nr_active; + spin_unlock(&rxnet->client_conn_cache_lock); + ASSERTCMP(nr_active, >=, 0); + + while (!list_empty(&graveyard)) { + conn = list_entry(graveyard.next, + struct rxrpc_connection, cache_link); + list_del_init(&conn->cache_link); + + rxrpc_put_connection(conn); + } + + _leave(" [culled]"); +} diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index df6624c140be..a1ceef4f5cd0 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -472,7 +472,7 @@ void rxrpc_process_connection(struct work_struct *work) /* go through the conn-level event packets, releasing the ref on this * connection that each one has when we've finished with it */ while ((skb = skb_dequeue(&conn->rx_queue))) { - rxrpc_see_skb(skb, rxrpc_skb_rx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); ret = rxrpc_process_event(conn, skb, &abort_code); switch (ret) { case -EPROTO: @@ -484,7 +484,7 @@ void rxrpc_process_connection(struct work_struct *work) goto requeue_and_leave; case -ECONNABORTED: default: - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); break; } } @@ -501,6 +501,6 @@ requeue_and_leave: protocol_error: if (rxrpc_abort_connection(conn, ret, abort_code) < 0) goto requeue_and_leave; - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); goto out; } diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 434ef392212b..ed05b6922132 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -398,7 +398,7 @@ void rxrpc_service_connection_reaper(struct work_struct *work) if (conn->state == RXRPC_CONN_SERVICE_PREALLOC) continue; - if (rxnet->live) { + if (rxnet->live && !conn->params.local->dead) { idle_timestamp = READ_ONCE(conn->idle_timestamp); expire_at = idle_timestamp + rxrpc_connection_expiry * HZ; if (conn->params.local->service_closed) diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 5bd6f1546e5c..d122c53c8697 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -196,15 +196,14 @@ send_extra_data: * Ping the other end to fill our RTT cache and to retrieve the rwind * and MTU parameters. */ -static void rxrpc_send_ping(struct rxrpc_call *call, struct sk_buff *skb, - int skew) +static void rxrpc_send_ping(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); ktime_t now = skb->tstamp; if (call->peer->rtt_usage < 3 || ktime_before(ktime_add_ms(call->peer->rtt_last_req, 1000), now)) - rxrpc_propose_ACK(call, RXRPC_ACK_PING, skew, sp->hdr.serial, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, sp->hdr.serial, true, true, rxrpc_propose_ack_ping_for_params); } @@ -234,7 +233,7 @@ static bool rxrpc_rotate_tx_window(struct rxrpc_call *call, rxrpc_seq_t to, ix = call->tx_hard_ack & RXRPC_RXTX_BUFF_MASK; skb = call->rxtx_buffer[ix]; annotation = call->rxtx_annotations[ix]; - rxrpc_see_skb(skb, rxrpc_skb_tx_rotated); + rxrpc_see_skb(skb, rxrpc_skb_rotated); call->rxtx_buffer[ix] = NULL; call->rxtx_annotations[ix] = 0; skb->next = list; @@ -259,7 +258,7 @@ static bool rxrpc_rotate_tx_window(struct rxrpc_call *call, rxrpc_seq_t to, skb = list; list = skb->next; skb_mark_not_on_list(skb); - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); } return rot_last; @@ -348,7 +347,7 @@ static bool rxrpc_receiving_reply(struct rxrpc_call *call) } /* - * Scan a jumbo packet to validate its structure and to work out how many + * Scan a data packet to validate its structure and to work out how many * subpackets it contains. * * A jumbo packet is a collection of consecutive packets glued together with @@ -359,16 +358,21 @@ static bool rxrpc_receiving_reply(struct rxrpc_call *call) * the last are RXRPC_JUMBO_DATALEN in size. The last subpacket may be of any * size. */ -static bool rxrpc_validate_jumbo(struct sk_buff *skb) +static bool rxrpc_validate_data(struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); unsigned int offset = sizeof(struct rxrpc_wire_header); unsigned int len = skb->len; - int nr_jumbo = 1; u8 flags = sp->hdr.flags; - do { - nr_jumbo++; + for (;;) { + if (flags & RXRPC_REQUEST_ACK) + __set_bit(sp->nr_subpackets, sp->rx_req_ack); + sp->nr_subpackets++; + + if (!(flags & RXRPC_JUMBO_PACKET)) + break; + if (len - offset < RXRPC_JUMBO_SUBPKTLEN) goto protocol_error; if (flags & RXRPC_LAST_PACKET) @@ -377,9 +381,10 @@ static bool rxrpc_validate_jumbo(struct sk_buff *skb) if (skb_copy_bits(skb, offset, &flags, 1) < 0) goto protocol_error; offset += sizeof(struct rxrpc_jumbo_header); - } while (flags & RXRPC_JUMBO_PACKET); + } - sp->nr_jumbo = nr_jumbo; + if (flags & RXRPC_LAST_PACKET) + sp->rx_flags |= RXRPC_SKB_INCL_LAST; return true; protocol_error: @@ -400,10 +405,10 @@ protocol_error: * (that information is encoded in the ACK packet). */ static void rxrpc_input_dup_data(struct rxrpc_call *call, rxrpc_seq_t seq, - u8 annotation, bool *_jumbo_bad) + bool is_jumbo, bool *_jumbo_bad) { /* Discard normal packets that are duplicates. */ - if (annotation == 0) + if (is_jumbo) return; /* Skip jumbo subpackets that are duplicates. When we've had three or @@ -417,30 +422,30 @@ static void rxrpc_input_dup_data(struct rxrpc_call *call, rxrpc_seq_t seq, } /* - * Process a DATA packet, adding the packet to the Rx ring. + * Process a DATA packet, adding the packet to the Rx ring. The caller's + * packet ref must be passed on or discarded. */ -static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb, - u16 skew) +static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); enum rxrpc_call_state state; - unsigned int offset = sizeof(struct rxrpc_wire_header); - unsigned int ix; + unsigned int j; rxrpc_serial_t serial = sp->hdr.serial, ack_serial = 0; - rxrpc_seq_t seq = sp->hdr.seq, hard_ack; - bool immediate_ack = false, jumbo_bad = false, queued; - u16 len; - u8 ack = 0, flags, annotation = 0; + rxrpc_seq_t seq0 = sp->hdr.seq, hard_ack; + bool immediate_ack = false, jumbo_bad = false; + u8 ack = 0; _enter("{%u,%u},{%u,%u}", - call->rx_hard_ack, call->rx_top, skb->len, seq); + call->rx_hard_ack, call->rx_top, skb->len, seq0); - _proto("Rx DATA %%%u { #%u f=%02x }", - sp->hdr.serial, seq, sp->hdr.flags); + _proto("Rx DATA %%%u { #%u f=%02x n=%u }", + sp->hdr.serial, seq0, sp->hdr.flags, sp->nr_subpackets); state = READ_ONCE(call->state); - if (state >= RXRPC_CALL_COMPLETE) + if (state >= RXRPC_CALL_COMPLETE) { + rxrpc_free_skb(skb, rxrpc_skb_freed); return; + } if (call->state == RXRPC_CALL_SERVER_RECV_REQUEST) { unsigned long timo = READ_ONCE(call->next_req_timo); @@ -465,156 +470,157 @@ static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb, !rxrpc_receiving_reply(call)) goto unlock; - call->ackr_prev_seq = seq; - + call->ackr_prev_seq = seq0; hard_ack = READ_ONCE(call->rx_hard_ack); - if (after(seq, hard_ack + call->rx_winsize)) { - ack = RXRPC_ACK_EXCEEDS_WINDOW; - ack_serial = serial; - goto ack; - } - flags = sp->hdr.flags; - if (flags & RXRPC_JUMBO_PACKET) { + if (sp->nr_subpackets > 1) { if (call->nr_jumbo_bad > 3) { ack = RXRPC_ACK_NOSPACE; ack_serial = serial; goto ack; } - annotation = 1; } -next_subpacket: - queued = false; - ix = seq & RXRPC_RXTX_BUFF_MASK; - len = skb->len; - if (flags & RXRPC_JUMBO_PACKET) - len = RXRPC_JUMBO_DATALEN; - - if (flags & RXRPC_LAST_PACKET) { - if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) && - seq != call->rx_top) { - rxrpc_proto_abort("LSN", call, seq); - goto unlock; - } - } else { - if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) && - after_eq(seq, call->rx_top)) { - rxrpc_proto_abort("LSA", call, seq); - goto unlock; + for (j = 0; j < sp->nr_subpackets; j++) { + rxrpc_serial_t serial = sp->hdr.serial + j; + rxrpc_seq_t seq = seq0 + j; + unsigned int ix = seq & RXRPC_RXTX_BUFF_MASK; + bool terminal = (j == sp->nr_subpackets - 1); + bool last = terminal && (sp->rx_flags & RXRPC_SKB_INCL_LAST); + u8 flags, annotation = j; + + _proto("Rx DATA+%u %%%u { #%x t=%u l=%u }", + j, serial, seq, terminal, last); + + if (last) { + if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) && + seq != call->rx_top) { + rxrpc_proto_abort("LSN", call, seq); + goto unlock; + } + } else { + if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) && + after_eq(seq, call->rx_top)) { + rxrpc_proto_abort("LSA", call, seq); + goto unlock; + } } - } - - trace_rxrpc_rx_data(call->debug_id, seq, serial, flags, annotation); - if (before_eq(seq, hard_ack)) { - ack = RXRPC_ACK_DUPLICATE; - ack_serial = serial; - goto skip; - } - if (flags & RXRPC_REQUEST_ACK && !ack) { - ack = RXRPC_ACK_REQUESTED; - ack_serial = serial; - } + flags = 0; + if (last) + flags |= RXRPC_LAST_PACKET; + if (!terminal) + flags |= RXRPC_JUMBO_PACKET; + if (test_bit(j, sp->rx_req_ack)) + flags |= RXRPC_REQUEST_ACK; + trace_rxrpc_rx_data(call->debug_id, seq, serial, flags, annotation); - if (call->rxtx_buffer[ix]) { - rxrpc_input_dup_data(call, seq, annotation, &jumbo_bad); - if (ack != RXRPC_ACK_DUPLICATE) { + if (before_eq(seq, hard_ack)) { ack = RXRPC_ACK_DUPLICATE; ack_serial = serial; + continue; } - immediate_ack = true; - goto skip; - } - - /* Queue the packet. We use a couple of memory barriers here as need - * to make sure that rx_top is perceived to be set after the buffer - * pointer and that the buffer pointer is set after the annotation and - * the skb data. - * - * Barriers against rxrpc_recvmsg_data() and rxrpc_rotate_rx_window() - * and also rxrpc_fill_out_ack(). - */ - rxrpc_get_skb(skb, rxrpc_skb_rx_got); - call->rxtx_annotations[ix] = annotation; - smp_wmb(); - call->rxtx_buffer[ix] = skb; - if (after(seq, call->rx_top)) { - smp_store_release(&call->rx_top, seq); - } else if (before(seq, call->rx_top)) { - /* Send an immediate ACK if we fill in a hole */ - if (!ack) { - ack = RXRPC_ACK_DELAY; - ack_serial = serial; - } - immediate_ack = true; - } - if (flags & RXRPC_LAST_PACKET) { - set_bit(RXRPC_CALL_RX_LAST, &call->flags); - trace_rxrpc_receive(call, rxrpc_receive_queue_last, serial, seq); - } else { - trace_rxrpc_receive(call, rxrpc_receive_queue, serial, seq); - } - queued = true; - if (after_eq(seq, call->rx_expect_next)) { - if (after(seq, call->rx_expect_next)) { - _net("OOS %u > %u", seq, call->rx_expect_next); - ack = RXRPC_ACK_OUT_OF_SEQUENCE; - ack_serial = serial; + if (call->rxtx_buffer[ix]) { + rxrpc_input_dup_data(call, seq, sp->nr_subpackets > 1, + &jumbo_bad); + if (ack != RXRPC_ACK_DUPLICATE) { + ack = RXRPC_ACK_DUPLICATE; + ack_serial = serial; + } + immediate_ack = true; + continue; } - call->rx_expect_next = seq + 1; - } -skip: - offset += len; - if (flags & RXRPC_JUMBO_PACKET) { - if (skb_copy_bits(skb, offset, &flags, 1) < 0) { - rxrpc_proto_abort("XJF", call, seq); - goto unlock; - } - offset += sizeof(struct rxrpc_jumbo_header); - seq++; - serial++; - annotation++; - if (flags & RXRPC_JUMBO_PACKET) - annotation |= RXRPC_RX_ANNO_JLAST; if (after(seq, hard_ack + call->rx_winsize)) { ack = RXRPC_ACK_EXCEEDS_WINDOW; ack_serial = serial; - if (!jumbo_bad) { - call->nr_jumbo_bad++; - jumbo_bad = true; + if (flags & RXRPC_JUMBO_PACKET) { + if (!jumbo_bad) { + call->nr_jumbo_bad++; + jumbo_bad = true; + } } + goto ack; } - _proto("Rx DATA Jumbo %%%u", serial); - goto next_subpacket; - } + if (flags & RXRPC_REQUEST_ACK && !ack) { + ack = RXRPC_ACK_REQUESTED; + ack_serial = serial; + } + + /* Queue the packet. We use a couple of memory barriers here as need + * to make sure that rx_top is perceived to be set after the buffer + * pointer and that the buffer pointer is set after the annotation and + * the skb data. + * + * Barriers against rxrpc_recvmsg_data() and rxrpc_rotate_rx_window() + * and also rxrpc_fill_out_ack(). + */ + if (!terminal) + rxrpc_get_skb(skb, rxrpc_skb_got); + call->rxtx_annotations[ix] = annotation; + smp_wmb(); + call->rxtx_buffer[ix] = skb; + if (after(seq, call->rx_top)) { + smp_store_release(&call->rx_top, seq); + } else if (before(seq, call->rx_top)) { + /* Send an immediate ACK if we fill in a hole */ + if (!ack) { + ack = RXRPC_ACK_DELAY; + ack_serial = serial; + } + immediate_ack = true; + } - if (queued && flags & RXRPC_LAST_PACKET && !ack) { - ack = RXRPC_ACK_DELAY; - ack_serial = serial; + if (terminal) { + /* From this point on, we're not allowed to touch the + * packet any longer as its ref now belongs to the Rx + * ring. + */ + skb = NULL; + } + + if (last) { + set_bit(RXRPC_CALL_RX_LAST, &call->flags); + if (!ack) { + ack = RXRPC_ACK_DELAY; + ack_serial = serial; + } + trace_rxrpc_receive(call, rxrpc_receive_queue_last, serial, seq); + } else { + trace_rxrpc_receive(call, rxrpc_receive_queue, serial, seq); + } + + if (after_eq(seq, call->rx_expect_next)) { + if (after(seq, call->rx_expect_next)) { + _net("OOS %u > %u", seq, call->rx_expect_next); + ack = RXRPC_ACK_OUT_OF_SEQUENCE; + ack_serial = serial; + } + call->rx_expect_next = seq + 1; + } } ack: if (ack) - rxrpc_propose_ACK(call, ack, skew, ack_serial, + rxrpc_propose_ACK(call, ack, ack_serial, immediate_ack, true, rxrpc_propose_ack_input_data); else - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, skew, serial, + rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, serial, false, true, rxrpc_propose_ack_input_data); - if (sp->hdr.seq == READ_ONCE(call->rx_hard_ack) + 1) { + if (seq0 == READ_ONCE(call->rx_hard_ack) + 1) { trace_rxrpc_notify_socket(call->debug_id, serial); rxrpc_notify_socket(call); } unlock: spin_unlock(&call->input_lock); + rxrpc_free_skb(skb, rxrpc_skb_freed); _leave(" [queued]"); } @@ -822,8 +828,7 @@ static void rxrpc_input_soft_acks(struct rxrpc_call *call, u8 *acks, * soft-ACK means that the packet may be discarded and retransmission * requested. A phase is complete when all packets are hard-ACK'd. */ -static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb, - u16 skew) +static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb) { struct rxrpc_ack_summary summary = { 0 }; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); @@ -867,11 +872,11 @@ static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb, if (buf.ack.reason == RXRPC_ACK_PING) { _proto("Rx ACK %%%u PING Request", sp->hdr.serial); rxrpc_propose_ACK(call, RXRPC_ACK_PING_RESPONSE, - skew, sp->hdr.serial, true, true, + sp->hdr.serial, true, true, rxrpc_propose_ack_respond_to_ping); } else if (sp->hdr.flags & RXRPC_REQUEST_ACK) { rxrpc_propose_ACK(call, RXRPC_ACK_REQUESTED, - skew, sp->hdr.serial, true, true, + sp->hdr.serial, true, true, rxrpc_propose_ack_respond_to_ack); } @@ -948,7 +953,7 @@ static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb, RXRPC_TX_ANNO_LAST && summary.nr_acks == call->tx_top - hard_ack && rxrpc_is_client_call(call)) - rxrpc_propose_ACK(call, RXRPC_ACK_PING, skew, sp->hdr.serial, + rxrpc_propose_ACK(call, RXRPC_ACK_PING, sp->hdr.serial, false, true, rxrpc_propose_ack_ping_for_lost_reply); @@ -1004,7 +1009,7 @@ static void rxrpc_input_abort(struct rxrpc_call *call, struct sk_buff *skb) * Process an incoming call packet. */ static void rxrpc_input_call_packet(struct rxrpc_call *call, - struct sk_buff *skb, u16 skew) + struct sk_buff *skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); unsigned long timo; @@ -1023,11 +1028,11 @@ static void rxrpc_input_call_packet(struct rxrpc_call *call, switch (sp->hdr.type) { case RXRPC_PACKET_TYPE_DATA: - rxrpc_input_data(call, skb, skew); - break; + rxrpc_input_data(call, skb); + goto no_free; case RXRPC_PACKET_TYPE_ACK: - rxrpc_input_ack(call, skb, skew); + rxrpc_input_ack(call, skb); break; case RXRPC_PACKET_TYPE_BUSY: @@ -1051,6 +1056,8 @@ static void rxrpc_input_call_packet(struct rxrpc_call *call, break; } + rxrpc_free_skb(skb, rxrpc_skb_freed); +no_free: _leave(""); } @@ -1108,8 +1115,12 @@ static void rxrpc_post_packet_to_local(struct rxrpc_local *local, { _enter("%p,%p", local, skb); - skb_queue_tail(&local->event_queue, skb); - rxrpc_queue_local(local); + if (rxrpc_get_local_maybe(local)) { + skb_queue_tail(&local->event_queue, skb); + rxrpc_queue_local(local); + } else { + rxrpc_free_skb(skb, rxrpc_skb_freed); + } } /* @@ -1119,8 +1130,12 @@ static void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb) { CHECK_SLAB_OKAY(&local->usage); - skb_queue_tail(&local->reject_queue, skb); - rxrpc_queue_local(local); + if (rxrpc_get_local_maybe(local)) { + skb_queue_tail(&local->reject_queue, skb); + rxrpc_queue_local(local); + } else { + rxrpc_free_skb(skb, rxrpc_skb_freed); + } } /* @@ -1173,7 +1188,6 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) struct rxrpc_peer *peer = NULL; struct rxrpc_sock *rx = NULL; unsigned int channel; - int skew = 0; _enter("%p", udp_sk); @@ -1184,7 +1198,7 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) if (skb->tstamp == 0) skb->tstamp = ktime_get_real(); - rxrpc_new_skb(skb, rxrpc_skb_rx_received); + rxrpc_new_skb(skb, rxrpc_skb_received); skb_pull(skb, sizeof(struct udphdr)); @@ -1201,7 +1215,7 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) static int lose; if ((lose++ & 7) == 7) { trace_rxrpc_rx_lose(sp); - rxrpc_free_skb(skb, rxrpc_skb_rx_lost); + rxrpc_free_skb(skb, rxrpc_skb_lost); return 0; } } @@ -1233,9 +1247,26 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) if (sp->hdr.callNumber == 0 || sp->hdr.seq == 0) goto bad_message; - if (sp->hdr.flags & RXRPC_JUMBO_PACKET && - !rxrpc_validate_jumbo(skb)) + if (!rxrpc_validate_data(skb)) goto bad_message; + + /* Unshare the packet so that it can be modified for in-place + * decryption. + */ + if (sp->hdr.securityIndex != 0) { + struct sk_buff *nskb = skb_unshare(skb, GFP_ATOMIC); + if (!nskb) { + rxrpc_eaten_skb(skb, rxrpc_skb_unshared_nomem); + goto out; + } + + if (nskb != skb) { + rxrpc_eaten_skb(skb, rxrpc_skb_received); + rxrpc_new_skb(skb, rxrpc_skb_unshared); + skb = nskb; + sp = rxrpc_skb(skb); + } + } break; case RXRPC_PACKET_TYPE_CHALLENGE: @@ -1301,15 +1332,8 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) goto out; } - /* Note the serial number skew here */ - skew = (int)sp->hdr.serial - (int)conn->hi_serial; - if (skew >= 0) { - if (skew > 0) - conn->hi_serial = sp->hdr.serial; - } else { - skew = -skew; - skew = min(skew, 65535); - } + if ((int)sp->hdr.serial - (int)conn->hi_serial > 0) + conn->hi_serial = sp->hdr.serial; /* Call-bound packets are routed by connection channel. */ channel = sp->hdr.cid & RXRPC_CHANNELMASK; @@ -1372,15 +1396,18 @@ int rxrpc_input_packet(struct sock *udp_sk, struct sk_buff *skb) call = rxrpc_new_incoming_call(local, rx, skb); if (!call) goto reject_packet; - rxrpc_send_ping(call, skb, skew); + rxrpc_send_ping(call, skb); mutex_unlock(&call->user_mutex); } - rxrpc_input_call_packet(call, skb, skew); - goto discard; + /* Process a call packet; this either discards or passes on the ref + * elsewhere. + */ + rxrpc_input_call_packet(call, skb); + goto out; discard: - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); out: trace_rxrpc_rx_done(0, 0); return 0; diff --git a/net/rxrpc/local_event.c b/net/rxrpc/local_event.c index e93a78f7c05e..3ce6d628cd75 100644 --- a/net/rxrpc/local_event.c +++ b/net/rxrpc/local_event.c @@ -90,7 +90,7 @@ void rxrpc_process_local_events(struct rxrpc_local *local) if (skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - rxrpc_see_skb(skb, rxrpc_skb_rx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); _debug("{%d},{%u}", local->debug_id, sp->hdr.type); switch (sp->hdr.type) { @@ -108,7 +108,7 @@ void rxrpc_process_local_events(struct rxrpc_local *local) break; } - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); } _leave(""); diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index b1c71bad510b..36587260cabd 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -79,6 +79,7 @@ static struct rxrpc_local *rxrpc_alloc_local(struct rxrpc_net *rxnet, local = kzalloc(sizeof(struct rxrpc_local), GFP_KERNEL); if (local) { atomic_set(&local->usage, 1); + atomic_set(&local->active_users, 1); local->rxnet = rxnet; INIT_LIST_HEAD(&local->link); INIT_WORK(&local->processor, rxrpc_local_processor); @@ -92,7 +93,7 @@ static struct rxrpc_local *rxrpc_alloc_local(struct rxrpc_net *rxnet, local->debug_id = atomic_inc_return(&rxrpc_debug_id); memcpy(&local->srx, srx, sizeof(*srx)); local->srx.srx_service = 0; - trace_rxrpc_local(local, rxrpc_local_new, 1, NULL); + trace_rxrpc_local(local->debug_id, rxrpc_local_new, 1, NULL); } _leave(" = %p", local); @@ -266,11 +267,8 @@ struct rxrpc_local *rxrpc_lookup_local(struct net *net, * bind the transport socket may still fail if we're attempting * to use a local address that the dying object is still using. */ - if (!rxrpc_get_local_maybe(local)) { - cursor = cursor->next; - list_del_init(&local->link); + if (!rxrpc_use_local(local)) break; - } age = "old"; goto found; @@ -284,7 +282,10 @@ struct rxrpc_local *rxrpc_lookup_local(struct net *net, if (ret < 0) goto sock_error; - list_add_tail(&local->link, cursor); + if (cursor != &rxnet->local_endpoints) + list_replace_init(cursor, &local->link); + else + list_add_tail(&local->link, cursor); age = "new"; found: @@ -320,7 +321,7 @@ struct rxrpc_local *rxrpc_get_local(struct rxrpc_local *local) int n; n = atomic_inc_return(&local->usage); - trace_rxrpc_local(local, rxrpc_local_got, n, here); + trace_rxrpc_local(local->debug_id, rxrpc_local_got, n, here); return local; } @@ -334,7 +335,8 @@ struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local) if (local) { int n = atomic_fetch_add_unless(&local->usage, 1, 0); if (n > 0) - trace_rxrpc_local(local, rxrpc_local_got, n + 1, here); + trace_rxrpc_local(local->debug_id, rxrpc_local_got, + n + 1, here); else local = NULL; } @@ -342,24 +344,18 @@ struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local) } /* - * Queue a local endpoint. + * Queue a local endpoint and pass the caller's reference to the work item. */ void rxrpc_queue_local(struct rxrpc_local *local) { const void *here = __builtin_return_address(0); + unsigned int debug_id = local->debug_id; + int n = atomic_read(&local->usage); if (rxrpc_queue_work(&local->processor)) - trace_rxrpc_local(local, rxrpc_local_queued, - atomic_read(&local->usage), here); -} - -/* - * A local endpoint reached its end of life. - */ -static void __rxrpc_put_local(struct rxrpc_local *local) -{ - _enter("%d", local->debug_id); - rxrpc_queue_work(&local->processor); + trace_rxrpc_local(debug_id, rxrpc_local_queued, n, here); + else + rxrpc_put_local(local); } /* @@ -372,10 +368,47 @@ void rxrpc_put_local(struct rxrpc_local *local) if (local) { n = atomic_dec_return(&local->usage); - trace_rxrpc_local(local, rxrpc_local_put, n, here); + trace_rxrpc_local(local->debug_id, rxrpc_local_put, n, here); if (n == 0) - __rxrpc_put_local(local); + call_rcu(&local->rcu, rxrpc_local_rcu); + } +} + +/* + * Start using a local endpoint. + */ +struct rxrpc_local *rxrpc_use_local(struct rxrpc_local *local) +{ + unsigned int au; + + local = rxrpc_get_local_maybe(local); + if (!local) + return NULL; + + au = atomic_fetch_add_unless(&local->active_users, 1, 0); + if (au == 0) { + rxrpc_put_local(local); + return NULL; + } + + return local; +} + +/* + * Cease using a local endpoint. Once the number of active users reaches 0, we + * start the closure of the transport in the work processor. + */ +void rxrpc_unuse_local(struct rxrpc_local *local) +{ + unsigned int au; + + if (local) { + au = atomic_dec_return(&local->active_users); + if (au == 0) + rxrpc_queue_local(local); + else + rxrpc_put_local(local); } } @@ -393,21 +426,14 @@ static void rxrpc_local_destroyer(struct rxrpc_local *local) _enter("%d", local->debug_id); - /* We can get a race between an incoming call packet queueing the - * processor again and the work processor starting the destruction - * process which will shut down the UDP socket. - */ - if (local->dead) { - _leave(" [already dead]"); - return; - } local->dead = true; mutex_lock(&rxnet->local_mutex); list_del_init(&local->link); mutex_unlock(&rxnet->local_mutex); - ASSERT(RB_EMPTY_ROOT(&local->client_conns)); + rxrpc_clean_up_local_conns(local); + rxrpc_service_connection_reaper(&rxnet->service_conn_reaper); ASSERT(!local->service); if (socket) { @@ -422,13 +448,11 @@ static void rxrpc_local_destroyer(struct rxrpc_local *local) */ rxrpc_purge_queue(&local->reject_queue); rxrpc_purge_queue(&local->event_queue); - - _debug("rcu local %d", local->debug_id); - call_rcu(&local->rcu, rxrpc_local_rcu); } /* - * Process events on an endpoint + * Process events on an endpoint. The work item carries a ref which + * we must release. */ static void rxrpc_local_processor(struct work_struct *work) { @@ -436,13 +460,15 @@ static void rxrpc_local_processor(struct work_struct *work) container_of(work, struct rxrpc_local, processor); bool again; - trace_rxrpc_local(local, rxrpc_local_processing, + trace_rxrpc_local(local->debug_id, rxrpc_local_processing, atomic_read(&local->usage), NULL); do { again = false; - if (atomic_read(&local->usage) == 0) - return rxrpc_local_destroyer(local); + if (atomic_read(&local->active_users) == 0) { + rxrpc_local_destroyer(local); + break; + } if (!skb_queue_empty(&local->reject_queue)) { rxrpc_reject_packets(local); @@ -454,6 +480,8 @@ static void rxrpc_local_processor(struct work_struct *work) again = true; } } while (again); + + rxrpc_put_local(local); } /* diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 948e3fe249ec..935bb60fff56 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -87,7 +87,7 @@ static size_t rxrpc_fill_out_ack(struct rxrpc_connection *conn, *_top = top; pkt->ack.bufferSpace = htons(8); - pkt->ack.maxSkew = htons(call->ackr_skew); + pkt->ack.maxSkew = htons(0); pkt->ack.firstPacket = htonl(hard_ack + 1); pkt->ack.previousPacket = htonl(call->ackr_prev_seq); pkt->ack.serial = htonl(serial); @@ -228,7 +228,6 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, if (ping) clear_bit(RXRPC_CALL_PINGING, &call->flags); rxrpc_propose_ACK(call, pkt->ack.reason, - ntohs(pkt->ack.maxSkew), ntohl(pkt->ack.serial), false, true, rxrpc_propose_ack_retry_tx); @@ -566,7 +565,7 @@ void rxrpc_reject_packets(struct rxrpc_local *local) memset(&whdr, 0, sizeof(whdr)); while ((skb = skb_dequeue(&local->reject_queue))) { - rxrpc_see_skb(skb, rxrpc_skb_rx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); sp = rxrpc_skb(skb); switch (skb->mark) { @@ -582,7 +581,7 @@ void rxrpc_reject_packets(struct rxrpc_local *local) ioc = 2; break; default: - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); continue; } @@ -607,7 +606,7 @@ void rxrpc_reject_packets(struct rxrpc_local *local) rxrpc_tx_point_reject); } - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); } _leave(""); diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 7666ec72d37e..c97ebdc043e4 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -163,11 +163,11 @@ void rxrpc_error_report(struct sock *sk) _leave("UDP socket errqueue empty"); return; } - rxrpc_new_skb(skb, rxrpc_skb_rx_received); + rxrpc_new_skb(skb, rxrpc_skb_received); serr = SKB_EXT_ERR(skb); if (!skb->len && serr->ee.ee_origin == SO_EE_ORIGIN_TIMESTAMPING) { _leave("UDP empty message"); - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); return; } @@ -177,7 +177,7 @@ void rxrpc_error_report(struct sock *sk) peer = NULL; if (!peer) { rcu_read_unlock(); - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); _leave(" [no peer]"); return; } @@ -189,7 +189,7 @@ void rxrpc_error_report(struct sock *sk) serr->ee.ee_code == ICMP_FRAG_NEEDED)) { rxrpc_adjust_mtu(peer, serr); rcu_read_unlock(); - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); rxrpc_put_peer(peer); _leave(" [MTU update]"); return; @@ -197,7 +197,7 @@ void rxrpc_error_report(struct sock *sk) rxrpc_store_error(peer, serr); rcu_read_unlock(); - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); rxrpc_put_peer(peer); _leave(""); diff --git a/net/rxrpc/protocol.h b/net/rxrpc/protocol.h index 99ce322d7caa..49bb972539aa 100644 --- a/net/rxrpc/protocol.h +++ b/net/rxrpc/protocol.h @@ -89,6 +89,15 @@ struct rxrpc_jumbo_header { #define RXRPC_JUMBO_DATALEN 1412 /* non-terminal jumbo packet data length */ #define RXRPC_JUMBO_SUBPKTLEN (RXRPC_JUMBO_DATALEN + sizeof(struct rxrpc_jumbo_header)) +/* + * The maximum number of subpackets that can possibly fit in a UDP packet is: + * + * ((max_IP - IP_hdr - UDP_hdr) / RXRPC_JUMBO_SUBPKTLEN) + 1 + * = ((65535 - 28 - 28) / 1416) + 1 + * = 46 non-terminal packets and 1 terminal packet. + */ +#define RXRPC_MAX_NR_JUMBO 47 + /*****************************************************************************/ /* * on-the-wire Rx ACK packet data payload diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index 5abf46cf9e6c..3b0becb12041 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -141,7 +141,7 @@ static void rxrpc_end_rx_phase(struct rxrpc_call *call, rxrpc_serial_t serial) ASSERTCMP(call->rx_hard_ack, ==, call->rx_top); if (call->state == RXRPC_CALL_CLIENT_RECV_REPLY) { - rxrpc_propose_ACK(call, RXRPC_ACK_IDLE, 0, serial, false, true, + rxrpc_propose_ACK(call, RXRPC_ACK_IDLE, serial, false, true, rxrpc_propose_ack_terminal_ack); //rxrpc_send_ack_packet(call, false, NULL); } @@ -159,7 +159,7 @@ static void rxrpc_end_rx_phase(struct rxrpc_call *call, rxrpc_serial_t serial) call->state = RXRPC_CALL_SERVER_ACK_REQUEST; call->expect_req_by = jiffies + MAX_JIFFY_OFFSET; write_unlock_bh(&call->state_lock); - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, 0, serial, false, true, + rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, serial, false, true, rxrpc_propose_ack_processing_op); break; default: @@ -177,7 +177,8 @@ static void rxrpc_rotate_rx_window(struct rxrpc_call *call) struct sk_buff *skb; rxrpc_serial_t serial; rxrpc_seq_t hard_ack, top; - u8 flags; + bool last = false; + u8 subpacket; int ix; _enter("%d", call->debug_id); @@ -189,30 +190,32 @@ static void rxrpc_rotate_rx_window(struct rxrpc_call *call) hard_ack++; ix = hard_ack & RXRPC_RXTX_BUFF_MASK; skb = call->rxtx_buffer[ix]; - rxrpc_see_skb(skb, rxrpc_skb_rx_rotated); + rxrpc_see_skb(skb, rxrpc_skb_rotated); sp = rxrpc_skb(skb); - flags = sp->hdr.flags; - serial = sp->hdr.serial; - if (call->rxtx_annotations[ix] & RXRPC_RX_ANNO_JUMBO) - serial += (call->rxtx_annotations[ix] & RXRPC_RX_ANNO_JUMBO) - 1; + + subpacket = call->rxtx_annotations[ix] & RXRPC_RX_ANNO_SUBPACKET; + serial = sp->hdr.serial + subpacket; + + if (subpacket == sp->nr_subpackets - 1 && + sp->rx_flags & RXRPC_SKB_INCL_LAST) + last = true; call->rxtx_buffer[ix] = NULL; call->rxtx_annotations[ix] = 0; /* Barrier against rxrpc_input_data(). */ smp_store_release(&call->rx_hard_ack, hard_ack); - rxrpc_free_skb(skb, rxrpc_skb_rx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); - _debug("%u,%u,%02x", hard_ack, top, flags); trace_rxrpc_receive(call, rxrpc_receive_rotate, serial, hard_ack); - if (flags & RXRPC_LAST_PACKET) { + if (last) { rxrpc_end_rx_phase(call, serial); } else { /* Check to see if there's an ACK that needs sending. */ if (after_eq(hard_ack, call->ackr_consumed + 2) || after_eq(top, call->ackr_seen + 2) || (hard_ack == top && after(hard_ack, call->ackr_consumed))) - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, 0, serial, + rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, serial, true, true, rxrpc_propose_ack_rotate_rx); if (call->ackr_reason && call->ackr_reason != RXRPC_ACK_DELAY) @@ -233,18 +236,19 @@ static int rxrpc_verify_packet(struct rxrpc_call *call, struct sk_buff *skb, struct rxrpc_skb_priv *sp = rxrpc_skb(skb); rxrpc_seq_t seq = sp->hdr.seq; u16 cksum = sp->hdr.cksum; + u8 subpacket = annotation & RXRPC_RX_ANNO_SUBPACKET; _enter(""); /* For all but the head jumbo subpacket, the security checksum is in a * jumbo header immediately prior to the data. */ - if ((annotation & RXRPC_RX_ANNO_JUMBO) > 1) { + if (subpacket > 0) { __be16 tmp; if (skb_copy_bits(skb, offset - 2, &tmp, 2) < 0) BUG(); cksum = ntohs(tmp); - seq += (annotation & RXRPC_RX_ANNO_JUMBO) - 1; + seq += subpacket; } return call->conn->security->verify_packet(call, skb, offset, len, @@ -265,19 +269,18 @@ static int rxrpc_locate_data(struct rxrpc_call *call, struct sk_buff *skb, u8 *_annotation, unsigned int *_offset, unsigned int *_len) { + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); unsigned int offset = sizeof(struct rxrpc_wire_header); unsigned int len; int ret; u8 annotation = *_annotation; + u8 subpacket = annotation & RXRPC_RX_ANNO_SUBPACKET; /* Locate the subpacket */ + offset += subpacket * RXRPC_JUMBO_SUBPKTLEN; len = skb->len - offset; - if ((annotation & RXRPC_RX_ANNO_JUMBO) > 0) { - offset += (((annotation & RXRPC_RX_ANNO_JUMBO) - 1) * - RXRPC_JUMBO_SUBPKTLEN); - len = (annotation & RXRPC_RX_ANNO_JLAST) ? - skb->len - offset : RXRPC_JUMBO_SUBPKTLEN; - } + if (subpacket < sp->nr_subpackets - 1) + len = RXRPC_JUMBO_DATALEN; if (!(annotation & RXRPC_RX_ANNO_VERIFIED)) { ret = rxrpc_verify_packet(call, skb, annotation, offset, len); @@ -303,6 +306,7 @@ static int rxrpc_recvmsg_data(struct socket *sock, struct rxrpc_call *call, { struct rxrpc_skb_priv *sp; struct sk_buff *skb; + rxrpc_serial_t serial; rxrpc_seq_t hard_ack, top, seq; size_t remain; bool last; @@ -336,12 +340,15 @@ static int rxrpc_recvmsg_data(struct socket *sock, struct rxrpc_call *call, break; } smp_rmb(); - rxrpc_see_skb(skb, rxrpc_skb_rx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); sp = rxrpc_skb(skb); - if (!(flags & MSG_PEEK)) + if (!(flags & MSG_PEEK)) { + serial = sp->hdr.serial; + serial += call->rxtx_annotations[ix] & RXRPC_RX_ANNO_SUBPACKET; trace_rxrpc_receive(call, rxrpc_receive_front, - sp->hdr.serial, seq); + serial, seq); + } if (msg) sock_recv_timestamp(msg, sock->sk, skb); diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index dbb109da1835..8d8aa3c230b5 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -223,10 +223,8 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, struct rxrpc_skb_priv *sp; struct rxrpc_crypt iv; struct scatterlist sg[16]; - struct sk_buff *trailer; unsigned int len; u16 check; - int nsg; int err; sp = rxrpc_skb(skb); @@ -250,15 +248,14 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, crypto_skcipher_encrypt(req); /* we want to encrypt the skbuff in-place */ - nsg = skb_cow_data(skb, 0, &trailer); - err = -ENOMEM; - if (nsg < 0 || nsg > 16) + err = -EMSGSIZE; + if (skb_shinfo(skb)->nr_frags > 16) goto out; len = data_size + call->conn->size_align - 1; len &= ~(call->conn->size_align - 1); - sg_init_table(sg, nsg); + sg_init_table(sg, ARRAY_SIZE(sg)); err = skb_to_sgvec(skb, sg, 0, len); if (unlikely(err < 0)) goto out; @@ -359,11 +356,10 @@ static int rxkad_verify_packet_1(struct rxrpc_call *call, struct sk_buff *skb, struct rxkad_level1_hdr sechdr; struct rxrpc_crypt iv; struct scatterlist sg[16]; - struct sk_buff *trailer; bool aborted; u32 data_size, buf; u16 check; - int nsg, ret; + int ret; _enter(""); @@ -376,11 +372,7 @@ static int rxkad_verify_packet_1(struct rxrpc_call *call, struct sk_buff *skb, /* Decrypt the skbuff in-place. TODO: We really want to decrypt * directly into the target buffer. */ - nsg = skb_cow_data(skb, 0, &trailer); - if (nsg < 0 || nsg > 16) - goto nomem; - - sg_init_table(sg, nsg); + sg_init_table(sg, ARRAY_SIZE(sg)); ret = skb_to_sgvec(skb, sg, offset, 8); if (unlikely(ret < 0)) return ret; @@ -428,10 +420,6 @@ protocol_error: if (aborted) rxrpc_send_abort_packet(call); return -EPROTO; - -nomem: - _leave(" = -ENOMEM"); - return -ENOMEM; } /* @@ -446,7 +434,6 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, struct rxkad_level2_hdr sechdr; struct rxrpc_crypt iv; struct scatterlist _sg[4], *sg; - struct sk_buff *trailer; bool aborted; u32 data_size, buf; u16 check; @@ -463,12 +450,11 @@ static int rxkad_verify_packet_2(struct rxrpc_call *call, struct sk_buff *skb, /* Decrypt the skbuff in-place. TODO: We really want to decrypt * directly into the target buffer. */ - nsg = skb_cow_data(skb, 0, &trailer); - if (nsg < 0) - goto nomem; - sg = _sg; - if (unlikely(nsg > 4)) { + nsg = skb_shinfo(skb)->nr_frags; + if (nsg <= 4) { + nsg = 4; + } else { sg = kmalloc_array(nsg, sizeof(*sg), GFP_NOIO); if (!sg) goto nomem; diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index bae14438f869..6a1547b270fe 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -176,7 +176,7 @@ static int rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call, skb->tstamp = ktime_get_real(); ix = seq & RXRPC_RXTX_BUFF_MASK; - rxrpc_get_skb(skb, rxrpc_skb_tx_got); + rxrpc_get_skb(skb, rxrpc_skb_got); call->rxtx_annotations[ix] = annotation; smp_wmb(); call->rxtx_buffer[ix] = skb; @@ -248,7 +248,7 @@ static int rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call, } out: - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); _leave(" = %d", ret); return ret; } @@ -289,7 +289,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, skb = call->tx_pending; call->tx_pending = NULL; - rxrpc_see_skb(skb, rxrpc_skb_tx_seen); + rxrpc_see_skb(skb, rxrpc_skb_seen); copied = 0; do { @@ -336,7 +336,9 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, if (!skb) goto maybe_error; - rxrpc_new_skb(skb, rxrpc_skb_tx_new); + sp = rxrpc_skb(skb); + sp->rx_flags |= RXRPC_SKB_TX_BUFFER; + rxrpc_new_skb(skb, rxrpc_skb_new); _debug("ALLOC SEND %p", skb); @@ -346,7 +348,6 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, skb_reserve(skb, call->conn->security_size); skb->len += call->conn->security_size; - sp = rxrpc_skb(skb); sp->remain = chunk; if (sp->remain > skb_tailroom(skb)) sp->remain = skb_tailroom(skb); @@ -439,7 +440,7 @@ out: return ret; call_terminated: - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); + rxrpc_free_skb(skb, rxrpc_skb_freed); _leave(" = %d", call->error); return call->error; diff --git a/net/rxrpc/skbuff.c b/net/rxrpc/skbuff.c index 9ad5045b7c2f..0348d2bf6f7d 100644 --- a/net/rxrpc/skbuff.c +++ b/net/rxrpc/skbuff.c @@ -14,7 +14,8 @@ #include <net/af_rxrpc.h> #include "ar-internal.h" -#define select_skb_count(op) (op >= rxrpc_skb_tx_cleaned ? &rxrpc_n_tx_skbs : &rxrpc_n_rx_skbs) +#define is_tx_skb(skb) (rxrpc_skb(skb)->rx_flags & RXRPC_SKB_TX_BUFFER) +#define select_skb_count(skb) (is_tx_skb(skb) ? &rxrpc_n_tx_skbs : &rxrpc_n_rx_skbs) /* * Note the allocation or reception of a socket buffer. @@ -22,8 +23,9 @@ void rxrpc_new_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) { const void *here = __builtin_return_address(0); - int n = atomic_inc_return(select_skb_count(op)); - trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, here); + int n = atomic_inc_return(select_skb_count(skb)); + trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, + rxrpc_skb(skb)->rx_flags, here); } /* @@ -33,8 +35,9 @@ void rxrpc_see_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) { const void *here = __builtin_return_address(0); if (skb) { - int n = atomic_read(select_skb_count(op)); - trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, here); + int n = atomic_read(select_skb_count(skb)); + trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, + rxrpc_skb(skb)->rx_flags, here); } } @@ -44,12 +47,23 @@ void rxrpc_see_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) void rxrpc_get_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) { const void *here = __builtin_return_address(0); - int n = atomic_inc_return(select_skb_count(op)); - trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, here); + int n = atomic_inc_return(select_skb_count(skb)); + trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, + rxrpc_skb(skb)->rx_flags, here); skb_get(skb); } /* + * Note the dropping of a ref on a socket buffer by the core. + */ +void rxrpc_eaten_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) +{ + const void *here = __builtin_return_address(0); + int n = atomic_inc_return(&rxrpc_n_rx_skbs); + trace_rxrpc_skb(skb, op, 0, n, 0, here); +} + +/* * Note the destruction of a socket buffer. */ void rxrpc_free_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) @@ -58,8 +72,9 @@ void rxrpc_free_skb(struct sk_buff *skb, enum rxrpc_skb_trace op) if (skb) { int n; CHECK_SLAB_OKAY(&skb->users); - n = atomic_dec_return(select_skb_count(op)); - trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, here); + n = atomic_dec_return(select_skb_count(skb)); + trace_rxrpc_skb(skb, op, refcount_read(&skb->users), n, + rxrpc_skb(skb)->rx_flags, here); kfree_skb(skb); } } @@ -72,9 +87,10 @@ void rxrpc_purge_queue(struct sk_buff_head *list) const void *here = __builtin_return_address(0); struct sk_buff *skb; while ((skb = skb_dequeue((list))) != NULL) { - int n = atomic_dec_return(select_skb_count(rxrpc_skb_rx_purged)); - trace_rxrpc_skb(skb, rxrpc_skb_rx_purged, - refcount_read(&skb->users), n, here); + int n = atomic_dec_return(select_skb_count(skb)); + trace_rxrpc_skb(skb, rxrpc_skb_purged, + refcount_read(&skb->users), n, + rxrpc_skb(skb)->rx_flags, here); kfree_skb(skb); } } diff --git a/net/sched/Kconfig b/net/sched/Kconfig index afd2ba157a13..b3faafeafab9 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -963,6 +963,19 @@ config NET_IFE_SKBTCINDEX tristate "Support to encoding decoding skb tcindex on IFE action" depends on NET_ACT_IFE +config NET_TC_SKB_EXT + bool "TC recirculation support" + depends on NET_CLS_ACT + default y if NET_CLS_ACT + select SKB_EXTENSIONS + + help + Say Y here to allow tc chain misses to continue in OvS datapath in + the correct recirc_id, and hardware chain misses to continue in + the correct chain in tc software datapath. + + Say N here if you won't be using tc<->ovs offload or tc chains offload. + endif # NET_SCHED config NET_SCH_FIFO diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index fd1f7e799e23..04b7bd4ec751 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -422,7 +422,7 @@ static __net_init int bpf_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, bpf_net_id); - return tc_action_net_init(tn, &act_bpf_ops); + return tc_action_net_init(net, tn, &act_bpf_ops); } static void __net_exit bpf_exit_net(struct list_head *net_list) diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 32ac04d77a45..2b43cacf82af 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -231,7 +231,7 @@ static __net_init int connmark_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, connmark_net_id); - return tc_action_net_init(tn, &act_connmark_ops); + return tc_action_net_init(net, tn, &act_connmark_ops); } static void __net_exit connmark_exit_net(struct list_head *net_list) diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 9b9288267a54..d3cfad88dc3a 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -714,7 +714,7 @@ static __net_init int csum_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, csum_net_id); - return tc_action_net_init(tn, &act_csum_ops); + return tc_action_net_init(net, tn, &act_csum_ops); } static void __net_exit csum_exit_net(struct list_head *net_list) diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c index 33a1a7406e87..cdd6f3818097 100644 --- a/net/sched/act_ct.c +++ b/net/sched/act_ct.c @@ -939,7 +939,7 @@ static __net_init int ct_init_net(struct net *net) tn->labels = true; } - return tc_action_net_init(&tn->tn, &act_ct_ops); + return tc_action_net_init(net, &tn->tn, &act_ct_ops); } static void __net_exit ct_exit_net(struct list_head *net_list) diff --git a/net/sched/act_ctinfo.c b/net/sched/act_ctinfo.c index 06ef74b74911..0dbcfd1dca7b 100644 --- a/net/sched/act_ctinfo.c +++ b/net/sched/act_ctinfo.c @@ -376,7 +376,7 @@ static __net_init int ctinfo_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, ctinfo_net_id); - return tc_action_net_init(tn, &act_ctinfo_ops); + return tc_action_net_init(net, tn, &act_ctinfo_ops); } static void __net_exit ctinfo_exit_net(struct list_head *net_list) diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 8f0140c6ca58..324f1d1f6d47 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -278,7 +278,7 @@ static __net_init int gact_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, gact_net_id); - return tc_action_net_init(tn, &act_gact_ops); + return tc_action_net_init(net, tn, &act_gact_ops); } static void __net_exit gact_exit_net(struct list_head *net_list) diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 92ee853d43e6..3a31e241c647 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -890,7 +890,7 @@ static __net_init int ife_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, ife_net_id); - return tc_action_net_init(tn, &act_ife_ops); + return tc_action_net_init(net, tn, &act_ife_ops); } static void __net_exit ife_exit_net(struct list_head *net_list) diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index ce2c30a591d2..214a03d405cf 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -61,12 +61,13 @@ static int ipt_init_target(struct net *net, struct xt_entry_target *t, return 0; } -static void ipt_destroy_target(struct xt_entry_target *t) +static void ipt_destroy_target(struct xt_entry_target *t, struct net *net) { struct xt_tgdtor_param par = { .target = t->u.kernel.target, .targinfo = t->data, .family = NFPROTO_IPV4, + .net = net, }; if (par.target->destroy != NULL) par.target->destroy(&par); @@ -78,7 +79,7 @@ static void tcf_ipt_release(struct tc_action *a) struct tcf_ipt *ipt = to_ipt(a); if (ipt->tcfi_t) { - ipt_destroy_target(ipt->tcfi_t); + ipt_destroy_target(ipt->tcfi_t, a->idrinfo->net); kfree(ipt->tcfi_t); } kfree(ipt->tcfi_tname); @@ -180,7 +181,7 @@ static int __tcf_ipt_init(struct net *net, unsigned int id, struct nlattr *nla, spin_lock_bh(&ipt->tcf_lock); if (ret != ACT_P_CREATED) { - ipt_destroy_target(ipt->tcfi_t); + ipt_destroy_target(ipt->tcfi_t, net); kfree(ipt->tcfi_tname); kfree(ipt->tcfi_t); } @@ -350,7 +351,7 @@ static __net_init int ipt_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, ipt_net_id); - return tc_action_net_init(tn, &act_ipt_ops); + return tc_action_net_init(net, tn, &act_ipt_ops); } static void __net_exit ipt_exit_net(struct list_head *net_list) @@ -399,7 +400,7 @@ static __net_init int xt_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, xt_net_id); - return tc_action_net_init(tn, &act_xt_ops); + return tc_action_net_init(net, tn, &act_xt_ops); } static void __net_exit xt_exit_net(struct list_head *net_list) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index be3f88dfc37e..9d1bf508075a 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -453,7 +453,7 @@ static __net_init int mirred_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, mirred_net_id); - return tc_action_net_init(tn, &act_mirred_ops); + return tc_action_net_init(net, tn, &act_mirred_ops); } static void __net_exit mirred_exit_net(struct list_head *net_list) diff --git a/net/sched/act_mpls.c b/net/sched/act_mpls.c index 0f299e3b618c..e168df0e008a 100644 --- a/net/sched/act_mpls.c +++ b/net/sched/act_mpls.c @@ -375,7 +375,7 @@ static __net_init int mpls_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, mpls_net_id); - return tc_action_net_init(tn, &act_mpls_ops); + return tc_action_net_init(net, tn, &act_mpls_ops); } static void __net_exit mpls_exit_net(struct list_head *net_list) diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 7b858c11b1b5..ea4c5359e7df 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -327,7 +327,7 @@ static __net_init int nat_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, nat_net_id); - return tc_action_net_init(tn, &act_nat_ops); + return tc_action_net_init(net, tn, &act_nat_ops); } static void __net_exit nat_exit_net(struct list_head *net_list) diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 17360c6faeaa..cdfaa79382a2 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -498,7 +498,7 @@ static __net_init int pedit_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, pedit_net_id); - return tc_action_net_init(tn, &act_pedit_ops); + return tc_action_net_init(net, tn, &act_pedit_ops); } static void __net_exit pedit_exit_net(struct list_head *net_list) diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 49cec3e64a4d..89c04c52af3d 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -40,6 +40,8 @@ static const struct nla_policy police_policy[TCA_POLICE_MAX + 1] = { [TCA_POLICE_PEAKRATE] = { .len = TC_RTAB_SIZE }, [TCA_POLICE_AVRATE] = { .type = NLA_U32 }, [TCA_POLICE_RESULT] = { .type = NLA_U32 }, + [TCA_POLICE_RATE64] = { .type = NLA_U64 }, + [TCA_POLICE_PEAKRATE64] = { .type = NLA_U64 }, }; static int tcf_police_init(struct net *net, struct nlattr *nla, @@ -58,6 +60,7 @@ static int tcf_police_init(struct net *net, struct nlattr *nla, struct tcf_police_params *new; bool exists = false; u32 index; + u64 rate64, prate64; if (nla == NULL) return -EINVAL; @@ -155,14 +158,18 @@ static int tcf_police_init(struct net *net, struct nlattr *nla, } if (R_tab) { new->rate_present = true; - psched_ratecfg_precompute(&new->rate, &R_tab->rate, 0); + rate64 = tb[TCA_POLICE_RATE64] ? + nla_get_u64(tb[TCA_POLICE_RATE64]) : 0; + psched_ratecfg_precompute(&new->rate, &R_tab->rate, rate64); qdisc_put_rtab(R_tab); } else { new->rate_present = false; } if (P_tab) { new->peak_present = true; - psched_ratecfg_precompute(&new->peak, &P_tab->rate, 0); + prate64 = tb[TCA_POLICE_PEAKRATE64] ? + nla_get_u64(tb[TCA_POLICE_PEAKRATE64]) : 0; + psched_ratecfg_precompute(&new->peak, &P_tab->rate, prate64); qdisc_put_rtab(P_tab); } else { new->peak_present = false; @@ -313,10 +320,22 @@ static int tcf_police_dump(struct sk_buff *skb, struct tc_action *a, lockdep_is_held(&police->tcf_lock)); opt.mtu = p->tcfp_mtu; opt.burst = PSCHED_NS2TICKS(p->tcfp_burst); - if (p->rate_present) + if (p->rate_present) { psched_ratecfg_getrate(&opt.rate, &p->rate); - if (p->peak_present) + if ((police->params->rate.rate_bytes_ps >= (1ULL << 32)) && + nla_put_u64_64bit(skb, TCA_POLICE_RATE64, + police->params->rate.rate_bytes_ps, + TCA_POLICE_PAD)) + goto nla_put_failure; + } + if (p->peak_present) { psched_ratecfg_getrate(&opt.peakrate, &p->peak); + if ((police->params->peak.rate_bytes_ps >= (1ULL << 32)) && + nla_put_u64_64bit(skb, TCA_POLICE_PEAKRATE64, + police->params->peak.rate_bytes_ps, + TCA_POLICE_PAD)) + goto nla_put_failure; + } if (nla_put(skb, TCA_POLICE_TBF, sizeof(opt), &opt)) goto nla_put_failure; if (p->tcfp_result && @@ -371,7 +390,7 @@ static __net_init int police_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, police_net_id); - return tc_action_net_init(tn, &act_police_ops); + return tc_action_net_init(net, tn, &act_police_ops); } static void __net_exit police_exit_net(struct list_head *net_list) diff --git a/net/sched/act_sample.c b/net/sched/act_sample.c index 595308d60133..10229124a992 100644 --- a/net/sched/act_sample.c +++ b/net/sched/act_sample.c @@ -102,13 +102,17 @@ static int tcf_sample_init(struct net *net, struct nlattr *nla, goto_ch = tcf_action_set_ctrlact(*a, parm->action, goto_ch); s->rate = rate; s->psample_group_num = psample_group_num; - RCU_INIT_POINTER(s->psample_group, psample_group); + rcu_swap_protected(s->psample_group, psample_group, + lockdep_is_held(&s->tcf_lock)); if (tb[TCA_SAMPLE_TRUNC_SIZE]) { s->truncate = true; s->trunc_size = nla_get_u32(tb[TCA_SAMPLE_TRUNC_SIZE]); } spin_unlock_bh(&s->tcf_lock); + + if (psample_group) + psample_group_put(psample_group); if (goto_ch) tcf_chain_put_by_act(goto_ch); @@ -265,7 +269,7 @@ static __net_init int sample_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, sample_net_id); - return tc_action_net_init(tn, &act_sample_ops); + return tc_action_net_init(net, tn, &act_sample_ops); } static void __net_exit sample_exit_net(struct list_head *net_list) diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 33aefa25b545..6120e56117ca 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -232,7 +232,7 @@ static __net_init int simp_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, simp_net_id); - return tc_action_net_init(tn, &act_simp_ops); + return tc_action_net_init(net, tn, &act_simp_ops); } static void __net_exit simp_exit_net(struct list_head *net_list) diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index b100870f02a6..6a8d3337c577 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -307,6 +307,17 @@ static int tcf_skbedit_search(struct net *net, struct tc_action **a, u32 index) return tcf_idr_search(tn, a, index); } +static size_t tcf_skbedit_get_fill_size(const struct tc_action *act) +{ + return nla_total_size(sizeof(struct tc_skbedit)) + + nla_total_size(sizeof(u32)) /* TCA_SKBEDIT_PRIORITY */ + + nla_total_size(sizeof(u16)) /* TCA_SKBEDIT_QUEUE_MAPPING */ + + nla_total_size(sizeof(u32)) /* TCA_SKBEDIT_MARK */ + + nla_total_size(sizeof(u16)) /* TCA_SKBEDIT_PTYPE */ + + nla_total_size(sizeof(u32)) /* TCA_SKBEDIT_MASK */ + + nla_total_size_64bit(sizeof(u64)); /* TCA_SKBEDIT_FLAGS */ +} + static struct tc_action_ops act_skbedit_ops = { .kind = "skbedit", .id = TCA_ID_SKBEDIT, @@ -316,6 +327,7 @@ static struct tc_action_ops act_skbedit_ops = { .init = tcf_skbedit_init, .cleanup = tcf_skbedit_cleanup, .walk = tcf_skbedit_walker, + .get_fill_size = tcf_skbedit_get_fill_size, .lookup = tcf_skbedit_search, .size = sizeof(struct tcf_skbedit), }; @@ -324,7 +336,7 @@ static __net_init int skbedit_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, skbedit_net_id); - return tc_action_net_init(tn, &act_skbedit_ops); + return tc_action_net_init(net, tn, &act_skbedit_ops); } static void __net_exit skbedit_exit_net(struct list_head *net_list) diff --git a/net/sched/act_skbmod.c b/net/sched/act_skbmod.c index 7da3518e18ef..888437f97ba6 100644 --- a/net/sched/act_skbmod.c +++ b/net/sched/act_skbmod.c @@ -287,7 +287,7 @@ static __net_init int skbmod_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, skbmod_net_id); - return tc_action_net_init(tn, &act_skbmod_ops); + return tc_action_net_init(net, tn, &act_skbmod_ops); } static void __net_exit skbmod_exit_net(struct list_head *net_list) diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c index 6d0debdc9b97..2f83a79f76aa 100644 --- a/net/sched/act_tunnel_key.c +++ b/net/sched/act_tunnel_key.c @@ -600,7 +600,7 @@ static __net_init int tunnel_key_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); - return tc_action_net_init(tn, &act_tunnel_key_ops); + return tc_action_net_init(net, tn, &act_tunnel_key_ops); } static void __net_exit tunnel_key_exit_net(struct list_head *net_list) diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index a3c9eea1ee8a..08aaf719a70f 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -301,6 +301,19 @@ static int tcf_vlan_walker(struct net *net, struct sk_buff *skb, return tcf_generic_walker(tn, skb, cb, type, ops, extack); } +static void tcf_vlan_stats_update(struct tc_action *a, u64 bytes, u32 packets, + u64 lastuse, bool hw) +{ + struct tcf_vlan *v = to_vlan(a); + struct tcf_t *tm = &v->tcf_tm; + + _bstats_cpu_update(this_cpu_ptr(a->cpu_bstats), bytes, packets); + if (hw) + _bstats_cpu_update(this_cpu_ptr(a->cpu_bstats_hw), + bytes, packets); + tm->lastuse = max_t(u64, tm->lastuse, lastuse); +} + static int tcf_vlan_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, vlan_net_id); @@ -325,6 +338,7 @@ static struct tc_action_ops act_vlan_ops = { .init = tcf_vlan_init, .cleanup = tcf_vlan_cleanup, .walk = tcf_vlan_walker, + .stats_update = tcf_vlan_stats_update, .get_fill_size = tcf_vlan_get_fill_size, .lookup = tcf_vlan_search, .size = sizeof(struct tcf_vlan), @@ -334,7 +348,7 @@ static __net_init int vlan_init_net(struct net *net) { struct tc_action_net *tn = net_generic(net, vlan_net_id); - return tc_action_net_init(tn, &act_vlan_ops); + return tc_action_net_init(net, tn, &act_vlan_ops); } static void __net_exit vlan_exit_net(struct list_head *net_list) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index e0d8b456e9f5..05c4fe1c3ca2 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -568,9 +568,11 @@ static void tc_indr_block_ing_cmd(struct net_device *dev, bo.block = &block->flow_block; + down_write(&block->cb_lock); cb(dev, cb_priv, TC_SETUP_BLOCK, &bo); tcf_block_setup(block, &bo); + up_write(&block->cb_lock); } static struct tcf_block *tc_dev_ingress_block(struct net_device *dev) @@ -627,7 +629,7 @@ static void tc_indr_block_call(struct tcf_block *block, static bool tcf_block_offload_in_use(struct tcf_block *block) { - return block->offloadcnt; + return atomic_read(&block->offloadcnt); } static int tcf_block_offload_cmd(struct tcf_block *block, @@ -661,6 +663,7 @@ static int tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q, struct net_device *dev = q->dev_queue->dev; int err; + down_write(&block->cb_lock); if (!dev->netdev_ops->ndo_setup_tc) goto no_offload_dev_inc; @@ -669,24 +672,31 @@ static int tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q, */ if (!tc_can_offload(dev) && tcf_block_offload_in_use(block)) { NL_SET_ERR_MSG(extack, "Bind to offloaded block failed as dev has offload disabled"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto err_unlock; } err = tcf_block_offload_cmd(block, dev, ei, FLOW_BLOCK_BIND, extack); if (err == -EOPNOTSUPP) goto no_offload_dev_inc; if (err) - return err; + goto err_unlock; tc_indr_block_call(block, dev, ei, FLOW_BLOCK_BIND, extack); + up_write(&block->cb_lock); return 0; no_offload_dev_inc: - if (tcf_block_offload_in_use(block)) - return -EOPNOTSUPP; + if (tcf_block_offload_in_use(block)) { + err = -EOPNOTSUPP; + goto err_unlock; + } + err = 0; block->nooffloaddevcnt++; tc_indr_block_call(block, dev, ei, FLOW_BLOCK_BIND, extack); - return 0; +err_unlock: + up_write(&block->cb_lock); + return err; } static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, @@ -695,6 +705,7 @@ static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, struct net_device *dev = q->dev_queue->dev; int err; + down_write(&block->cb_lock); tc_indr_block_call(block, dev, ei, FLOW_BLOCK_UNBIND, NULL); if (!dev->netdev_ops->ndo_setup_tc) @@ -702,10 +713,12 @@ static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, err = tcf_block_offload_cmd(block, dev, ei, FLOW_BLOCK_UNBIND, NULL); if (err == -EOPNOTSUPP) goto no_offload_dev_dec; + up_write(&block->cb_lock); return; no_offload_dev_dec: WARN_ON(block->nooffloaddevcnt-- == 0); + up_write(&block->cb_lock); } static int @@ -820,6 +833,7 @@ static struct tcf_block *tcf_block_create(struct net *net, struct Qdisc *q, return ERR_PTR(-ENOMEM); } mutex_init(&block->lock); + init_rwsem(&block->cb_lock); flow_block_init(&block->flow_block); INIT_LIST_HEAD(&block->chain_list); INIT_LIST_HEAD(&block->owner_list); @@ -1355,6 +1369,8 @@ tcf_block_playback_offloads(struct tcf_block *block, flow_setup_cb_t *cb, struct tcf_proto *tp, *tp_prev; int err; + lockdep_assert_held(&block->cb_lock); + for (chain = __tcf_get_next_chain(block, NULL); chain; chain_prev = chain, @@ -1393,6 +1409,8 @@ static int tcf_block_bind(struct tcf_block *block, struct flow_block_cb *block_cb, *next; int err, i = 0; + lockdep_assert_held(&block->cb_lock); + list_for_each_entry(block_cb, &bo->cb_list, list) { err = tcf_block_playback_offloads(block, block_cb->cb, block_cb->cb_priv, true, @@ -1400,6 +1418,8 @@ static int tcf_block_bind(struct tcf_block *block, bo->extack); if (err) goto err_unroll; + if (!bo->unlocked_driver_cb) + block->lockeddevcnt++; i++; } @@ -1415,6 +1435,8 @@ err_unroll: block_cb->cb_priv, false, tcf_block_offload_in_use(block), NULL); + if (!bo->unlocked_driver_cb) + block->lockeddevcnt--; } flow_block_cb_free(block_cb); } @@ -1427,6 +1449,8 @@ static void tcf_block_unbind(struct tcf_block *block, { struct flow_block_cb *block_cb, *next; + lockdep_assert_held(&block->cb_lock); + list_for_each_entry_safe(block_cb, next, &bo->cb_list, list) { tcf_block_playback_offloads(block, block_cb->cb, block_cb->cb_priv, false, @@ -1434,6 +1458,8 @@ static void tcf_block_unbind(struct tcf_block *block, NULL); list_del(&block_cb->list); flow_block_cb_free(block_cb); + if (!bo->unlocked_driver_cb) + block->lockeddevcnt--; } } @@ -1488,6 +1514,18 @@ reclassify: goto reset; } else if (unlikely(TC_ACT_EXT_CMP(err, TC_ACT_GOTO_CHAIN))) { first_tp = res->goto_tp; + +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + { + struct tc_skb_ext *ext; + + ext = skb_ext_add(skb, TC_SKB_EXT); + if (WARN_ON_ONCE(!ext)) + return TC_ACT_SHOT; + + ext->chain = err & TC_ACT_EXT_VAL_MASK; + } +#endif goto reset; } #endif @@ -2980,17 +3018,61 @@ int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts) } EXPORT_SYMBOL(tcf_exts_dump_stats); -int tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type, - void *type_data, bool err_stop) +static void tcf_block_offload_inc(struct tcf_block *block, u32 *flags) +{ + if (*flags & TCA_CLS_FLAGS_IN_HW) + return; + *flags |= TCA_CLS_FLAGS_IN_HW; + atomic_inc(&block->offloadcnt); +} + +static void tcf_block_offload_dec(struct tcf_block *block, u32 *flags) +{ + if (!(*flags & TCA_CLS_FLAGS_IN_HW)) + return; + *flags &= ~TCA_CLS_FLAGS_IN_HW; + atomic_dec(&block->offloadcnt); +} + +static void tc_cls_offload_cnt_update(struct tcf_block *block, + struct tcf_proto *tp, u32 *cnt, + u32 *flags, u32 diff, bool add) +{ + lockdep_assert_held(&block->cb_lock); + + spin_lock(&tp->lock); + if (add) { + if (!*cnt) + tcf_block_offload_inc(block, flags); + *cnt += diff; + } else { + *cnt -= diff; + if (!*cnt) + tcf_block_offload_dec(block, flags); + } + spin_unlock(&tp->lock); +} + +static void +tc_cls_offload_cnt_reset(struct tcf_block *block, struct tcf_proto *tp, + u32 *cnt, u32 *flags) +{ + lockdep_assert_held(&block->cb_lock); + + spin_lock(&tp->lock); + tcf_block_offload_dec(block, flags); + *cnt = 0; + spin_unlock(&tp->lock); +} + +static int +__tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type, + void *type_data, bool err_stop) { struct flow_block_cb *block_cb; int ok_count = 0; int err; - /* Make sure all netdevs sharing this block are offload-capable. */ - if (block->nooffloaddevcnt && err_stop) - return -EOPNOTSUPP; - list_for_each_entry(block_cb, &block->flow_block.cb_list, list) { err = block_cb->cb(type, type_data, block_cb->cb_priv); if (err) { @@ -3002,17 +3084,235 @@ int tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type, } return ok_count; } + +int tc_setup_cb_call(struct tcf_block *block, enum tc_setup_type type, + void *type_data, bool err_stop, bool rtnl_held) +{ + bool take_rtnl = READ_ONCE(block->lockeddevcnt) && !rtnl_held; + int ok_count; + +retry: + if (take_rtnl) + rtnl_lock(); + down_read(&block->cb_lock); + /* Need to obtain rtnl lock if block is bound to devs that require it. + * In block bind code cb_lock is obtained while holding rtnl, so we must + * obtain the locks in same order here. + */ + if (!rtnl_held && !take_rtnl && block->lockeddevcnt) { + up_read(&block->cb_lock); + take_rtnl = true; + goto retry; + } + + ok_count = __tc_setup_cb_call(block, type, type_data, err_stop); + + up_read(&block->cb_lock); + if (take_rtnl) + rtnl_unlock(); + return ok_count; +} EXPORT_SYMBOL(tc_setup_cb_call); +/* Non-destructive filter add. If filter that wasn't already in hardware is + * successfully offloaded, increment block offloads counter. On failure, + * previously offloaded filter is considered to be intact and offloads counter + * is not decremented. + */ + +int tc_setup_cb_add(struct tcf_block *block, struct tcf_proto *tp, + enum tc_setup_type type, void *type_data, bool err_stop, + u32 *flags, unsigned int *in_hw_count, bool rtnl_held) +{ + bool take_rtnl = READ_ONCE(block->lockeddevcnt) && !rtnl_held; + int ok_count; + +retry: + if (take_rtnl) + rtnl_lock(); + down_read(&block->cb_lock); + /* Need to obtain rtnl lock if block is bound to devs that require it. + * In block bind code cb_lock is obtained while holding rtnl, so we must + * obtain the locks in same order here. + */ + if (!rtnl_held && !take_rtnl && block->lockeddevcnt) { + up_read(&block->cb_lock); + take_rtnl = true; + goto retry; + } + + /* Make sure all netdevs sharing this block are offload-capable. */ + if (block->nooffloaddevcnt && err_stop) { + ok_count = -EOPNOTSUPP; + goto err_unlock; + } + + ok_count = __tc_setup_cb_call(block, type, type_data, err_stop); + if (ok_count < 0) + goto err_unlock; + + if (tp->ops->hw_add) + tp->ops->hw_add(tp, type_data); + if (ok_count > 0) + tc_cls_offload_cnt_update(block, tp, in_hw_count, flags, + ok_count, true); +err_unlock: + up_read(&block->cb_lock); + if (take_rtnl) + rtnl_unlock(); + return ok_count < 0 ? ok_count : 0; +} +EXPORT_SYMBOL(tc_setup_cb_add); + +/* Destructive filter replace. If filter that wasn't already in hardware is + * successfully offloaded, increment block offload counter. On failure, + * previously offloaded filter is considered to be destroyed and offload counter + * is decremented. + */ + +int tc_setup_cb_replace(struct tcf_block *block, struct tcf_proto *tp, + enum tc_setup_type type, void *type_data, bool err_stop, + u32 *old_flags, unsigned int *old_in_hw_count, + u32 *new_flags, unsigned int *new_in_hw_count, + bool rtnl_held) +{ + bool take_rtnl = READ_ONCE(block->lockeddevcnt) && !rtnl_held; + int ok_count; + +retry: + if (take_rtnl) + rtnl_lock(); + down_read(&block->cb_lock); + /* Need to obtain rtnl lock if block is bound to devs that require it. + * In block bind code cb_lock is obtained while holding rtnl, so we must + * obtain the locks in same order here. + */ + if (!rtnl_held && !take_rtnl && block->lockeddevcnt) { + up_read(&block->cb_lock); + take_rtnl = true; + goto retry; + } + + /* Make sure all netdevs sharing this block are offload-capable. */ + if (block->nooffloaddevcnt && err_stop) { + ok_count = -EOPNOTSUPP; + goto err_unlock; + } + + tc_cls_offload_cnt_reset(block, tp, old_in_hw_count, old_flags); + if (tp->ops->hw_del) + tp->ops->hw_del(tp, type_data); + + ok_count = __tc_setup_cb_call(block, type, type_data, err_stop); + if (ok_count < 0) + goto err_unlock; + + if (tp->ops->hw_add) + tp->ops->hw_add(tp, type_data); + if (ok_count > 0) + tc_cls_offload_cnt_update(block, tp, new_in_hw_count, + new_flags, ok_count, true); +err_unlock: + up_read(&block->cb_lock); + if (take_rtnl) + rtnl_unlock(); + return ok_count < 0 ? ok_count : 0; +} +EXPORT_SYMBOL(tc_setup_cb_replace); + +/* Destroy filter and decrement block offload counter, if filter was previously + * offloaded. + */ + +int tc_setup_cb_destroy(struct tcf_block *block, struct tcf_proto *tp, + enum tc_setup_type type, void *type_data, bool err_stop, + u32 *flags, unsigned int *in_hw_count, bool rtnl_held) +{ + bool take_rtnl = READ_ONCE(block->lockeddevcnt) && !rtnl_held; + int ok_count; + +retry: + if (take_rtnl) + rtnl_lock(); + down_read(&block->cb_lock); + /* Need to obtain rtnl lock if block is bound to devs that require it. + * In block bind code cb_lock is obtained while holding rtnl, so we must + * obtain the locks in same order here. + */ + if (!rtnl_held && !take_rtnl && block->lockeddevcnt) { + up_read(&block->cb_lock); + take_rtnl = true; + goto retry; + } + + ok_count = __tc_setup_cb_call(block, type, type_data, err_stop); + + tc_cls_offload_cnt_reset(block, tp, in_hw_count, flags); + if (tp->ops->hw_del) + tp->ops->hw_del(tp, type_data); + + up_read(&block->cb_lock); + if (take_rtnl) + rtnl_unlock(); + return ok_count < 0 ? ok_count : 0; +} +EXPORT_SYMBOL(tc_setup_cb_destroy); + +int tc_setup_cb_reoffload(struct tcf_block *block, struct tcf_proto *tp, + bool add, flow_setup_cb_t *cb, + enum tc_setup_type type, void *type_data, + void *cb_priv, u32 *flags, unsigned int *in_hw_count) +{ + int err = cb(type, type_data, cb_priv); + + if (err) { + if (add && tc_skip_sw(*flags)) + return err; + } else { + tc_cls_offload_cnt_update(block, tp, in_hw_count, flags, 1, + add); + } + + return 0; +} +EXPORT_SYMBOL(tc_setup_cb_reoffload); + +void tc_cleanup_flow_action(struct flow_action *flow_action) +{ + struct flow_action_entry *entry; + int i; + + flow_action_for_each(i, entry, flow_action) { + switch (entry->id) { + case FLOW_ACTION_REDIRECT: + case FLOW_ACTION_MIRRED: + case FLOW_ACTION_REDIRECT_INGRESS: + case FLOW_ACTION_MIRRED_INGRESS: + if (entry->dev) + dev_put(entry->dev); + break; + case FLOW_ACTION_TUNNEL_ENCAP: + kfree(entry->tunnel); + break; + default: + break; + } + } +} +EXPORT_SYMBOL(tc_cleanup_flow_action); + int tc_setup_flow_action(struct flow_action *flow_action, - const struct tcf_exts *exts) + const struct tcf_exts *exts, bool rtnl_held) { const struct tc_action *act; - int i, j, k; + int i, j, k, err = 0; if (!exts) return 0; + if (!rtnl_held) + rtnl_lock(); + j = 0; tcf_exts_for_each_action(i, act, exts) { struct flow_action_entry *entry; @@ -3030,15 +3330,23 @@ int tc_setup_flow_action(struct flow_action *flow_action, } else if (is_tcf_mirred_egress_redirect(act)) { entry->id = FLOW_ACTION_REDIRECT; entry->dev = tcf_mirred_dev(act); + if (entry->dev) + dev_hold(entry->dev); } else if (is_tcf_mirred_egress_mirror(act)) { entry->id = FLOW_ACTION_MIRRED; entry->dev = tcf_mirred_dev(act); + if (entry->dev) + dev_hold(entry->dev); } else if (is_tcf_mirred_ingress_redirect(act)) { entry->id = FLOW_ACTION_REDIRECT_INGRESS; entry->dev = tcf_mirred_dev(act); + if (entry->dev) + dev_hold(entry->dev); } else if (is_tcf_mirred_ingress_mirror(act)) { entry->id = FLOW_ACTION_MIRRED_INGRESS; entry->dev = tcf_mirred_dev(act); + if (entry->dev) + dev_hold(entry->dev); } else if (is_tcf_vlan(act)) { switch (tcf_vlan_action(act)) { case TCA_VLAN_ACT_PUSH: @@ -3057,11 +3365,16 @@ int tc_setup_flow_action(struct flow_action *flow_action, entry->vlan.prio = tcf_vlan_push_prio(act); break; default: + err = -EOPNOTSUPP; goto err_out; } } else if (is_tcf_tunnel_set(act)) { entry->id = FLOW_ACTION_TUNNEL_ENCAP; - entry->tunnel = tcf_tunnel_info(act); + entry->tunnel = tcf_tunnel_info_copy(act); + if (!entry->tunnel) { + err = -ENOMEM; + goto err_out; + } } else if (is_tcf_tunnel_release(act)) { entry->id = FLOW_ACTION_TUNNEL_DECAP; } else if (is_tcf_pedit(act)) { @@ -3074,6 +3387,7 @@ int tc_setup_flow_action(struct flow_action *flow_action, entry->id = FLOW_ACTION_ADD; break; default: + err = -EOPNOTSUPP; goto err_out; } entry->mangle.htype = tcf_pedit_htype(act, k); @@ -3132,15 +3446,22 @@ int tc_setup_flow_action(struct flow_action *flow_action, entry->id = FLOW_ACTION_PTYPE; entry->ptype = tcf_skbedit_ptype(act); } else { + err = -EOPNOTSUPP; goto err_out; } if (!is_tcf_pedit(act)) j++; } - return 0; + err_out: - return -EOPNOTSUPP; + if (!rtnl_held) + rtnl_unlock(); + + if (err) + tc_cleanup_flow_action(flow_action); + + return err; } EXPORT_SYMBOL(tc_setup_flow_action); diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 3f7a9c02b70c..bf10bdaf5012 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -163,17 +163,19 @@ static int cls_bpf_offload_cmd(struct tcf_proto *tp, struct cls_bpf_prog *prog, cls_bpf.exts_integrated = obj->exts_integrated; if (oldprog) - tcf_block_offload_dec(block, &oldprog->gen_flags); + err = tc_setup_cb_replace(block, tp, TC_SETUP_CLSBPF, &cls_bpf, + skip_sw, &oldprog->gen_flags, + &oldprog->in_hw_count, + &prog->gen_flags, &prog->in_hw_count, + true); + else + err = tc_setup_cb_add(block, tp, TC_SETUP_CLSBPF, &cls_bpf, + skip_sw, &prog->gen_flags, + &prog->in_hw_count, true); - err = tc_setup_cb_call(block, TC_SETUP_CLSBPF, &cls_bpf, skip_sw); - if (prog) { - if (err < 0) { - cls_bpf_offload_cmd(tp, oldprog, prog, extack); - return err; - } else if (err > 0) { - prog->in_hw_count = err; - tcf_block_offload_inc(block, &prog->gen_flags); - } + if (prog && err) { + cls_bpf_offload_cmd(tp, oldprog, prog, extack); + return err; } if (prog && skip_sw && !(prog->gen_flags & TCA_CLS_FLAGS_IN_HW)) @@ -230,7 +232,7 @@ static void cls_bpf_offload_update_stats(struct tcf_proto *tp, cls_bpf.name = prog->bpf_name; cls_bpf.exts_integrated = prog->exts_integrated; - tc_setup_cb_call(block, TC_SETUP_CLSBPF, &cls_bpf, false); + tc_setup_cb_call(block, TC_SETUP_CLSBPF, &cls_bpf, false, true); } static int cls_bpf_init(struct tcf_proto *tp) @@ -673,15 +675,11 @@ static int cls_bpf_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb cls_bpf.name = prog->bpf_name; cls_bpf.exts_integrated = prog->exts_integrated; - err = cb(TC_SETUP_CLSBPF, &cls_bpf, cb_priv); - if (err) { - if (add && tc_skip_sw(prog->gen_flags)) - return err; - continue; - } - - tc_cls_offload_cnt_update(block, &prog->in_hw_count, - &prog->gen_flags, add); + err = tc_setup_cb_reoffload(block, tp, add, cb, TC_SETUP_CLSBPF, + &cls_bpf, cb_priv, &prog->gen_flags, + &prog->in_hw_count); + if (err) + return err; } return 0; diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 054123742e32..74221e3351c3 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -412,41 +412,27 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f, struct tcf_block *block = tp->chain->block; struct flow_cls_offload cls_flower = {}; - if (!rtnl_held) - rtnl_lock(); - tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, extack); cls_flower.command = FLOW_CLS_DESTROY; cls_flower.cookie = (unsigned long) f; - tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false); - spin_lock(&tp->lock); - list_del_init(&f->hw_list); - tcf_block_offload_dec(block, &f->flags); - spin_unlock(&tp->lock); + tc_setup_cb_destroy(block, tp, TC_SETUP_CLSFLOWER, &cls_flower, false, + &f->flags, &f->in_hw_count, rtnl_held); - if (!rtnl_held) - rtnl_unlock(); } static int fl_hw_replace_filter(struct tcf_proto *tp, struct cls_fl_filter *f, bool rtnl_held, struct netlink_ext_ack *extack) { - struct cls_fl_head *head = fl_head_dereference(tp); struct tcf_block *block = tp->chain->block; struct flow_cls_offload cls_flower = {}; bool skip_sw = tc_skip_sw(f->flags); int err = 0; - if (!rtnl_held) - rtnl_lock(); - cls_flower.rule = flow_rule_alloc(tcf_exts_num_actions(&f->exts)); - if (!cls_flower.rule) { - err = -ENOMEM; - goto errout; - } + if (!cls_flower.rule) + return -ENOMEM; tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, extack); cls_flower.command = FLOW_CLS_REPLACE; @@ -456,43 +442,31 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, cls_flower.rule->match.key = &f->mkey; cls_flower.classid = f->res.classid; - err = tc_setup_flow_action(&cls_flower.rule->action, &f->exts); + err = tc_setup_flow_action(&cls_flower.rule->action, &f->exts, + rtnl_held); if (err) { kfree(cls_flower.rule); - if (skip_sw) + if (skip_sw) { NL_SET_ERR_MSG_MOD(extack, "Failed to setup flow action"); - else - err = 0; - goto errout; + return err; + } + return 0; } - err = tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, skip_sw); + err = tc_setup_cb_add(block, tp, TC_SETUP_CLSFLOWER, &cls_flower, + skip_sw, &f->flags, &f->in_hw_count, rtnl_held); + tc_cleanup_flow_action(&cls_flower.rule->action); kfree(cls_flower.rule); - if (err < 0) { - fl_hw_destroy_filter(tp, f, true, NULL); - goto errout; - } else if (err > 0) { - f->in_hw_count = err; - err = 0; - spin_lock(&tp->lock); - tcf_block_offload_inc(block, &f->flags); - spin_unlock(&tp->lock); - } - - if (skip_sw && !(f->flags & TCA_CLS_FLAGS_IN_HW)) { - err = -EINVAL; - goto errout; + if (err) { + fl_hw_destroy_filter(tp, f, rtnl_held, NULL); + return err; } - spin_lock(&tp->lock); - list_add(&f->hw_list, &head->hw_filters); - spin_unlock(&tp->lock); -errout: - if (!rtnl_held) - rtnl_unlock(); + if (skip_sw && !(f->flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; - return err; + return 0; } static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f, @@ -501,22 +475,17 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f, struct tcf_block *block = tp->chain->block; struct flow_cls_offload cls_flower = {}; - if (!rtnl_held) - rtnl_lock(); - tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, NULL); cls_flower.command = FLOW_CLS_STATS; cls_flower.cookie = (unsigned long) f; cls_flower.classid = f->res.classid; - tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false); + tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false, + rtnl_held); tcf_exts_stats_update(&f->exts, cls_flower.stats.bytes, cls_flower.stats.pkts, cls_flower.stats.lastused); - - if (!rtnl_held) - rtnl_unlock(); } static void __fl_put(struct cls_fl_filter *f) @@ -1831,7 +1800,8 @@ static int fl_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb, cls_flower.rule->match.mask = &f->mask->key; cls_flower.rule->match.key = &f->mkey; - err = tc_setup_flow_action(&cls_flower.rule->action, &f->exts); + err = tc_setup_flow_action(&cls_flower.rule->action, &f->exts, + true); if (err) { kfree(cls_flower.rule); if (tc_skip_sw(f->flags)) { @@ -1844,21 +1814,17 @@ static int fl_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb, cls_flower.classid = f->res.classid; - err = cb(TC_SETUP_CLSFLOWER, &cls_flower, cb_priv); + err = tc_setup_cb_reoffload(block, tp, add, cb, + TC_SETUP_CLSFLOWER, &cls_flower, + cb_priv, &f->flags, + &f->in_hw_count); + tc_cleanup_flow_action(&cls_flower.rule->action); kfree(cls_flower.rule); if (err) { - if (add && tc_skip_sw(f->flags)) { - __fl_put(f); - return err; - } - goto next_flow; + __fl_put(f); + return err; } - - spin_lock(&tp->lock); - tc_cls_offload_cnt_update(block, &f->in_hw_count, &f->flags, - add); - spin_unlock(&tp->lock); next_flow: __fl_put(f); } @@ -1866,6 +1832,30 @@ next_flow: return 0; } +static void fl_hw_add(struct tcf_proto *tp, void *type_data) +{ + struct flow_cls_offload *cls_flower = type_data; + struct cls_fl_filter *f = + (struct cls_fl_filter *) cls_flower->cookie; + struct cls_fl_head *head = fl_head_dereference(tp); + + spin_lock(&tp->lock); + list_add(&f->hw_list, &head->hw_filters); + spin_unlock(&tp->lock); +} + +static void fl_hw_del(struct tcf_proto *tp, void *type_data) +{ + struct flow_cls_offload *cls_flower = type_data; + struct cls_fl_filter *f = + (struct cls_fl_filter *) cls_flower->cookie; + + spin_lock(&tp->lock); + if (!list_empty(&f->hw_list)) + list_del_init(&f->hw_list); + spin_unlock(&tp->lock); +} + static int fl_hw_create_tmplt(struct tcf_chain *chain, struct fl_flow_tmplt *tmplt) { @@ -1886,7 +1876,7 @@ static int fl_hw_create_tmplt(struct tcf_chain *chain, /* We don't care if driver (any of them) fails to handle this * call. It serves just as a hint for it. */ - tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false); + tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false, true); kfree(cls_flower.rule); return 0; @@ -1902,7 +1892,7 @@ static void fl_hw_destroy_tmplt(struct tcf_chain *chain, cls_flower.command = FLOW_CLS_TMPLT_DESTROY; cls_flower.cookie = (unsigned long) tmplt; - tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false); + tc_setup_cb_call(block, TC_SETUP_CLSFLOWER, &cls_flower, false, true); } static void *fl_tmplt_create(struct net *net, struct tcf_chain *chain, @@ -2526,6 +2516,8 @@ static struct tcf_proto_ops cls_fl_ops __read_mostly = { .delete = fl_delete, .walk = fl_walk, .reoffload = fl_reoffload, + .hw_add = fl_hw_add, + .hw_del = fl_hw_del, .dump = fl_dump, .bind_class = fl_bind_class, .tmplt_create = fl_tmplt_create, diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index 455ea2793f9b..7fc2eb62aa98 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -75,8 +75,8 @@ static void mall_destroy_hw_filter(struct tcf_proto *tp, cls_mall.command = TC_CLSMATCHALL_DESTROY; cls_mall.cookie = cookie; - tc_setup_cb_call(block, TC_SETUP_CLSMATCHALL, &cls_mall, false); - tcf_block_offload_dec(block, &head->flags); + tc_setup_cb_destroy(block, tp, TC_SETUP_CLSMATCHALL, &cls_mall, false, + &head->flags, &head->in_hw_count, true); } static int mall_replace_hw_filter(struct tcf_proto *tp, @@ -97,7 +97,7 @@ static int mall_replace_hw_filter(struct tcf_proto *tp, cls_mall.command = TC_CLSMATCHALL_REPLACE; cls_mall.cookie = cookie; - err = tc_setup_flow_action(&cls_mall.rule->action, &head->exts); + err = tc_setup_flow_action(&cls_mall.rule->action, &head->exts, true); if (err) { kfree(cls_mall.rule); mall_destroy_hw_filter(tp, head, cookie, NULL); @@ -109,15 +109,14 @@ static int mall_replace_hw_filter(struct tcf_proto *tp, return err; } - err = tc_setup_cb_call(block, TC_SETUP_CLSMATCHALL, &cls_mall, skip_sw); + err = tc_setup_cb_add(block, tp, TC_SETUP_CLSMATCHALL, &cls_mall, + skip_sw, &head->flags, &head->in_hw_count, true); + tc_cleanup_flow_action(&cls_mall.rule->action); kfree(cls_mall.rule); - if (err < 0) { + if (err) { mall_destroy_hw_filter(tp, head, cookie, NULL); return err; - } else if (err > 0) { - head->in_hw_count = err; - tcf_block_offload_inc(block, &head->flags); } if (skip_sw && !(head->flags & TCA_CLS_FLAGS_IN_HW)) @@ -302,7 +301,7 @@ static int mall_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb, TC_CLSMATCHALL_REPLACE : TC_CLSMATCHALL_DESTROY; cls_mall.cookie = (unsigned long)head; - err = tc_setup_flow_action(&cls_mall.rule->action, &head->exts); + err = tc_setup_flow_action(&cls_mall.rule->action, &head->exts, true); if (err) { kfree(cls_mall.rule); if (add && tc_skip_sw(head->flags)) { @@ -312,16 +311,14 @@ static int mall_reoffload(struct tcf_proto *tp, bool add, flow_setup_cb_t *cb, return 0; } - err = cb(TC_SETUP_CLSMATCHALL, &cls_mall, cb_priv); + err = tc_setup_cb_reoffload(block, tp, add, cb, TC_SETUP_CLSMATCHALL, + &cls_mall, cb_priv, &head->flags, + &head->in_hw_count); + tc_cleanup_flow_action(&cls_mall.rule->action); kfree(cls_mall.rule); - if (err) { - if (add && tc_skip_sw(head->flags)) - return err; - return 0; - } - - tc_cls_offload_cnt_update(block, &head->in_hw_count, &head->flags, add); + if (err) + return err; return 0; } @@ -337,7 +334,7 @@ static void mall_stats_hw_filter(struct tcf_proto *tp, cls_mall.command = TC_CLSMATCHALL_STATS; cls_mall.cookie = cookie; - tc_setup_cb_call(block, TC_SETUP_CLSMATCHALL, &cls_mall, false); + tc_setup_cb_call(block, TC_SETUP_CLSMATCHALL, &cls_mall, false, true); tcf_exts_stats_update(&head->exts, cls_mall.stats.bytes, cls_mall.stats.pkts, cls_mall.stats.lastused); diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 8614088edd1b..a0e6fac613de 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -480,7 +480,7 @@ static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, cls_u32.hnode.handle = h->handle; cls_u32.hnode.prio = h->prio; - tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, false); + tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, false, true); } static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, @@ -498,7 +498,7 @@ static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, cls_u32.hnode.handle = h->handle; cls_u32.hnode.prio = h->prio; - err = tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, skip_sw); + err = tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, skip_sw, true); if (err < 0) { u32_clear_hw_hnode(tp, h, NULL); return err; @@ -522,8 +522,8 @@ static void u32_remove_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, cls_u32.command = TC_CLSU32_DELETE_KNODE; cls_u32.knode.handle = n->handle; - tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, false); - tcf_block_offload_dec(block, &n->flags); + tc_setup_cb_destroy(block, tp, TC_SETUP_CLSU32, &cls_u32, false, + &n->flags, &n->in_hw_count, true); } static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, @@ -552,13 +552,11 @@ static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, if (n->ht_down) cls_u32.knode.link_handle = ht->handle; - err = tc_setup_cb_call(block, TC_SETUP_CLSU32, &cls_u32, skip_sw); - if (err < 0) { + err = tc_setup_cb_add(block, tp, TC_SETUP_CLSU32, &cls_u32, skip_sw, + &n->flags, &n->in_hw_count, true); + if (err) { u32_remove_hw_knode(tp, n, NULL); return err; - } else if (err > 0) { - n->in_hw_count = err; - tcf_block_offload_inc(block, &n->flags); } if (skip_sw && !(n->flags & TCA_CLS_FLAGS_IN_HW)) @@ -1201,14 +1199,11 @@ static int u32_reoffload_knode(struct tcf_proto *tp, struct tc_u_knode *n, cls_u32.knode.link_handle = ht->handle; } - err = cb(TC_SETUP_CLSU32, &cls_u32, cb_priv); - if (err) { - if (add && tc_skip_sw(n->flags)) - return err; - return 0; - } - - tc_cls_offload_cnt_update(block, &n->in_hw_count, &n->flags, add); + err = tc_setup_cb_reoffload(block, tp, add, cb, TC_SETUP_CLSU32, + &cls_u32, cb_priv, &n->flags, + &n->in_hw_count); + if (err) + return err; return 0; } diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c index 732e109c3055..93b58fde99b7 100644 --- a/net/sched/sch_cbs.c +++ b/net/sched/sch_cbs.c @@ -181,11 +181,6 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch) s64 credits; int len; - if (atomic64_read(&q->port_rate) == -1) { - WARN_ONCE(1, "cbs: dequeue() called with unknown port rate."); - return NULL; - } - if (q->credits < 0) { credits = timediff_to_credits(now - q->last, q->idleslope); @@ -303,11 +298,19 @@ static int cbs_enable_offload(struct net_device *dev, struct cbs_sched_data *q, static void cbs_set_port_rate(struct net_device *dev, struct cbs_sched_data *q) { struct ethtool_link_ksettings ecmd; - int port_rate = -1; + int speed = SPEED_10; + int port_rate; + int err; + + err = __ethtool_get_link_ksettings(dev, &ecmd); + if (err < 0) + goto skip; + + if (ecmd.base.speed != SPEED_UNKNOWN) + speed = ecmd.base.speed; - if (!__ethtool_get_link_ksettings(dev, &ecmd) && - ecmd.base.speed != SPEED_UNKNOWN) - port_rate = ecmd.base.speed * 1000 * BYTES_PER_KBIT; +skip: + port_rate = speed * 1000 * BYTES_PER_KBIT; atomic64_set(&q->port_rate, port_rate); netdev_dbg(dev, "cbs: set %s's port_rate to: %lld, linkspeed: %d\n", diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 11c03cf4aa74..137db1cbde85 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -624,8 +624,12 @@ static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, err = skb_array_produce(q, skb); - if (unlikely(err)) - return qdisc_drop_cpu(skb, qdisc, to_free); + if (unlikely(err)) { + if (qdisc_is_percpu_stats(qdisc)) + return qdisc_drop_cpu(skb, qdisc, to_free); + else + return qdisc_drop(skb, qdisc, to_free); + } qdisc_update_stats_at_enqueue(qdisc, pkt_len); return NET_XMIT_SUCCESS; @@ -688,11 +692,14 @@ static void pfifo_fast_reset(struct Qdisc *qdisc) kfree_skb(skb); } - for_each_possible_cpu(i) { - struct gnet_stats_queue *q = per_cpu_ptr(qdisc->cpu_qstats, i); + if (qdisc_is_percpu_stats(qdisc)) { + for_each_possible_cpu(i) { + struct gnet_stats_queue *q; - q->backlog = 0; - q->qlen = 0; + q = per_cpu_ptr(qdisc->cpu_qstats, i); + q->backlog = 0; + q->qlen = 0; + } } } diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index 046fd2c102b4..84b863e2bdbd 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -477,11 +477,6 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch) u32 gate_mask; int i; - if (atomic64_read(&q->picos_per_byte) == -1) { - WARN_ONCE(1, "taprio: dequeue() called with unknown picos per byte."); - return NULL; - } - rcu_read_lock(); entry = rcu_dereference(q->current_entry); /* if there's no entry, it means that the schedule didn't @@ -954,12 +949,20 @@ static void taprio_set_picos_per_byte(struct net_device *dev, struct taprio_sched *q) { struct ethtool_link_ksettings ecmd; - int picos_per_byte = -1; + int speed = SPEED_10; + int picos_per_byte; + int err; - if (!__ethtool_get_link_ksettings(dev, &ecmd) && - ecmd.base.speed != SPEED_UNKNOWN) - picos_per_byte = div64_s64(NSEC_PER_SEC * 1000LL * 8, - ecmd.base.speed * 1000 * 1000); + err = __ethtool_get_link_ksettings(dev, &ecmd); + if (err < 0) + goto skip; + + if (ecmd.base.speed != SPEED_UNKNOWN) + speed = ecmd.base.speed; + +skip: + picos_per_byte = div64_s64(NSEC_PER_SEC * 1000LL * 8, + speed * 1000 * 1000); atomic64_set(&q->picos_per_byte, picos_per_byte); netdev_dbg(dev, "taprio: set %s's picos_per_byte to: %lld, linkspeed: %d\n", @@ -1191,7 +1194,8 @@ unlock: spin_unlock_bh(qdisc_lock(sch)); free_sched: - kfree(new_admin); + if (new_admin) + call_rcu(&new_admin->rcu, taprio_free_sched_cb); return err; } @@ -1244,6 +1248,10 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt, */ q->clockid = -1; + spin_lock(&taprio_list_lock); + list_add(&q->taprio_list, &taprio_list); + spin_unlock(&taprio_list_lock); + if (sch->parent != TC_H_ROOT) return -EOPNOTSUPP; @@ -1261,10 +1269,6 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt, if (!opt) return -EINVAL; - spin_lock(&taprio_list_lock); - list_add(&q->taprio_list, &taprio_list); - spin_unlock(&taprio_list_lock); - for (i = 0; i < dev->num_tx_queues; i++) { struct netdev_queue *dev_queue; struct Qdisc *qdisc; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 5010cce52c93..d2ffc9a0ba3a 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -54,7 +54,6 @@ static struct sctp_association *sctp_association_init( const struct sock *sk, enum sctp_scope scope, gfp_t gfp) { - struct net *net = sock_net(sk); struct sctp_sock *sp; struct sctp_paramhdr *p; int i; @@ -214,14 +213,6 @@ static struct sctp_association *sctp_association_init( asoc->peer.sack_needed = 1; asoc->peer.sack_generation = 1; - /* Assume that the peer will tell us if he recognizes ASCONF - * as part of INIT exchange. - * The sctp_addip_noauth option is there for backward compatibility - * and will revert old behavior. - */ - if (net->sctp.addip_noauth) - asoc->peer.asconf_capable = 1; - /* Create an input queue. */ sctp_inq_init(&asoc->base.inqueue); sctp_inq_set_th_handler(&asoc->base.inqueue, sctp_assoc_bh_rcv); diff --git a/net/sctp/auth.c b/net/sctp/auth.c index de4c78d4a21e..4278764d82b8 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -389,7 +389,7 @@ int sctp_auth_asoc_init_active_key(struct sctp_association *asoc, gfp_t gfp) /* If we don't support AUTH, or peer is not capable * we don't need to do anything. */ - if (!asoc->ep->auth_enable || !asoc->peer.auth_capable) + if (!asoc->peer.auth_capable) return 0; /* If the key_id is non-zero and we couldn't find an @@ -675,7 +675,7 @@ int sctp_auth_send_cid(enum sctp_cid chunk, const struct sctp_association *asoc) if (!asoc) return 0; - if (!asoc->ep->auth_enable || !asoc->peer.auth_capable) + if (!asoc->peer.auth_capable) return 0; return __sctp_auth_cid(chunk, asoc->peer.peer_chunks); @@ -687,7 +687,7 @@ int sctp_auth_recv_cid(enum sctp_cid chunk, const struct sctp_association *asoc) if (!asoc) return 0; - if (!asoc->ep->auth_enable) + if (!asoc->peer.auth_capable) return 0; return __sctp_auth_cid(chunk, @@ -831,10 +831,15 @@ int sctp_auth_set_key(struct sctp_endpoint *ep, /* Try to find the given key id to see if * we are doing a replace, or adding a new key */ - if (asoc) + if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; sh_keys = &asoc->endpoint_shared_keys; - else + } else { + if (!ep->auth_enable) + return -EACCES; sh_keys = &ep->endpoint_shared_keys; + } key_for_each(shkey, sh_keys) { if (shkey->key_id == auth_key->sca_keynumber) { @@ -875,10 +880,15 @@ int sctp_auth_set_active_key(struct sctp_endpoint *ep, int found = 0; /* The key identifier MUST correst to an existing key */ - if (asoc) + if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; sh_keys = &asoc->endpoint_shared_keys; - else + } else { + if (!ep->auth_enable) + return -EACCES; sh_keys = &ep->endpoint_shared_keys; + } key_for_each(key, sh_keys) { if (key->key_id == key_id) { @@ -911,11 +921,15 @@ int sctp_auth_del_key_id(struct sctp_endpoint *ep, * The key identifier MUST correst to an existing key */ if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; if (asoc->active_key_id == key_id) return -EINVAL; sh_keys = &asoc->endpoint_shared_keys; } else { + if (!ep->auth_enable) + return -EACCES; if (ep->active_key_id == key_id) return -EINVAL; @@ -950,11 +964,15 @@ int sctp_auth_deact_key_id(struct sctp_endpoint *ep, * The key identifier MUST correst to an existing key */ if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; if (asoc->active_key_id == key_id) return -EINVAL; sh_keys = &asoc->endpoint_shared_keys; } else { + if (!ep->auth_enable) + return -EACCES; if (ep->active_key_id == key_id) return -EINVAL; @@ -989,3 +1007,72 @@ int sctp_auth_deact_key_id(struct sctp_endpoint *ep, return 0; } + +int sctp_auth_init(struct sctp_endpoint *ep, gfp_t gfp) +{ + int err = -ENOMEM; + + /* Allocate space for HMACS and CHUNKS authentication + * variables. There are arrays that we encode directly + * into parameters to make the rest of the operations easier. + */ + if (!ep->auth_hmacs_list) { + struct sctp_hmac_algo_param *auth_hmacs; + + auth_hmacs = kzalloc(struct_size(auth_hmacs, hmac_ids, + SCTP_AUTH_NUM_HMACS), gfp); + if (!auth_hmacs) + goto nomem; + /* Initialize the HMACS parameter. + * SCTP-AUTH: Section 3.3 + * Every endpoint supporting SCTP chunk authentication MUST + * support the HMAC based on the SHA-1 algorithm. + */ + auth_hmacs->param_hdr.type = SCTP_PARAM_HMAC_ALGO; + auth_hmacs->param_hdr.length = + htons(sizeof(struct sctp_paramhdr) + 2); + auth_hmacs->hmac_ids[0] = htons(SCTP_AUTH_HMAC_ID_SHA1); + ep->auth_hmacs_list = auth_hmacs; + } + + if (!ep->auth_chunk_list) { + struct sctp_chunks_param *auth_chunks; + + auth_chunks = kzalloc(sizeof(*auth_chunks) + + SCTP_NUM_CHUNK_TYPES, gfp); + if (!auth_chunks) + goto nomem; + /* Initialize the CHUNKS parameter */ + auth_chunks->param_hdr.type = SCTP_PARAM_CHUNKS; + auth_chunks->param_hdr.length = + htons(sizeof(struct sctp_paramhdr)); + ep->auth_chunk_list = auth_chunks; + } + + /* Allocate and initialize transorms arrays for supported + * HMACs. + */ + err = sctp_auth_init_hmacs(ep, gfp); + if (err) + goto nomem; + + return 0; + +nomem: + /* Free all allocations */ + kfree(ep->auth_hmacs_list); + kfree(ep->auth_chunk_list); + ep->auth_hmacs_list = NULL; + ep->auth_chunk_list = NULL; + return err; +} + +void sctp_auth_free(struct sctp_endpoint *ep) +{ + kfree(ep->auth_hmacs_list); + kfree(ep->auth_chunk_list); + ep->auth_hmacs_list = NULL; + ep->auth_chunk_list = NULL; + sctp_auth_destroy_hmacs(ep->auth_hmacs); + ep->auth_hmacs = NULL; +} diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 69cebb2c998b..ea53049d1db6 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -43,62 +43,21 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, gfp_t gfp) { struct net *net = sock_net(sk); - struct sctp_hmac_algo_param *auth_hmacs = NULL; - struct sctp_chunks_param *auth_chunks = NULL; struct sctp_shared_key *null_key; - int err; ep->digest = kzalloc(SCTP_SIGNATURE_SIZE, gfp); if (!ep->digest) return NULL; + ep->asconf_enable = net->sctp.addip_enable; ep->auth_enable = net->sctp.auth_enable; if (ep->auth_enable) { - /* Allocate space for HMACS and CHUNKS authentication - * variables. There are arrays that we encode directly - * into parameters to make the rest of the operations easier. - */ - auth_hmacs = kzalloc(struct_size(auth_hmacs, hmac_ids, - SCTP_AUTH_NUM_HMACS), gfp); - if (!auth_hmacs) - goto nomem; - - auth_chunks = kzalloc(sizeof(*auth_chunks) + - SCTP_NUM_CHUNK_TYPES, gfp); - if (!auth_chunks) + if (sctp_auth_init(ep, gfp)) goto nomem; - - /* Initialize the HMACS parameter. - * SCTP-AUTH: Section 3.3 - * Every endpoint supporting SCTP chunk authentication MUST - * support the HMAC based on the SHA-1 algorithm. - */ - auth_hmacs->param_hdr.type = SCTP_PARAM_HMAC_ALGO; - auth_hmacs->param_hdr.length = - htons(sizeof(struct sctp_paramhdr) + 2); - auth_hmacs->hmac_ids[0] = htons(SCTP_AUTH_HMAC_ID_SHA1); - - /* Initialize the CHUNKS parameter */ - auth_chunks->param_hdr.type = SCTP_PARAM_CHUNKS; - auth_chunks->param_hdr.length = - htons(sizeof(struct sctp_paramhdr)); - - /* If the Add-IP functionality is enabled, we must - * authenticate, ASCONF and ASCONF-ACK chunks - */ - if (net->sctp.addip_enable) { - auth_chunks->chunks[0] = SCTP_CID_ASCONF; - auth_chunks->chunks[1] = SCTP_CID_ASCONF_ACK; - auth_chunks->param_hdr.length = - htons(sizeof(struct sctp_paramhdr) + 2); + if (ep->asconf_enable) { + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF); + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF_ACK); } - - /* Allocate and initialize transorms arrays for supported - * HMACs. - */ - err = sctp_auth_init_hmacs(ep, gfp); - if (err) - goto nomem; } /* Initialize the base structure. */ @@ -145,10 +104,9 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, /* Add the null key to the endpoint shared keys list and * set the hmcas and chunks pointers. */ - ep->auth_hmacs_list = auth_hmacs; - ep->auth_chunk_list = auth_chunks; ep->prsctp_enable = net->sctp.prsctp_enable; ep->reconf_enable = net->sctp.reconf_enable; + ep->ecn_enable = net->sctp.ecn_enable; /* Remember who we are attached to. */ ep->base.sk = sk; @@ -157,11 +115,8 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, return ep; nomem_shkey: - sctp_auth_destroy_hmacs(ep->auth_hmacs); + sctp_auth_free(ep); nomem: - /* Free all allocations */ - kfree(auth_hmacs); - kfree(auth_chunks); kfree(ep->digest); return NULL; @@ -244,11 +199,7 @@ static void sctp_endpoint_destroy(struct sctp_endpoint *ep) * chunks and hmacs arrays that were allocated */ sctp_auth_destroy_keys(&ep->endpoint_shared_keys); - kfree(ep->auth_hmacs_list); - kfree(ep->auth_chunk_list); - - /* AUTH - Free any allocated HMAC transform containers */ - sctp_auth_destroy_hmacs(ep->auth_hmacs); + sctp_auth_free(ep); /* Cleanup. */ sctp_inq_free(&ep->base.inqueue); diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2d47adcb4cbe..b48ffe845c31 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1254,6 +1254,9 @@ static int __net_init sctp_defaults_init(struct net *net) /* Disable AUTH by default. */ net->sctp.auth_enable = 0; + /* Enable ECN by default. */ + net->sctp.ecn_enable = 1; + /* Set SCOPE policy to enabled */ net->sctp.scope_policy = SCTP_SCOPE_POLICY_ENABLE; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 36bd8a6e82df..e41ed2e0ae7d 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -207,7 +207,6 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, const struct sctp_bind_addr *bp, gfp_t gfp, int vparam_len) { - struct net *net = sock_net(asoc->base.sk); struct sctp_supported_ext_param ext_param; struct sctp_adaptation_ind_param aiparam; struct sctp_paramhdr *auth_chunks = NULL; @@ -245,7 +244,9 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, chunksize = sizeof(init) + addrs_len; chunksize += SCTP_PAD4(SCTP_SAT_LEN(num_types)); - chunksize += sizeof(ecap_param); + + if (asoc->ep->ecn_enable) + chunksize += sizeof(ecap_param); if (asoc->ep->prsctp_enable) chunksize += sizeof(prsctp_param); @@ -255,7 +256,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, * the ASCONF,the ASCONF-ACK, and the AUTH chunks in its INIT and * INIT-ACK parameters. */ - if (net->sctp.addip_enable) { + if (asoc->ep->asconf_enable) { extensions[num_ext] = SCTP_CID_ASCONF; extensions[num_ext+1] = SCTP_CID_ASCONF_ACK; num_ext += 2; @@ -336,7 +337,8 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, sctp_addto_chunk(retval, sizeof(sat), &sat); sctp_addto_chunk(retval, num_types * sizeof(__u16), &types); - sctp_addto_chunk(retval, sizeof(ecap_param), &ecap_param); + if (asoc->ep->ecn_enable) + sctp_addto_chunk(retval, sizeof(ecap_param), &ecap_param); /* Add the supported extensions parameter. Be nice and add this * fist before addiding the parameters for the extensions themselves @@ -1964,7 +1966,9 @@ static int sctp_process_hn_param(const struct sctp_association *asoc, return 0; } -static int sctp_verify_ext_param(struct net *net, union sctp_params param) +static int sctp_verify_ext_param(struct net *net, + const struct sctp_endpoint *ep, + union sctp_params param) { __u16 num_ext = ntohs(param.p->length) - sizeof(struct sctp_paramhdr); int have_asconf = 0; @@ -1991,7 +1995,7 @@ static int sctp_verify_ext_param(struct net *net, union sctp_params param) if (net->sctp.addip_noauth) return 1; - if (net->sctp.addip_enable && !have_auth && have_asconf) + if (ep->asconf_enable && !have_auth && have_asconf) return 0; return 1; @@ -2001,7 +2005,6 @@ static void sctp_process_ext_param(struct sctp_association *asoc, union sctp_params param) { __u16 num_ext = ntohs(param.p->length) - sizeof(struct sctp_paramhdr); - struct net *net = sock_net(asoc->base.sk); int i; for (i = 0; i < num_ext; i++) { @@ -2023,7 +2026,7 @@ static void sctp_process_ext_param(struct sctp_association *asoc, break; case SCTP_CID_ASCONF: case SCTP_CID_ASCONF_ACK: - if (net->sctp.addip_enable) + if (asoc->ep->asconf_enable) asoc->peer.asconf_capable = 1; break; case SCTP_CID_I_DATA: @@ -2145,12 +2148,12 @@ static enum sctp_ierror sctp_verify_param(struct net *net, break; case SCTP_PARAM_SUPPORTED_EXT: - if (!sctp_verify_ext_param(net, param)) + if (!sctp_verify_ext_param(net, ep, param)) return SCTP_IERROR_ABORT; break; case SCTP_PARAM_SET_PRIMARY: - if (net->sctp.addip_enable) + if (ep->asconf_enable) break; goto fallthrough; @@ -2597,15 +2600,20 @@ do_addr_param: break; case SCTP_PARAM_ECN_CAPABLE: - asoc->peer.ecn_capable = 1; - break; + if (asoc->ep->ecn_enable) { + asoc->peer.ecn_capable = 1; + break; + } + /* Fall Through */ + goto fall_through; + case SCTP_PARAM_ADAPTATION_LAYER_IND: asoc->peer.adaptation_ind = ntohl(param.aind->adaptation_ind); break; case SCTP_PARAM_SET_PRIMARY: - if (!net->sctp.addip_enable) + if (!ep->asconf_enable) goto fall_through; addr_param = param.v + sizeof(struct sctp_addip_param); diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index a554d6d15d1b..1cf5bb5b73c4 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -546,7 +546,7 @@ static void sctp_do_8_2_transport_strike(struct sctp_cmd_seq *commands, */ if (net->sctp.pf_enable && (transport->state == SCTP_ACTIVE) && - (asoc->pf_retrans < transport->pathmaxrxt) && + (transport->error_count < transport->pathmaxrxt) && (transport->error_count > asoc->pf_retrans)) { sctp_assoc_control_transport(asoc, transport, diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 2c244b29a199..0c21c52fc408 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -3721,7 +3721,8 @@ enum sctp_disposition sctp_sf_do_asconf(struct net *net, * is received unauthenticated it MUST be silently discarded as * described in [I-D.ietf-tsvwg-sctp-auth]. */ - if (!net->sctp.addip_noauth && !chunk->auth) + if (!asoc->peer.asconf_capable || + (!net->sctp.addip_noauth && !chunk->auth)) return sctp_sf_discard_chunk(net, ep, asoc, type, arg, commands); @@ -3863,7 +3864,8 @@ enum sctp_disposition sctp_sf_do_asconf_ack(struct net *net, * is received unauthenticated it MUST be silently discarded as * described in [I-D.ietf-tsvwg-sctp-auth]. */ - if (!net->sctp.addip_noauth && !asconf_ack->auth) + if (!asoc->peer.asconf_capable || + (!net->sctp.addip_noauth && !asconf_ack->auth)) return sctp_sf_discard_chunk(net, ep, asoc, type, arg, commands); diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 61ed9c6e3be3..88ea87f4f0e7 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -976,26 +976,22 @@ static const struct sctp_sm_table_entry *sctp_chunk_event_lookup( if (cid <= SCTP_CID_BASE_MAX) return &chunk_event_table[cid][state]; - if (net->sctp.prsctp_enable) { - if (cid == SCTP_CID_FWD_TSN || cid == SCTP_CID_I_FWD_TSN) - return &prsctp_chunk_event_table[0][state]; - } + switch ((u16)cid) { + case SCTP_CID_FWD_TSN: + case SCTP_CID_I_FWD_TSN: + return &prsctp_chunk_event_table[0][state]; - if (net->sctp.addip_enable) { - if (cid == SCTP_CID_ASCONF) - return &addip_chunk_event_table[0][state]; + case SCTP_CID_ASCONF: + return &addip_chunk_event_table[0][state]; - if (cid == SCTP_CID_ASCONF_ACK) - return &addip_chunk_event_table[1][state]; - } + case SCTP_CID_ASCONF_ACK: + return &addip_chunk_event_table[1][state]; - if (net->sctp.reconf_enable) - if (cid == SCTP_CID_RECONF) - return &reconf_chunk_event_table[0][state]; + case SCTP_CID_RECONF: + return &reconf_chunk_event_table[0][state]; - if (net->sctp.auth_enable) { - if (cid == SCTP_CID_AUTH) - return &auth_chunk_event_table[0][state]; + case SCTP_CID_AUTH: + return &auth_chunk_event_table[0][state]; } return &chunk_event_table_unknown[state]; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 12503e16fa96..3e50a9712fb1 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -524,7 +524,6 @@ static int sctp_send_asconf_add_ip(struct sock *sk, struct sockaddr *addrs, int addrcnt) { - struct net *net = sock_net(sk); struct sctp_sock *sp; struct sctp_endpoint *ep; struct sctp_association *asoc; @@ -539,12 +538,12 @@ static int sctp_send_asconf_add_ip(struct sock *sk, int i; int retval = 0; - if (!net->sctp.addip_enable) - return retval; - sp = sctp_sk(sk); ep = sp->ep; + if (!ep->asconf_enable) + return retval; + pr_debug("%s: sk:%p, addrs:%p, addrcnt:%d\n", __func__, sk, addrs, addrcnt); @@ -727,7 +726,6 @@ static int sctp_send_asconf_del_ip(struct sock *sk, struct sockaddr *addrs, int addrcnt) { - struct net *net = sock_net(sk); struct sctp_sock *sp; struct sctp_endpoint *ep; struct sctp_association *asoc; @@ -743,12 +741,12 @@ static int sctp_send_asconf_del_ip(struct sock *sk, int stored = 0; chunk = NULL; - if (!net->sctp.addip_enable) - return retval; - sp = sctp_sk(sk); ep = sp->ep; + if (!ep->asconf_enable) + return retval; + pr_debug("%s: sk:%p, addrs:%p, addrcnt:%d\n", __func__, sk, addrs, addrcnt); @@ -3330,7 +3328,6 @@ static int sctp_setsockopt_maxseg(struct sock *sk, char __user *optval, unsigned static int sctp_setsockopt_peer_primary_addr(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); struct sctp_sock *sp; struct sctp_association *asoc = NULL; struct sctp_setpeerprim prim; @@ -3340,7 +3337,7 @@ static int sctp_setsockopt_peer_primary_addr(struct sock *sk, char __user *optva sp = sctp_sk(sk); - if (!net->sctp.addip_enable) + if (!sp->ep->asconf_enable) return -EPERM; if (optlen != sizeof(struct sctp_setpeerprim)) @@ -3690,9 +3687,6 @@ static int sctp_setsockopt_auth_key(struct sock *sk, struct sctp_association *asoc; int ret = -EINVAL; - if (!ep->auth_enable) - return -EACCES; - if (optlen <= sizeof(struct sctp_authkey)) return -EINVAL; /* authkey->sca_keylength is u16, so optlen can't be bigger than @@ -3759,9 +3753,6 @@ static int sctp_setsockopt_active_key(struct sock *sk, struct sctp_authkeyid val; int ret = 0; - if (!ep->auth_enable) - return -EACCES; - if (optlen != sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -3813,9 +3804,6 @@ static int sctp_setsockopt_del_key(struct sock *sk, struct sctp_authkeyid val; int ret = 0; - if (!ep->auth_enable) - return -EACCES; - if (optlen != sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -3866,9 +3854,6 @@ static int sctp_setsockopt_deactivate_key(struct sock *sk, char __user *optval, struct sctp_authkeyid val; int ret = 0; - if (!ep->auth_enable) - return -EACCES; - if (optlen != sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -4499,6 +4484,110 @@ static int sctp_setsockopt_event(struct sock *sk, char __user *optval, return retval; } +static int sctp_setsockopt_asconf_supported(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + struct sctp_endpoint *ep; + int retval = -EINVAL; + + if (optlen != sizeof(params)) + goto out; + + if (copy_from_user(¶ms, optval, optlen)) { + retval = -EFAULT; + goto out; + } + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) + goto out; + + ep = sctp_sk(sk)->ep; + ep->asconf_enable = !!params.assoc_value; + + if (ep->asconf_enable && ep->auth_enable) { + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF); + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF_ACK); + } + + retval = 0; + +out: + return retval; +} + +static int sctp_setsockopt_auth_supported(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + struct sctp_endpoint *ep; + int retval = -EINVAL; + + if (optlen != sizeof(params)) + goto out; + + if (copy_from_user(¶ms, optval, optlen)) { + retval = -EFAULT; + goto out; + } + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) + goto out; + + ep = sctp_sk(sk)->ep; + if (params.assoc_value) { + retval = sctp_auth_init(ep, GFP_KERNEL); + if (retval) + goto out; + if (ep->asconf_enable) { + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF); + sctp_auth_ep_add_chunkid(ep, SCTP_CID_ASCONF_ACK); + } + } + + ep->auth_enable = !!params.assoc_value; + retval = 0; + +out: + return retval; +} + +static int sctp_setsockopt_ecn_supported(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EINVAL; + + if (optlen != sizeof(params)) + goto out; + + if (copy_from_user(¶ms, optval, optlen)) { + retval = -EFAULT; + goto out; + } + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) + goto out; + + sctp_sk(sk)->ep->ecn_enable = !!params.assoc_value; + retval = 0; + +out: + return retval; +} + /* API 6.2 setsockopt(), getsockopt() * * Applications use setsockopt() and getsockopt() to set or retrieve @@ -4699,6 +4788,15 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, case SCTP_EVENT: retval = sctp_setsockopt_event(sk, optval, optlen); break; + case SCTP_ASCONF_SUPPORTED: + retval = sctp_setsockopt_asconf_supported(sk, optval, optlen); + break; + case SCTP_AUTH_SUPPORTED: + retval = sctp_setsockopt_auth_supported(sk, optval, optlen); + break; + case SCTP_ECN_SUPPORTED: + retval = sctp_setsockopt_ecn_supported(sk, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; @@ -6836,9 +6934,6 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, struct sctp_authkeyid val; struct sctp_association *asoc; - if (!ep->auth_enable) - return -EACCES; - if (len < sizeof(struct sctp_authkeyid)) return -EINVAL; @@ -6850,10 +6945,15 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, if (!asoc && val.scact_assoc_id && sctp_style(sk, UDP)) return -EINVAL; - if (asoc) + if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; val.scact_keynumber = asoc->active_key_id; - else + } else { + if (!ep->auth_enable) + return -EACCES; val.scact_keynumber = ep->active_key_id; + } if (put_user(len, optlen)) return -EFAULT; @@ -6866,7 +6966,6 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authchunks __user *p = (void __user *)optval; struct sctp_authchunks val; struct sctp_association *asoc; @@ -6874,9 +6973,6 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, u32 num_chunks = 0; char __user *to; - if (!ep->auth_enable) - return -EACCES; - if (len < sizeof(struct sctp_authchunks)) return -EINVAL; @@ -6888,6 +6984,9 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, if (!asoc) return -EINVAL; + if (!asoc->peer.auth_capable) + return -EACCES; + ch = asoc->peer.peer_chunks; if (!ch) goto num; @@ -6919,9 +7018,6 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, u32 num_chunks = 0; char __user *to; - if (!ep->auth_enable) - return -EACCES; - if (len < sizeof(struct sctp_authchunks)) return -EINVAL; @@ -6934,8 +7030,15 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, sctp_style(sk, UDP)) return -EINVAL; - ch = asoc ? (struct sctp_chunks_param *)asoc->c.auth_chunks - : ep->auth_chunk_list; + if (asoc) { + if (!asoc->peer.auth_capable) + return -EACCES; + ch = (struct sctp_chunks_param *)asoc->c.auth_chunks; + } else { + if (!ep->auth_enable) + return -EACCES; + ch = ep->auth_chunk_list; + } if (!ch) goto num; @@ -7678,6 +7781,123 @@ static int sctp_getsockopt_event(struct sock *sk, int len, char __user *optval, return 0; } +static int sctp_getsockopt_asconf_supported(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(params)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) { + retval = -EINVAL; + goto out; + } + + params.assoc_value = asoc ? asoc->peer.asconf_capable + : sctp_sk(sk)->ep->asconf_enable; + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, ¶ms, len)) + goto out; + + retval = 0; + +out: + return retval; +} + +static int sctp_getsockopt_auth_supported(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(params)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) { + retval = -EINVAL; + goto out; + } + + params.assoc_value = asoc ? asoc->peer.auth_capable + : sctp_sk(sk)->ep->auth_enable; + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, ¶ms, len)) + goto out; + + retval = 0; + +out: + return retval; +} + +static int sctp_getsockopt_ecn_supported(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(params)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (!asoc && params.assoc_id != SCTP_FUTURE_ASSOC && + sctp_style(sk, UDP)) { + retval = -EINVAL; + goto out; + } + + params.assoc_value = asoc ? asoc->peer.ecn_capable + : sctp_sk(sk)->ep->ecn_enable; + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, ¶ms, len)) + goto out; + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -7879,6 +8099,17 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, case SCTP_EVENT: retval = sctp_getsockopt_event(sk, len, optval, optlen); break; + case SCTP_ASCONF_SUPPORTED: + retval = sctp_getsockopt_asconf_supported(sk, len, optval, + optlen); + break; + case SCTP_AUTH_SUPPORTED: + retval = sctp_getsockopt_auth_supported(sk, len, optval, + optlen); + break; + case SCTP_ECN_SUPPORTED: + retval = sctp_getsockopt_ecn_supported(sk, len, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; diff --git a/net/sctp/stream.c b/net/sctp/stream.c index 25946604af85..e83cdaa2ab76 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -316,6 +316,7 @@ int sctp_send_reset_streams(struct sctp_association *asoc, nstr_list[i] = htons(str_list[i]); if (out && !sctp_stream_outq_is_empty(stream, str_nums, nstr_list)) { + kfree(nstr_list); retval = -EAGAIN; goto out; } diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 1250751bca1b..238cf1737576 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -278,6 +278,13 @@ static struct ctl_table sctp_net_table[] = { .proc_handler = proc_dointvec, }, { + .procname = "ecn_enable", + .data = &init_net.sctp.ecn_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { .procname = "addr_scope_policy", .data = &init_net.sctp.scope_policy, .maxlen = sizeof(int), diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index f0de323d15d6..6c8f09c1ce51 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -76,13 +76,11 @@ static int smc_tx_wait(struct smc_sock *smc, int flags) DEFINE_WAIT_FUNC(wait, woken_wake_function); struct smc_connection *conn = &smc->conn; struct sock *sk = &smc->sk; - bool noblock; long timeo; int rc = 0; /* similar to sk_stream_wait_memory */ timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT); - noblock = timeo ? false : true; add_wait_queue(sk_sleep(sk), &wait); while (1) { sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk); @@ -97,8 +95,8 @@ static int smc_tx_wait(struct smc_sock *smc, int flags) break; } if (!timeo) { - if (noblock) - set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + /* ensure EPOLLOUT is subsequently generated */ + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); rc = -EAGAIN; break; } diff --git a/net/sunrpc/auth_gss/gss_krb5_keys.c b/net/sunrpc/auth_gss/gss_krb5_keys.c index 550fdf18d3b3..3b7f721c023b 100644 --- a/net/sunrpc/auth_gss/gss_krb5_keys.c +++ b/net/sunrpc/auth_gss/gss_krb5_keys.c @@ -228,14 +228,11 @@ u32 krb5_derive_key(const struct gss_krb5_enctype *gk5e, ret = 0; err_free_raw: - memset(rawkey, 0, keybytes); - kfree(rawkey); + kzfree(rawkey); err_free_out: - memset(outblockdata, 0, blocksize); - kfree(outblockdata); + kzfree(outblockdata); err_free_in: - memset(inblockdata, 0, blocksize); - kfree(inblockdata); + kzfree(inblockdata); err_free_cipher: crypto_free_sync_skcipher(cipher); err_return: diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index d8679b6027e9..a07b516e503a 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1970,6 +1970,7 @@ call_bind(struct rpc_task *task) static void call_bind_status(struct rpc_task *task) { + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; int status = -EIO; if (rpc_task_transmitted(task)) { @@ -1977,14 +1978,15 @@ call_bind_status(struct rpc_task *task) return; } - if (task->tk_status >= 0) { - dprint_status(task); + dprint_status(task); + trace_rpc_bind_status(task); + if (task->tk_status >= 0) + goto out_next; + if (xprt_bound(xprt)) { task->tk_status = 0; - task->tk_action = call_connect; - return; + goto out_next; } - trace_rpc_bind_status(task); switch (task->tk_status) { case -ENOMEM: dprintk("RPC: %5u rpcbind out of memory\n", task->tk_pid); @@ -2003,6 +2005,9 @@ call_bind_status(struct rpc_task *task) task->tk_rebind_retry--; rpc_delay(task, 3*HZ); goto retry_timeout; + case -ENOBUFS: + rpc_delay(task, HZ >> 2); + goto retry_timeout; case -EAGAIN: goto retry_timeout; case -ETIMEDOUT: @@ -2026,7 +2031,6 @@ call_bind_status(struct rpc_task *task) case -ENETDOWN: case -EHOSTUNREACH: case -ENETUNREACH: - case -ENOBUFS: case -EPIPE: dprintk("RPC: %5u remote rpcbind unreachable: %d\n", task->tk_pid, task->tk_status); @@ -2043,7 +2047,9 @@ call_bind_status(struct rpc_task *task) rpc_call_rpcerror(task, status); return; - +out_next: + task->tk_action = call_connect; + return; retry_timeout: task->tk_status = 0; task->tk_action = call_bind; @@ -2090,6 +2096,7 @@ call_connect(struct rpc_task *task) static void call_connect_status(struct rpc_task *task) { + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; struct rpc_clnt *clnt = task->tk_client; int status = task->tk_status; @@ -2099,8 +2106,17 @@ call_connect_status(struct rpc_task *task) } dprint_status(task); - trace_rpc_connect_status(task); + + if (task->tk_status == 0) { + clnt->cl_stats->netreconn++; + goto out_next; + } + if (xprt_connected(xprt)) { + task->tk_status = 0; + goto out_next; + } + task->tk_status = 0; switch (status) { case -ECONNREFUSED: @@ -2117,8 +2133,6 @@ call_connect_status(struct rpc_task *task) case -ENETDOWN: case -ENETUNREACH: case -EHOSTUNREACH: - case -EADDRINUSE: - case -ENOBUFS: case -EPIPE: xprt_conditional_disconnect(task->tk_rqstp->rq_xprt, task->tk_rqstp->rq_connect_cookie); @@ -2127,17 +2141,20 @@ call_connect_status(struct rpc_task *task) /* retry with existing socket, after a delay */ rpc_delay(task, 3*HZ); /* fall through */ + case -EADDRINUSE: case -ENOTCONN: case -EAGAIN: case -ETIMEDOUT: goto out_retry; - case 0: - clnt->cl_stats->netreconn++; - task->tk_action = call_transmit; - return; + case -ENOBUFS: + rpc_delay(task, HZ >> 2); + goto out_retry; } rpc_call_rpcerror(task, status); return; +out_next: + task->tk_action = call_transmit; + return; out_retry: /* Check for timeouts before looping back to call_bind */ task->tk_action = call_bind; @@ -2365,7 +2382,7 @@ call_status(struct rpc_task *task) case -ECONNABORTED: case -ENOTCONN: rpc_force_rebind(clnt); - /* fall through */ + break; case -EADDRINUSE: rpc_delay(task, 3*HZ); /* fall through */ diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 783748dc5e6f..2e71f5455c6c 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -1408,13 +1408,6 @@ xprt_request_transmit(struct rpc_rqst *req, struct rpc_task *snd_task) status = -EBADMSG; goto out_dequeue; } - if (task->tk_ops->rpc_call_prepare_transmit) { - task->tk_ops->rpc_call_prepare_transmit(task, - task->tk_calldata); - status = task->tk_status; - if (status < 0) - goto out_dequeue; - } if (RPC_SIGNALLED(task)) { status = -ERESTARTSYS; goto out_dequeue; diff --git a/net/tipc/addr.c b/net/tipc/addr.c index b88d48d00913..0f1eaed1bd1b 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -75,6 +75,7 @@ void tipc_set_node_addr(struct net *net, u32 addr) tipc_set_node_id(net, node_id); } tn->trial_addr = addr; + tn->addr_trial_end = jiffies; pr_info("32-bit node address hash set to %x\n", addr); } diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 34f3e5641438..6ef1abdd525f 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -185,7 +185,7 @@ static void tipc_bcbase_xmit(struct net *net, struct sk_buff_head *xmitq) } /* We have to transmit across all bearers */ - skb_queue_head_init(&_xmitq); + __skb_queue_head_init(&_xmitq); for (bearer_id = 0; bearer_id < MAX_BEARERS; bearer_id++) { if (!bb->dests[bearer_id]) continue; @@ -256,7 +256,7 @@ static int tipc_bcast_xmit(struct net *net, struct sk_buff_head *pkts, struct sk_buff_head xmitq; int rc = 0; - skb_queue_head_init(&xmitq); + __skb_queue_head_init(&xmitq); tipc_bcast_lock(net); if (tipc_link_bc_peers(l)) rc = tipc_link_xmit(l, pkts, &xmitq); @@ -286,7 +286,7 @@ static int tipc_rcast_xmit(struct net *net, struct sk_buff_head *pkts, u32 dnode, selector; selector = msg_link_selector(buf_msg(skb_peek(pkts))); - skb_queue_head_init(&_pkts); + __skb_queue_head_init(&_pkts); list_for_each_entry_safe(dst, tmp, &dests->list, list) { dnode = dst->node; @@ -344,7 +344,7 @@ static int tipc_mcast_send_sync(struct net *net, struct sk_buff *skb, msg_set_size(_hdr, MCAST_H_SIZE); msg_set_is_rcast(_hdr, !msg_is_rcast(hdr)); - skb_queue_head_init(&tmpq); + __skb_queue_head_init(&tmpq); __skb_queue_tail(&tmpq, _skb); if (method->rcast) tipc_bcast_xmit(net, &tmpq, cong_link_cnt); @@ -378,7 +378,7 @@ int tipc_mcast_xmit(struct net *net, struct sk_buff_head *pkts, int rc = 0; skb_queue_head_init(&inputq); - skb_queue_head_init(&localq); + __skb_queue_head_init(&localq); /* Clone packets before they are consumed by next call */ if (dests->local && !tipc_msg_reassemble(pkts, &localq)) { diff --git a/net/tipc/group.c b/net/tipc/group.c index 5f98d38bcf08..89257e2a980d 100644 --- a/net/tipc/group.c +++ b/net/tipc/group.c @@ -199,7 +199,7 @@ void tipc_group_join(struct net *net, struct tipc_group *grp, int *sk_rcvbuf) struct tipc_member *m, *tmp; struct sk_buff_head xmitq; - skb_queue_head_init(&xmitq); + __skb_queue_head_init(&xmitq); rbtree_postorder_for_each_entry_safe(m, tmp, tree, tree_node) { tipc_group_proto_xmit(grp, m, GRP_JOIN_MSG, &xmitq); tipc_group_update_member(m, 0); @@ -435,7 +435,7 @@ bool tipc_group_cong(struct tipc_group *grp, u32 dnode, u32 dport, return true; if (state == MBR_PENDING && adv == ADV_IDLE) return true; - skb_queue_head_init(&xmitq); + __skb_queue_head_init(&xmitq); tipc_group_proto_xmit(grp, m, GRP_ADV_MSG, &xmitq); tipc_node_distr_xmit(grp->net, &xmitq); return true; diff --git a/net/tipc/link.c b/net/tipc/link.c index dd3155b14654..6cc75ffd9e2c 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -106,8 +106,6 @@ struct tipc_stats { * @transmitq: queue for sent, non-acked messages * @backlogq: queue for messages waiting to be sent * @snt_nxt: next sequence number to use for outbound messages - * @prev_from: sequence number of most previous retransmission request - * @stale_limit: time when repeated identical retransmits must force link reset * @ackers: # of peers that needs to ack each packet before it can be released * @acked: # last packet acked by a certain peer. Used for broadcast. * @rcv_nxt: next sequence number to expect for inbound messages @@ -164,9 +162,7 @@ struct tipc_link { u16 limit; } backlog[5]; u16 snd_nxt; - u16 prev_from; u16 window; - unsigned long stale_limit; /* Reception */ u16 rcv_nxt; @@ -959,7 +955,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list, pr_warn("Too large msg, purging xmit list %d %d %d %d %d!\n", skb_queue_len(list), msg_user(hdr), msg_type(hdr), msg_size(hdr), mtu); - skb_queue_purge(list); + __skb_queue_purge(list); return -EMSGSIZE; } @@ -988,7 +984,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list, if (likely(skb_queue_len(transmq) < maxwin)) { _skb = skb_clone(skb, GFP_ATOMIC); if (!_skb) { - skb_queue_purge(list); + __skb_queue_purge(list); return -ENOBUFS; } __skb_dequeue(list); @@ -1063,47 +1059,53 @@ static void tipc_link_advance_backlog(struct tipc_link *l, * link_retransmit_failure() - Detect repeated retransmit failures * @l: tipc link sender * @r: tipc link receiver (= l in case of unicast) - * @from: seqno of the 1st packet in retransmit request * @rc: returned code * * Return: true if the repeated retransmit failures happens, otherwise * false */ static bool link_retransmit_failure(struct tipc_link *l, struct tipc_link *r, - u16 from, int *rc) + int *rc) { struct sk_buff *skb = skb_peek(&l->transmq); struct tipc_msg *hdr; if (!skb) return false; - hdr = buf_msg(skb); - /* Detect repeated retransmit failures on same packet */ - if (r->prev_from != from) { - r->prev_from = from; - r->stale_limit = jiffies + msecs_to_jiffies(r->tolerance); - } else if (time_after(jiffies, r->stale_limit)) { - pr_warn("Retransmission failure on link <%s>\n", l->name); - link_print(l, "State of link "); - pr_info("Failed msg: usr %u, typ %u, len %u, err %u\n", - msg_user(hdr), msg_type(hdr), msg_size(hdr), - msg_errcode(hdr)); - pr_info("sqno %u, prev: %x, src: %x\n", - msg_seqno(hdr), msg_prevnode(hdr), msg_orignode(hdr)); - - trace_tipc_list_dump(&l->transmq, true, "retrans failure!"); - trace_tipc_link_dump(l, TIPC_DUMP_NONE, "retrans failure!"); - trace_tipc_link_dump(r, TIPC_DUMP_NONE, "retrans failure!"); + if (!TIPC_SKB_CB(skb)->retr_cnt) + return false; - if (link_is_bc_sndlink(l)) - *rc = TIPC_LINK_DOWN_EVT; + if (!time_after(jiffies, TIPC_SKB_CB(skb)->retr_stamp + + msecs_to_jiffies(r->tolerance))) + return false; + + hdr = buf_msg(skb); + if (link_is_bc_sndlink(l) && !less(r->acked, msg_seqno(hdr))) + return false; + pr_warn("Retransmission failure on link <%s>\n", l->name); + link_print(l, "State of link "); + pr_info("Failed msg: usr %u, typ %u, len %u, err %u\n", + msg_user(hdr), msg_type(hdr), msg_size(hdr), msg_errcode(hdr)); + pr_info("sqno %u, prev: %x, dest: %x\n", + msg_seqno(hdr), msg_prevnode(hdr), msg_destnode(hdr)); + pr_info("retr_stamp %d, retr_cnt %d\n", + jiffies_to_msecs(TIPC_SKB_CB(skb)->retr_stamp), + TIPC_SKB_CB(skb)->retr_cnt); + + trace_tipc_list_dump(&l->transmq, true, "retrans failure!"); + trace_tipc_link_dump(l, TIPC_DUMP_NONE, "retrans failure!"); + trace_tipc_link_dump(r, TIPC_DUMP_NONE, "retrans failure!"); + + if (link_is_bc_sndlink(l)) { + r->state = LINK_RESET; + *rc = TIPC_LINK_DOWN_EVT; + } else { *rc = tipc_link_fsm_evt(l, LINK_FAILURE_EVT); - return true; } - return false; + return true; } /* tipc_link_bc_retrans() - retransmit zero or more packets @@ -1129,7 +1131,7 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r, trace_tipc_link_retrans(r, from, to, &l->transmq); - if (link_retransmit_failure(l, r, from, &rc)) + if (link_retransmit_failure(l, r, &rc)) return rc; skb_queue_walk(&l->transmq, skb) { @@ -1138,11 +1140,10 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r, continue; if (more(msg_seqno(hdr), to)) break; - if (link_is_bc_sndlink(l)) { - if (time_before(jiffies, TIPC_SKB_CB(skb)->nxt_retr)) - continue; - TIPC_SKB_CB(skb)->nxt_retr = TIPC_BC_RETR_LIM; - } + + if (time_before(jiffies, TIPC_SKB_CB(skb)->nxt_retr)) + continue; + TIPC_SKB_CB(skb)->nxt_retr = TIPC_BC_RETR_LIM; _skb = __pskb_copy(skb, LL_MAX_HEADER + MIN_H_SIZE, GFP_ATOMIC); if (!_skb) return 0; @@ -1152,6 +1153,10 @@ static int tipc_link_bc_retrans(struct tipc_link *l, struct tipc_link *r, _skb->priority = TC_PRIO_CONTROL; __skb_queue_tail(xmitq, _skb); l->stats.retransmitted++; + + /* Increase actual retrans counter & mark first time */ + if (!TIPC_SKB_CB(skb)->retr_cnt++) + TIPC_SKB_CB(skb)->retr_stamp = jiffies; } return 0; } @@ -1393,12 +1398,10 @@ static int tipc_link_advance_transmq(struct tipc_link *l, u16 acked, u16 gap, struct tipc_msg *hdr; u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1; u16 ack = l->rcv_nxt - 1; + bool passed = false; u16 seqno, n = 0; int rc = 0; - if (gap && link_retransmit_failure(l, l, acked + 1, &rc)) - return rc; - skb_queue_walk_safe(&l->transmq, skb, tmp) { seqno = buf_seqno(skb); @@ -1408,12 +1411,17 @@ next_gap_ack: __skb_unlink(skb, &l->transmq); kfree_skb(skb); } else if (less_eq(seqno, acked + gap)) { - /* retransmit skb */ + /* First, check if repeated retrans failures occurs? */ + if (!passed && link_retransmit_failure(l, l, &rc)) + return rc; + passed = true; + + /* retransmit skb if unrestricted*/ if (time_before(jiffies, TIPC_SKB_CB(skb)->nxt_retr)) continue; TIPC_SKB_CB(skb)->nxt_retr = TIPC_UC_RETR_TIME; - - _skb = __pskb_copy(skb, MIN_H_SIZE, GFP_ATOMIC); + _skb = __pskb_copy(skb, LL_MAX_HEADER + MIN_H_SIZE, + GFP_ATOMIC); if (!_skb) continue; hdr = buf_msg(_skb); @@ -1422,6 +1430,10 @@ next_gap_ack: _skb->priority = TC_PRIO_CONTROL; __skb_queue_tail(xmitq, _skb); l->stats.retransmitted++; + + /* Increase actual retrans counter & mark first time */ + if (!TIPC_SKB_CB(skb)->retr_cnt++) + TIPC_SKB_CB(skb)->retr_stamp = jiffies; } else { /* retry with Gap ACK blocks if any */ if (!ga || n >= ga->gack_cnt) @@ -1668,7 +1680,7 @@ void tipc_link_create_dummy_tnl_msg(struct tipc_link *l, struct sk_buff *skb; u32 dnode = l->addr; - skb_queue_head_init(&tnlq); + __skb_queue_head_init(&tnlq); skb = tipc_msg_create(TUNNEL_PROTOCOL, FAILOVER_MSG, INT_H_SIZE, BASIC_H_SIZE, dnode, onode, 0, 0, 0); @@ -1708,9 +1720,9 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl, if (!tnl) return; - skb_queue_head_init(&tnlq); - skb_queue_head_init(&tmpxq); - skb_queue_head_init(&frags); + __skb_queue_head_init(&tnlq); + __skb_queue_head_init(&tmpxq); + __skb_queue_head_init(&frags); /* At least one packet required for safe algorithm => add dummy */ skb = tipc_msg_create(TIPC_LOW_IMPORTANCE, TIPC_DIRECT_MSG, @@ -1720,7 +1732,7 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl, pr_warn("%sunable to create tunnel packet\n", link_co_err); return; } - skb_queue_tail(&tnlq, skb); + __skb_queue_tail(&tnlq, skb); tipc_link_xmit(l, &tnlq, &tmpxq); __skb_queue_purge(&tmpxq); @@ -2681,7 +2693,7 @@ int tipc_link_dump(struct tipc_link *l, u16 dqueues, char *buf) i += scnprintf(buf + i, sz - i, " %x", l->peer_caps); i += scnprintf(buf + i, sz - i, " %u", l->silent_intv_cnt); i += scnprintf(buf + i, sz - i, " %u", l->rst_cnt); - i += scnprintf(buf + i, sz - i, " %u", l->prev_from); + i += scnprintf(buf + i, sz - i, " %u", 0); i += scnprintf(buf + i, sz - i, " %u", 0); i += scnprintf(buf + i, sz - i, " %u", l->acked); diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 1c8c8dd32a4e..0daa6f04ca81 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -102,13 +102,15 @@ struct plist; #define TIPC_MEDIA_INFO_OFFSET 5 struct tipc_skb_cb { - u32 bytes_read; - u32 orig_member; struct sk_buff *tail; unsigned long nxt_retr; - bool validated; + unsigned long retr_stamp; + u32 bytes_read; + u32 orig_member; u16 chain_imp; u16 ackers; + u16 retr_cnt; + bool validated; }; #define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 44abc8e9c990..61219f0b9677 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -190,7 +190,7 @@ void tipc_named_node_up(struct net *net, u32 dnode) struct name_table *nt = tipc_name_table(net); struct sk_buff_head head; - skb_queue_head_init(&head); + __skb_queue_head_init(&head); read_lock_bh(&nt->cluster_scope_lock); named_distribute(net, &head, dnode, &nt->cluster_scope); diff --git a/net/tipc/node.c b/net/tipc/node.c index 1bdcf0fc1a4d..c8f6177dd5a2 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1444,13 +1444,14 @@ int tipc_node_xmit(struct net *net, struct sk_buff_head *list, if (in_own_node(net, dnode)) { tipc_loopback_trace(net, list); + spin_lock_init(&list->lock); tipc_sk_rcv(net, list); return 0; } n = tipc_node_find(net, dnode); if (unlikely(!n)) { - skb_queue_purge(list); + __skb_queue_purge(list); return -EHOSTUNREACH; } @@ -1459,7 +1460,7 @@ int tipc_node_xmit(struct net *net, struct sk_buff_head *list, if (unlikely(bearer_id == INVALID_BEARER_ID)) { tipc_node_read_unlock(n); tipc_node_put(n); - skb_queue_purge(list); + __skb_queue_purge(list); return -EHOSTUNREACH; } @@ -1491,7 +1492,7 @@ int tipc_node_xmit_skb(struct net *net, struct sk_buff *skb, u32 dnode, { struct sk_buff_head head; - skb_queue_head_init(&head); + __skb_queue_head_init(&head); __skb_queue_tail(&head, skb); tipc_node_xmit(net, &head, dnode, selector); return 0; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 83ae41d7e554..3b9f8cc328f5 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -809,7 +809,7 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq, msg_set_nameupper(hdr, seq->upper); /* Build message as chain of buffers */ - skb_queue_head_init(&pkts); + __skb_queue_head_init(&pkts); rc = tipc_msg_build(hdr, msg, 0, dlen, mtu, &pkts); /* Send message if build was successful */ @@ -853,7 +853,7 @@ static int tipc_send_group_msg(struct net *net, struct tipc_sock *tsk, msg_set_grp_bc_seqno(hdr, bc_snd_nxt); /* Build message as chain of buffers */ - skb_queue_head_init(&pkts); + __skb_queue_head_init(&pkts); mtu = tipc_node_get_mtu(net, dnode, tsk->portid); rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); if (unlikely(rc != dlen)) @@ -1058,7 +1058,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m, msg_set_grp_bc_ack_req(hdr, ack); /* Build message as chain of buffers */ - skb_queue_head_init(&pkts); + __skb_queue_head_init(&pkts); rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); if (unlikely(rc != dlen)) return rc; @@ -1387,7 +1387,7 @@ static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen) if (unlikely(rc)) return rc; - skb_queue_head_init(&pkts); + __skb_queue_head_init(&pkts); mtu = tipc_node_get_mtu(net, dnode, tsk->portid); rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts); if (unlikely(rc != dlen)) @@ -1445,7 +1445,7 @@ static int __tipc_sendstream(struct socket *sock, struct msghdr *m, size_t dlen) int send, sent = 0; int rc = 0; - skb_queue_head_init(&pkts); + __skb_queue_head_init(&pkts); if (unlikely(dlen > INT_MAX)) return -EMSGSIZE; @@ -1805,7 +1805,7 @@ static int tipc_recvmsg(struct socket *sock, struct msghdr *m, /* Send group flow control advertisement when applicable */ if (tsk->group && msg_in_group(hdr) && !grp_evt) { - skb_queue_head_init(&xmitq); + __skb_queue_head_init(&xmitq); tipc_group_update_rcv_win(tsk->group, tsk_blocks(hlen + dlen), msg_orignode(hdr), msg_origport(hdr), &xmitq); @@ -2674,7 +2674,7 @@ static void tipc_sk_timeout(struct timer_list *t) struct sk_buff_head list; int rc = 0; - skb_queue_head_init(&list); + __skb_queue_head_init(&list); bh_lock_sock(sk); /* Try again later if socket is busy */ diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index d184230665eb..41c106e45f01 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -61,7 +61,7 @@ static void tls_device_free_ctx(struct tls_context *ctx) if (ctx->rx_conf == TLS_HW) kfree(tls_offload_ctx_rx(ctx)); - tls_ctx_free(ctx); + tls_ctx_free(NULL, ctx); } static void tls_device_gc_task(struct work_struct *work) @@ -159,12 +159,8 @@ static void tls_icsk_clean_acked(struct sock *sk, u32 acked_seq) spin_lock_irqsave(&ctx->lock, flags); info = ctx->retransmit_hint; - if (info && !before(acked_seq, info->end_seq)) { + if (info && !before(acked_seq, info->end_seq)) ctx->retransmit_hint = NULL; - list_del(&info->list); - destroy_record(info); - deleted_records++; - } list_for_each_entry_safe(info, temp, &ctx->records_list, list) { if (before(acked_seq, info->end_seq)) @@ -373,9 +369,9 @@ static int tls_push_data(struct sock *sk, struct tls_context *tls_ctx = tls_get_ctx(sk); struct tls_prot_info *prot = &tls_ctx->prot_info; struct tls_offload_context_tx *ctx = tls_offload_ctx_tx(tls_ctx); - int tls_push_record_flags = flags | MSG_SENDPAGE_NOTLAST; int more = flags & (MSG_SENDPAGE_NOTLAST | MSG_MORE); struct tls_record_info *record = ctx->open_record; + int tls_push_record_flags; struct page_frag *pfrag; size_t orig_size = size; u32 max_open_record_len; @@ -390,6 +386,9 @@ static int tls_push_data(struct sock *sk, if (sk->sk_err) return -sk->sk_err; + flags |= MSG_SENDPAGE_DECRYPTED; + tls_push_record_flags = flags | MSG_SENDPAGE_NOTLAST; + timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT); if (tls_is_partially_sent_record(tls_ctx)) { rc = tls_push_partial_record(sk, tls_ctx, flags); @@ -576,7 +575,9 @@ void tls_device_write_space(struct sock *sk, struct tls_context *ctx) gfp_t sk_allocation = sk->sk_allocation; sk->sk_allocation = GFP_ATOMIC; - tls_push_partial_record(sk, ctx, MSG_DONTWAIT | MSG_NOSIGNAL); + tls_push_partial_record(sk, ctx, + MSG_DONTWAIT | MSG_NOSIGNAL | + MSG_SENDPAGE_DECRYPTED); sk->sk_allocation = sk_allocation; } } @@ -833,22 +834,18 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx) struct net_device *netdev; char *iv, *rec_seq; struct sk_buff *skb; - int rc = -EINVAL; __be64 rcd_sn; + int rc; if (!ctx) - goto out; + return -EINVAL; - if (ctx->priv_ctx_tx) { - rc = -EEXIST; - goto out; - } + if (ctx->priv_ctx_tx) + return -EEXIST; start_marker_record = kmalloc(sizeof(*start_marker_record), GFP_KERNEL); - if (!start_marker_record) { - rc = -ENOMEM; - goto out; - } + if (!start_marker_record) + return -ENOMEM; offload_ctx = kzalloc(TLS_OFFLOAD_CONTEXT_SIZE_TX, GFP_KERNEL); if (!offload_ctx) { @@ -934,17 +931,11 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx) if (skb) TCP_SKB_CB(skb)->eor = 1; - /* We support starting offload on multiple sockets - * concurrently, so we only need a read lock here. - * This lock must precede get_netdev_for_sock to prevent races between - * NETDEV_DOWN and setsockopt. - */ - down_read(&device_offload_lock); netdev = get_netdev_for_sock(sk); if (!netdev) { pr_err_ratelimited("%s: netdev not found\n", __func__); rc = -EINVAL; - goto release_lock; + goto disable_cad; } if (!(netdev->features & NETIF_F_HW_TLS_TX)) { @@ -955,10 +946,15 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx) /* Avoid offloading if the device is down * We don't want to offload new flows after * the NETDEV_DOWN event + * + * device_offload_lock is taken in tls_devices's NETDEV_DOWN + * handler thus protecting from the device going down before + * ctx was added to tls_device_list. */ + down_read(&device_offload_lock); if (!(netdev->flags & IFF_UP)) { rc = -EINVAL; - goto release_netdev; + goto release_lock; } ctx->priv_ctx_tx = offload_ctx; @@ -966,9 +962,10 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx) &ctx->crypto_send.info, tcp_sk(sk)->write_seq); if (rc) - goto release_netdev; + goto release_lock; tls_device_attach(ctx, sk, netdev); + up_read(&device_offload_lock); /* following this assignment tls_is_sk_tx_device_offloaded * will return true and the context might be accessed @@ -976,13 +973,14 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx) */ smp_store_release(&sk->sk_validate_xmit_skb, tls_validate_xmit_skb); dev_put(netdev); - up_read(&device_offload_lock); - goto out; -release_netdev: - dev_put(netdev); + return 0; + release_lock: up_read(&device_offload_lock); +release_netdev: + dev_put(netdev); +disable_cad: clean_acked_data_disable(inet_csk(sk)); crypto_free_aead(offload_ctx->aead_send); free_rec_seq: @@ -994,7 +992,6 @@ free_offload_ctx: ctx->priv_ctx_tx = NULL; free_marker_record: kfree(start_marker_record); -out: return rc; } @@ -1007,17 +1004,10 @@ int tls_set_device_offload_rx(struct sock *sk, struct tls_context *ctx) if (ctx->crypto_recv.info.version != TLS_1_2_VERSION) return -EOPNOTSUPP; - /* We support starting offload on multiple sockets - * concurrently, so we only need a read lock here. - * This lock must precede get_netdev_for_sock to prevent races between - * NETDEV_DOWN and setsockopt. - */ - down_read(&device_offload_lock); netdev = get_netdev_for_sock(sk); if (!netdev) { pr_err_ratelimited("%s: netdev not found\n", __func__); - rc = -EINVAL; - goto release_lock; + return -EINVAL; } if (!(netdev->features & NETIF_F_HW_TLS_RX)) { @@ -1028,16 +1018,21 @@ int tls_set_device_offload_rx(struct sock *sk, struct tls_context *ctx) /* Avoid offloading if the device is down * We don't want to offload new flows after * the NETDEV_DOWN event + * + * device_offload_lock is taken in tls_devices's NETDEV_DOWN + * handler thus protecting from the device going down before + * ctx was added to tls_device_list. */ + down_read(&device_offload_lock); if (!(netdev->flags & IFF_UP)) { rc = -EINVAL; - goto release_netdev; + goto release_lock; } context = kzalloc(TLS_OFFLOAD_CONTEXT_SIZE_RX, GFP_KERNEL); if (!context) { rc = -ENOMEM; - goto release_netdev; + goto release_lock; } context->resync_nh_reset = 1; @@ -1053,7 +1048,11 @@ int tls_set_device_offload_rx(struct sock *sk, struct tls_context *ctx) goto free_sw_resources; tls_device_attach(ctx, sk, netdev); - goto release_netdev; + up_read(&device_offload_lock); + + dev_put(netdev); + + return 0; free_sw_resources: up_read(&device_offload_lock); @@ -1061,10 +1060,10 @@ free_sw_resources: down_read(&device_offload_lock); release_ctx: ctx->priv_ctx_rx = NULL; -release_netdev: - dev_put(netdev); release_lock: up_read(&device_offload_lock); +release_netdev: + dev_put(netdev); return rc; } diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index 9cbbae606ced..ac88877dcade 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -39,6 +39,7 @@ #include <linux/netdevice.h> #include <linux/sched/signal.h> #include <linux/inetdevice.h> +#include <linux/inet_diag.h> #include <net/tls.h> @@ -251,14 +252,26 @@ static void tls_write_space(struct sock *sk) ctx->sk_write_space(sk); } -void tls_ctx_free(struct tls_context *ctx) +/** + * tls_ctx_free() - free TLS ULP context + * @sk: socket to with @ctx is attached + * @ctx: TLS context structure + * + * Free TLS context. If @sk is %NULL caller guarantees that the socket + * to which @ctx was attached has no outstanding references. + */ +void tls_ctx_free(struct sock *sk, struct tls_context *ctx) { if (!ctx) return; memzero_explicit(&ctx->crypto_send, sizeof(ctx->crypto_send)); memzero_explicit(&ctx->crypto_recv, sizeof(ctx->crypto_recv)); - kfree(ctx); + + if (sk) + kfree_rcu(ctx, rcu); + else + kfree(ctx); } static void tls_sk_proto_cleanup(struct sock *sk, @@ -273,19 +286,14 @@ static void tls_sk_proto_cleanup(struct sock *sk, kfree(ctx->tx.rec_seq); kfree(ctx->tx.iv); tls_sw_release_resources_tx(sk); -#ifdef CONFIG_TLS_DEVICE } else if (ctx->tx_conf == TLS_HW) { tls_device_free_resources_tx(sk); -#endif } if (ctx->rx_conf == TLS_SW) tls_sw_release_resources_rx(sk); - -#ifdef CONFIG_TLS_DEVICE - if (ctx->rx_conf == TLS_HW) + else if (ctx->rx_conf == TLS_HW) tls_device_offload_cleanup_rx(sk); -#endif } static void tls_sk_proto_close(struct sock *sk, long timeout) @@ -306,8 +314,10 @@ static void tls_sk_proto_close(struct sock *sk, long timeout) write_lock_bh(&sk->sk_callback_lock); if (free_ctx) - icsk->icsk_ulp_data = NULL; + rcu_assign_pointer(icsk->icsk_ulp_data, NULL); sk->sk_prot = ctx->sk_proto; + if (sk->sk_write_space == tls_write_space) + sk->sk_write_space = ctx->sk_write_space; write_unlock_bh(&sk->sk_callback_lock); release_sock(sk); if (ctx->tx_conf == TLS_SW) @@ -316,10 +326,10 @@ static void tls_sk_proto_close(struct sock *sk, long timeout) tls_sw_strparser_done(ctx); if (ctx->rx_conf == TLS_SW) tls_sw_free_ctx_rx(ctx); - ctx->sk_proto_close(sk, timeout); + ctx->sk_proto->close(sk, timeout); if (free_ctx) - tls_ctx_free(ctx); + tls_ctx_free(sk, ctx); } static int do_tls_getsockopt_tx(struct sock *sk, char __user *optval, @@ -436,7 +446,8 @@ static int tls_getsockopt(struct sock *sk, int level, int optname, struct tls_context *ctx = tls_get_ctx(sk); if (level != SOL_TLS) - return ctx->getsockopt(sk, level, optname, optval, optlen); + return ctx->sk_proto->getsockopt(sk, level, + optname, optval, optlen); return do_tls_getsockopt(sk, optname, optval, optlen); } @@ -521,26 +532,18 @@ static int do_tls_setsockopt_conf(struct sock *sk, char __user *optval, } if (tx) { -#ifdef CONFIG_TLS_DEVICE rc = tls_set_device_offload(sk, ctx); conf = TLS_HW; if (rc) { -#else - { -#endif rc = tls_set_sw_offload(sk, ctx, 1); if (rc) goto err_crypto_info; conf = TLS_SW; } } else { -#ifdef CONFIG_TLS_DEVICE rc = tls_set_device_offload_rx(sk, ctx); conf = TLS_HW; if (rc) { -#else - { -#endif rc = tls_set_sw_offload(sk, ctx, 0); if (rc) goto err_crypto_info; @@ -594,7 +597,8 @@ static int tls_setsockopt(struct sock *sk, int level, int optname, struct tls_context *ctx = tls_get_ctx(sk); if (level != SOL_TLS) - return ctx->setsockopt(sk, level, optname, optval, optlen); + return ctx->sk_proto->setsockopt(sk, level, optname, optval, + optlen); return do_tls_setsockopt(sk, optname, optval, optlen); } @@ -608,11 +612,8 @@ static struct tls_context *create_ctx(struct sock *sk) if (!ctx) return NULL; - icsk->icsk_ulp_data = ctx; - ctx->setsockopt = sk->sk_prot->setsockopt; - ctx->getsockopt = sk->sk_prot->getsockopt; - ctx->sk_proto_close = sk->sk_prot->close; - ctx->unhash = sk->sk_prot->unhash; + rcu_assign_pointer(icsk->icsk_ulp_data, ctx); + ctx->sk_proto = sk->sk_prot; return ctx; } @@ -649,8 +650,8 @@ static void tls_hw_sk_destruct(struct sock *sk) ctx->sk_destruct(sk); /* Free ctx */ - tls_ctx_free(ctx); - icsk->icsk_ulp_data = NULL; + rcu_assign_pointer(icsk->icsk_ulp_data, NULL); + tls_ctx_free(sk, ctx); } static int tls_hw_prot(struct sock *sk) @@ -668,9 +669,6 @@ static int tls_hw_prot(struct sock *sk) spin_unlock_bh(&device_spinlock); tls_build_proto(sk); - ctx->hash = sk->sk_prot->hash; - ctx->unhash = sk->sk_prot->unhash; - ctx->sk_proto_close = sk->sk_prot->close; ctx->sk_destruct = sk->sk_destruct; sk->sk_destruct = tls_hw_sk_destruct; ctx->rx_conf = TLS_HW_RECORD; @@ -702,7 +700,7 @@ static void tls_hw_unhash(struct sock *sk) } } spin_unlock_bh(&device_spinlock); - ctx->unhash(sk); + ctx->sk_proto->unhash(sk); } static int tls_hw_hash(struct sock *sk) @@ -711,7 +709,7 @@ static int tls_hw_hash(struct sock *sk) struct tls_device *dev; int err; - err = ctx->hash(sk); + err = ctx->sk_proto->hash(sk); spin_lock_bh(&device_spinlock); list_for_each_entry(dev, &device_list, dev_list) { if (dev->hash) { @@ -801,7 +799,6 @@ static int tls_init(struct sock *sk) ctx->tx_conf = TLS_BASE; ctx->rx_conf = TLS_BASE; - ctx->sk_proto = sk->sk_prot; update_sk_prot(sk, ctx); out: write_unlock_bh(&sk->sk_callback_lock); @@ -813,12 +810,71 @@ static void tls_update(struct sock *sk, struct proto *p) struct tls_context *ctx; ctx = tls_get_ctx(sk); - if (likely(ctx)) { - ctx->sk_proto_close = p->close; + if (likely(ctx)) ctx->sk_proto = p; - } else { + else sk->sk_prot = p; - } +} + +static int tls_get_info(const struct sock *sk, struct sk_buff *skb) +{ + u16 version, cipher_type; + struct tls_context *ctx; + struct nlattr *start; + int err; + + start = nla_nest_start_noflag(skb, INET_ULP_INFO_TLS); + if (!start) + return -EMSGSIZE; + + rcu_read_lock(); + ctx = rcu_dereference(inet_csk(sk)->icsk_ulp_data); + if (!ctx) { + err = 0; + goto nla_failure; + } + version = ctx->prot_info.version; + if (version) { + err = nla_put_u16(skb, TLS_INFO_VERSION, version); + if (err) + goto nla_failure; + } + cipher_type = ctx->prot_info.cipher_type; + if (cipher_type) { + err = nla_put_u16(skb, TLS_INFO_CIPHER, cipher_type); + if (err) + goto nla_failure; + } + err = nla_put_u16(skb, TLS_INFO_TXCONF, tls_user_config(ctx, true)); + if (err) + goto nla_failure; + + err = nla_put_u16(skb, TLS_INFO_RXCONF, tls_user_config(ctx, false)); + if (err) + goto nla_failure; + + rcu_read_unlock(); + nla_nest_end(skb, start); + return 0; + +nla_failure: + rcu_read_unlock(); + nla_nest_cancel(skb, start); + return err; +} + +static size_t tls_get_info_size(const struct sock *sk) +{ + size_t size = 0; + + size += nla_total_size(0) + /* INET_ULP_INFO_TLS */ + nla_total_size(sizeof(u16)) + /* TLS_INFO_VERSION */ + nla_total_size(sizeof(u16)) + /* TLS_INFO_CIPHER */ + nla_total_size(sizeof(u16)) + /* TLS_INFO_RXCONF */ + nla_total_size(sizeof(u16)) + /* TLS_INFO_TXCONF */ + 0; + + return size; } void tls_register_device(struct tls_device *device) @@ -842,6 +898,8 @@ static struct tcp_ulp_ops tcp_tls_ulp_ops __read_mostly = { .owner = THIS_MODULE, .init = tls_init, .update = tls_update, + .get_info = tls_get_info, + .get_info_size = tls_get_info_size, }; static int __init tls_register(void) @@ -849,9 +907,7 @@ static int __init tls_register(void) tls_sw_proto_ops = inet_stream_ops; tls_sw_proto_ops.splice_read = tls_sw_splice_read; -#ifdef CONFIG_TLS_DEVICE tls_device_init(); -#endif tcp_register_ulp(&tcp_tls_ulp_ops); return 0; @@ -860,9 +916,7 @@ static int __init tls_register(void) static void __exit tls_unregister(void) { tcp_unregister_ulp(&tcp_tls_ulp_ops); -#ifdef CONFIG_TLS_DEVICE tls_device_cleanup(); -#endif } module_init(tls_register); diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 91d21b048a9b..c2b5e0d2ba1a 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1489,13 +1489,12 @@ static int decrypt_skb_update(struct sock *sk, struct sk_buff *skb, int pad, err = 0; if (!ctx->decrypted) { -#ifdef CONFIG_TLS_DEVICE if (tls_ctx->rx_conf == TLS_HW) { err = tls_device_decrypted(sk, skb); if (err < 0) return err; } -#endif + /* Still not decrypted after tls_device */ if (!ctx->decrypted) { err = decrypt_internal(sk, skb, dest, NULL, chunk, zc, @@ -2014,10 +2013,9 @@ static int tls_read_size(struct strparser *strp, struct sk_buff *skb) ret = -EINVAL; goto read_failure; } -#ifdef CONFIG_TLS_DEVICE + tls_device_rx_resync_new_rec(strp->sk, data_len + TLS_HEADER_SIZE, TCP_SKB_CB(skb)->seq + rxm->offset); -#endif return data_len + TLS_HEADER_SIZE; read_failure: diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 94cc0fa3e848..5bb70c692b1e 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -307,8 +307,13 @@ virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, spin_unlock_bh(&vvs->rx_lock); - /* We send a credit update only when the space available seen - * by the transmitter is less than VIRTIO_VSOCK_MAX_PKT_BUF_SIZE + /* To reduce the number of credit update messages, + * don't update credits as long as lots of space is available. + * Note: the limit chosen here is arbitrary. Setting the limit + * too high causes extra messages. Too low causes transmitter + * stalls. As stalls are in theory more expensive than extra + * messages, we set the limit to a high value. TODO: experiment + * with different values. */ if (free_space < VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) { virtio_transport_send_credit_update(vsk, diff --git a/net/wireless/chan.c b/net/wireless/chan.c index 7dc1bbd0888f..e851cafd8e2f 100644 --- a/net/wireless/chan.c +++ b/net/wireless/chan.c @@ -14,6 +14,11 @@ #include "core.h" #include "rdev-ops.h" +static bool cfg80211_valid_60g_freq(u32 freq) +{ + return freq >= 58320 && freq <= 70200; +} + void cfg80211_chandef_create(struct cfg80211_chan_def *chandef, struct ieee80211_channel *chan, enum nl80211_channel_type chan_type) @@ -23,6 +28,8 @@ void cfg80211_chandef_create(struct cfg80211_chan_def *chandef, chandef->chan = chan; chandef->center_freq2 = 0; + chandef->edmg.bw_config = 0; + chandef->edmg.channels = 0; switch (chan_type) { case NL80211_CHAN_NO_HT: @@ -47,6 +54,91 @@ void cfg80211_chandef_create(struct cfg80211_chan_def *chandef, } EXPORT_SYMBOL(cfg80211_chandef_create); +static bool cfg80211_edmg_chandef_valid(const struct cfg80211_chan_def *chandef) +{ + int max_contiguous = 0; + int num_of_enabled = 0; + int contiguous = 0; + int i; + + if (!chandef->edmg.channels || !chandef->edmg.bw_config) + return false; + + if (!cfg80211_valid_60g_freq(chandef->chan->center_freq)) + return false; + + for (i = 0; i < 6; i++) { + if (chandef->edmg.channels & BIT(i)) { + contiguous++; + num_of_enabled++; + } else { + contiguous = 0; + } + + max_contiguous = max(contiguous, max_contiguous); + } + /* basic verification of edmg configuration according to + * IEEE P802.11ay/D4.0 section 9.4.2.251 + */ + /* check bw_config against contiguous edmg channels */ + switch (chandef->edmg.bw_config) { + case IEEE80211_EDMG_BW_CONFIG_4: + case IEEE80211_EDMG_BW_CONFIG_8: + case IEEE80211_EDMG_BW_CONFIG_12: + if (max_contiguous < 1) + return false; + break; + case IEEE80211_EDMG_BW_CONFIG_5: + case IEEE80211_EDMG_BW_CONFIG_9: + case IEEE80211_EDMG_BW_CONFIG_13: + if (max_contiguous < 2) + return false; + break; + case IEEE80211_EDMG_BW_CONFIG_6: + case IEEE80211_EDMG_BW_CONFIG_10: + case IEEE80211_EDMG_BW_CONFIG_14: + if (max_contiguous < 3) + return false; + break; + case IEEE80211_EDMG_BW_CONFIG_7: + case IEEE80211_EDMG_BW_CONFIG_11: + case IEEE80211_EDMG_BW_CONFIG_15: + if (max_contiguous < 4) + return false; + break; + + default: + return false; + } + + /* check bw_config against aggregated (non contiguous) edmg channels */ + switch (chandef->edmg.bw_config) { + case IEEE80211_EDMG_BW_CONFIG_4: + case IEEE80211_EDMG_BW_CONFIG_5: + case IEEE80211_EDMG_BW_CONFIG_6: + case IEEE80211_EDMG_BW_CONFIG_7: + break; + case IEEE80211_EDMG_BW_CONFIG_8: + case IEEE80211_EDMG_BW_CONFIG_9: + case IEEE80211_EDMG_BW_CONFIG_10: + case IEEE80211_EDMG_BW_CONFIG_11: + if (num_of_enabled < 2) + return false; + break; + case IEEE80211_EDMG_BW_CONFIG_12: + case IEEE80211_EDMG_BW_CONFIG_13: + case IEEE80211_EDMG_BW_CONFIG_14: + case IEEE80211_EDMG_BW_CONFIG_15: + if (num_of_enabled < 4 || max_contiguous < 2) + return false; + break; + default: + return false; + } + + return true; +} + bool cfg80211_chandef_valid(const struct cfg80211_chan_def *chandef) { u32 control_freq; @@ -112,6 +204,10 @@ bool cfg80211_chandef_valid(const struct cfg80211_chan_def *chandef) return false; } + if (cfg80211_chandef_is_edmg(chandef) && + !cfg80211_edmg_chandef_valid(chandef)) + return false; + return true; } EXPORT_SYMBOL(cfg80211_chandef_valid); @@ -721,12 +817,66 @@ static bool cfg80211_secondary_chans_ok(struct wiphy *wiphy, return true; } +/* check if the operating channels are valid and supported */ +static bool cfg80211_edmg_usable(struct wiphy *wiphy, u8 edmg_channels, + enum ieee80211_edmg_bw_config edmg_bw_config, + int primary_channel, + struct ieee80211_edmg *edmg_cap) +{ + struct ieee80211_channel *chan; + int i, freq; + int channels_counter = 0; + + if (!edmg_channels && !edmg_bw_config) + return true; + + if ((!edmg_channels && edmg_bw_config) || + (edmg_channels && !edmg_bw_config)) + return false; + + if (!(edmg_channels & BIT(primary_channel - 1))) + return false; + + /* 60GHz channels 1..6 */ + for (i = 0; i < 6; i++) { + if (!(edmg_channels & BIT(i))) + continue; + + if (!(edmg_cap->channels & BIT(i))) + return false; + + channels_counter++; + + freq = ieee80211_channel_to_frequency(i + 1, + NL80211_BAND_60GHZ); + chan = ieee80211_get_channel(wiphy, freq); + if (!chan || chan->flags & IEEE80211_CHAN_DISABLED) + return false; + } + + /* IEEE802.11 allows max 4 channels */ + if (channels_counter > 4) + return false; + + /* check bw_config is a subset of what driver supports + * (see IEEE P802.11ay/D4.0 section 9.4.2.251, Table 13) + */ + if ((edmg_bw_config % 4) > (edmg_cap->bw_config % 4)) + return false; + + if (edmg_bw_config > edmg_cap->bw_config) + return false; + + return true; +} + bool cfg80211_chandef_usable(struct wiphy *wiphy, const struct cfg80211_chan_def *chandef, u32 prohibited_flags) { struct ieee80211_sta_ht_cap *ht_cap; struct ieee80211_sta_vht_cap *vht_cap; + struct ieee80211_edmg *edmg_cap; u32 width, control_freq, cap; if (WARN_ON(!cfg80211_chandef_valid(chandef))) @@ -734,6 +884,15 @@ bool cfg80211_chandef_usable(struct wiphy *wiphy, ht_cap = &wiphy->bands[chandef->chan->band]->ht_cap; vht_cap = &wiphy->bands[chandef->chan->band]->vht_cap; + edmg_cap = &wiphy->bands[chandef->chan->band]->edmg_cap; + + if (edmg_cap->channels && + !cfg80211_edmg_usable(wiphy, + chandef->edmg.channels, + chandef->edmg.bw_config, + chandef->chan->hw_value, + edmg_cap)) + return false; control_freq = chandef->chan->center_freq; @@ -894,7 +1053,8 @@ static bool cfg80211_ir_permissive_chan(struct wiphy *wiphy, if (chan == other_chan) return true; - if (chan->band != NL80211_BAND_5GHZ) + if (chan->band != NL80211_BAND_5GHZ && + chan->band != NL80211_BAND_6GHZ) continue; r1 = cfg80211_get_unii(chan->center_freq); diff --git a/net/wireless/ibss.c b/net/wireless/ibss.c index d1743e6abc34..ae8fe66a9bb8 100644 --- a/net/wireless/ibss.c +++ b/net/wireless/ibss.c @@ -104,13 +104,19 @@ int __cfg80211_join_ibss(struct cfg80211_registered_device *rdev, * use the mandatory rate set for 11b or * 11a for maximum compatibility. */ - struct ieee80211_supported_band *sband = - rdev->wiphy.bands[params->chandef.chan->band]; + struct ieee80211_supported_band *sband; + enum nl80211_band band; + u32 flag; int j; - u32 flag = params->chandef.chan->band == NL80211_BAND_5GHZ ? - IEEE80211_RATE_MANDATORY_A : - IEEE80211_RATE_MANDATORY_B; + band = params->chandef.chan->band; + if (band == NL80211_BAND_5GHZ || + band == NL80211_BAND_6GHZ) + flag = IEEE80211_RATE_MANDATORY_A; + else + flag = IEEE80211_RATE_MANDATORY_B; + + sband = rdev->wiphy.bands[band]; for (j = 0; j < sband->n_bitrates; j++) { if (sband->bitrates[j].flags & flag) params->basic_rates |= BIT(j); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 92e06482563c..4565d7385884 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -298,6 +298,13 @@ const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_WIPHY_FREQ] = { .type = NLA_U32 }, [NL80211_ATTR_WIPHY_CHANNEL_TYPE] = { .type = NLA_U32 }, + [NL80211_ATTR_WIPHY_EDMG_CHANNELS] = NLA_POLICY_RANGE(NLA_U8, + NL80211_EDMG_CHANNELS_MIN, + NL80211_EDMG_CHANNELS_MAX), + [NL80211_ATTR_WIPHY_EDMG_BW_CONFIG] = NLA_POLICY_RANGE(NLA_U8, + NL80211_EDMG_BW_CONFIG_MIN, + NL80211_EDMG_BW_CONFIG_MAX), + [NL80211_ATTR_CHANNEL_WIDTH] = { .type = NLA_U32 }, [NL80211_ATTR_CENTER_FREQ1] = { .type = NLA_U32 }, [NL80211_ATTR_CENTER_FREQ2] = { .type = NLA_U32 }, @@ -677,6 +684,7 @@ static const struct nla_policy nl80211_match_band_rssi_policy[NUM_NL80211_BANDS] = { [NL80211_BAND_2GHZ] = { .type = NLA_S32 }, [NL80211_BAND_5GHZ] = { .type = NLA_S32 }, + [NL80211_BAND_6GHZ] = { .type = NLA_S32 }, [NL80211_BAND_60GHZ] = { .type = NLA_S32 }, }; @@ -1573,6 +1581,15 @@ static int nl80211_send_band_rateinfo(struct sk_buff *msg, nla_nest_end(msg, nl_iftype_data); } + /* add EDMG info */ + if (sband->edmg_cap.channels && + (nla_put_u8(msg, NL80211_BAND_ATTR_EDMG_CHANNELS, + sband->edmg_cap.channels) || + nla_put_u8(msg, NL80211_BAND_ATTR_EDMG_BW_CONFIG, + sband->edmg_cap.bw_config))) + + return -ENOBUFS; + /* add bitrates */ nl_rates = nla_nest_start_noflag(msg, NL80211_BAND_ATTR_RATES); if (!nl_rates) @@ -2676,6 +2693,18 @@ int nl80211_parse_chandef(struct cfg80211_registered_device *rdev, nla_get_u32(attrs[NL80211_ATTR_CENTER_FREQ2]); } + if (info->attrs[NL80211_ATTR_WIPHY_EDMG_CHANNELS]) { + chandef->edmg.channels = + nla_get_u8(info->attrs[NL80211_ATTR_WIPHY_EDMG_CHANNELS]); + + if (info->attrs[NL80211_ATTR_WIPHY_EDMG_BW_CONFIG]) + chandef->edmg.bw_config = + nla_get_u8(info->attrs[NL80211_ATTR_WIPHY_EDMG_BW_CONFIG]); + } else { + chandef->edmg.bw_config = 0; + chandef->edmg.channels = 0; + } + if (!cfg80211_chandef_valid(chandef)) { NL_SET_ERR_MSG(extack, "invalid channel definition"); return -EINVAL; @@ -5031,6 +5060,7 @@ static int nl80211_send_station(struct sk_buff *msg, u32 cmd, u32 portid, PUT_SINFO(CONNECTED_TIME, connected_time, u32); PUT_SINFO(INACTIVE_TIME, inactive_time, u32); + PUT_SINFO_U64(ASSOC_AT_BOOTTIME, assoc_at); if (sinfo->filled & (BIT_ULL(NL80211_STA_INFO_RX_BYTES) | BIT_ULL(NL80211_STA_INFO_RX_BYTES64)) && @@ -9892,6 +9922,15 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) return -EINVAL; } + if (info->attrs[NL80211_ATTR_WIPHY_EDMG_CHANNELS]) { + connect.edmg.channels = + nla_get_u8(info->attrs[NL80211_ATTR_WIPHY_EDMG_CHANNELS]); + + if (info->attrs[NL80211_ATTR_WIPHY_EDMG_BW_CONFIG]) + connect.edmg.bw_config = + nla_get_u8(info->attrs[NL80211_ATTR_WIPHY_EDMG_BW_CONFIG]); + } + if (connect.privacy && info->attrs[NL80211_ATTR_KEYS]) { connkeys = nl80211_parse_connkeys(rdev, info, NULL); if (IS_ERR(connkeys)) diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 4831ad745f91..5311d0ae2454 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -2788,7 +2788,7 @@ static void reg_process_pending_hints(void) /* When last_request->processed becomes true this will be rescheduled */ if (lr && !lr->processed) { - reg_process_hint(lr); + pr_debug("Pending regulatory request, waiting for it to be processed...\n"); return; } @@ -3806,8 +3806,9 @@ void wiphy_regulatory_deregister(struct wiphy *wiphy) } /* - * See http://www.fcc.gov/document/5-ghz-unlicensed-spectrum-unii, for - * UNII band definitions + * See FCC notices for UNII band definitions + * 5GHz: https://www.fcc.gov/document/5-ghz-unlicensed-spectrum-unii + * 6GHz: https://www.fcc.gov/document/fcc-proposes-more-spectrum-unlicensed-use-0 */ int cfg80211_get_unii(int freq) { @@ -3831,6 +3832,22 @@ int cfg80211_get_unii(int freq) if (freq > 5725 && freq <= 5825) return 4; + /* UNII-5 */ + if (freq > 5925 && freq <= 6425) + return 5; + + /* UNII-6 */ + if (freq > 6425 && freq <= 6525) + return 6; + + /* UNII-7 */ + if (freq > 6525 && freq <= 6875) + return 7; + + /* UNII-8 */ + if (freq > 6875 && freq <= 7125) + return 8; + return -EINVAL; } diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 4fbb91a511ae..d98ad2b3143b 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2446,10 +2446,11 @@ TRACE_EVENT(rdev_set_mcast_rate, sizeof(int) * NUM_NL80211_BANDS); ), TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", " - "mcast_rates [2.4GHz=0x%x, 5.2GHz=0x%x, 60GHz=0x%x]", + "mcast_rates [2.4GHz=0x%x, 5.2GHz=0x%x, 6GHz=0x%x, 60GHz=0x%x]", WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->mcast_rate[NL80211_BAND_2GHZ], __entry->mcast_rate[NL80211_BAND_5GHZ], + __entry->mcast_rate[NL80211_BAND_6GHZ], __entry->mcast_rate[NL80211_BAND_60GHZ]) ); diff --git a/net/wireless/util.c b/net/wireless/util.c index d0e35b7b9e35..92cb2cbb179b 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -91,6 +91,11 @@ int ieee80211_channel_to_frequency(int chan, enum nl80211_band band) else return 5000 + chan * 5; break; + case NL80211_BAND_6GHZ: + /* see 802.11ax D4.1 27.3.22.2 */ + if (chan <= 253) + return 5940 + chan * 5; + break; case NL80211_BAND_60GHZ: if (chan < 7) return 56160 + chan * 2160; @@ -111,8 +116,11 @@ int ieee80211_frequency_to_channel(int freq) return (freq - 2407) / 5; else if (freq >= 4910 && freq <= 4980) return (freq - 4000) / 5; - else if (freq <= 45000) /* DMG band lower limit */ + else if (freq < 5940) return (freq - 5000) / 5; + else if (freq <= 45000) /* DMG band lower limit */ + /* see 802.11ax D4.1 27.3.22.2 */ + return (freq - 5940) / 5; else if (freq >= 58320 && freq <= 70200) return (freq - 56160) / 2160; else @@ -148,6 +156,7 @@ static void set_mandatory_flags_band(struct ieee80211_supported_band *sband) switch (sband->band) { case NL80211_BAND_5GHZ: + case NL80211_BAND_6GHZ: want = 3; for (i = 0; i < sband->n_bitrates; i++) { if (sband->bitrates[i].bitrate == 60 || @@ -233,25 +242,30 @@ int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev, switch (params->cipher) { case WLAN_CIPHER_SUITE_TKIP: + /* Extended Key ID can only be used with CCMP/GCMP ciphers */ + if ((pairwise && key_idx) || + params->mode != NL80211_KEY_RX_TX) + return -EINVAL; + break; case WLAN_CIPHER_SUITE_CCMP: case WLAN_CIPHER_SUITE_CCMP_256: case WLAN_CIPHER_SUITE_GCMP: case WLAN_CIPHER_SUITE_GCMP_256: - /* IEEE802.11-2016 allows only 0 and - when using Extended Key - * ID - 1 as index for pairwise keys. + /* IEEE802.11-2016 allows only 0 and - when supporting + * Extended Key ID - 1 as index for pairwise keys. * @NL80211_KEY_NO_TX is only allowed for pairwise keys when * the driver supports Extended Key ID. * @NL80211_KEY_SET_TX can't be set when installing and * validating a key. */ - if (params->mode == NL80211_KEY_NO_TX) { - if (!wiphy_ext_feature_isset(&rdev->wiphy, - NL80211_EXT_FEATURE_EXT_KEY_ID)) - return -EINVAL; - else if (!pairwise || key_idx < 0 || key_idx > 1) + if ((params->mode == NL80211_KEY_NO_TX && !pairwise) || + params->mode == NL80211_KEY_SET_TX) + return -EINVAL; + if (wiphy_ext_feature_isset(&rdev->wiphy, + NL80211_EXT_FEATURE_EXT_KEY_ID)) { + if (pairwise && (key_idx < 0 || key_idx > 1)) return -EINVAL; - } else if ((pairwise && key_idx) || - params->mode == NL80211_KEY_SET_TX) { + } else if (pairwise && key_idx) { return -EINVAL; } break; @@ -1034,7 +1048,7 @@ static u32 cfg80211_calculate_bitrate_ht(struct rate_info *rate) return (bitrate + 50000) / 100000; } -static u32 cfg80211_calculate_bitrate_60g(struct rate_info *rate) +static u32 cfg80211_calculate_bitrate_dmg(struct rate_info *rate) { static const u32 __mcs2bitrate[] = { /* control PHY */ @@ -1081,6 +1095,40 @@ static u32 cfg80211_calculate_bitrate_60g(struct rate_info *rate) return __mcs2bitrate[rate->mcs]; } +static u32 cfg80211_calculate_bitrate_edmg(struct rate_info *rate) +{ + static const u32 __mcs2bitrate[] = { + /* control PHY */ + [0] = 275, + /* SC PHY */ + [1] = 3850, + [2] = 7700, + [3] = 9625, + [4] = 11550, + [5] = 12512, /* 1251.25 mbps */ + [6] = 13475, + [7] = 15400, + [8] = 19250, + [9] = 23100, + [10] = 25025, + [11] = 26950, + [12] = 30800, + [13] = 38500, + [14] = 46200, + [15] = 50050, + [16] = 53900, + [17] = 57750, + [18] = 69300, + [19] = 75075, + [20] = 80850, + }; + + if (WARN_ON_ONCE(rate->mcs >= ARRAY_SIZE(__mcs2bitrate))) + return 0; + + return __mcs2bitrate[rate->mcs] * rate->n_bonded_ch; +} + static u32 cfg80211_calculate_bitrate_vht(struct rate_info *rate) { static const u32 base[4][10] = { @@ -1253,8 +1301,10 @@ u32 cfg80211_calculate_bitrate(struct rate_info *rate) { if (rate->flags & RATE_INFO_FLAGS_MCS) return cfg80211_calculate_bitrate_ht(rate); - if (rate->flags & RATE_INFO_FLAGS_60G) - return cfg80211_calculate_bitrate_60g(rate); + if (rate->flags & RATE_INFO_FLAGS_DMG) + return cfg80211_calculate_bitrate_dmg(rate); + if (rate->flags & RATE_INFO_FLAGS_EDMG) + return cfg80211_calculate_bitrate_edmg(rate); if (rate->flags & RATE_INFO_FLAGS_VHT_MCS) return cfg80211_calculate_bitrate_vht(rate); if (rate->flags & RATE_INFO_FLAGS_HE_MCS) @@ -1466,6 +1516,9 @@ bool ieee80211_operating_class_to_band(u8 operating_class, case 128 ... 130: *band = NL80211_BAND_5GHZ; return true; + case 131 ... 135: + *band = NL80211_BAND_6GHZ; + return true; case 81: case 82: case 83: diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c index e997b263a0dd..947b8ff0227e 100644 --- a/net/xdp/xdp_umem.c +++ b/net/xdp/xdp_umem.c @@ -415,7 +415,7 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) umem->pages = kcalloc(umem->npgs, sizeof(*umem->pages), GFP_KERNEL); if (!umem->pages) { err = -ENOMEM; - goto out_account; + goto out_pin; } err = xdp_umem_map_pages(umem); @@ -424,6 +424,8 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) kfree(umem->pages); +out_pin: + xdp_umem_unpin_pages(umem); out_account: xdp_umem_unaccount_pages(umem); return err; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 8ca637a72697..ec94f5795ea4 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3269,7 +3269,7 @@ decode_session4(struct sk_buff *skb, struct flowi *fl, bool reverse) struct flowi4 *fl4 = &fl->u.ip4; int oif = 0; - if (skb_dst(skb)) + if (skb_dst(skb) && skb_dst(skb)->dev) oif = skb_dst(skb)->dev->ifindex; memset(fl4, 0, sizeof(struct flowi4)); @@ -3387,7 +3387,7 @@ decode_session6(struct sk_buff *skb, struct flowi *fl, bool reverse) nexthdr = nh[nhoff]; - if (skb_dst(skb)) + if (skb_dst(skb) && skb_dst(skb)->dev) oif = skb_dst(skb)->dev->ifindex; memset(fl6, 0, sizeof(struct flowi6)); |