diff options
Diffstat (limited to 'net')
481 files changed, 14000 insertions, 40871 deletions
diff --git a/net/Kconfig b/net/Kconfig index 7d57ef34b79c..9dba2715919d 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -166,13 +166,6 @@ menuconfig NETFILTER if NETFILTER -config NETFILTER_DEBUG - bool "Network packet filtering debugging" - depends on NETFILTER - help - You can say Y here if you want to get additional messages useful in - debugging the netfilter code. - config NETFILTER_ADVANCED bool "Advanced netfilter configuration" depends on NETFILTER @@ -235,6 +228,7 @@ source "net/openvswitch/Kconfig" source "net/vmw_vsock/Kconfig" source "net/netlink/Kconfig" source "net/mpls/Kconfig" +source "net/nsh/Kconfig" source "net/hsr/Kconfig" source "net/switchdev/Kconfig" source "net/l3mdev/Kconfig" @@ -301,6 +295,18 @@ config BPF_JIT /proc/sys/net/core/bpf_jit_harden (optional) /proc/sys/net/core/bpf_jit_kallsyms (optional) +config BPF_STREAM_PARSER + bool "enable BPF STREAM_PARSER" + depends on BPF_SYSCALL + select STREAM_PARSER + ---help--- + Enabling this allows a stream parser to be used with + BPF_MAP_TYPE_SOCKMAP. + + BPF_MAP_TYPE_SOCKMAP provides a map type to use with network sockets. + It can be used to enforce socket policy, implement socket redirects, + etc. + config NET_FLOW_LIMIT bool depends on RPS @@ -364,7 +370,6 @@ endmenu source "net/ax25/Kconfig" source "net/can/Kconfig" -source "net/irda/Kconfig" source "net/bluetooth/Kconfig" source "net/rxrpc/Kconfig" source "net/kcm/Kconfig" diff --git a/net/Makefile b/net/Makefile index bed80fa398b7..ae2fe2283d2f 100644 --- a/net/Makefile +++ b/net/Makefile @@ -31,7 +31,6 @@ obj-$(CONFIG_NETROM) += netrom/ obj-$(CONFIG_ROSE) += rose/ obj-$(CONFIG_AX25) += ax25/ obj-$(CONFIG_CAN) += can/ -obj-$(CONFIG_IRDA) += irda/ obj-$(CONFIG_BT) += bluetooth/ obj-$(CONFIG_SUNRPC) += sunrpc/ obj-$(CONFIG_AF_RXRPC) += rxrpc/ @@ -76,6 +75,7 @@ obj-$(CONFIG_NET_IFE) += ife/ obj-$(CONFIG_OPENVSWITCH) += openvswitch/ obj-$(CONFIG_VSOCKETS) += vmw_vsock/ obj-$(CONFIG_MPLS) += mpls/ +obj-$(CONFIG_NET_NSH) += nsh/ obj-$(CONFIG_HSR) += hsr/ ifneq ($(CONFIG_NET_SWITCHDEV),) obj-y += switchdev/ diff --git a/net/atm/clip.c b/net/atm/clip.c index f271a7bcf5b2..65f706e4344c 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -617,7 +617,7 @@ static void atmarpd_close(struct atm_vcc *vcc) module_put(THIS_MODULE); } -static struct atmdev_ops atmarpd_dev_ops = { +static const struct atmdev_ops atmarpd_dev_ops = { .close = atmarpd_close }; diff --git a/net/atm/lec.c b/net/atm/lec.c index 093fe8707731..a3d93a1bb133 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -486,7 +486,7 @@ static void lec_atm_close(struct atm_vcc *vcc) module_put(THIS_MODULE); } -static struct atmdev_ops lecdev_ops = { +static const struct atmdev_ops lecdev_ops = { .close = lec_atm_close, .send = lec_atm_send }; diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 680a4b9095a1..5677147209e8 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -779,7 +779,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) netif_rx(new_skb); } -static struct atmdev_ops mpc_ops = { /* only send is required */ +static const struct atmdev_ops mpc_ops = { /* only send is required */ .close = mpoad_close, .send = msg_from_mpoad }; diff --git a/net/atm/signaling.c b/net/atm/signaling.c index 983c3a21a133..0a20f6e953ac 100644 --- a/net/atm/signaling.c +++ b/net/atm/signaling.c @@ -217,7 +217,7 @@ static void sigd_close(struct atm_vcc *vcc) read_unlock(&vcc_sklist_lock); } -static struct atmdev_ops sigd_dev_ops = { +static const struct atmdev_ops sigd_dev_ops = { .close = sigd_close, .send = sigd_send }; diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index a3501173e200..83ba5483455a 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -729,11 +729,9 @@ static void batadv_iv_ogm_aggregate(struct batadv_forw_packet *forw_packet_aggr, const unsigned char *packet_buff, int packet_len, bool direct_link) { - unsigned char *skb_buff; unsigned long new_direct_link_flag; - skb_buff = skb_put_data(forw_packet_aggr->skb, packet_buff, - packet_len); + skb_put_data(forw_packet_aggr->skb, packet_buff, packet_len); forw_packet_aggr->packet_len += packet_len; forw_packet_aggr->num_packets++; @@ -1281,7 +1279,7 @@ static bool batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node, batadv_ogm_packet->tq = combined_tq; batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "bidirectional: orig = %-15pM neigh = %-15pM => own_bcast = %2i, real recv = %2i, local tq: %3i, asym_penalty: %3i, iface_penalty: %3i, total tq: %3i, if_incoming = %s, if_outgoing = %s\n", + "bidirectional: orig = %pM neigh = %pM => own_bcast = %2i, real recv = %2i, local tq: %3i, asym_penalty: %3i, iface_penalty: %3i, total tq: %3i, if_incoming = %s, if_outgoing = %s\n", orig_node->orig, orig_neigh_node->orig, total_count, neigh_rq_count, tq_own, tq_asym_penalty, tq_iface_penalty, batadv_ogm_packet->tq, if_incoming->net_dev->name, diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 1e3dc374bfde..8be61734fc43 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -137,7 +137,7 @@ static void batadv_v_ogm_send(struct work_struct *work) struct batadv_priv *bat_priv; struct batadv_ogm2_packet *ogm_packet; struct sk_buff *skb, *skb_tmp; - unsigned char *ogm_buff, *pkt_buff; + unsigned char *ogm_buff; int ogm_buff_len; u16 tvlv_len = 0; int ret; @@ -166,7 +166,7 @@ static void batadv_v_ogm_send(struct work_struct *work) goto reschedule; skb_reserve(skb, ETH_HLEN); - pkt_buff = skb_put_data(skb, ogm_buff, ogm_buff_len); + skb_put_data(skb, ogm_buff, ogm_buff_len); ogm_packet = (struct batadv_ogm2_packet *)skb->data; ogm_packet->seqno = htonl(atomic_read(&bat_priv->bat_v.ogm_seqno)); @@ -200,7 +200,7 @@ static void batadv_v_ogm_send(struct work_struct *work) type = "unknown"; } - batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "OGM2 from ourselve on %s surpressed: %s\n", + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "OGM2 from ourselves on %s suppressed: %s\n", hard_iface->net_dev->name, type); batadv_hardif_put(hard_iface); @@ -683,18 +683,18 @@ static void batadv_v_ogm_process(const struct sk_buff *skb, int ogm_offset, ogm_throughput = ntohl(ogm_packet->throughput); batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Received OGM2 packet via NB: %pM, IF: %s [%pM] (from OG: %pM, seqno %u, troughput %u, TTL %u, V %u, tvlv_len %u)\n", + "Received OGM2 packet via NB: %pM, IF: %s [%pM] (from OG: %pM, seqno %u, throughput %u, TTL %u, V %u, tvlv_len %u)\n", ethhdr->h_source, if_incoming->net_dev->name, if_incoming->net_dev->dev_addr, ogm_packet->orig, ntohl(ogm_packet->seqno), ogm_throughput, ogm_packet->ttl, ogm_packet->version, ntohs(ogm_packet->tvlv_len)); - /* If the troughput metric is 0, immediately drop the packet. No need to - * create orig_node / neigh_node for an unusable route. + /* If the throughput metric is 0, immediately drop the packet. No need + * to create orig_node / neigh_node for an unusable route. */ if (ogm_throughput == 0) { batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Drop packet: originator packet with troughput metric of 0\n"); + "Drop packet: originator packet with throughput metric of 0\n"); return; } @@ -762,7 +762,7 @@ static void batadv_v_ogm_process(const struct sk_buff *skb, int ogm_offset, type = "unknown"; } - batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "OGM2 packet from %pM on %s surpressed: %s\n", + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "OGM2 packet from %pM on %s suppressed: %s\n", ogm_packet->orig, hard_iface->net_dev->name, type); diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index 6930d6b50f99..b6cfa78e9381 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -834,7 +834,7 @@ int batadv_dat_cache_seq_print_text(struct seq_file *seq, void *offset) last_seen_msecs = last_seen_msecs % 60000; last_seen_secs = last_seen_msecs / 1000; - seq_printf(seq, " * %15pI4 %14pM %4i %6i:%02i\n", + seq_printf(seq, " * %15pI4 %pM %4i %6i:%02i\n", &dat_entry->ip, dat_entry->mac_addr, batadv_print_vid(dat_entry->vid), last_seen_mins, last_seen_secs); diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 2be8f1f46529..05cc7637c064 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -24,7 +24,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2017.2" +#define BATADV_SOURCE_VERSION "2017.3" #endif /* B.A.T.M.A.N. parameters */ diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index d239a9d72ac3..054a65e6eb68 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -911,7 +911,7 @@ static void batadv_send_outstanding_bcast_packet(struct work_struct *work) type = "unknown"; } - batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "BCAST packet from orig %pM on %s surpressed: %s\n", + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "BCAST packet from orig %pM on %s suppressed: %s\n", bcast_packet->orig, hard_iface->net_dev->name, type); diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index e1133bc634b5..8a3ce79b1307 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -1549,9 +1549,41 @@ batadv_tt_global_entry_has_orig(const struct batadv_tt_global_entry *entry, return found; } +/** + * batadv_tt_global_sync_flags - update TT sync flags + * @tt_global: the TT global entry to update sync flags in + * + * Updates the sync flag bits in the tt_global flag attribute with a logical + * OR of all sync flags from any of its TT orig entries. + */ +static void +batadv_tt_global_sync_flags(struct batadv_tt_global_entry *tt_global) +{ + struct batadv_tt_orig_list_entry *orig_entry; + const struct hlist_head *head; + u16 flags = BATADV_NO_FLAGS; + + rcu_read_lock(); + head = &tt_global->orig_list; + hlist_for_each_entry_rcu(orig_entry, head, list) + flags |= orig_entry->flags; + rcu_read_unlock(); + + flags |= tt_global->common.flags & (~BATADV_TT_SYNC_MASK); + tt_global->common.flags = flags; +} + +/** + * batadv_tt_global_orig_entry_add - add or update a TT orig entry + * @tt_global: the TT global entry to add an orig entry in + * @orig_node: the originator to add an orig entry for + * @ttvn: translation table version number of this changeset + * @flags: TT sync flags + */ static void batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global, - struct batadv_orig_node *orig_node, int ttvn) + struct batadv_orig_node *orig_node, int ttvn, + u8 flags) { struct batadv_tt_orig_list_entry *orig_entry; @@ -1561,7 +1593,8 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global, * was added during a "temporary client detection" */ orig_entry->ttvn = ttvn; - goto out; + orig_entry->flags = flags; + goto sync_flags; } orig_entry = kmem_cache_zalloc(batadv_tt_orig_cache, GFP_ATOMIC); @@ -1573,6 +1606,7 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global, batadv_tt_global_size_inc(orig_node, tt_global->common.vid); orig_entry->orig_node = orig_node; orig_entry->ttvn = ttvn; + orig_entry->flags = flags; kref_init(&orig_entry->refcount); spin_lock_bh(&tt_global->list_lock); @@ -1582,6 +1616,8 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global, spin_unlock_bh(&tt_global->list_lock); atomic_inc(&tt_global->orig_list_count); +sync_flags: + batadv_tt_global_sync_flags(tt_global); out: if (orig_entry) batadv_tt_orig_list_entry_put(orig_entry); @@ -1703,10 +1739,10 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv, } /* the change can carry possible "attribute" flags like the - * TT_CLIENT_WIFI, therefore they have to be copied in the + * TT_CLIENT_TEMP, therefore they have to be copied in the * client entry */ - common->flags |= flags; + common->flags |= flags & (~BATADV_TT_SYNC_MASK); /* If there is the BATADV_TT_CLIENT_ROAM flag set, there is only * one originator left in the list and we previously received a @@ -1723,7 +1759,8 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv, } add_orig_entry: /* add the new orig_entry (if needed) or update it */ - batadv_tt_global_orig_entry_add(tt_global_entry, orig_node, ttvn); + batadv_tt_global_orig_entry_add(tt_global_entry, orig_node, ttvn, + flags & BATADV_TT_SYNC_MASK); batadv_dbg(BATADV_DBG_TT, bat_priv, "Creating new global tt entry: %pM (vid: %d, via %pM)\n", @@ -1946,6 +1983,7 @@ batadv_tt_global_dump_subentry(struct sk_buff *msg, u32 portid, u32 seq, struct batadv_tt_orig_list_entry *orig, bool best) { + u16 flags = (common->flags & (~BATADV_TT_SYNC_MASK)) | orig->flags; void *hdr; struct batadv_orig_node_vlan *vlan; u8 last_ttvn; @@ -1975,7 +2013,7 @@ batadv_tt_global_dump_subentry(struct sk_buff *msg, u32 portid, u32 seq, nla_put_u8(msg, BATADV_ATTR_TT_LAST_TTVN, last_ttvn) || nla_put_u32(msg, BATADV_ATTR_TT_CRC32, crc) || nla_put_u16(msg, BATADV_ATTR_TT_VID, common->vid) || - nla_put_u32(msg, BATADV_ATTR_TT_FLAGS, common->flags)) + nla_put_u32(msg, BATADV_ATTR_TT_FLAGS, flags)) goto nla_put_failure; if (best && nla_put_flag(msg, BATADV_ATTR_FLAG_BEST)) @@ -2589,6 +2627,7 @@ static u32 batadv_tt_global_crc(struct batadv_priv *bat_priv, unsigned short vid) { struct batadv_hashtable *hash = bat_priv->tt.global_hash; + struct batadv_tt_orig_list_entry *tt_orig; struct batadv_tt_common_entry *tt_common; struct batadv_tt_global_entry *tt_global; struct hlist_head *head; @@ -2627,8 +2666,9 @@ static u32 batadv_tt_global_crc(struct batadv_priv *bat_priv, /* find out if this global entry is announced by this * originator */ - if (!batadv_tt_global_entry_has_orig(tt_global, - orig_node)) + tt_orig = batadv_tt_global_orig_entry_find(tt_global, + orig_node); + if (!tt_orig) continue; /* use network order to read the VID: this ensures that @@ -2640,10 +2680,12 @@ static u32 batadv_tt_global_crc(struct batadv_priv *bat_priv, /* compute the CRC on flags that have to be kept in sync * among nodes */ - flags = tt_common->flags & BATADV_TT_SYNC_MASK; + flags = tt_orig->flags; crc_tmp = crc32c(crc_tmp, &flags, sizeof(flags)); crc ^= crc32c(crc_tmp, tt_common->addr, ETH_ALEN); + + batadv_tt_orig_list_entry_put(tt_orig); } rcu_read_unlock(); } diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index ea43a6449247..a62795868794 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1260,6 +1260,7 @@ struct batadv_tt_global_entry { * struct batadv_tt_orig_list_entry - orig node announcing a non-mesh client * @orig_node: pointer to orig node announcing this non-mesh client * @ttvn: translation table version number which added the non-mesh client + * @flags: per orig entry TT sync flags * @list: list node for batadv_tt_global_entry::orig_list * @refcount: number of contexts the object is used * @rcu: struct used for freeing in an RCU-safe manner @@ -1267,6 +1268,7 @@ struct batadv_tt_global_entry { struct batadv_tt_orig_list_entry { struct batadv_orig_node *orig_node; u8 ttvn; + u8 flags; struct hlist_node list; struct kref refcount; struct rcu_head rcu; diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index ab3b654b05cc..4e2576fc0c59 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -273,9 +273,6 @@ static int iphc_decompress(struct sk_buff *skb, struct net_device *netdev, struct lowpan_peer *peer) { const u8 *saddr; - struct lowpan_btle_dev *dev; - - dev = lowpan_btle_dev(netdev); saddr = peer->lladdr; @@ -618,12 +615,8 @@ static void ifup(struct net_device *netdev) static void ifdown(struct net_device *netdev) { - int err; - rtnl_lock(); - err = dev_close(netdev); - if (err < 0) - BT_INFO("iface %s cannot be closed (%d)", netdev->name, err); + dev_close(netdev); rtnl_unlock(); } diff --git a/net/bluetooth/Kconfig b/net/bluetooth/Kconfig index 68f951b3e85a..c18115d22f00 100644 --- a/net/bluetooth/Kconfig +++ b/net/bluetooth/Kconfig @@ -45,6 +45,11 @@ config BT_BREDR bool "Bluetooth Classic (BR/EDR) features" depends on BT default y + help + Bluetooth Classic includes support for Basic Rate (BR) + available with Bluetooth version 1.0b or later and support + for Enhanced Data Rate (EDR) available with Bluetooth + version 2.0 or later. source "net/bluetooth/rfcomm/Kconfig" @@ -58,11 +63,18 @@ config BT_HS bool "Bluetooth High Speed (HS) features" depends on BT_BREDR default y + help + Bluetooth High Speed includes support for off-loading + Bluetooth connections via 802.11 (wifi) physical layer + available with Bluetooth version 3.0 or later. config BT_LE bool "Bluetooth Low Energy (LE) features" depends on BT default y + help + Bluetooth Low Energy includes support low-energy physical + layer available with Bluetooth version 4.0 or later. config BT_6LOWPAN tristate "Bluetooth 6LoWPAN support" @@ -114,4 +126,14 @@ config BT_DEBUGFS Provide extensive information about internal Bluetooth states in debugfs. +config BT_LEGACY_IOCTL + bool "Enable legacy ioctl interfaces" + depends on BT && BT_BREDR + default y + help + Enable support for legacy ioctl interfaces. This is only needed + for old and deprecated applications using direct ioctl calls for + controller management. Since Linux 3.4 all configuration and + setup is done via mgmt interface and this is no longer needed. + source "drivers/bluetooth/Kconfig" diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 65d734c165bd..0bad296fe0af 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -878,6 +878,7 @@ static int hci_sock_release(struct socket *sock) return 0; } +#ifdef CONFIG_BT_LEGACY_IOCTL static int hci_sock_blacklist_add(struct hci_dev *hdev, void __user *arg) { bdaddr_t bdaddr; @@ -1049,6 +1050,7 @@ done: release_sock(sk); return err; } +#endif static int hci_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) @@ -1969,7 +1971,11 @@ static const struct proto_ops hci_sock_ops = { .getname = hci_sock_getname, .sendmsg = hci_sock_sendmsg, .recvmsg = hci_sock_recvmsg, +#ifdef CONFIG_BT_LEGACY_IOCTL .ioctl = hci_sock_ioctl, +#else + .ioctl = sock_no_ioctl, +#endif .poll = datagram_poll, .listen = sock_no_listen, .shutdown = sock_no_shutdown, diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index ca7a35ebaefb..aa300f3a0d51 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -13,7 +13,7 @@ static void bt_link_release(struct device *dev) kfree(conn); } -static struct device_type bt_link = { +static const struct device_type bt_link = { .name = "link", .release = bt_link_release, }; @@ -86,7 +86,7 @@ static void bt_host_release(struct device *dev) module_put(THIS_MODULE); } -static struct device_type bt_host = { +static const struct device_type bt_host = { .name = "host", .release = bt_host_release, }; diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 002743ea509c..8112893037bd 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -734,7 +734,7 @@ static void hidp_stop(struct hid_device *hid) hid->claimed = 0; } -static struct hid_ll_driver hidp_hid_driver = { +struct hid_ll_driver hidp_hid_driver = { .parse = hidp_parse, .start = hidp_start, .stop = hidp_stop, @@ -743,6 +743,7 @@ static struct hid_ll_driver hidp_hid_driver = { .raw_request = hidp_raw_request, .output_report = hidp_output_report, }; +EXPORT_SYMBOL_GPL(hidp_hid_driver); /* This function sets up the hid device. It does not add it to the HID system. That is done in hidp_add_connection(). */ diff --git a/net/bluetooth/lib.c b/net/bluetooth/lib.c index aa4cf64e32a6..63e65d9b4b24 100644 --- a/net/bluetooth/lib.c +++ b/net/bluetooth/lib.c @@ -30,10 +30,10 @@ #include <net/bluetooth/bluetooth.h> -void baswap(bdaddr_t *dst, bdaddr_t *src) +void baswap(bdaddr_t *dst, const bdaddr_t *src) { - unsigned char *d = (unsigned char *) dst; - unsigned char *s = (unsigned char *) src; + const unsigned char *s = (const unsigned char *)src; + unsigned char *d = (unsigned char *)dst; unsigned int i; for (i = 0; i < 6; i++) diff --git a/net/bluetooth/selftest.c b/net/bluetooth/selftest.c index ee92c925ecc5..34a1227f4391 100644 --- a/net/bluetooth/selftest.c +++ b/net/bluetooth/selftest.c @@ -164,7 +164,7 @@ static int __init test_ecdh_sample(const u8 priv_a[32], const u8 priv_b[32], ret = -EINVAL; out: - kfree(dhkey_a); + kfree(tmp); return ret; } diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index f0f3447e8aa4..f6b6a92f1c48 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -34,11 +34,11 @@ static struct lock_class_key bridge_netdev_addr_lock_key; netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); - const unsigned char *dest = skb->data; struct net_bridge_fdb_entry *dst; struct net_bridge_mdb_entry *mdst; struct pcpu_sw_netstats *brstats = this_cpu_ptr(br->stats); const struct nf_br_ops *nf_ops; + const unsigned char *dest; u16 vid = 0; rcu_read_lock(); @@ -53,6 +53,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) brstats->tx_bytes += skb->len; u64_stats_update_end(&brstats->syncp); + br_switchdev_frame_unmark(skb); BR_INPUT_SKB_CB(skb)->brdev = dev; skb_reset_mac_header(skb); @@ -61,6 +62,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid)) goto out; + dest = eth_hdr(skb)->h_dest; if (is_broadcast_ether_addr(dest)) { br_flood(br, skb, BR_PKT_BROADCAST, false, true); } else if (is_multicast_ether_addr(dest)) { diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index a5e4a736a984..4ea5c8bbe286 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -25,6 +25,7 @@ #include <asm/unaligned.h> #include <linux/if_vlan.h> #include <net/switchdev.h> +#include <trace/events/bridge.h> #include "br_private.h" static struct kmem_cache *br_fdb_cache __read_mostly; @@ -169,29 +170,13 @@ static void fdb_del_hw_addr(struct net_bridge *br, const unsigned char *addr) } } -static void fdb_del_external_learn(struct net_bridge_fdb_entry *f) -{ - struct switchdev_obj_port_fdb fdb = { - .obj = { - .orig_dev = f->dst->dev, - .id = SWITCHDEV_OBJ_ID_PORT_FDB, - .flags = SWITCHDEV_F_DEFER, - }, - .vid = f->vlan_id, - }; - - ether_addr_copy(fdb.addr, f->addr.addr); - switchdev_port_obj_del(f->dst->dev, &fdb.obj); -} - static void fdb_delete(struct net_bridge *br, struct net_bridge_fdb_entry *f) { + trace_fdb_delete(br, f); + if (f->is_static) fdb_del_hw_addr(br, f->addr.addr); - if (f->added_by_external_learn) - fdb_del_external_learn(f); - hlist_del_init_rcu(&f->hlist); fdb_notify(br, f, RTM_DELNEIGH); call_rcu(&f->rcu, fdb_rcu_free); @@ -598,8 +583,10 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, fdb->updated = now; if (unlikely(added_by_user)) fdb->added_by_user = 1; - if (unlikely(fdb_modified)) + if (unlikely(fdb_modified)) { + trace_br_fdb_update(br, source, addr, vid, added_by_user); fdb_notify(br, fdb, RTM_NEWNEIGH); + } } } else { spin_lock(&br->hash_lock); @@ -608,6 +595,7 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, if (fdb) { if (unlikely(added_by_user)) fdb->added_by_user = 1; + trace_br_fdb_update(br, source, addr, vid, added_by_user); fdb_notify(br, fdb, RTM_NEWNEIGH); } } @@ -888,6 +876,8 @@ int br_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], struct net_bridge *br = NULL; int err = 0; + trace_br_fdb_add(ndm, dev, addr, vid, nlh_flags); + if (!(ndm->ndm_state & (NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE))) { pr_info("bridge: RTM_NEWNEIGH with invalid state %#x\n", ndm->ndm_state); return -EINVAL; @@ -1084,6 +1074,8 @@ int br_fdb_external_learn_add(struct net_bridge *br, struct net_bridge_port *p, bool modified = false; int err = 0; + trace_br_fdb_external_learn_add(br, p, addr, vid); + spin_lock_bh(&br->hash_lock); head = &br->hash[br_mac_hash(addr, vid)]; diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 013f2290bfa5..7637f58c1226 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -131,11 +131,11 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { struct net_bridge_port *p = br_port_get_rcu(skb->dev); - const unsigned char *dest = eth_hdr(skb)->h_dest; enum br_pkt_type pkt_type = BR_PKT_UNICAST; struct net_bridge_fdb_entry *dst = NULL; struct net_bridge_mdb_entry *mdst; bool local_rcv, mcast_hit = false; + const unsigned char *dest; struct net_bridge *br; u16 vid = 0; @@ -153,6 +153,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb br_fdb_update(br, p, eth_hdr(skb)->h_source, vid, false); local_rcv = !!(br->dev->flags & IFF_PROMISC); + dest = eth_hdr(skb)->h_dest; if (is_multicast_ether_addr(dest)) { /* by definition the broadcast is also a multicast address */ if (is_broadcast_ether_addr(dest)) { diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index a0b11e7d67d9..ca01def49af0 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -713,9 +713,9 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh, void br_mdb_init(void) { - rtnl_register(PF_BRIDGE, RTM_GETMDB, NULL, br_mdb_dump, NULL); - rtnl_register(PF_BRIDGE, RTM_NEWMDB, br_mdb_add, NULL, NULL); - rtnl_register(PF_BRIDGE, RTM_DELMDB, br_mdb_del, NULL, NULL); + rtnl_register(PF_BRIDGE, RTM_GETMDB, NULL, br_mdb_dump, 0); + rtnl_register(PF_BRIDGE, RTM_NEWMDB, br_mdb_add, NULL, 0); + rtnl_register(PF_BRIDGE, RTM_DELMDB, br_mdb_del, NULL, 0); } void br_mdb_uninit(void) diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index 2261e5194c82..c2eea1b8737a 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -887,7 +887,7 @@ EXPORT_SYMBOL_GPL(br_netfilter_enable); /* For br_nf_post_routing, we need (prio = NF_BR_PRI_LAST), because * br_dev_queue_push_xmit is called afterwards */ -static struct nf_hook_ops br_nf_ops[] __read_mostly = { +static const struct nf_hook_ops br_nf_ops[] = { { .hook = br_nf_pre_routing, .pf = NFPROTO_BRIDGE, @@ -985,22 +985,25 @@ int br_nf_hook_thresh(unsigned int hook, struct net *net, int (*okfn)(struct net *, struct sock *, struct sk_buff *)) { - struct nf_hook_entry *elem; + const struct nf_hook_entries *e; struct nf_hook_state state; + struct nf_hook_ops **ops; + unsigned int i; int ret; - for (elem = rcu_dereference(net->nf.hooks[NFPROTO_BRIDGE][hook]); - elem && nf_hook_entry_priority(elem) <= NF_BR_PRI_BRNF; - elem = rcu_dereference(elem->next)) - ; - - if (!elem) + e = rcu_dereference(net->nf.hooks[NFPROTO_BRIDGE][hook]); + if (!e) return okfn(net, sk, skb); + ops = nf_hook_entries_get_hook_ops(e); + for (i = 0; i < e->num_hook_entries && + ops[i]->priority <= NF_BR_PRI_BRNF; i++) + ; + nf_hook_state_init(&state, hook, NFPROTO_BRIDGE, indev, outdev, sk, net, okfn); - ret = nf_hook_slow(skb, &state, elem); + ret = nf_hook_slow(skb, &state, e, i); if (ret == 1) ret = okfn(net, sk, skb); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index fd9ee73e0a6d..e870cfc85b14 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -1091,6 +1091,11 @@ int br_switchdev_set_port_flag(struct net_bridge_port *p, unsigned long mask); void br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type); + +static inline void br_switchdev_frame_unmark(struct sk_buff *skb) +{ + skb->offload_fwd_mark = 0; +} #else static inline int nbp_switchdev_mark_set(struct net_bridge_port *p) { @@ -1119,6 +1124,10 @@ static inline void br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type) { } + +static inline void br_switchdev_frame_unmark(struct sk_buff *skb) +{ +} #endif /* CONFIG_NET_SWITCHDEV */ #endif diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index 181a44d0f1da..f6b1c7de059d 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -115,7 +115,7 @@ br_switchdev_fdb_call_notifiers(bool adding, const unsigned char *mac, void br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type) { - if (!fdb->added_by_user) + if (!fdb->added_by_user || !fdb->dst) return; switch (type) { diff --git a/net/bridge/netfilter/ebt_ip.c b/net/bridge/netfilter/ebt_ip.c index d06968bdf5ec..2b46c50abce0 100644 --- a/net/bridge/netfilter/ebt_ip.c +++ b/net/bridge/netfilter/ebt_ip.c @@ -64,14 +64,14 @@ ebt_ip_mt(const struct sk_buff *skb, struct xt_action_param *par) if (NF_INVF(info, EBT_IP_DPORT, dst < info->dport[0] || dst > info->dport[1])) - return false; + return false; } if (info->bitmask & EBT_IP_SPORT) { u32 src = ntohs(pptr->src); if (NF_INVF(info, EBT_IP_SPORT, src < info->sport[0] || src > info->sport[1])) - return false; + return false; } } return true; diff --git a/net/bridge/netfilter/ebt_ip6.c b/net/bridge/netfilter/ebt_ip6.c index 4617491be41e..2a5a52a53ec4 100644 --- a/net/bridge/netfilter/ebt_ip6.c +++ b/net/bridge/netfilter/ebt_ip6.c @@ -89,7 +89,7 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) if (NF_INVF(info, EBT_IP6_SPORT, src < info->sport[0] || src > info->sport[1])) - return false; + return false; } if ((info->bitmask & EBT_IP6_ICMP6) && NF_INVF(info, EBT_IP6_ICMP6, diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index f22ef7c21913..45a00dbdbcad 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -70,7 +70,7 @@ ebt_out_hook(void *priv, struct sk_buff *skb, return ebt_do_table(skb, state, state->net->xt.frame_filter); } -static struct nf_hook_ops ebt_ops_filter[] __read_mostly = { +static const struct nf_hook_ops ebt_ops_filter[] = { { .hook = ebt_in_hook, .pf = NFPROTO_BRIDGE, diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 2f7a4f314406..57cd5bb154e7 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -48,7 +48,7 @@ static int check(const struct ebt_table_info *info, unsigned int valid_hooks) return 0; } -static struct ebt_table frame_nat = { +static const struct ebt_table frame_nat = { .name = "nat", .table = &initial_table, .valid_hooks = NAT_VALID_HOOKS, @@ -70,7 +70,7 @@ ebt_nat_out(void *priv, struct sk_buff *skb, return ebt_do_table(skb, state, state->net->xt.frame_nat); } -static struct nf_hook_ops ebt_ops_nat[] __read_mostly = { +static const struct nf_hook_ops ebt_ops_nat[] = { { .hook = ebt_nat_out, .pf = NFPROTO_BRIDGE, diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 9c6e619f452b..83951f978445 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -252,13 +252,11 @@ unsigned int ebt_do_table(struct sk_buff *skb, } if (verdict == EBT_RETURN) { letsreturn: -#ifdef CONFIG_NETFILTER_DEBUG - if (sp == 0) { - BUGPRINT("RETURN on base chain"); + if (WARN(sp == 0, "RETURN on base chain")) { /* act like this is EBT_CONTINUE */ goto letscontinue; } -#endif + sp--; /* put all the local variables right */ i = cs[sp].n; @@ -271,26 +269,24 @@ letsreturn: } if (verdict == EBT_CONTINUE) goto letscontinue; -#ifdef CONFIG_NETFILTER_DEBUG - if (verdict < 0) { - BUGPRINT("bogus standard verdict\n"); + + if (WARN(verdict < 0, "bogus standard verdict\n")) { read_unlock_bh(&table->lock); return NF_DROP; } -#endif + /* jump to a udc */ cs[sp].n = i + 1; cs[sp].chaininfo = chaininfo; cs[sp].e = ebt_next_entry(point); i = 0; chaininfo = (struct ebt_entries *) (base + verdict); -#ifdef CONFIG_NETFILTER_DEBUG - if (chaininfo->distinguisher) { - BUGPRINT("jump to non-chain\n"); + + if (WARN(chaininfo->distinguisher, "jump to non-chain\n")) { read_unlock_bh(&table->lock); return NF_DROP; } -#endif + nentries = chaininfo->nentries; point = (struct ebt_entry *)chaininfo->data; counter_base = cb_base + chaininfo->counter_offset; @@ -1069,15 +1065,10 @@ static int do_replace_finish(struct net *net, struct ebt_replace *repl, #ifdef CONFIG_AUDIT if (audit_enabled) { - struct audit_buffer *ab; - - ab = audit_log_start(current->audit_context, GFP_KERNEL, - AUDIT_NETFILTER_CFG); - if (ab) { - audit_log_format(ab, "table=%s family=%u entries=%u", - repl->name, AF_BRIDGE, repl->nentries); - audit_log_end(ab); - } + audit_log(current->audit_context, GFP_KERNEL, + AUDIT_NETFILTER_CFG, + "table=%s family=%u entries=%u", + repl->name, AF_BRIDGE, repl->nentries); } #endif return ret; diff --git a/net/can/gw.c b/net/can/gw.c index 29748d844c3f..73a02af4b5d7 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -1031,15 +1031,15 @@ static __init int cgw_module_init(void) notifier.notifier_call = cgw_notifier; register_netdevice_notifier(¬ifier); - if (__rtnl_register(PF_CAN, RTM_GETROUTE, NULL, cgw_dump_jobs, NULL)) { + if (__rtnl_register(PF_CAN, RTM_GETROUTE, NULL, cgw_dump_jobs, 0)) { unregister_netdevice_notifier(¬ifier); kmem_cache_destroy(cgw_cache); return -ENOBUFS; } /* Only the first call to __rtnl_register can fail */ - __rtnl_register(PF_CAN, RTM_NEWROUTE, cgw_create_job, NULL, NULL); - __rtnl_register(PF_CAN, RTM_DELROUTE, cgw_remove_job, NULL, NULL); + __rtnl_register(PF_CAN, RTM_NEWROUTE, cgw_create_job, NULL, 0); + __rtnl_register(PF_CAN, RTM_DELROUTE, cgw_remove_job, NULL, 0); return 0; } diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c index 746b145bfd11..417df675c71b 100644 --- a/net/ceph/crush/mapper.c +++ b/net/ceph/crush/mapper.c @@ -306,7 +306,7 @@ static __u32 *get_choose_arg_weights(const struct crush_bucket_straw2 *bucket, const struct crush_choose_arg *arg, int position) { - if (!arg || !arg->weight_set || arg->weight_set_size == 0) + if (!arg || !arg->weight_set) return bucket->item_weights; if (position >= arg->weight_set_size) diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index 0c31035bbfee..a67298c7e0cd 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -1287,10 +1287,10 @@ static void prepare_write_message(struct ceph_connection *con) if (m->needs_out_seq) { m->hdr.seq = cpu_to_le64(++con->out_seq); m->needs_out_seq = false; - } - if (con->ops->reencode_message) - con->ops->reencode_message(m); + if (con->ops->reencode_message) + con->ops->reencode_message(m); + } dout("prepare_write_message %p seq %lld type %d len %d+%d+%zd\n", m, con->out_seq, le16_to_cpu(m->hdr.type), @@ -3203,8 +3203,10 @@ static struct ceph_msg_data *ceph_msg_data_create(enum ceph_msg_data_type type) return NULL; data = kmem_cache_zalloc(ceph_msg_data_cache, GFP_NOFS); - if (data) - data->type = type; + if (!data) + return NULL; + + data->type = type; INIT_LIST_HEAD(&data->links); return data; diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 86a9737d8e3f..dcfbdd74dfd1 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -1337,6 +1337,8 @@ static enum calc_target_result calc_target(struct ceph_osd_client *osdc, bool legacy_change; bool split = false; bool sort_bitwise = ceph_osdmap_flag(osdc, CEPH_OSDMAP_SORTBITWISE); + bool recovery_deletes = ceph_osdmap_flag(osdc, + CEPH_OSDMAP_RECOVERY_DELETES); enum calc_target_result ct_res; int ret; @@ -1399,6 +1401,8 @@ static enum calc_target_result calc_target(struct ceph_osd_client *osdc, pi->pg_num, t->sort_bitwise, sort_bitwise, + t->recovery_deletes, + recovery_deletes, &last_pgid)) force_resend = true; @@ -1421,6 +1425,7 @@ static enum calc_target_result calc_target(struct ceph_osd_client *osdc, t->pg_num = pi->pg_num; t->pg_num_mask = pi->pg_num_mask; t->sort_bitwise = sort_bitwise; + t->recovery_deletes = recovery_deletes; t->osd = acting.primary; } @@ -1918,10 +1923,12 @@ static void encode_request_partial(struct ceph_osd_request *req, } ceph_encode_32(&p, req->r_attempts); /* retry_attempt */ - BUG_ON(p != end - 8); /* space for features */ + BUG_ON(p > end - 8); /* space for features */ msg->hdr.version = cpu_to_le16(8); /* MOSDOp v8 */ /* front_len is finalized in encode_request_finish() */ + msg->front.iov_len = p - msg->front.iov_base; + msg->hdr.front_len = cpu_to_le32(msg->front.iov_len); msg->hdr.data_len = cpu_to_le32(data_len); /* * The header "data_off" is a hint to the receiver allowing it @@ -1937,11 +1944,12 @@ static void encode_request_partial(struct ceph_osd_request *req, static void encode_request_finish(struct ceph_msg *msg) { void *p = msg->front.iov_base; + void *const partial_end = p + msg->front.iov_len; void *const end = p + msg->front_alloc_len; if (CEPH_HAVE_FEATURE(msg->con->peer_features, RESEND_ON_SPLIT)) { /* luminous OSD -- encode features and be done */ - p = end - 8; + p = partial_end; ceph_encode_64(&p, msg->con->peer_features); } else { struct { @@ -1984,7 +1992,7 @@ static void encode_request_finish(struct ceph_msg *msg) oid_len = p - oid; tail = p; - tail_len = (end - p) - 8; + tail_len = partial_end - p; p = msg->front.iov_base; ceph_encode_copy(&p, &head.client_inc, sizeof(head.client_inc)); @@ -5310,7 +5318,10 @@ static int invalidate_authorizer(struct ceph_connection *con) static void osd_reencode_message(struct ceph_msg *msg) { - encode_request_finish(msg); + int type = le16_to_cpu(msg->hdr.type); + + if (type == CEPH_MSG_OSD_OP) + encode_request_finish(msg); } static int osd_sign_message(struct ceph_msg *msg) diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index 864789c5974e..f358d0bfa76b 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -295,6 +295,10 @@ static int decode_choose_args(void **p, void *end, struct crush_map *c) ret = decode_choose_arg(p, end, arg); if (ret) goto fail; + + if (arg->ids_size && + arg->ids_size != c->buckets[bucket_index]->size) + goto e_inval; } insert_choose_arg_map(&c->choose_args, arg_map); @@ -338,7 +342,7 @@ static void crush_finalize(struct crush_map *c) static struct crush_map *crush_decode(void *pbyval, void *end) { struct crush_map *c; - int err = -EINVAL; + int err; int i, j; void **p = &pbyval; void *start = pbyval; @@ -407,7 +411,6 @@ static struct crush_map *crush_decode(void *pbyval, void *end) size = sizeof(struct crush_bucket_straw2); break; default: - err = -EINVAL; goto bad; } BUG_ON(size == 0); @@ -439,31 +442,31 @@ static struct crush_map *crush_decode(void *pbyval, void *end) err = crush_decode_uniform_bucket(p, end, (struct crush_bucket_uniform *)b); if (err < 0) - goto bad; + goto fail; break; case CRUSH_BUCKET_LIST: err = crush_decode_list_bucket(p, end, (struct crush_bucket_list *)b); if (err < 0) - goto bad; + goto fail; break; case CRUSH_BUCKET_TREE: err = crush_decode_tree_bucket(p, end, (struct crush_bucket_tree *)b); if (err < 0) - goto bad; + goto fail; break; case CRUSH_BUCKET_STRAW: err = crush_decode_straw_bucket(p, end, (struct crush_bucket_straw *)b); if (err < 0) - goto bad; + goto fail; break; case CRUSH_BUCKET_STRAW2: err = crush_decode_straw2_bucket(p, end, (struct crush_bucket_straw2 *)b); if (err < 0) - goto bad; + goto fail; break; } } @@ -474,7 +477,6 @@ static struct crush_map *crush_decode(void *pbyval, void *end) u32 yes; struct crush_rule *r; - err = -EINVAL; ceph_decode_32_safe(p, end, yes, bad); if (!yes) { dout("crush_decode NO rule %d off %x %p to %p\n", @@ -489,7 +491,6 @@ static struct crush_map *crush_decode(void *pbyval, void *end) /* len */ ceph_decode_32_safe(p, end, yes, bad); #if BITS_PER_LONG == 32 - err = -EINVAL; if (yes > (ULONG_MAX - sizeof(*r)) / sizeof(struct crush_rule_step)) goto bad; @@ -557,7 +558,7 @@ static struct crush_map *crush_decode(void *pbyval, void *end) if (*p != end) { err = decode_choose_args(p, end, c); if (err) - goto bad; + goto fail; } done: @@ -567,10 +568,14 @@ done: badmem: err = -ENOMEM; -bad: +fail: dout("crush_decode fail %d\n", err); crush_destroy(c); return ERR_PTR(err); + +bad: + err = -EINVAL; + goto fail; } int ceph_pg_compare(const struct ceph_pg *lhs, const struct ceph_pg *rhs) @@ -1399,7 +1404,7 @@ static struct ceph_pg_mapping *__decode_pg_upmap_items(void **p, void *end, return ERR_PTR(-EINVAL); ceph_decode_need(p, end, 2 * len * sizeof(u32), e_inval); - pg = kzalloc(sizeof(*pg) + 2 * len * sizeof(u32), GFP_NOIO); + pg = alloc_pg_mapping(2 * len * sizeof(u32)); if (!pg) return ERR_PTR(-ENOMEM); @@ -1544,7 +1549,7 @@ static int osdmap_decode(void **p, void *end, struct ceph_osdmap *map) if (struct_v >= 3) { /* erasure_code_profiles */ ceph_decode_skip_map_of_map(p, end, string, string, string, - bad); + e_inval); } if (struct_v >= 4) { @@ -1825,9 +1830,9 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, if (struct_v >= 3) { /* new_erasure_code_profiles */ ceph_decode_skip_map_of_map(p, end, string, string, string, - bad); + e_inval); /* old_erasure_code_profiles */ - ceph_decode_skip_set(p, end, string, bad); + ceph_decode_skip_set(p, end, string, e_inval); } if (struct_v >= 4) { @@ -2077,6 +2082,8 @@ bool ceph_is_new_interval(const struct ceph_osds *old_acting, u32 new_pg_num, bool old_sort_bitwise, bool new_sort_bitwise, + bool old_recovery_deletes, + bool new_recovery_deletes, const struct ceph_pg *pgid) { return !osds_equal(old_acting, new_acting) || @@ -2084,7 +2091,8 @@ bool ceph_is_new_interval(const struct ceph_osds *old_acting, old_size != new_size || old_min_size != new_min_size || ceph_pg_is_split(pgid, old_pg_num, new_pg_num) || - old_sort_bitwise != new_sort_bitwise; + old_sort_bitwise != new_sort_bitwise || + old_recovery_deletes != new_recovery_deletes; } static int calc_pg_rank(int osd, const struct ceph_osds *acting) @@ -2300,10 +2308,17 @@ static u32 raw_pg_to_pps(struct ceph_pg_pool_info *pi, } } +/* + * Magic value used for a "default" fallback choose_args, used if the + * crush_choose_arg_map passed to do_crush() does not exist. If this + * also doesn't exist, fall back to canonical weights. + */ +#define CEPH_DEFAULT_CHOOSE_ARGS -1 + static int do_crush(struct ceph_osdmap *map, int ruleno, int x, int *result, int result_max, const __u32 *weight, int weight_max, - u64 choose_args_index) + s64 choose_args_index) { struct crush_choose_arg_map *arg_map; int r; @@ -2312,6 +2327,9 @@ static int do_crush(struct ceph_osdmap *map, int ruleno, int x, arg_map = lookup_choose_arg_map(&map->crush->choose_args, choose_args_index); + if (!arg_map) + arg_map = lookup_choose_arg_map(&map->crush->choose_args, + CEPH_DEFAULT_CHOOSE_ARGS); mutex_lock(&map->crush_workspace_mutex); r = crush_do_rule(map->crush, ruleno, x, result, result_max, @@ -2422,40 +2440,23 @@ static void apply_upmap(struct ceph_osdmap *osdmap, for (i = 0; i < pg->pg_upmap.len; i++) raw->osds[i] = pg->pg_upmap.osds[i]; raw->size = pg->pg_upmap.len; - return; + /* check and apply pg_upmap_items, if any */ } pg = lookup_pg_mapping(&osdmap->pg_upmap_items, pgid); if (pg) { - /* - * Note: this approach does not allow a bidirectional swap, - * e.g., [[1,2],[2,1]] applied to [0,1,2] -> [0,2,1]. - */ - for (i = 0; i < pg->pg_upmap_items.len; i++) { - int from = pg->pg_upmap_items.from_to[i][0]; - int to = pg->pg_upmap_items.from_to[i][1]; - int pos = -1; - bool exists = false; - - /* make sure replacement doesn't already appear */ - for (j = 0; j < raw->size; j++) { - int osd = raw->osds[j]; - - if (osd == to) { - exists = true; + for (i = 0; i < raw->size; i++) { + for (j = 0; j < pg->pg_upmap_items.len; j++) { + int from = pg->pg_upmap_items.from_to[j][0]; + int to = pg->pg_upmap_items.from_to[j][1]; + + if (from == raw->osds[i]) { + if (!(to != CRUSH_ITEM_NONE && + to < osdmap->max_osd && + osdmap->osd_weight[to] == 0)) + raw->osds[i] = to; break; } - /* ignore mapping if target is marked out */ - if (osd == from && pos < 0 && - !(to != CRUSH_ITEM_NONE && - to < osdmap->max_osd && - osdmap->osd_weight[to] == 0)) { - pos = j; - } - } - if (!exists && pos >= 0) { - raw->osds[pos] = to; - return; } } } diff --git a/net/core/Makefile b/net/core/Makefile index 79f9479e9658..56d771a887b6 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -9,9 +9,9 @@ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o obj-y += dev.o ethtool.o dev_addr_lists.o dst.o netevent.o \ neighbour.o rtnetlink.o utils.o link_watch.o filter.o \ - sock_diag.o dev_ioctl.o tso.o sock_reuseport.o + sock_diag.o dev_ioctl.o tso.o sock_reuseport.o \ + fib_notifier.o -obj-$(CONFIG_XFRM) += flow.o obj-y += net-sysfs.o obj-$(CONFIG_PROC_FS) += net-procfs.o obj-$(CONFIG_NET_PKTGEN) += pktgen.o diff --git a/net/core/datagram.c b/net/core/datagram.c index ee5647bd91b3..f7fb7e3f2acf 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -169,14 +169,20 @@ struct sk_buff *__skb_try_recv_from_queue(struct sock *sk, int *peeked, int *off, int *err, struct sk_buff **last) { + bool peek_at_off = false; struct sk_buff *skb; - int _off = *off; + int _off = 0; + + if (unlikely(flags & MSG_PEEK && *off >= 0)) { + peek_at_off = true; + _off = *off; + } *last = queue->prev; skb_queue_walk(queue, skb) { if (flags & MSG_PEEK) { - if (_off >= skb->len && (skb->len || _off || - skb->peeked)) { + if (peek_at_off && _off >= skb->len && + (_off || skb->peeked)) { _off -= skb->len; continue; } @@ -356,7 +362,7 @@ int __sk_queue_drop_skb(struct sock *sk, struct sk_buff_head *sk_queue, if (flags & MSG_PEEK) { err = -ENOENT; spin_lock_bh(&sk_queue->lock); - if (skb == skb_peek(sk_queue)) { + if (skb->next) { __skb_unlink(skb, sk_queue); refcount_dec(&skb->users); if (destructor) @@ -573,27 +579,12 @@ fault: } EXPORT_SYMBOL(skb_copy_datagram_from_iter); -/** - * zerocopy_sg_from_iter - Build a zerocopy datagram from an iov_iter - * @skb: buffer to copy - * @from: the source to copy from - * - * The function will first copy up to headlen, and then pin the userspace - * pages and build frags through them. - * - * Returns 0, -EFAULT or -EMSGSIZE. - */ -int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *from) +int __zerocopy_sg_from_iter(struct sock *sk, struct sk_buff *skb, + struct iov_iter *from, size_t length) { - int len = iov_iter_count(from); - int copy = min_t(int, skb_headlen(skb), len); - int frag = 0; + int frag = skb_shinfo(skb)->nr_frags; - /* copy up to skb headlen */ - if (skb_copy_datagram_from_iter(skb, 0, from, copy)) - return -EFAULT; - - while (iov_iter_count(from)) { + while (length && iov_iter_count(from)) { struct page *pages[MAX_SKB_FRAGS]; size_t start; ssize_t copied; @@ -603,18 +594,24 @@ int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *from) if (frag == MAX_SKB_FRAGS) return -EMSGSIZE; - copied = iov_iter_get_pages(from, pages, ~0U, + copied = iov_iter_get_pages(from, pages, length, MAX_SKB_FRAGS - frag, &start); if (copied < 0) return -EFAULT; iov_iter_advance(from, copied); + length -= copied; truesize = PAGE_ALIGN(copied + start); skb->data_len += copied; skb->len += copied; skb->truesize += truesize; - refcount_add(truesize, &skb->sk->sk_wmem_alloc); + if (sk && sk->sk_type == SOCK_STREAM) { + sk->sk_wmem_queued += truesize; + sk_mem_charge(sk, truesize); + } else { + refcount_add(truesize, &skb->sk->sk_wmem_alloc); + } while (copied) { int size = min_t(int, copied, PAGE_SIZE - start); skb_fill_page_desc(skb, frag++, pages[n], start, size); @@ -625,6 +622,28 @@ int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *from) } return 0; } +EXPORT_SYMBOL(__zerocopy_sg_from_iter); + +/** + * zerocopy_sg_from_iter - Build a zerocopy datagram from an iov_iter + * @skb: buffer to copy + * @from: the source to copy from + * + * The function will first copy up to headlen, and then pin the userspace + * pages and build frags through them. + * + * Returns 0, -EFAULT or -EMSGSIZE. + */ +int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *from) +{ + int copy = min_t(int, skb_headlen(skb), iov_iter_count(from)); + + /* copy up to skb headlen */ + if (skb_copy_datagram_from_iter(skb, 0, from, copy)) + return -EFAULT; + + return __zerocopy_sg_from_iter(NULL, skb, from, ~0U); +} EXPORT_SYMBOL(zerocopy_sg_from_iter); static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, diff --git a/net/core/dev.c b/net/core/dev.c index 8515f8fe0460..6f845e4fec17 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -144,6 +144,7 @@ #include <linux/netfilter_ingress.h> #include <linux/crash_dump.h> #include <linux/sctp.h> +#include <net/udp_tunnel.h> #include "net-sysfs.h" @@ -1413,7 +1414,7 @@ int dev_open(struct net_device *dev) } EXPORT_SYMBOL(dev_open); -static int __dev_close_many(struct list_head *head) +static void __dev_close_many(struct list_head *head) { struct net_device *dev; @@ -1455,23 +1456,18 @@ static int __dev_close_many(struct list_head *head) dev->flags &= ~IFF_UP; netpoll_poll_enable(dev); } - - return 0; } -static int __dev_close(struct net_device *dev) +static void __dev_close(struct net_device *dev) { - int retval; LIST_HEAD(single); list_add(&dev->close_list, &single); - retval = __dev_close_many(&single); + __dev_close_many(&single); list_del(&single); - - return retval; } -int dev_close_many(struct list_head *head, bool unlink) +void dev_close_many(struct list_head *head, bool unlink) { struct net_device *dev, *tmp; @@ -1488,8 +1484,6 @@ int dev_close_many(struct list_head *head, bool unlink) if (unlink) list_del_init(&dev->close_list); } - - return 0; } EXPORT_SYMBOL(dev_close_many); @@ -1502,7 +1496,7 @@ EXPORT_SYMBOL(dev_close_many); * is then deactivated and finally a %NETDEV_DOWN is sent to the notifier * chain. */ -int dev_close(struct net_device *dev) +void dev_close(struct net_device *dev) { if (dev->flags & IFF_UP) { LIST_HEAD(single); @@ -1511,7 +1505,6 @@ int dev_close(struct net_device *dev) dev_close_many(&single, true); list_del(&single); } - return 0; } EXPORT_SYMBOL(dev_close); @@ -1860,7 +1853,7 @@ static inline int deliver_skb(struct sk_buff *skb, struct packet_type *pt_prev, struct net_device *orig_dev) { - if (unlikely(skb_orphan_frags(skb, GFP_ATOMIC))) + if (unlikely(skb_orphan_frags_rx(skb, GFP_ATOMIC))) return -ENOMEM; refcount_inc(&skb->users); return pt_prev->func(skb, skb->dev, pt_prev, orig_dev); @@ -2738,8 +2731,7 @@ EXPORT_SYMBOL(skb_mac_gso_segment); static inline bool skb_needs_check(struct sk_buff *skb, bool tx_path) { if (tx_path) - return skb->ip_summed != CHECKSUM_PARTIAL && - skb->ip_summed != CHECKSUM_NONE; + return skb->ip_summed != CHECKSUM_PARTIAL; return skb->ip_summed == CHECKSUM_NONE; } @@ -3865,6 +3857,121 @@ drop: return NET_RX_DROP; } +static u32 netif_receive_generic_xdp(struct sk_buff *skb, + struct bpf_prog *xdp_prog) +{ + struct xdp_buff xdp; + u32 act = XDP_DROP; + void *orig_data; + int hlen, off; + u32 mac_len; + + /* Reinjected packets coming from act_mirred or similar should + * not get XDP generic processing. + */ + if (skb_cloned(skb)) + return XDP_PASS; + + if (skb_linearize(skb)) + goto do_drop; + + /* The XDP program wants to see the packet starting at the MAC + * header. + */ + mac_len = skb->data - skb_mac_header(skb); + hlen = skb_headlen(skb) + mac_len; + xdp.data = skb->data - mac_len; + xdp.data_end = xdp.data + hlen; + xdp.data_hard_start = skb->data - skb_headroom(skb); + orig_data = xdp.data; + + act = bpf_prog_run_xdp(xdp_prog, &xdp); + + off = xdp.data - orig_data; + if (off > 0) + __skb_pull(skb, off); + else if (off < 0) + __skb_push(skb, -off); + + switch (act) { + case XDP_REDIRECT: + case XDP_TX: + __skb_push(skb, mac_len); + /* fall through */ + case XDP_PASS: + break; + + default: + bpf_warn_invalid_xdp_action(act); + /* fall through */ + case XDP_ABORTED: + trace_xdp_exception(skb->dev, xdp_prog, act); + /* fall through */ + case XDP_DROP: + do_drop: + kfree_skb(skb); + break; + } + + return act; +} + +/* When doing generic XDP we have to bypass the qdisc layer and the + * network taps in order to match in-driver-XDP behavior. + */ +void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog) +{ + struct net_device *dev = skb->dev; + struct netdev_queue *txq; + bool free_skb = true; + int cpu, rc; + + txq = netdev_pick_tx(dev, skb, NULL); + cpu = smp_processor_id(); + HARD_TX_LOCK(dev, txq, cpu); + if (!netif_xmit_stopped(txq)) { + rc = netdev_start_xmit(skb, dev, txq, 0); + if (dev_xmit_complete(rc)) + free_skb = false; + } + HARD_TX_UNLOCK(dev, txq); + if (free_skb) { + trace_xdp_exception(dev, xdp_prog, XDP_TX); + kfree_skb(skb); + } +} +EXPORT_SYMBOL_GPL(generic_xdp_tx); + +static struct static_key generic_xdp_needed __read_mostly; + +int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff *skb) +{ + if (xdp_prog) { + u32 act = netif_receive_generic_xdp(skb, xdp_prog); + int err; + + if (act != XDP_PASS) { + switch (act) { + case XDP_REDIRECT: + err = xdp_do_generic_redirect(skb->dev, skb, + xdp_prog); + if (err) + goto out_redir; + /* fallthru to submit skb */ + case XDP_TX: + generic_xdp_tx(skb, xdp_prog); + break; + } + return XDP_DROP; + } + } + return XDP_PASS; +out_redir: + kfree_skb(skb); + return XDP_DROP; +} +EXPORT_SYMBOL_GPL(do_xdp_generic); + static int netif_rx_internal(struct sk_buff *skb) { int ret; @@ -3872,6 +3979,19 @@ static int netif_rx_internal(struct sk_buff *skb) net_timestamp_check(netdev_tstamp_prequeue, skb); trace_netif_rx(skb); + + if (static_key_false(&generic_xdp_needed)) { + int ret = do_xdp_generic(rcu_dereference(skb->dev->xdp_prog), + skb); + + /* Consider XDP consuming the packet a success from + * the netdev point of view we do not want to count + * this as an error. + */ + if (ret != XDP_PASS) + return NET_RX_SUCCESS; + } + #ifdef CONFIG_RPS if (static_key_false(&rps_needed)) { struct rps_dev_flow voidflow, *rflow = &voidflow; @@ -4292,7 +4412,7 @@ skip_classify: } if (pt_prev) { - if (unlikely(skb_orphan_frags(skb, GFP_ATOMIC))) + if (unlikely(skb_orphan_frags_rx(skb, GFP_ATOMIC))) goto drop; else ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev); @@ -4338,8 +4458,6 @@ static int __netif_receive_skb(struct sk_buff *skb) return ret; } -static struct static_key generic_xdp_needed __read_mostly; - static int generic_xdp_install(struct net_device *dev, struct netdev_xdp *xdp) { struct bpf_prog *old = rtnl_dereference(dev->xdp_prog); @@ -4373,89 +4491,6 @@ static int generic_xdp_install(struct net_device *dev, struct netdev_xdp *xdp) return ret; } -static u32 netif_receive_generic_xdp(struct sk_buff *skb, - struct bpf_prog *xdp_prog) -{ - struct xdp_buff xdp; - u32 act = XDP_DROP; - void *orig_data; - int hlen, off; - u32 mac_len; - - /* Reinjected packets coming from act_mirred or similar should - * not get XDP generic processing. - */ - if (skb_cloned(skb)) - return XDP_PASS; - - if (skb_linearize(skb)) - goto do_drop; - - /* The XDP program wants to see the packet starting at the MAC - * header. - */ - mac_len = skb->data - skb_mac_header(skb); - hlen = skb_headlen(skb) + mac_len; - xdp.data = skb->data - mac_len; - xdp.data_end = xdp.data + hlen; - xdp.data_hard_start = skb->data - skb_headroom(skb); - orig_data = xdp.data; - - act = bpf_prog_run_xdp(xdp_prog, &xdp); - - off = xdp.data - orig_data; - if (off > 0) - __skb_pull(skb, off); - else if (off < 0) - __skb_push(skb, -off); - - switch (act) { - case XDP_TX: - __skb_push(skb, mac_len); - /* fall through */ - case XDP_PASS: - break; - - default: - bpf_warn_invalid_xdp_action(act); - /* fall through */ - case XDP_ABORTED: - trace_xdp_exception(skb->dev, xdp_prog, act); - /* fall through */ - case XDP_DROP: - do_drop: - kfree_skb(skb); - break; - } - - return act; -} - -/* When doing generic XDP we have to bypass the qdisc layer and the - * network taps in order to match in-driver-XDP behavior. - */ -static void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog) -{ - struct net_device *dev = skb->dev; - struct netdev_queue *txq; - bool free_skb = true; - int cpu, rc; - - txq = netdev_pick_tx(dev, skb, NULL); - cpu = smp_processor_id(); - HARD_TX_LOCK(dev, txq, cpu); - if (!netif_xmit_stopped(txq)) { - rc = netdev_start_xmit(skb, dev, txq, 0); - if (dev_xmit_complete(rc)) - free_skb = false; - } - HARD_TX_UNLOCK(dev, txq); - if (free_skb) { - trace_xdp_exception(dev, xdp_prog, XDP_TX); - kfree_skb(skb); - } -} - static int netif_receive_skb_internal(struct sk_buff *skb) { int ret; @@ -4468,17 +4503,12 @@ static int netif_receive_skb_internal(struct sk_buff *skb) rcu_read_lock(); if (static_key_false(&generic_xdp_needed)) { - struct bpf_prog *xdp_prog = rcu_dereference(skb->dev->xdp_prog); + int ret = do_xdp_generic(rcu_dereference(skb->dev->xdp_prog), + skb); - if (xdp_prog) { - u32 act = netif_receive_generic_xdp(skb, xdp_prog); - - if (act != XDP_PASS) { - rcu_read_unlock(); - if (act == XDP_TX) - generic_xdp_tx(skb, xdp_prog); - return NET_RX_DROP; - } + if (ret != XDP_PASS) { + rcu_read_unlock(); + return NET_RX_DROP; } } @@ -5289,6 +5319,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock) * Ideally, a new ndo_busy_poll_stop() could avoid another round. */ rc = napi->poll(napi, BUSY_POLL_BUDGET); + trace_napi_poll(napi, rc, BUSY_POLL_BUDGET); netpoll_poll_unlock(have_poll_lock); if (rc == BUSY_POLL_BUDGET) __napi_schedule(napi); @@ -5667,12 +5698,13 @@ EXPORT_SYMBOL(netdev_has_upper_dev_all_rcu); * Find out if a device is linked to an upper device and return true in case * it is. The caller must hold the RTNL lock. */ -static bool netdev_has_any_upper_dev(struct net_device *dev) +bool netdev_has_any_upper_dev(struct net_device *dev) { ASSERT_RTNL(); return !list_empty(&dev->adj_list.upper); } +EXPORT_SYMBOL(netdev_has_any_upper_dev); /** * netdev_master_upper_dev_get - Get master upper device @@ -6689,8 +6721,12 @@ int __dev_change_flags(struct net_device *dev, unsigned int flags) */ ret = 0; - if ((old_flags ^ flags) & IFF_UP) - ret = ((old_flags & IFF_UP) ? __dev_close : __dev_open)(dev); + if ((old_flags ^ flags) & IFF_UP) { + if (old_flags & IFF_UP) + __dev_close(dev); + else + ret = __dev_open(dev); + } if ((flags ^ dev->gflags) & IFF_PROMISC) { int inc = (flags & IFF_PROMISC) ? 1 : -1; @@ -7235,24 +7271,6 @@ static netdev_features_t netdev_fix_features(struct net_device *dev, features &= ~NETIF_F_GSO; } - /* UFO needs SG and checksumming */ - if (features & NETIF_F_UFO) { - /* maybe split UFO into V4 and V6? */ - if (!(features & NETIF_F_HW_CSUM) && - ((features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) != - (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) { - netdev_dbg(dev, - "Dropping NETIF_F_UFO since no checksum offload features.\n"); - features &= ~NETIF_F_UFO; - } - - if (!(features & NETIF_F_SG)) { - netdev_dbg(dev, - "Dropping NETIF_F_UFO since no NETIF_F_SG feature.\n"); - features &= ~NETIF_F_UFO; - } - } - /* GSO partial features require GSO partial be set */ if ((features & dev->gso_partial_features) && !(features & NETIF_F_GSO_PARTIAL)) { @@ -7313,8 +7331,27 @@ sync_lower: netdev_for_each_lower_dev(dev, lower, iter) netdev_sync_lower_features(dev, lower, features); - if (!err) + if (!err) { + netdev_features_t diff = features ^ dev->features; + + if (diff & NETIF_F_RX_UDP_TUNNEL_PORT) { + /* udp_tunnel_{get,drop}_rx_info both need + * NETIF_F_RX_UDP_TUNNEL_PORT enabled on the + * device, or they won't do anything. + * Thus we need to update dev->features + * *before* calling udp_tunnel_get_rx_info, + * but *after* calling udp_tunnel_drop_rx_info. + */ + if (features & NETIF_F_RX_UDP_TUNNEL_PORT) { + dev->features = features; + udp_tunnel_get_rx_info(dev); + } else { + udp_tunnel_drop_rx_info(dev); + } + } + dev->features = features; + } return err < 0 ? 0 : 1; } @@ -7516,6 +7553,12 @@ int register_netdevice(struct net_device *dev) */ dev->hw_features |= NETIF_F_SOFT_FEATURES; dev->features |= NETIF_F_SOFT_FEATURES; + + if (dev->netdev_ops->ndo_udp_tunnel_add) { + dev->features |= NETIF_F_RX_UDP_TUNNEL_PORT; + dev->hw_features |= NETIF_F_RX_UDP_TUNNEL_PORT; + } + dev->wanted_features = dev->features & dev->hw_features; if (!(dev->flags & IFF_LOOPBACK)) diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 82fd4c9c4a1b..709a4e6fb447 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -28,6 +28,7 @@ static int dev_ifname(struct net *net, struct ifreq __user *arg) if (copy_from_user(&ifr, arg, sizeof(struct ifreq))) return -EFAULT; + ifr.ifr_name[IFNAMSIZ-1] = 0; error = netdev_get_name(net, ifr.ifr_name, ifr.ifr_ifindex); if (error) @@ -262,6 +263,8 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, unsigned int cmd) return dev_set_mtu(dev, ifr->ifr_mtu); case SIOCSIFHWADDR: + if (dev->addr_len > sizeof(struct sockaddr)) + return -EINVAL; return dev_set_mac_address(dev, &ifr->ifr_hwaddr); case SIOCSIFHWBROADCAST: @@ -424,6 +427,8 @@ int dev_ioctl(struct net *net, unsigned int cmd, void __user *arg) if (copy_from_user(&iwr, arg, sizeof(iwr))) return -EFAULT; + iwr.ifr_name[sizeof(iwr.ifr_name) - 1] = 0; + return wext_handle_ioctl(net, &iwr, cmd, arg); } diff --git a/net/core/devlink.c b/net/core/devlink.c index a0adfc31a3fe..7d430c1d9c3e 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -29,6 +29,57 @@ #define CREATE_TRACE_POINTS #include <trace/events/devlink.h> +static struct devlink_dpipe_field devlink_dpipe_fields_ethernet[] = { + { + .name = "destination mac", + .id = DEVLINK_DPIPE_FIELD_ETHERNET_DST_MAC, + .bitwidth = 48, + }, +}; + +struct devlink_dpipe_header devlink_dpipe_header_ethernet = { + .name = "ethernet", + .id = DEVLINK_DPIPE_HEADER_ETHERNET, + .fields = devlink_dpipe_fields_ethernet, + .fields_count = ARRAY_SIZE(devlink_dpipe_fields_ethernet), + .global = true, +}; +EXPORT_SYMBOL(devlink_dpipe_header_ethernet); + +static struct devlink_dpipe_field devlink_dpipe_fields_ipv4[] = { + { + .name = "destination ip", + .id = DEVLINK_DPIPE_FIELD_IPV4_DST_IP, + .bitwidth = 32, + }, +}; + +struct devlink_dpipe_header devlink_dpipe_header_ipv4 = { + .name = "ipv4", + .id = DEVLINK_DPIPE_HEADER_IPV4, + .fields = devlink_dpipe_fields_ipv4, + .fields_count = ARRAY_SIZE(devlink_dpipe_fields_ipv4), + .global = true, +}; +EXPORT_SYMBOL(devlink_dpipe_header_ipv4); + +static struct devlink_dpipe_field devlink_dpipe_fields_ipv6[] = { + { + .name = "destination ip", + .id = DEVLINK_DPIPE_FIELD_IPV6_DST_IP, + .bitwidth = 128, + }, +}; + +struct devlink_dpipe_header devlink_dpipe_header_ipv6 = { + .name = "ipv6", + .id = DEVLINK_DPIPE_HEADER_IPV6, + .fields = devlink_dpipe_fields_ipv6, + .fields_count = ARRAY_SIZE(devlink_dpipe_fields_ipv6), + .global = true, +}; +EXPORT_SYMBOL(devlink_dpipe_header_ipv6); + EXPORT_TRACEPOINT_SYMBOL_GPL(devlink_hwmsg); static LIST_HEAD(devlink_list); @@ -1613,13 +1664,15 @@ static int devlink_dpipe_table_put(struct sk_buff *skb, struct devlink_dpipe_table *table) { struct nlattr *table_attr; + u64 table_size; + table_size = table->table_ops->size_get(table->priv); table_attr = nla_nest_start(skb, DEVLINK_ATTR_DPIPE_TABLE); if (!table_attr) return -EMSGSIZE; if (nla_put_string(skb, DEVLINK_ATTR_DPIPE_TABLE_NAME, table->name) || - nla_put_u64_64bit(skb, DEVLINK_ATTR_DPIPE_TABLE_SIZE, table->size, + nla_put_u64_64bit(skb, DEVLINK_ATTR_DPIPE_TABLE_SIZE, table_size, DEVLINK_ATTR_PAD)) goto nla_put_failure; if (nla_put_u8(skb, DEVLINK_ATTR_DPIPE_TABLE_COUNTERS_ENABLED, @@ -1960,6 +2013,28 @@ int devlink_dpipe_entry_ctx_close(struct devlink_dpipe_dump_ctx *dump_ctx) } EXPORT_SYMBOL_GPL(devlink_dpipe_entry_ctx_close); +void devlink_dpipe_entry_clear(struct devlink_dpipe_entry *entry) + +{ + unsigned int value_count, value_index; + struct devlink_dpipe_value *value; + + value = entry->action_values; + value_count = entry->action_values_count; + for (value_index = 0; value_index < value_count; value_index++) { + kfree(value[value_index].value); + kfree(value[value_index].mask); + } + + value = entry->match_values; + value_count = entry->match_values_count; + for (value_index = 0; value_index < value_count; value_index++) { + kfree(value[value_index].value); + kfree(value[value_index].mask); + } +} +EXPORT_SYMBOL(devlink_dpipe_entry_clear); + static int devlink_dpipe_entries_fill(struct genl_info *info, enum devlink_command cmd, int flags, struct devlink_dpipe_table *table) @@ -2684,20 +2759,21 @@ EXPORT_SYMBOL_GPL(devlink_dpipe_table_counter_enabled); * @table_name: table name * @table_ops: table ops * @priv: priv - * @size: size * @counter_control_extern: external control for counters */ int devlink_dpipe_table_register(struct devlink *devlink, const char *table_name, struct devlink_dpipe_table_ops *table_ops, - void *priv, u64 size, - bool counter_control_extern) + void *priv, bool counter_control_extern) { struct devlink_dpipe_table *table; if (devlink_dpipe_table_find(&devlink->dpipe_table_list, table_name)) return -EEXIST; + if (WARN_ON(!table_ops->size_get)) + return -EINVAL; + table = kzalloc(sizeof(*table), GFP_KERNEL); if (!table) return -ENOMEM; @@ -2705,7 +2781,6 @@ int devlink_dpipe_table_register(struct devlink *devlink, table->name = table_name; table->table_ops = table_ops; table->priv = priv; - table->size = size; table->counter_control_extern = counter_control_extern; mutex_lock(&devlink_mutex); diff --git a/net/core/dst.c b/net/core/dst.c index 00aa972ad1a1..a6c47da7d0f8 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -55,7 +55,7 @@ const struct dst_metrics dst_default_metrics = { * We really want to avoid false sharing on this variable, and catch * any writes on it. */ - .refcnt = ATOMIC_INIT(1), + .refcnt = REFCOUNT_INIT(1), }; void dst_init(struct dst_entry *dst, struct dst_ops *ops, @@ -213,7 +213,7 @@ u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old) struct dst_metrics *old_p = (struct dst_metrics *)__DST_METRICS_PTR(old); unsigned long prev, new; - atomic_set(&p->refcnt, 1); + refcount_set(&p->refcnt, 1); memcpy(p->metrics, old_p->metrics, sizeof(p->metrics)); new = (unsigned long) p; @@ -225,7 +225,7 @@ u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old) if (prev & DST_METRICS_READ_ONLY) p = NULL; } else if (prev & DST_METRICS_REFCOUNTED) { - if (atomic_dec_and_test(&old_p->refcnt)) + if (refcount_dec_and_test(&old_p->refcnt)) kfree(old_p); } } @@ -299,7 +299,8 @@ EXPORT_SYMBOL_GPL(metadata_dst_alloc); void metadata_dst_free(struct metadata_dst *md_dst) { #ifdef CONFIG_DST_CACHE - dst_cache_destroy(&md_dst->u.tun_info.dst_cache); + if (md_dst->type == METADATA_IP_TUNNEL) + dst_cache_destroy(&md_dst->u.tun_info.dst_cache); #endif kfree(md_dst); } diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 674b6c9cec18..6a582ae4c5d9 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -76,7 +76,6 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] [NETIF_F_LRO_BIT] = "rx-lro", [NETIF_F_TSO_BIT] = "tx-tcp-segmentation", - [NETIF_F_UFO_BIT] = "tx-udp-fragmentation", [NETIF_F_GSO_ROBUST_BIT] = "tx-gso-robust", [NETIF_F_TSO_ECN_BIT] = "tx-tcp-ecn-segmentation", [NETIF_F_TSO_MANGLEID_BIT] = "tx-tcp-mangleid-segmentation", @@ -106,6 +105,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] [NETIF_F_HW_TC_BIT] = "hw-tc-offload", [NETIF_F_HW_ESP_BIT] = "esp-hw-offload", [NETIF_F_HW_ESP_TX_CSUM_BIT] = "esp-tx-csum-hw-offload", + [NETIF_F_RX_UDP_TUNNEL_PORT_BIT] = "rx-udp_tunnel-port-offload", }; static const char @@ -299,9 +299,6 @@ static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd) case ETHTOOL_GTSO: case ETHTOOL_STSO: return NETIF_F_ALL_TSO; - case ETHTOOL_GUFO: - case ETHTOOL_SUFO: - return NETIF_F_UFO; case ETHTOOL_GGSO: case ETHTOOL_SGSO: return NETIF_F_GSO; @@ -2515,6 +2512,33 @@ static int set_phy_tunable(struct net_device *dev, void __user *useraddr) return ret; } +static int ethtool_get_fecparam(struct net_device *dev, void __user *useraddr) +{ + struct ethtool_fecparam fecparam = { ETHTOOL_GFECPARAM }; + + if (!dev->ethtool_ops->get_fecparam) + return -EOPNOTSUPP; + + dev->ethtool_ops->get_fecparam(dev, &fecparam); + + if (copy_to_user(useraddr, &fecparam, sizeof(fecparam))) + return -EFAULT; + return 0; +} + +static int ethtool_set_fecparam(struct net_device *dev, void __user *useraddr) +{ + struct ethtool_fecparam fecparam; + + if (!dev->ethtool_ops->set_fecparam) + return -EOPNOTSUPP; + + if (copy_from_user(&fecparam, useraddr, sizeof(fecparam))) + return -EFAULT; + + return dev->ethtool_ops->set_fecparam(dev, &fecparam); +} + /* The main entry point in this file. Called from net/core/dev_ioctl.c */ int dev_ethtool(struct net *net, struct ifreq *ifr) @@ -2555,7 +2579,6 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GPHYSTATS: case ETHTOOL_GTSO: case ETHTOOL_GPERMADDR: - case ETHTOOL_GUFO: case ETHTOOL_GGSO: case ETHTOOL_GGRO: case ETHTOOL_GFLAGS: @@ -2574,6 +2597,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GTUNABLE: case ETHTOOL_PHY_GTUNABLE: case ETHTOOL_GLINKSETTINGS: + case ETHTOOL_GFECPARAM: break; default: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) @@ -2723,7 +2747,6 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GRXCSUM: case ETHTOOL_GSG: case ETHTOOL_GTSO: - case ETHTOOL_GUFO: case ETHTOOL_GGSO: case ETHTOOL_GGRO: rc = ethtool_get_one_feature(dev, useraddr, ethcmd); @@ -2732,7 +2755,6 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_SRXCSUM: case ETHTOOL_SSG: case ETHTOOL_STSO: - case ETHTOOL_SUFO: case ETHTOOL_SGSO: case ETHTOOL_SGRO: rc = ethtool_set_one_feature(dev, useraddr, ethcmd); @@ -2785,6 +2807,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_PHY_STUNABLE: rc = set_phy_tunable(dev, useraddr); break; + case ETHTOOL_GFECPARAM: + rc = ethtool_get_fecparam(dev, useraddr); + break; + case ETHTOOL_SFECPARAM: + rc = ethtool_set_fecparam(dev, useraddr); + break; default: rc = -EOPNOTSUPP; } diff --git a/net/core/fib_notifier.c b/net/core/fib_notifier.c new file mode 100644 index 000000000000..4fc202dbdfb6 --- /dev/null +++ b/net/core/fib_notifier.c @@ -0,0 +1,173 @@ +#include <linux/rtnetlink.h> +#include <linux/notifier.h> +#include <linux/rcupdate.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/net_namespace.h> +#include <net/fib_notifier.h> + +static ATOMIC_NOTIFIER_HEAD(fib_chain); + +int call_fib_notifier(struct notifier_block *nb, struct net *net, + enum fib_event_type event_type, + struct fib_notifier_info *info) +{ + info->net = net; + return nb->notifier_call(nb, event_type, info); +} +EXPORT_SYMBOL(call_fib_notifier); + +int call_fib_notifiers(struct net *net, enum fib_event_type event_type, + struct fib_notifier_info *info) +{ + info->net = net; + return atomic_notifier_call_chain(&fib_chain, event_type, info); +} +EXPORT_SYMBOL(call_fib_notifiers); + +static unsigned int fib_seq_sum(void) +{ + struct fib_notifier_ops *ops; + unsigned int fib_seq = 0; + struct net *net; + + rtnl_lock(); + for_each_net(net) { + list_for_each_entry(ops, &net->fib_notifier_ops, list) { + if (!try_module_get(ops->owner)) + continue; + fib_seq += ops->fib_seq_read(net); + module_put(ops->owner); + } + } + rtnl_unlock(); + + return fib_seq; +} + +static int fib_net_dump(struct net *net, struct notifier_block *nb) +{ + struct fib_notifier_ops *ops; + + list_for_each_entry_rcu(ops, &net->fib_notifier_ops, list) { + int err; + + if (!try_module_get(ops->owner)) + continue; + err = ops->fib_dump(net, nb); + module_put(ops->owner); + if (err) + return err; + } + + return 0; +} + +static bool fib_dump_is_consistent(struct notifier_block *nb, + void (*cb)(struct notifier_block *nb), + unsigned int fib_seq) +{ + atomic_notifier_chain_register(&fib_chain, nb); + if (fib_seq == fib_seq_sum()) + return true; + atomic_notifier_chain_unregister(&fib_chain, nb); + if (cb) + cb(nb); + return false; +} + +#define FIB_DUMP_MAX_RETRIES 5 +int register_fib_notifier(struct notifier_block *nb, + void (*cb)(struct notifier_block *nb)) +{ + int retries = 0; + int err; + + do { + unsigned int fib_seq = fib_seq_sum(); + struct net *net; + + rcu_read_lock(); + for_each_net_rcu(net) { + err = fib_net_dump(net, nb); + if (err) + goto err_fib_net_dump; + } + rcu_read_unlock(); + + if (fib_dump_is_consistent(nb, cb, fib_seq)) + return 0; + } while (++retries < FIB_DUMP_MAX_RETRIES); + + return -EBUSY; + +err_fib_net_dump: + rcu_read_unlock(); + return err; +} +EXPORT_SYMBOL(register_fib_notifier); + +int unregister_fib_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&fib_chain, nb); +} +EXPORT_SYMBOL(unregister_fib_notifier); + +static int __fib_notifier_ops_register(struct fib_notifier_ops *ops, + struct net *net) +{ + struct fib_notifier_ops *o; + + list_for_each_entry(o, &net->fib_notifier_ops, list) + if (ops->family == o->family) + return -EEXIST; + list_add_tail_rcu(&ops->list, &net->fib_notifier_ops); + return 0; +} + +struct fib_notifier_ops * +fib_notifier_ops_register(const struct fib_notifier_ops *tmpl, struct net *net) +{ + struct fib_notifier_ops *ops; + int err; + + ops = kmemdup(tmpl, sizeof(*ops), GFP_KERNEL); + if (!ops) + return ERR_PTR(-ENOMEM); + + err = __fib_notifier_ops_register(ops, net); + if (err) + goto err_register; + + return ops; + +err_register: + kfree(ops); + return ERR_PTR(err); +} +EXPORT_SYMBOL(fib_notifier_ops_register); + +void fib_notifier_ops_unregister(struct fib_notifier_ops *ops) +{ + list_del_rcu(&ops->list); + kfree_rcu(ops, rcu); +} +EXPORT_SYMBOL(fib_notifier_ops_unregister); + +static int __net_init fib_notifier_net_init(struct net *net) +{ + INIT_LIST_HEAD(&net->fib_notifier_ops); + return 0; +} + +static struct pernet_operations fib_notifier_net_ops = { + .init = fib_notifier_net_init, +}; + +static int __init fib_notifier_init(void) +{ + return register_pernet_subsys(&fib_notifier_net_ops); +} + +subsys_initcall(fib_notifier_init); diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index a0093e1b0235..9a6d97c1d810 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -299,6 +299,67 @@ out: } EXPORT_SYMBOL_GPL(fib_rules_lookup); +static int call_fib_rule_notifier(struct notifier_block *nb, struct net *net, + enum fib_event_type event_type, + struct fib_rule *rule, int family) +{ + struct fib_rule_notifier_info info = { + .info.family = family, + .rule = rule, + }; + + return call_fib_notifier(nb, net, event_type, &info.info); +} + +static int call_fib_rule_notifiers(struct net *net, + enum fib_event_type event_type, + struct fib_rule *rule, + struct fib_rules_ops *ops) +{ + struct fib_rule_notifier_info info = { + .info.family = ops->family, + .rule = rule, + }; + + ops->fib_rules_seq++; + return call_fib_notifiers(net, event_type, &info.info); +} + +/* Called with rcu_read_lock() */ +int fib_rules_dump(struct net *net, struct notifier_block *nb, int family) +{ + struct fib_rules_ops *ops; + struct fib_rule *rule; + + ops = lookup_rules_ops(net, family); + if (!ops) + return -EAFNOSUPPORT; + list_for_each_entry_rcu(rule, &ops->rules_list, list) + call_fib_rule_notifier(nb, net, FIB_EVENT_RULE_ADD, rule, + family); + rules_ops_put(ops); + + return 0; +} +EXPORT_SYMBOL_GPL(fib_rules_dump); + +unsigned int fib_rules_seq_read(struct net *net, int family) +{ + unsigned int fib_rules_seq; + struct fib_rules_ops *ops; + + ASSERT_RTNL(); + + ops = lookup_rules_ops(net, family); + if (!ops) + return 0; + fib_rules_seq = ops->fib_rules_seq; + rules_ops_put(ops); + + return fib_rules_seq; +} +EXPORT_SYMBOL_GPL(fib_rules_seq_read); + static int validate_rulemsg(struct fib_rule_hdr *frh, struct nlattr **tb, struct fib_rules_ops *ops) { @@ -400,6 +461,7 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh, err = -ENOMEM; goto errout; } + refcount_set(&rule->refcnt, 1); rule->fr_net = net; rule->pref = tb[FRA_PRIORITY] ? nla_get_u32(tb[FRA_PRIORITY]) @@ -517,8 +579,6 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh, last = r; } - refcount_set(&rule->refcnt, 1); - if (last) list_add_rcu(&rule->list, &last->list); else @@ -549,6 +609,7 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh, if (rule->tun_id) ip_tunnel_need_metadata(); + call_fib_rule_notifiers(net, FIB_EVENT_RULE_ADD, rule, ops); notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).portid); flush_route_cache(ops); rules_ops_put(ops); @@ -688,6 +749,7 @@ int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh, } } + call_fib_rule_notifiers(net, FIB_EVENT_RULE_DEL, rule, ops); notify_rule_change(RTM_DELRULE, rule, ops, nlh, NETLINK_CB(skb).portid); fib_rule_put(rule); @@ -964,9 +1026,9 @@ static struct pernet_operations fib_rules_net_ops = { static int __init fib_rules_init(void) { int err; - rtnl_register(PF_UNSPEC, RTM_NEWRULE, fib_nl_newrule, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELRULE, fib_nl_delrule, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_GETRULE, NULL, fib_nl_dumprule, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWRULE, fib_nl_newrule, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELRULE, fib_nl_delrule, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETRULE, NULL, fib_nl_dumprule, 0); err = register_pernet_subsys(&fib_rules_net_ops); if (err < 0) diff --git a/net/core/filter.c b/net/core/filter.c index c7f737058d89..5912c738a7b2 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -55,6 +55,7 @@ #include <net/sock_reuseport.h> #include <net/busy_poll.h> #include <net/tcp.h> +#include <linux/bpf_trace.h> /** * sk_filter_trim_cap - run a packet through a socket filter @@ -513,14 +514,27 @@ do_pass: break; } - /* Convert JEQ into JNE when 'jump_true' is next insn. */ - if (fp->jt == 0 && BPF_OP(fp->code) == BPF_JEQ) { - insn->code = BPF_JMP | BPF_JNE | bpf_src; + /* Convert some jumps when 'jump_true' is next insn. */ + if (fp->jt == 0) { + switch (BPF_OP(fp->code)) { + case BPF_JEQ: + insn->code = BPF_JMP | BPF_JNE | bpf_src; + break; + case BPF_JGT: + insn->code = BPF_JMP | BPF_JLE | bpf_src; + break; + case BPF_JGE: + insn->code = BPF_JMP | BPF_JLT | bpf_src; + break; + default: + goto jmp_rest; + } + target = i + fp->jf + 1; BPF_EMIT_JMP; break; } - +jmp_rest: /* Other jumps are mapped into two insns: Jxx and JA. */ target = i + fp->jt + 1; insn->code = BPF_JMP | BPF_OP(fp->code) | bpf_src; @@ -1778,6 +1792,8 @@ static const struct bpf_func_proto bpf_clone_redirect_proto = { struct redirect_info { u32 ifindex; u32 flags; + struct bpf_map *map; + struct bpf_map *map_to_flush; }; static DEFINE_PER_CPU(struct redirect_info, redirect_info); @@ -1791,6 +1807,7 @@ BPF_CALL_2(bpf_redirect, u32, ifindex, u64, flags) ri->ifindex = ifindex; ri->flags = flags; + ri->map = NULL; return TC_ACT_REDIRECT; } @@ -1818,6 +1835,45 @@ static const struct bpf_func_proto bpf_redirect_proto = { .arg2_type = ARG_ANYTHING, }; +BPF_CALL_3(bpf_sk_redirect_map, struct bpf_map *, map, u32, key, u64, flags) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + + if (unlikely(flags)) + return SK_ABORTED; + + ri->ifindex = key; + ri->flags = flags; + ri->map = map; + + return SK_REDIRECT; +} + +struct sock *do_sk_redirect_map(void) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + struct sock *sk = NULL; + + if (ri->map) { + sk = __sock_map_lookup_elem(ri->map, ri->ifindex); + + ri->ifindex = 0; + ri->map = NULL; + /* we do not clear flags for future lookup */ + } + + return sk; +} + +static const struct bpf_func_proto bpf_sk_redirect_map_proto = { + .func = bpf_sk_redirect_map, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_CONST_MAP_PTR, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, +}; + BPF_CALL_1(bpf_get_cgroup_classid, const struct sk_buff *, skb) { return task_get_classid(skb); @@ -2024,8 +2080,8 @@ static int bpf_skb_proto_4_to_6(struct sk_buff *skb) return ret; if (skb_is_gso(skb)) { - /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV4 needs to - * be changed into SKB_GSO_TCPV6. + /* SKB_GSO_TCPV4 needs to be changed into + * SKB_GSO_TCPV6. */ if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) { skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV4; @@ -2060,8 +2116,8 @@ static int bpf_skb_proto_6_to_4(struct sk_buff *skb) return ret; if (skb_is_gso(skb)) { - /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV6 needs to - * be changed into SKB_GSO_TCPV4. + /* SKB_GSO_TCPV6 needs to be changed into + * SKB_GSO_TCPV4. */ if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) { skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV6; @@ -2248,7 +2304,7 @@ static int bpf_skb_adjust_net(struct sk_buff *skb, s32 len_diff) bpf_skb_net_grow(skb, len_diff_abs); bpf_compute_data_end(skb); - return 0; + return ret; } BPF_CALL_4(bpf_skb_adjust_room, struct sk_buff *, skb, s32, len_diff, @@ -2412,6 +2468,180 @@ static const struct bpf_func_proto bpf_xdp_adjust_head_proto = { .arg2_type = ARG_ANYTHING, }; +static int __bpf_tx_xdp(struct net_device *dev, + struct bpf_map *map, + struct xdp_buff *xdp, + u32 index) +{ + int err; + + if (!dev->netdev_ops->ndo_xdp_xmit) { + return -EOPNOTSUPP; + } + + err = dev->netdev_ops->ndo_xdp_xmit(dev, xdp); + if (err) + return err; + if (map) + __dev_map_insert_ctx(map, index); + else + dev->netdev_ops->ndo_xdp_flush(dev); + return 0; +} + +void xdp_do_flush_map(void) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + struct bpf_map *map = ri->map_to_flush; + + ri->map_to_flush = NULL; + if (map) + __dev_map_flush(map); +} +EXPORT_SYMBOL_GPL(xdp_do_flush_map); + +static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp, + struct bpf_prog *xdp_prog) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + struct bpf_map *map = ri->map; + u32 index = ri->ifindex; + struct net_device *fwd; + int err; + + ri->ifindex = 0; + ri->map = NULL; + + fwd = __dev_map_lookup_elem(map, index); + if (!fwd) { + err = -EINVAL; + goto err; + } + if (ri->map_to_flush && ri->map_to_flush != map) + xdp_do_flush_map(); + + err = __bpf_tx_xdp(fwd, map, xdp, index); + if (unlikely(err)) + goto err; + + ri->map_to_flush = map; + _trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index); + return 0; +err: + _trace_xdp_redirect_map_err(dev, xdp_prog, fwd, map, index, err); + return err; +} + +int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp, + struct bpf_prog *xdp_prog) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + struct net_device *fwd; + u32 index = ri->ifindex; + int err; + + if (ri->map) + return xdp_do_redirect_map(dev, xdp, xdp_prog); + + fwd = dev_get_by_index_rcu(dev_net(dev), index); + ri->ifindex = 0; + if (unlikely(!fwd)) { + err = -EINVAL; + goto err; + } + + err = __bpf_tx_xdp(fwd, NULL, xdp, 0); + if (unlikely(err)) + goto err; + + _trace_xdp_redirect(dev, xdp_prog, index); + return 0; +err: + _trace_xdp_redirect_err(dev, xdp_prog, index, err); + return err; +} +EXPORT_SYMBOL_GPL(xdp_do_redirect); + +int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb, + struct bpf_prog *xdp_prog) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + u32 index = ri->ifindex; + struct net_device *fwd; + unsigned int len; + int err = 0; + + fwd = dev_get_by_index_rcu(dev_net(dev), index); + ri->ifindex = 0; + if (unlikely(!fwd)) { + err = -EINVAL; + goto err; + } + + if (unlikely(!(fwd->flags & IFF_UP))) { + err = -ENETDOWN; + goto err; + } + + len = fwd->mtu + fwd->hard_header_len + VLAN_HLEN; + if (skb->len > len) { + err = -EMSGSIZE; + goto err; + } + + skb->dev = fwd; + _trace_xdp_redirect(dev, xdp_prog, index); + return 0; +err: + _trace_xdp_redirect_err(dev, xdp_prog, index, err); + return err; +} +EXPORT_SYMBOL_GPL(xdp_do_generic_redirect); + +BPF_CALL_2(bpf_xdp_redirect, u32, ifindex, u64, flags) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + + if (unlikely(flags)) + return XDP_ABORTED; + + ri->ifindex = ifindex; + ri->flags = flags; + + return XDP_REDIRECT; +} + +static const struct bpf_func_proto bpf_xdp_redirect_proto = { + .func = bpf_xdp_redirect, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_ANYTHING, + .arg2_type = ARG_ANYTHING, +}; + +BPF_CALL_3(bpf_xdp_redirect_map, struct bpf_map *, map, u32, ifindex, u64, flags) +{ + struct redirect_info *ri = this_cpu_ptr(&redirect_info); + + if (unlikely(flags)) + return XDP_ABORTED; + + ri->ifindex = ifindex; + ri->flags = flags; + ri->map = map; + + return XDP_REDIRECT; +} + +static const struct bpf_func_proto bpf_xdp_redirect_map_proto = { + .func = bpf_xdp_redirect_map, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_CONST_MAP_PTR, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, +}; + bool bpf_helper_changes_pkt_data(void *func) { if (func == bpf_skb_vlan_push || @@ -2836,15 +3066,12 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, sk->sk_prot->setsockopt == tcp_setsockopt) { if (optname == TCP_CONGESTION) { char name[TCP_CA_NAME_MAX]; + bool reinit = bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN; strncpy(name, optval, min_t(long, optlen, TCP_CA_NAME_MAX-1)); name[TCP_CA_NAME_MAX-1] = 0; - ret = tcp_set_congestion_control(sk, name, false); - if (!ret && bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN) - /* replacing an existing ca */ - tcp_reinit_congestion_control(sk, - inet_csk(sk)->icsk_ca_ops); + ret = tcp_set_congestion_control(sk, name, false, reinit); } else { struct tcp_sock *tp = tcp_sk(sk); @@ -2872,7 +3099,6 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, ret = -EINVAL; } } - ret = -EINVAL; #endif } else { ret = -EINVAL; @@ -2920,6 +3146,20 @@ bpf_base_func_proto(enum bpf_func_id func_id) } static const struct bpf_func_proto * +sock_filter_func_proto(enum bpf_func_id func_id) +{ + switch (func_id) { + /* inet and inet6 sockets are created in a process + * context so there is always a valid uid/gid + */ + case BPF_FUNC_get_current_uid_gid: + return &bpf_get_current_uid_gid_proto; + default: + return bpf_base_func_proto(func_id); + } +} + +static const struct bpf_func_proto * sk_filter_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -3011,6 +3251,10 @@ xdp_func_proto(enum bpf_func_id func_id) return &bpf_get_smp_processor_id_proto; case BPF_FUNC_xdp_adjust_head: return &bpf_xdp_adjust_head_proto; + case BPF_FUNC_redirect: + return &bpf_xdp_redirect_proto; + case BPF_FUNC_redirect_map: + return &bpf_xdp_redirect_map_proto; default: return bpf_base_func_proto(func_id); } @@ -3049,6 +3293,32 @@ static const struct bpf_func_proto * switch (func_id) { case BPF_FUNC_setsockopt: return &bpf_setsockopt_proto; + case BPF_FUNC_sock_map_update: + return &bpf_sock_map_update_proto; + default: + return bpf_base_func_proto(func_id); + } +} + +static const struct bpf_func_proto *sk_skb_func_proto(enum bpf_func_id func_id) +{ + switch (func_id) { + case BPF_FUNC_skb_store_bytes: + return &bpf_skb_store_bytes_proto; + case BPF_FUNC_skb_load_bytes: + return &bpf_skb_load_bytes_proto; + case BPF_FUNC_skb_pull_data: + return &bpf_skb_pull_data_proto; + case BPF_FUNC_skb_change_tail: + return &bpf_skb_change_tail_proto; + case BPF_FUNC_skb_change_head: + return &bpf_skb_change_head_proto; + case BPF_FUNC_get_socket_cookie: + return &bpf_get_socket_cookie_proto; + case BPF_FUNC_get_socket_uid: + return &bpf_get_socket_uid_proto; + case BPF_FUNC_sk_redirect_map: + return &bpf_sk_redirect_map_proto; default: return bpf_base_func_proto(func_id); } @@ -3106,6 +3376,10 @@ static bool bpf_skb_is_valid_access(int off, int size, enum bpf_access_type type if (off + size > offsetofend(struct __sk_buff, cb[4])) return false; break; + case bpf_ctx_range_till(struct __sk_buff, remote_ip6[0], remote_ip6[3]): + case bpf_ctx_range_till(struct __sk_buff, local_ip6[0], local_ip6[3]): + case bpf_ctx_range_till(struct __sk_buff, remote_ip4, remote_ip4): + case bpf_ctx_range_till(struct __sk_buff, local_ip4, local_ip4): case bpf_ctx_range(struct __sk_buff, data): case bpf_ctx_range(struct __sk_buff, data_end): if (size != size_default) @@ -3134,6 +3408,7 @@ static bool sk_filter_is_valid_access(int off, int size, case bpf_ctx_range(struct __sk_buff, tc_classid): case bpf_ctx_range(struct __sk_buff, data): case bpf_ctx_range(struct __sk_buff, data_end): + case bpf_ctx_range_till(struct __sk_buff, family, local_port): return false; } @@ -3155,6 +3430,7 @@ static bool lwt_is_valid_access(int off, int size, { switch (off) { case bpf_ctx_range(struct __sk_buff, tc_classid): + case bpf_ctx_range_till(struct __sk_buff, family, local_port): return false; } @@ -3188,6 +3464,8 @@ static bool sock_filter_is_valid_access(int off, int size, if (type == BPF_WRITE) { switch (off) { case offsetof(struct bpf_sock, bound_dev_if): + case offsetof(struct bpf_sock, mark): + case offsetof(struct bpf_sock, priority): break; default: return false; @@ -3205,8 +3483,8 @@ static bool sock_filter_is_valid_access(int off, int size, return true; } -static int tc_cls_act_prologue(struct bpf_insn *insn_buf, bool direct_write, - const struct bpf_prog *prog) +static int bpf_unclone_prologue(struct bpf_insn *insn_buf, bool direct_write, + const struct bpf_prog *prog, int drop_verdict) { struct bpf_insn *insn = insn_buf; @@ -3233,7 +3511,7 @@ static int tc_cls_act_prologue(struct bpf_insn *insn_buf, bool direct_write, * return TC_ACT_SHOT; */ *insn++ = BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2); - *insn++ = BPF_ALU32_IMM(BPF_MOV, BPF_REG_0, TC_ACT_SHOT); + *insn++ = BPF_ALU32_IMM(BPF_MOV, BPF_REG_0, drop_verdict); *insn++ = BPF_EXIT_INSN(); /* restore: */ @@ -3244,6 +3522,12 @@ static int tc_cls_act_prologue(struct bpf_insn *insn_buf, bool direct_write, return insn - insn_buf; } +static int tc_cls_act_prologue(struct bpf_insn *insn_buf, bool direct_write, + const struct bpf_prog *prog) +{ + return bpf_unclone_prologue(insn_buf, direct_write, prog, TC_ACT_SHOT); +} + static bool tc_cls_act_is_valid_access(int off, int size, enum bpf_access_type type, struct bpf_insn_access_aux *info) @@ -3268,6 +3552,8 @@ static bool tc_cls_act_is_valid_access(int off, int size, case bpf_ctx_range(struct __sk_buff, data_end): info->reg_type = PTR_TO_PACKET_END; break; + case bpf_ctx_range_till(struct __sk_buff, family, local_port): + return false; } return bpf_skb_is_valid_access(off, size, type, info); @@ -3340,6 +3626,41 @@ static bool sock_ops_is_valid_access(int off, int size, return __is_valid_sock_ops_access(off, size); } +static int sk_skb_prologue(struct bpf_insn *insn_buf, bool direct_write, + const struct bpf_prog *prog) +{ + return bpf_unclone_prologue(insn_buf, direct_write, prog, SK_DROP); +} + +static bool sk_skb_is_valid_access(int off, int size, + enum bpf_access_type type, + struct bpf_insn_access_aux *info) +{ + if (type == BPF_WRITE) { + switch (off) { + case bpf_ctx_range(struct __sk_buff, mark): + case bpf_ctx_range(struct __sk_buff, tc_index): + case bpf_ctx_range(struct __sk_buff, priority): + break; + default: + return false; + } + } + + switch (off) { + case bpf_ctx_range(struct __sk_buff, tc_classid): + return false; + case bpf_ctx_range(struct __sk_buff, data): + info->reg_type = PTR_TO_PACKET; + break; + case bpf_ctx_range(struct __sk_buff, data_end): + info->reg_type = PTR_TO_PACKET_END; + break; + } + + return bpf_skb_is_valid_access(off, size, type, info); +} + static u32 bpf_convert_ctx_access(enum bpf_access_type type, const struct bpf_insn *si, struct bpf_insn *insn_buf, @@ -3505,6 +3826,7 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type, bpf_target_off(struct sk_buff, tc_index, 2, target_size)); #else + *target_size = 2; if (type == BPF_WRITE) *insn++ = BPF_MOV64_REG(si->dst_reg, si->dst_reg); else @@ -3520,9 +3842,110 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type, *insn++ = BPF_JMP_IMM(BPF_JGE, si->dst_reg, MIN_NAPI_ID, 1); *insn++ = BPF_MOV64_IMM(si->dst_reg, 0); #else + *target_size = 4; *insn++ = BPF_MOV64_IMM(si->dst_reg, 0); #endif break; + case offsetof(struct __sk_buff, family): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, skc_family) != 2); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->dst_reg, + bpf_target_off(struct sock_common, + skc_family, + 2, target_size)); + break; + case offsetof(struct __sk_buff, remote_ip4): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, skc_daddr) != 4); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->dst_reg, + bpf_target_off(struct sock_common, + skc_daddr, + 4, target_size)); + break; + case offsetof(struct __sk_buff, local_ip4): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, + skc_rcv_saddr) != 4); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->dst_reg, + bpf_target_off(struct sock_common, + skc_rcv_saddr, + 4, target_size)); + break; + case offsetof(struct __sk_buff, remote_ip6[0]) ... + offsetof(struct __sk_buff, remote_ip6[3]): +#if IS_ENABLED(CONFIG_IPV6) + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, + skc_v6_daddr.s6_addr32[0]) != 4); + + off = si->off; + off -= offsetof(struct __sk_buff, remote_ip6[0]); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->dst_reg, + offsetof(struct sock_common, + skc_v6_daddr.s6_addr32[0]) + + off); +#else + *insn++ = BPF_MOV32_IMM(si->dst_reg, 0); +#endif + break; + case offsetof(struct __sk_buff, local_ip6[0]) ... + offsetof(struct __sk_buff, local_ip6[3]): +#if IS_ENABLED(CONFIG_IPV6) + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, + skc_v6_rcv_saddr.s6_addr32[0]) != 4); + + off = si->off; + off -= offsetof(struct __sk_buff, local_ip6[0]); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->dst_reg, + offsetof(struct sock_common, + skc_v6_rcv_saddr.s6_addr32[0]) + + off); +#else + *insn++ = BPF_MOV32_IMM(si->dst_reg, 0); +#endif + break; + + case offsetof(struct __sk_buff, remote_port): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, skc_dport) != 2); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->dst_reg, + bpf_target_off(struct sock_common, + skc_dport, + 2, target_size)); +#ifndef __BIG_ENDIAN_BITFIELD + *insn++ = BPF_ALU32_IMM(BPF_LSH, si->dst_reg, 16); +#endif + break; + + case offsetof(struct __sk_buff, local_port): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock_common, skc_num) != 2); + + *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct sk_buff, sk), + si->dst_reg, si->src_reg, + offsetof(struct sk_buff, sk)); + *insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->dst_reg, + bpf_target_off(struct sock_common, + skc_num, 2, target_size)); + break; } return insn - insn_buf; @@ -3547,6 +3970,28 @@ static u32 sock_filter_convert_ctx_access(enum bpf_access_type type, offsetof(struct sock, sk_bound_dev_if)); break; + case offsetof(struct bpf_sock, mark): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock, sk_mark) != 4); + + if (type == BPF_WRITE) + *insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg, + offsetof(struct sock, sk_mark)); + else + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg, + offsetof(struct sock, sk_mark)); + break; + + case offsetof(struct bpf_sock, priority): + BUILD_BUG_ON(FIELD_SIZEOF(struct sock, sk_priority) != 4); + + if (type == BPF_WRITE) + *insn++ = BPF_STX_MEM(BPF_W, si->dst_reg, si->src_reg, + offsetof(struct sock, sk_priority)); + else + *insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->src_reg, + offsetof(struct sock, sk_priority)); + break; + case offsetof(struct bpf_sock, family): BUILD_BUG_ON(FIELD_SIZEOF(struct sock, sk_family) != 2); @@ -3796,7 +4241,7 @@ const struct bpf_verifier_ops lwt_xmit_prog_ops = { }; const struct bpf_verifier_ops cg_sock_prog_ops = { - .get_func_proto = bpf_base_func_proto, + .get_func_proto = sock_filter_func_proto, .is_valid_access = sock_filter_is_valid_access, .convert_ctx_access = sock_filter_convert_ctx_access, }; @@ -3807,6 +4252,13 @@ const struct bpf_verifier_ops sock_ops_prog_ops = { .convert_ctx_access = sock_ops_convert_ctx_access, }; +const struct bpf_verifier_ops sk_skb_prog_ops = { + .get_func_proto = sk_skb_func_proto, + .is_valid_access = sk_skb_is_valid_access, + .convert_ctx_access = bpf_convert_ctx_access, + .gen_prologue = sk_skb_prologue, +}; + int sk_detach_filter(struct sock *sk) { int ret = -ENOENT; diff --git a/net/core/flow.c b/net/core/flow.c deleted file mode 100644 index f7f5d1932a27..000000000000 --- a/net/core/flow.c +++ /dev/null @@ -1,516 +0,0 @@ -/* flow.c: Generic flow cache. - * - * Copyright (C) 2003 Alexey N. Kuznetsov (kuznet@ms2.inr.ac.ru) - * Copyright (C) 2003 David S. Miller (davem@redhat.com) - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/list.h> -#include <linux/jhash.h> -#include <linux/interrupt.h> -#include <linux/mm.h> -#include <linux/random.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/smp.h> -#include <linux/completion.h> -#include <linux/percpu.h> -#include <linux/bitops.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/cpumask.h> -#include <linux/mutex.h> -#include <net/flow.h> -#include <linux/atomic.h> -#include <linux/security.h> -#include <net/net_namespace.h> - -struct flow_cache_entry { - union { - struct hlist_node hlist; - struct list_head gc_list; - } u; - struct net *net; - u16 family; - u8 dir; - u32 genid; - struct flowi key; - struct flow_cache_object *object; -}; - -struct flow_flush_info { - struct flow_cache *cache; - atomic_t cpuleft; - struct completion completion; -}; - -static struct kmem_cache *flow_cachep __read_mostly; - -#define flow_cache_hash_size(cache) (1U << (cache)->hash_shift) -#define FLOW_HASH_RND_PERIOD (10 * 60 * HZ) - -static void flow_cache_new_hashrnd(unsigned long arg) -{ - struct flow_cache *fc = (void *) arg; - int i; - - for_each_possible_cpu(i) - per_cpu_ptr(fc->percpu, i)->hash_rnd_recalc = 1; - - fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; - add_timer(&fc->rnd_timer); -} - -static int flow_entry_valid(struct flow_cache_entry *fle, - struct netns_xfrm *xfrm) -{ - if (atomic_read(&xfrm->flow_cache_genid) != fle->genid) - return 0; - if (fle->object && !fle->object->ops->check(fle->object)) - return 0; - return 1; -} - -static void flow_entry_kill(struct flow_cache_entry *fle, - struct netns_xfrm *xfrm) -{ - if (fle->object) - fle->object->ops->delete(fle->object); - kmem_cache_free(flow_cachep, fle); -} - -static void flow_cache_gc_task(struct work_struct *work) -{ - struct list_head gc_list; - struct flow_cache_entry *fce, *n; - struct netns_xfrm *xfrm = container_of(work, struct netns_xfrm, - flow_cache_gc_work); - - INIT_LIST_HEAD(&gc_list); - spin_lock_bh(&xfrm->flow_cache_gc_lock); - list_splice_tail_init(&xfrm->flow_cache_gc_list, &gc_list); - spin_unlock_bh(&xfrm->flow_cache_gc_lock); - - list_for_each_entry_safe(fce, n, &gc_list, u.gc_list) { - flow_entry_kill(fce, xfrm); - atomic_dec(&xfrm->flow_cache_gc_count); - } -} - -static void flow_cache_queue_garbage(struct flow_cache_percpu *fcp, - unsigned int deleted, - struct list_head *gc_list, - struct netns_xfrm *xfrm) -{ - if (deleted) { - atomic_add(deleted, &xfrm->flow_cache_gc_count); - fcp->hash_count -= deleted; - spin_lock_bh(&xfrm->flow_cache_gc_lock); - list_splice_tail(gc_list, &xfrm->flow_cache_gc_list); - spin_unlock_bh(&xfrm->flow_cache_gc_lock); - schedule_work(&xfrm->flow_cache_gc_work); - } -} - -static void __flow_cache_shrink(struct flow_cache *fc, - struct flow_cache_percpu *fcp, - unsigned int shrink_to) -{ - struct flow_cache_entry *fle; - struct hlist_node *tmp; - LIST_HEAD(gc_list); - unsigned int deleted = 0; - struct netns_xfrm *xfrm = container_of(fc, struct netns_xfrm, - flow_cache_global); - unsigned int i; - - for (i = 0; i < flow_cache_hash_size(fc); i++) { - unsigned int saved = 0; - - hlist_for_each_entry_safe(fle, tmp, - &fcp->hash_table[i], u.hlist) { - if (saved < shrink_to && - flow_entry_valid(fle, xfrm)) { - saved++; - } else { - deleted++; - hlist_del(&fle->u.hlist); - list_add_tail(&fle->u.gc_list, &gc_list); - } - } - } - - flow_cache_queue_garbage(fcp, deleted, &gc_list, xfrm); -} - -static void flow_cache_shrink(struct flow_cache *fc, - struct flow_cache_percpu *fcp) -{ - unsigned int shrink_to = fc->low_watermark / flow_cache_hash_size(fc); - - __flow_cache_shrink(fc, fcp, shrink_to); -} - -static void flow_new_hash_rnd(struct flow_cache *fc, - struct flow_cache_percpu *fcp) -{ - get_random_bytes(&fcp->hash_rnd, sizeof(u32)); - fcp->hash_rnd_recalc = 0; - __flow_cache_shrink(fc, fcp, 0); -} - -static u32 flow_hash_code(struct flow_cache *fc, - struct flow_cache_percpu *fcp, - const struct flowi *key, - unsigned int keysize) -{ - const u32 *k = (const u32 *) key; - const u32 length = keysize * sizeof(flow_compare_t) / sizeof(u32); - - return jhash2(k, length, fcp->hash_rnd) - & (flow_cache_hash_size(fc) - 1); -} - -/* I hear what you're saying, use memcmp. But memcmp cannot make - * important assumptions that we can here, such as alignment. - */ -static int flow_key_compare(const struct flowi *key1, const struct flowi *key2, - unsigned int keysize) -{ - const flow_compare_t *k1, *k1_lim, *k2; - - k1 = (const flow_compare_t *) key1; - k1_lim = k1 + keysize; - - k2 = (const flow_compare_t *) key2; - - do { - if (*k1++ != *k2++) - return 1; - } while (k1 < k1_lim); - - return 0; -} - -struct flow_cache_object * -flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, - flow_resolve_t resolver, void *ctx) -{ - struct flow_cache *fc = &net->xfrm.flow_cache_global; - struct flow_cache_percpu *fcp; - struct flow_cache_entry *fle, *tfle; - struct flow_cache_object *flo; - unsigned int keysize; - unsigned int hash; - - local_bh_disable(); - fcp = this_cpu_ptr(fc->percpu); - - fle = NULL; - flo = NULL; - - keysize = flow_key_size(family); - if (!keysize) - goto nocache; - - /* Packet really early in init? Making flow_cache_init a - * pre-smp initcall would solve this. --RR */ - if (!fcp->hash_table) - goto nocache; - - if (fcp->hash_rnd_recalc) - flow_new_hash_rnd(fc, fcp); - - hash = flow_hash_code(fc, fcp, key, keysize); - hlist_for_each_entry(tfle, &fcp->hash_table[hash], u.hlist) { - if (tfle->net == net && - tfle->family == family && - tfle->dir == dir && - flow_key_compare(key, &tfle->key, keysize) == 0) { - fle = tfle; - break; - } - } - - if (unlikely(!fle)) { - if (fcp->hash_count > fc->high_watermark) - flow_cache_shrink(fc, fcp); - - if (atomic_read(&net->xfrm.flow_cache_gc_count) > - 2 * num_online_cpus() * fc->high_watermark) { - flo = ERR_PTR(-ENOBUFS); - goto ret_object; - } - - fle = kmem_cache_alloc(flow_cachep, GFP_ATOMIC); - if (fle) { - fle->net = net; - fle->family = family; - fle->dir = dir; - memcpy(&fle->key, key, keysize * sizeof(flow_compare_t)); - fle->object = NULL; - hlist_add_head(&fle->u.hlist, &fcp->hash_table[hash]); - fcp->hash_count++; - } - } else if (likely(fle->genid == atomic_read(&net->xfrm.flow_cache_genid))) { - flo = fle->object; - if (!flo) - goto ret_object; - flo = flo->ops->get(flo); - if (flo) - goto ret_object; - } else if (fle->object) { - flo = fle->object; - flo->ops->delete(flo); - fle->object = NULL; - } - -nocache: - flo = NULL; - if (fle) { - flo = fle->object; - fle->object = NULL; - } - flo = resolver(net, key, family, dir, flo, ctx); - if (fle) { - fle->genid = atomic_read(&net->xfrm.flow_cache_genid); - if (!IS_ERR(flo)) - fle->object = flo; - else - fle->genid--; - } else { - if (!IS_ERR_OR_NULL(flo)) - flo->ops->delete(flo); - } -ret_object: - local_bh_enable(); - return flo; -} -EXPORT_SYMBOL(flow_cache_lookup); - -static void flow_cache_flush_tasklet(unsigned long data) -{ - struct flow_flush_info *info = (void *)data; - struct flow_cache *fc = info->cache; - struct flow_cache_percpu *fcp; - struct flow_cache_entry *fle; - struct hlist_node *tmp; - LIST_HEAD(gc_list); - unsigned int deleted = 0; - struct netns_xfrm *xfrm = container_of(fc, struct netns_xfrm, - flow_cache_global); - unsigned int i; - - fcp = this_cpu_ptr(fc->percpu); - for (i = 0; i < flow_cache_hash_size(fc); i++) { - hlist_for_each_entry_safe(fle, tmp, - &fcp->hash_table[i], u.hlist) { - if (flow_entry_valid(fle, xfrm)) - continue; - - deleted++; - hlist_del(&fle->u.hlist); - list_add_tail(&fle->u.gc_list, &gc_list); - } - } - - flow_cache_queue_garbage(fcp, deleted, &gc_list, xfrm); - - if (atomic_dec_and_test(&info->cpuleft)) - complete(&info->completion); -} - -/* - * Return whether a cpu needs flushing. Conservatively, we assume - * the presence of any entries means the core may require flushing, - * since the flow_cache_ops.check() function may assume it's running - * on the same core as the per-cpu cache component. - */ -static int flow_cache_percpu_empty(struct flow_cache *fc, int cpu) -{ - struct flow_cache_percpu *fcp; - unsigned int i; - - fcp = per_cpu_ptr(fc->percpu, cpu); - for (i = 0; i < flow_cache_hash_size(fc); i++) - if (!hlist_empty(&fcp->hash_table[i])) - return 0; - return 1; -} - -static void flow_cache_flush_per_cpu(void *data) -{ - struct flow_flush_info *info = data; - struct tasklet_struct *tasklet; - - tasklet = &this_cpu_ptr(info->cache->percpu)->flush_tasklet; - tasklet->data = (unsigned long)info; - tasklet_schedule(tasklet); -} - -void flow_cache_flush(struct net *net) -{ - struct flow_flush_info info; - cpumask_var_t mask; - int i, self; - - /* Track which cpus need flushing to avoid disturbing all cores. */ - if (!alloc_cpumask_var(&mask, GFP_KERNEL)) - return; - cpumask_clear(mask); - - /* Don't want cpus going down or up during this. */ - get_online_cpus(); - mutex_lock(&net->xfrm.flow_flush_sem); - info.cache = &net->xfrm.flow_cache_global; - for_each_online_cpu(i) - if (!flow_cache_percpu_empty(info.cache, i)) - cpumask_set_cpu(i, mask); - atomic_set(&info.cpuleft, cpumask_weight(mask)); - if (atomic_read(&info.cpuleft) == 0) - goto done; - - init_completion(&info.completion); - - local_bh_disable(); - self = cpumask_test_and_clear_cpu(smp_processor_id(), mask); - on_each_cpu_mask(mask, flow_cache_flush_per_cpu, &info, 0); - if (self) - flow_cache_flush_tasklet((unsigned long)&info); - local_bh_enable(); - - wait_for_completion(&info.completion); - -done: - mutex_unlock(&net->xfrm.flow_flush_sem); - put_online_cpus(); - free_cpumask_var(mask); -} - -static void flow_cache_flush_task(struct work_struct *work) -{ - struct netns_xfrm *xfrm = container_of(work, struct netns_xfrm, - flow_cache_flush_work); - struct net *net = container_of(xfrm, struct net, xfrm); - - flow_cache_flush(net); -} - -void flow_cache_flush_deferred(struct net *net) -{ - schedule_work(&net->xfrm.flow_cache_flush_work); -} - -static int flow_cache_cpu_prepare(struct flow_cache *fc, int cpu) -{ - struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); - unsigned int sz = sizeof(struct hlist_head) * flow_cache_hash_size(fc); - - if (!fcp->hash_table) { - fcp->hash_table = kzalloc_node(sz, GFP_KERNEL, cpu_to_node(cpu)); - if (!fcp->hash_table) { - pr_err("NET: failed to allocate flow cache sz %u\n", sz); - return -ENOMEM; - } - fcp->hash_rnd_recalc = 1; - fcp->hash_count = 0; - tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0); - } - return 0; -} - -static int flow_cache_cpu_up_prep(unsigned int cpu, struct hlist_node *node) -{ - struct flow_cache *fc = hlist_entry_safe(node, struct flow_cache, node); - - return flow_cache_cpu_prepare(fc, cpu); -} - -static int flow_cache_cpu_dead(unsigned int cpu, struct hlist_node *node) -{ - struct flow_cache *fc = hlist_entry_safe(node, struct flow_cache, node); - struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu); - - __flow_cache_shrink(fc, fcp, 0); - return 0; -} - -int flow_cache_init(struct net *net) -{ - int i; - struct flow_cache *fc = &net->xfrm.flow_cache_global; - - if (!flow_cachep) - flow_cachep = kmem_cache_create("flow_cache", - sizeof(struct flow_cache_entry), - 0, SLAB_PANIC, NULL); - spin_lock_init(&net->xfrm.flow_cache_gc_lock); - INIT_LIST_HEAD(&net->xfrm.flow_cache_gc_list); - INIT_WORK(&net->xfrm.flow_cache_gc_work, flow_cache_gc_task); - INIT_WORK(&net->xfrm.flow_cache_flush_work, flow_cache_flush_task); - mutex_init(&net->xfrm.flow_flush_sem); - atomic_set(&net->xfrm.flow_cache_gc_count, 0); - - fc->hash_shift = 10; - fc->low_watermark = 2 * flow_cache_hash_size(fc); - fc->high_watermark = 4 * flow_cache_hash_size(fc); - - fc->percpu = alloc_percpu(struct flow_cache_percpu); - if (!fc->percpu) - return -ENOMEM; - - if (cpuhp_state_add_instance(CPUHP_NET_FLOW_PREPARE, &fc->node)) - goto err; - - setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd, - (unsigned long) fc); - fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD; - add_timer(&fc->rnd_timer); - - return 0; - -err: - for_each_possible_cpu(i) { - struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, i); - kfree(fcp->hash_table); - fcp->hash_table = NULL; - } - - free_percpu(fc->percpu); - fc->percpu = NULL; - - return -ENOMEM; -} -EXPORT_SYMBOL(flow_cache_init); - -void flow_cache_fini(struct net *net) -{ - int i; - struct flow_cache *fc = &net->xfrm.flow_cache_global; - - del_timer_sync(&fc->rnd_timer); - - cpuhp_state_remove_instance_nocalls(CPUHP_NET_FLOW_PREPARE, &fc->node); - - for_each_possible_cpu(i) { - struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, i); - kfree(fcp->hash_table); - fcp->hash_table = NULL; - } - - free_percpu(fc->percpu); - fc->percpu = NULL; -} -EXPORT_SYMBOL(flow_cache_fini); - -void __init flow_cache_hp_init(void) -{ - int ret; - - ret = cpuhp_setup_state_multi(CPUHP_NET_FLOW_PREPARE, - "net/flow:prepare", - flow_cache_cpu_up_prep, - flow_cache_cpu_dead); - WARN_ON(ret < 0); -} diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index fc5fc4594c90..0a977373d003 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -4,6 +4,7 @@ #include <linux/ip.h> #include <linux/ipv6.h> #include <linux/if_vlan.h> +#include <net/dsa.h> #include <net/ip.h> #include <net/ipv6.h> #include <net/gre.h> @@ -114,12 +115,6 @@ __be32 __skb_flow_get_ports(const struct sk_buff *skb, int thoff, u8 ip_proto, } EXPORT_SYMBOL(__skb_flow_get_ports); -enum flow_dissect_ret { - FLOW_DISSECT_RET_OUT_GOOD, - FLOW_DISSECT_RET_OUT_BAD, - FLOW_DISSECT_RET_OUT_PROTO_AGAIN, -}; - static enum flow_dissect_ret __skb_flow_dissect_mpls(const struct sk_buff *skb, struct flow_dissector *flow_dissector, @@ -340,7 +335,7 @@ __skb_flow_dissect_gre(const struct sk_buff *skb, if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) return FLOW_DISSECT_RET_OUT_GOOD; - return FLOW_DISSECT_RET_OUT_PROTO_AGAIN; + return FLOW_DISSECT_RET_PROTO_AGAIN; } static void @@ -401,6 +396,18 @@ __skb_flow_dissect_ipv6(const struct sk_buff *skb, key_ip->ttl = iph->hop_limit; } +/* Maximum number of protocol headers that can be parsed in + * __skb_flow_dissect + */ +#define MAX_FLOW_DISSECT_HDRS 15 + +static bool skb_flow_dissect_allowed(int *num_hdrs) +{ + ++*num_hdrs; + + return (*num_hdrs <= MAX_FLOW_DISSECT_HDRS); +} + /** * __skb_flow_dissect - extract the flow_keys struct and return it * @skb: sk_buff to extract the flow from, can be NULL if the rest are specified @@ -430,7 +437,9 @@ bool __skb_flow_dissect(const struct sk_buff *skb, struct flow_dissector_key_icmp *key_icmp; struct flow_dissector_key_tags *key_tags; struct flow_dissector_key_vlan *key_vlan; + enum flow_dissect_ret fdret; bool skip_vlan = false; + int num_hdrs = 0; u8 ip_proto = 0; bool ret; @@ -440,6 +449,19 @@ bool __skb_flow_dissect(const struct sk_buff *skb, skb->vlan_proto : skb->protocol; nhoff = skb_network_offset(skb); hlen = skb_headlen(skb); +#if IS_ENABLED(CONFIG_NET_DSA) + if (unlikely(skb->dev && netdev_uses_dsa(skb->dev))) { + const struct dsa_device_ops *ops; + int offset; + + ops = skb->dev->dsa_ptr->tag_ops; + if (ops->flow_dissect && + !ops->flow_dissect(skb, &proto, &offset)) { + hlen -= offset; + nhoff += offset; + } + } +#endif } /* It is ensured by skb_flow_dissector_init() that control key will @@ -468,14 +490,19 @@ bool __skb_flow_dissect(const struct sk_buff *skb, } proto_again: + fdret = FLOW_DISSECT_RET_CONTINUE; + switch (proto) { case htons(ETH_P_IP): { const struct iphdr *iph; struct iphdr _iph; -ip: + iph = __skb_header_pointer(skb, nhoff, sizeof(_iph), data, hlen, &_iph); - if (!iph || iph->ihl < 5) - goto out_bad; + if (!iph || iph->ihl < 5) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } + nhoff += iph->ihl * 4; ip_proto = iph->protocol; @@ -495,19 +522,25 @@ ip: key_control->flags |= FLOW_DIS_IS_FRAGMENT; if (iph->frag_off & htons(IP_OFFSET)) { - goto out_good; + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; } else { key_control->flags |= FLOW_DIS_FIRST_FRAG; - if (!(flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG)) - goto out_good; + if (!(flags & + FLOW_DISSECTOR_F_PARSE_1ST_FRAG)) { + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; + } } } __skb_flow_dissect_ipv4(skb, flow_dissector, target_container, data, iph); - if (flags & FLOW_DISSECTOR_F_STOP_AT_L3) - goto out_good; + if (flags & FLOW_DISSECTOR_F_STOP_AT_L3) { + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; + } break; } @@ -515,10 +548,11 @@ ip: const struct ipv6hdr *iph; struct ipv6hdr _iph; -ipv6: iph = __skb_header_pointer(skb, nhoff, sizeof(_iph), data, hlen, &_iph); - if (!iph) - goto out_bad; + if (!iph) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } ip_proto = iph->nexthdr; nhoff += sizeof(struct ipv6hdr); @@ -547,15 +581,17 @@ ipv6: target_container); key_tags->flow_label = ntohl(flow_label); } - if (flags & FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL) - goto out_good; + if (flags & FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL) { + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; + } } __skb_flow_dissect_ipv6(skb, flow_dissector, target_container, data, iph); if (flags & FLOW_DISSECTOR_F_STOP_AT_L3) - goto out_good; + fdret = FLOW_DISSECT_RET_OUT_GOOD; break; } @@ -571,12 +607,17 @@ ipv6: if (!vlan_tag_present || eth_type_vlan(skb->protocol)) { vlan = __skb_header_pointer(skb, nhoff, sizeof(_vlan), data, hlen, &_vlan); - if (!vlan) - goto out_bad; + if (!vlan) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } + proto = vlan->h_vlan_encapsulated_proto; nhoff += sizeof(*vlan); - if (skip_vlan) - goto proto_again; + if (skip_vlan) { + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; + } } skip_vlan = true; @@ -599,7 +640,8 @@ ipv6: } } - goto proto_again; + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; } case htons(ETH_P_PPP_SES): { struct { @@ -607,18 +649,27 @@ ipv6: __be16 proto; } *hdr, _hdr; hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr); - if (!hdr) - goto out_bad; + if (!hdr) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } + proto = hdr->proto; nhoff += PPPOE_SES_HLEN; switch (proto) { case htons(PPP_IP): - goto ip; + proto = htons(ETH_P_IP); + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; case htons(PPP_IPV6): - goto ipv6; + proto = htons(ETH_P_IPV6); + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; default: - goto out_bad; + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; } + break; } case htons(ETH_P_TIPC): { struct { @@ -626,8 +677,10 @@ ipv6: __be32 srcnode; } *hdr, _hdr; hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr); - if (!hdr) - goto out_bad; + if (!hdr) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_TIPC_ADDRS)) { @@ -637,56 +690,64 @@ ipv6: key_addrs->tipcaddrs.srcnode = hdr->srcnode; key_control->addr_type = FLOW_DISSECTOR_KEY_TIPC_ADDRS; } - goto out_good; + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; } case htons(ETH_P_MPLS_UC): case htons(ETH_P_MPLS_MC): -mpls: - switch (__skb_flow_dissect_mpls(skb, flow_dissector, + fdret = __skb_flow_dissect_mpls(skb, flow_dissector, target_container, data, - nhoff, hlen)) { - case FLOW_DISSECT_RET_OUT_GOOD: - goto out_good; - case FLOW_DISSECT_RET_OUT_BAD: - default: - goto out_bad; - } + nhoff, hlen); + break; case htons(ETH_P_FCOE): - if ((hlen - nhoff) < FCOE_HEADER_LEN) - goto out_bad; + if ((hlen - nhoff) < FCOE_HEADER_LEN) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } nhoff += FCOE_HEADER_LEN; - goto out_good; + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; case htons(ETH_P_ARP): case htons(ETH_P_RARP): - switch (__skb_flow_dissect_arp(skb, flow_dissector, + fdret = __skb_flow_dissect_arp(skb, flow_dissector, target_container, data, - nhoff, hlen)) { - case FLOW_DISSECT_RET_OUT_GOOD: - goto out_good; - case FLOW_DISSECT_RET_OUT_BAD: - default: - goto out_bad; - } + nhoff, hlen); + break; + + default: + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } + + /* Process result of proto processing */ + switch (fdret) { + case FLOW_DISSECT_RET_OUT_GOOD: + goto out_good; + case FLOW_DISSECT_RET_PROTO_AGAIN: + if (skb_flow_dissect_allowed(&num_hdrs)) + goto proto_again; + goto out_good; + case FLOW_DISSECT_RET_CONTINUE: + case FLOW_DISSECT_RET_IPPROTO_AGAIN: + break; + case FLOW_DISSECT_RET_OUT_BAD: default: goto out_bad; } ip_proto_again: + fdret = FLOW_DISSECT_RET_CONTINUE; + switch (ip_proto) { case IPPROTO_GRE: - switch (__skb_flow_dissect_gre(skb, key_control, flow_dissector, + fdret = __skb_flow_dissect_gre(skb, key_control, flow_dissector, target_container, data, - &proto, &nhoff, &hlen, flags)) { - case FLOW_DISSECT_RET_OUT_GOOD: - goto out_good; - case FLOW_DISSECT_RET_OUT_BAD: - goto out_bad; - case FLOW_DISSECT_RET_OUT_PROTO_AGAIN: - goto proto_again; - } + &proto, &nhoff, &hlen, flags); + break; + case NEXTHDR_HOP: case NEXTHDR_ROUTING: case NEXTHDR_DEST: { @@ -697,13 +758,16 @@ ip_proto_again: opthdr = __skb_header_pointer(skb, nhoff, sizeof(_opthdr), data, hlen, &_opthdr); - if (!opthdr) - goto out_bad; + if (!opthdr) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } ip_proto = opthdr[0]; nhoff += (opthdr[1] + 1) << 3; - goto ip_proto_again; + fdret = FLOW_DISSECT_RET_IPPROTO_AGAIN; + break; } case NEXTHDR_FRAGMENT: { struct frag_hdr _fh, *fh; @@ -714,8 +778,10 @@ ip_proto_again: fh = __skb_header_pointer(skb, nhoff, sizeof(_fh), data, hlen, &_fh); - if (!fh) - goto out_bad; + if (!fh) { + fdret = FLOW_DISSECT_RET_OUT_BAD; + break; + } key_control->flags |= FLOW_DIS_IS_FRAGMENT; @@ -724,34 +790,50 @@ ip_proto_again: if (!(fh->frag_off & htons(IP6_OFFSET))) { key_control->flags |= FLOW_DIS_FIRST_FRAG; - if (flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG) - goto ip_proto_again; + if (flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG) { + fdret = FLOW_DISSECT_RET_IPPROTO_AGAIN; + break; + } } - goto out_good; + + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; } case IPPROTO_IPIP: proto = htons(ETH_P_IP); key_control->flags |= FLOW_DIS_ENCAPSULATION; - if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) - goto out_good; + if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) { + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; + } + + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; - goto ip; case IPPROTO_IPV6: proto = htons(ETH_P_IPV6); key_control->flags |= FLOW_DIS_ENCAPSULATION; - if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) - goto out_good; + if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) { + fdret = FLOW_DISSECT_RET_OUT_GOOD; + break; + } + + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; + - goto ipv6; case IPPROTO_MPLS: proto = htons(ETH_P_MPLS_UC); - goto mpls; + fdret = FLOW_DISSECT_RET_PROTO_AGAIN; + break; + case IPPROTO_TCP: __skb_flow_dissect_tcp(skb, flow_dissector, target_container, data, nhoff, hlen); break; + default: break; } @@ -773,6 +855,24 @@ ip_proto_again: key_icmp->icmp = skb_flow_get_be16(skb, nhoff, data, hlen); } + /* Process result of IP proto processing */ + switch (fdret) { + case FLOW_DISSECT_RET_PROTO_AGAIN: + if (skb_flow_dissect_allowed(&num_hdrs)) + goto proto_again; + break; + case FLOW_DISSECT_RET_IPPROTO_AGAIN: + if (skb_flow_dissect_allowed(&num_hdrs)) + goto ip_proto_again; + break; + case FLOW_DISSECT_RET_OUT_GOOD: + case FLOW_DISSECT_RET_CONTINUE: + break; + case FLOW_DISSECT_RET_OUT_BAD: + default: + goto out_bad; + } + out_good: ret = true; @@ -998,51 +1098,6 @@ __u32 skb_get_hash_perturb(const struct sk_buff *skb, u32 perturb) } EXPORT_SYMBOL(skb_get_hash_perturb); -__u32 __skb_get_hash_flowi6(struct sk_buff *skb, const struct flowi6 *fl6) -{ - struct flow_keys keys; - - memset(&keys, 0, sizeof(keys)); - - memcpy(&keys.addrs.v6addrs.src, &fl6->saddr, - sizeof(keys.addrs.v6addrs.src)); - memcpy(&keys.addrs.v6addrs.dst, &fl6->daddr, - sizeof(keys.addrs.v6addrs.dst)); - keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; - keys.ports.src = fl6->fl6_sport; - keys.ports.dst = fl6->fl6_dport; - keys.keyid.keyid = fl6->fl6_gre_key; - keys.tags.flow_label = (__force u32)fl6->flowlabel; - keys.basic.ip_proto = fl6->flowi6_proto; - - __skb_set_sw_hash(skb, flow_hash_from_keys(&keys), - flow_keys_have_l4(&keys)); - - return skb->hash; -} -EXPORT_SYMBOL(__skb_get_hash_flowi6); - -__u32 __skb_get_hash_flowi4(struct sk_buff *skb, const struct flowi4 *fl4) -{ - struct flow_keys keys; - - memset(&keys, 0, sizeof(keys)); - - keys.addrs.v4addrs.src = fl4->saddr; - keys.addrs.v4addrs.dst = fl4->daddr; - keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; - keys.ports.src = fl4->fl4_sport; - keys.ports.dst = fl4->fl4_dport; - keys.keyid.keyid = fl4->fl4_gre_key; - keys.basic.ip_proto = fl4->flowi4_proto; - - __skb_set_sw_hash(skb, flow_hash_from_keys(&keys), - flow_keys_have_l4(&keys)); - - return skb->hash; -} -EXPORT_SYMBOL(__skb_get_hash_flowi4); - u32 __skb_get_poff(const struct sk_buff *skb, void *data, const struct flow_keys *keys, int hlen) { diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c index d9cb3532f1dd..0b171756453c 100644 --- a/net/core/lwtunnel.c +++ b/net/core/lwtunnel.c @@ -44,6 +44,8 @@ static const char *lwtunnel_encap_str(enum lwtunnel_encap_types encap_type) return "SEG6"; case LWTUNNEL_ENCAP_BPF: return "BPF"; + case LWTUNNEL_ENCAP_SEG6_LOCAL: + return "SEG6LOCAL"; case LWTUNNEL_ENCAP_IP6: case LWTUNNEL_ENCAP_IP: case LWTUNNEL_ENCAP_NONE: @@ -65,7 +67,7 @@ struct lwtunnel_state *lwtunnel_state_alloc(int encap_len) return lws; } -EXPORT_SYMBOL(lwtunnel_state_alloc); +EXPORT_SYMBOL_GPL(lwtunnel_state_alloc); static const struct lwtunnel_encap_ops __rcu * lwtun_encaps[LWTUNNEL_ENCAP_MAX + 1] __read_mostly; @@ -80,7 +82,7 @@ int lwtunnel_encap_add_ops(const struct lwtunnel_encap_ops *ops, &lwtun_encaps[num], NULL, ops) ? 0 : -1; } -EXPORT_SYMBOL(lwtunnel_encap_add_ops); +EXPORT_SYMBOL_GPL(lwtunnel_encap_add_ops); int lwtunnel_encap_del_ops(const struct lwtunnel_encap_ops *ops, unsigned int encap_type) @@ -99,7 +101,7 @@ int lwtunnel_encap_del_ops(const struct lwtunnel_encap_ops *ops, return ret; } -EXPORT_SYMBOL(lwtunnel_encap_del_ops); +EXPORT_SYMBOL_GPL(lwtunnel_encap_del_ops); int lwtunnel_build_state(u16 encap_type, struct nlattr *encap, unsigned int family, @@ -138,7 +140,7 @@ int lwtunnel_build_state(u16 encap_type, return ret; } -EXPORT_SYMBOL(lwtunnel_build_state); +EXPORT_SYMBOL_GPL(lwtunnel_build_state); int lwtunnel_valid_encap_type(u16 encap_type, struct netlink_ext_ack *extack) { @@ -175,7 +177,7 @@ int lwtunnel_valid_encap_type(u16 encap_type, struct netlink_ext_ack *extack) return ret; } -EXPORT_SYMBOL(lwtunnel_valid_encap_type); +EXPORT_SYMBOL_GPL(lwtunnel_valid_encap_type); int lwtunnel_valid_encap_type_attr(struct nlattr *attr, int remaining, struct netlink_ext_ack *extack) @@ -205,7 +207,7 @@ int lwtunnel_valid_encap_type_attr(struct nlattr *attr, int remaining, return 0; } -EXPORT_SYMBOL(lwtunnel_valid_encap_type_attr); +EXPORT_SYMBOL_GPL(lwtunnel_valid_encap_type_attr); void lwtstate_free(struct lwtunnel_state *lws) { @@ -219,7 +221,7 @@ void lwtstate_free(struct lwtunnel_state *lws) } module_put(ops->owner); } -EXPORT_SYMBOL(lwtstate_free); +EXPORT_SYMBOL_GPL(lwtstate_free); int lwtunnel_fill_encap(struct sk_buff *skb, struct lwtunnel_state *lwtstate) { @@ -259,7 +261,7 @@ nla_put_failure: return (ret == -EOPNOTSUPP ? 0 : ret); } -EXPORT_SYMBOL(lwtunnel_fill_encap); +EXPORT_SYMBOL_GPL(lwtunnel_fill_encap); int lwtunnel_get_encap_size(struct lwtunnel_state *lwtstate) { @@ -281,7 +283,7 @@ int lwtunnel_get_encap_size(struct lwtunnel_state *lwtstate) return ret; } -EXPORT_SYMBOL(lwtunnel_get_encap_size); +EXPORT_SYMBOL_GPL(lwtunnel_get_encap_size); int lwtunnel_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b) { @@ -309,7 +311,7 @@ int lwtunnel_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b) return ret; } -EXPORT_SYMBOL(lwtunnel_cmp_encap); +EXPORT_SYMBOL_GPL(lwtunnel_cmp_encap); int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb) { @@ -343,7 +345,7 @@ drop: return ret; } -EXPORT_SYMBOL(lwtunnel_output); +EXPORT_SYMBOL_GPL(lwtunnel_output); int lwtunnel_xmit(struct sk_buff *skb) { @@ -378,7 +380,7 @@ drop: return ret; } -EXPORT_SYMBOL(lwtunnel_xmit); +EXPORT_SYMBOL_GPL(lwtunnel_xmit); int lwtunnel_input(struct sk_buff *skb) { @@ -412,4 +414,4 @@ drop: return ret; } -EXPORT_SYMBOL(lwtunnel_input); +EXPORT_SYMBOL_GPL(lwtunnel_input); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index d0713627deb6..16a1a4c4eb57 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -3261,13 +3261,13 @@ EXPORT_SYMBOL(neigh_sysctl_unregister); static int __init neigh_init(void) { - rtnl_register(PF_UNSPEC, RTM_NEWNEIGH, neigh_add, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELNEIGH, neigh_delete, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_GETNEIGH, NULL, neigh_dump_info, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWNEIGH, neigh_add, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELNEIGH, neigh_delete, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETNEIGH, NULL, neigh_dump_info, 0); rtnl_register(PF_UNSPEC, RTM_GETNEIGHTBL, NULL, neightbl_dump_info, - NULL); - rtnl_register(PF_UNSPEC, RTM_SETNEIGHTBL, neightbl_set, NULL, NULL); + 0); + rtnl_register(PF_UNSPEC, RTM_SETNEIGHTBL, neightbl_set, NULL, 0); return 0; } diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index b4f9922b6f23..927a6dcbad96 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -97,7 +97,8 @@ static ssize_t netdev_store(struct device *dev, struct device_attribute *attr, return restart_syscall(); if (dev_isalive(netdev)) { - if ((ret = (*set)(netdev, new)) == 0) + ret = (*set)(netdev, new); + if (ret == 0) ret = len; } rtnl_unlock(); @@ -160,6 +161,7 @@ static ssize_t broadcast_show(struct device *dev, struct device_attribute *attr, char *buf) { struct net_device *ndev = to_net_dev(dev); + if (dev_isalive(ndev)) return sysfs_format_mac(buf, ndev->broadcast, ndev->addr_len); return -EINVAL; @@ -170,7 +172,7 @@ static int change_carrier(struct net_device *dev, unsigned long new_carrier) { if (!netif_running(dev)) return -EINVAL; - return dev_change_carrier(dev, (bool) new_carrier); + return dev_change_carrier(dev, (bool)new_carrier); } static ssize_t carrier_store(struct device *dev, struct device_attribute *attr, @@ -183,9 +185,10 @@ static ssize_t carrier_show(struct device *dev, struct device_attribute *attr, char *buf) { struct net_device *netdev = to_net_dev(dev); - if (netif_running(netdev)) { + + if (netif_running(netdev)) return sprintf(buf, fmt_dec, !!netif_carrier_ok(netdev)); - } + return -EINVAL; } static DEVICE_ATTR_RW(carrier); @@ -290,6 +293,7 @@ static ssize_t carrier_changes_show(struct device *dev, char *buf) { struct net_device *netdev = to_net_dev(dev); + return sprintf(buf, fmt_dec, atomic_read(&netdev->carrier_changes)); } @@ -299,7 +303,7 @@ static DEVICE_ATTR_RO(carrier_changes); static int change_mtu(struct net_device *dev, unsigned long new_mtu) { - return dev_set_mtu(dev, (int) new_mtu); + return dev_set_mtu(dev, (int)new_mtu); } static ssize_t mtu_store(struct device *dev, struct device_attribute *attr, @@ -311,7 +315,7 @@ NETDEVICE_SHOW_RW(mtu, fmt_dec); static int change_flags(struct net_device *dev, unsigned long new_flags) { - return dev_change_flags(dev, (unsigned int) new_flags); + return dev_change_flags(dev, (unsigned int)new_flags); } static ssize_t flags_store(struct device *dev, struct device_attribute *attr, @@ -362,8 +366,8 @@ static int change_gro_flush_timeout(struct net_device *dev, unsigned long val) } static ssize_t gro_flush_timeout_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t len) + struct device_attribute *attr, + const char *buf, size_t len) { if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -412,7 +416,7 @@ static DEVICE_ATTR_RW(ifalias); static int change_group(struct net_device *dev, unsigned long new_group) { - dev_set_group(dev, (int) new_group); + dev_set_group(dev, (int)new_group); return 0; } @@ -426,7 +430,7 @@ static DEVICE_ATTR(netdev_group, S_IRUGO | S_IWUSR, group_show, group_store); static int change_proto_down(struct net_device *dev, unsigned long proto_down) { - return dev_change_proto_down(dev, (bool) proto_down); + return dev_change_proto_down(dev, (bool)proto_down); } static ssize_t proto_down_store(struct device *dev, @@ -508,7 +512,7 @@ static ssize_t phys_switch_id_show(struct device *dev, } static DEVICE_ATTR_RO(phys_switch_id); -static struct attribute *net_class_attrs[] = { +static struct attribute *net_class_attrs[] __ro_after_init = { &dev_attr_netdev_group.attr, &dev_attr_type.attr, &dev_attr_dev_id.attr, @@ -549,14 +553,14 @@ static ssize_t netstat_show(const struct device *d, ssize_t ret = -EINVAL; WARN_ON(offset > sizeof(struct rtnl_link_stats64) || - offset % sizeof(u64) != 0); + offset % sizeof(u64) != 0); read_lock(&dev_base_lock); if (dev_isalive(dev)) { struct rtnl_link_stats64 temp; const struct rtnl_link_stats64 *stats = dev_get_stats(dev, &temp); - ret = sprintf(buf, fmt_u64, *(u64 *)(((u8 *) stats) + offset)); + ret = sprintf(buf, fmt_u64, *(u64 *)(((u8 *)stats) + offset)); } read_unlock(&dev_base_lock); return ret; @@ -565,7 +569,7 @@ static ssize_t netstat_show(const struct device *d, /* generate a read-only statistics attribute */ #define NETSTAT_ENTRY(name) \ static ssize_t name##_show(struct device *d, \ - struct device_attribute *attr, char *buf) \ + struct device_attribute *attr, char *buf) \ { \ return netstat_show(d, attr, buf, \ offsetof(struct rtnl_link_stats64, name)); \ @@ -597,7 +601,7 @@ NETSTAT_ENTRY(rx_compressed); NETSTAT_ENTRY(tx_compressed); NETSTAT_ENTRY(rx_nohandler); -static struct attribute *netstat_attrs[] = { +static struct attribute *netstat_attrs[] __ro_after_init = { &dev_attr_rx_packets.attr, &dev_attr_tx_packets.attr, &dev_attr_rx_bytes.attr, @@ -625,7 +629,6 @@ static struct attribute *netstat_attrs[] = { NULL }; - static const struct attribute_group netstat_group = { .name = "statistics", .attrs = netstat_attrs, @@ -647,33 +650,33 @@ static const struct attribute_group wireless_group = { #endif /* CONFIG_SYSFS */ #ifdef CONFIG_SYSFS -#define to_rx_queue_attr(_attr) container_of(_attr, \ - struct rx_queue_attribute, attr) +#define to_rx_queue_attr(_attr) \ + container_of(_attr, struct rx_queue_attribute, attr) #define to_rx_queue(obj) container_of(obj, struct netdev_rx_queue, kobj) static ssize_t rx_queue_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { - struct rx_queue_attribute *attribute = to_rx_queue_attr(attr); + const struct rx_queue_attribute *attribute = to_rx_queue_attr(attr); struct netdev_rx_queue *queue = to_rx_queue(kobj); if (!attribute->show) return -EIO; - return attribute->show(queue, attribute, buf); + return attribute->show(queue, buf); } static ssize_t rx_queue_attr_store(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) { - struct rx_queue_attribute *attribute = to_rx_queue_attr(attr); + const struct rx_queue_attribute *attribute = to_rx_queue_attr(attr); struct netdev_rx_queue *queue = to_rx_queue(kobj); if (!attribute->store) return -EIO; - return attribute->store(queue, attribute, buf, count); + return attribute->store(queue, buf, count); } static const struct sysfs_ops rx_queue_sysfs_ops = { @@ -682,8 +685,7 @@ static const struct sysfs_ops rx_queue_sysfs_ops = { }; #ifdef CONFIG_RPS -static ssize_t show_rps_map(struct netdev_rx_queue *queue, - struct rx_queue_attribute *attribute, char *buf) +static ssize_t show_rps_map(struct netdev_rx_queue *queue, char *buf) { struct rps_map *map; cpumask_var_t mask; @@ -706,8 +708,7 @@ static ssize_t show_rps_map(struct netdev_rx_queue *queue, } static ssize_t store_rps_map(struct netdev_rx_queue *queue, - struct rx_queue_attribute *attribute, - const char *buf, size_t len) + const char *buf, size_t len) { struct rps_map *old_map, *map; cpumask_var_t mask; @@ -727,8 +728,8 @@ static ssize_t store_rps_map(struct netdev_rx_queue *queue, } map = kzalloc(max_t(unsigned int, - RPS_MAP_SIZE(cpumask_weight(mask)), L1_CACHE_BYTES), - GFP_KERNEL); + RPS_MAP_SIZE(cpumask_weight(mask)), L1_CACHE_BYTES), + GFP_KERNEL); if (!map) { free_cpumask_var(mask); return -ENOMEM; @@ -738,9 +739,9 @@ static ssize_t store_rps_map(struct netdev_rx_queue *queue, for_each_cpu_and(cpu, mask, cpu_online_mask) map->cpus[i++] = cpu; - if (i) + if (i) { map->len = i; - else { + } else { kfree(map); map = NULL; } @@ -765,7 +766,6 @@ static ssize_t store_rps_map(struct netdev_rx_queue *queue, } static ssize_t show_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, - struct rx_queue_attribute *attr, char *buf) { struct rps_dev_flow_table *flow_table; @@ -788,8 +788,7 @@ static void rps_dev_flow_table_release(struct rcu_head *rcu) } static ssize_t store_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, - struct rx_queue_attribute *attr, - const char *buf, size_t len) + const char *buf, size_t len) { unsigned long mask, count; struct rps_dev_flow_table *table, *old_table; @@ -831,8 +830,9 @@ static ssize_t store_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, table->mask = mask; for (count = 0; count <= mask; count++) table->flows[count].cpu = RPS_NO_CPU; - } else + } else { table = NULL; + } spin_lock(&rps_dev_flow_lock); old_table = rcu_dereference_protected(queue->rps_flow_table, @@ -846,16 +846,15 @@ static ssize_t store_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, return len; } -static struct rx_queue_attribute rps_cpus_attribute = - __ATTR(rps_cpus, S_IRUGO | S_IWUSR, show_rps_map, store_rps_map); - +static struct rx_queue_attribute rps_cpus_attribute __ro_after_init + = __ATTR(rps_cpus, S_IRUGO | S_IWUSR, show_rps_map, store_rps_map); -static struct rx_queue_attribute rps_dev_flow_table_cnt_attribute = - __ATTR(rps_flow_cnt, S_IRUGO | S_IWUSR, - show_rps_dev_flow_table_cnt, store_rps_dev_flow_table_cnt); +static struct rx_queue_attribute rps_dev_flow_table_cnt_attribute __ro_after_init + = __ATTR(rps_flow_cnt, S_IRUGO | S_IWUSR, + show_rps_dev_flow_table_cnt, store_rps_dev_flow_table_cnt); #endif /* CONFIG_RPS */ -static struct attribute *rx_queue_default_attrs[] = { +static struct attribute *rx_queue_default_attrs[] __ro_after_init = { #ifdef CONFIG_RPS &rps_cpus_attribute.attr, &rps_dev_flow_table_cnt_attribute.attr, @@ -870,7 +869,6 @@ static void rx_queue_release(struct kobject *kobj) struct rps_map *map; struct rps_dev_flow_table *flow_table; - map = rcu_dereference_protected(queue->rps_map, 1); if (map) { RCU_INIT_POINTER(queue->rps_map, NULL); @@ -900,7 +898,7 @@ static const void *rx_queue_namespace(struct kobject *kobj) return ns; } -static struct kobj_type rx_queue_ktype = { +static struct kobj_type rx_queue_ktype __ro_after_init = { .sysfs_ops = &rx_queue_sysfs_ops, .release = rx_queue_release, .default_attrs = rx_queue_default_attrs, @@ -915,23 +913,22 @@ static int rx_queue_add_kobject(struct net_device *dev, int index) kobj->kset = dev->queues_kset; error = kobject_init_and_add(kobj, &rx_queue_ktype, NULL, - "rx-%u", index); + "rx-%u", index); if (error) - goto exit; + return error; if (dev->sysfs_rx_queue_group) { error = sysfs_create_group(kobj, dev->sysfs_rx_queue_group); - if (error) - goto exit; + if (error) { + kobject_put(kobj); + return error; + } } kobject_uevent(kobj, KOBJ_ADD); dev_hold(queue->dev); return error; -exit: - kobject_put(kobj); - return error; } #endif /* CONFIG_SYSFS */ @@ -976,39 +973,40 @@ net_rx_queue_update_kobjects(struct net_device *dev, int old_num, int new_num) */ struct netdev_queue_attribute { struct attribute attr; - ssize_t (*show)(struct netdev_queue *queue, - struct netdev_queue_attribute *attr, char *buf); + ssize_t (*show)(struct netdev_queue *queue, char *buf); ssize_t (*store)(struct netdev_queue *queue, - struct netdev_queue_attribute *attr, const char *buf, size_t len); + const char *buf, size_t len); }; -#define to_netdev_queue_attr(_attr) container_of(_attr, \ - struct netdev_queue_attribute, attr) +#define to_netdev_queue_attr(_attr) \ + container_of(_attr, struct netdev_queue_attribute, attr) #define to_netdev_queue(obj) container_of(obj, struct netdev_queue, kobj) static ssize_t netdev_queue_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { - struct netdev_queue_attribute *attribute = to_netdev_queue_attr(attr); + const struct netdev_queue_attribute *attribute + = to_netdev_queue_attr(attr); struct netdev_queue *queue = to_netdev_queue(kobj); if (!attribute->show) return -EIO; - return attribute->show(queue, attribute, buf); + return attribute->show(queue, buf); } static ssize_t netdev_queue_attr_store(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) { - struct netdev_queue_attribute *attribute = to_netdev_queue_attr(attr); + const struct netdev_queue_attribute *attribute + = to_netdev_queue_attr(attr); struct netdev_queue *queue = to_netdev_queue(kobj); if (!attribute->store) return -EIO; - return attribute->store(queue, attribute, buf, count); + return attribute->store(queue, buf, count); } static const struct sysfs_ops netdev_queue_sysfs_ops = { @@ -1016,9 +1014,7 @@ static const struct sysfs_ops netdev_queue_sysfs_ops = { .store = netdev_queue_attr_store, }; -static ssize_t show_trans_timeout(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, - char *buf) +static ssize_t tx_timeout_show(struct netdev_queue *queue, char *buf) { unsigned long trans_timeout; @@ -1040,8 +1036,7 @@ static unsigned int get_netdev_queue_index(struct netdev_queue *queue) return i; } -static ssize_t show_traffic_class(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, +static ssize_t traffic_class_show(struct netdev_queue *queue, char *buf) { struct net_device *dev = queue->dev; @@ -1055,16 +1050,14 @@ static ssize_t show_traffic_class(struct netdev_queue *queue, } #ifdef CONFIG_XPS -static ssize_t show_tx_maxrate(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, +static ssize_t tx_maxrate_show(struct netdev_queue *queue, char *buf) { return sprintf(buf, "%lu\n", queue->tx_maxrate); } -static ssize_t set_tx_maxrate(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, - const char *buf, size_t len) +static ssize_t tx_maxrate_store(struct netdev_queue *queue, + const char *buf, size_t len) { struct net_device *dev = queue->dev; int err, index = get_netdev_queue_index(queue); @@ -1089,16 +1082,15 @@ static ssize_t set_tx_maxrate(struct netdev_queue *queue, return err; } -static struct netdev_queue_attribute queue_tx_maxrate = - __ATTR(tx_maxrate, S_IRUGO | S_IWUSR, - show_tx_maxrate, set_tx_maxrate); +static struct netdev_queue_attribute queue_tx_maxrate __ro_after_init + = __ATTR_RW(tx_maxrate); #endif -static struct netdev_queue_attribute queue_trans_timeout = - __ATTR(tx_timeout, S_IRUGO, show_trans_timeout, NULL); +static struct netdev_queue_attribute queue_trans_timeout __ro_after_init + = __ATTR_RO(tx_timeout); -static struct netdev_queue_attribute queue_traffic_class = - __ATTR(traffic_class, S_IRUGO, show_traffic_class, NULL); +static struct netdev_queue_attribute queue_traffic_class __ro_after_init + = __ATTR_RO(traffic_class); #ifdef CONFIG_BQL /* @@ -1115,9 +1107,9 @@ static ssize_t bql_set(const char *buf, const size_t count, unsigned int value; int err; - if (!strcmp(buf, "max") || !strcmp(buf, "max\n")) + if (!strcmp(buf, "max") || !strcmp(buf, "max\n")) { value = DQL_MAX_LIMIT; - else { + } else { err = kstrtouint(buf, 10, &value); if (err < 0) return err; @@ -1131,7 +1123,6 @@ static ssize_t bql_set(const char *buf, const size_t count, } static ssize_t bql_show_hold_time(struct netdev_queue *queue, - struct netdev_queue_attribute *attr, char *buf) { struct dql *dql = &queue->dql; @@ -1140,7 +1131,6 @@ static ssize_t bql_show_hold_time(struct netdev_queue *queue, } static ssize_t bql_set_hold_time(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, const char *buf, size_t len) { struct dql *dql = &queue->dql; @@ -1156,12 +1146,11 @@ static ssize_t bql_set_hold_time(struct netdev_queue *queue, return len; } -static struct netdev_queue_attribute bql_hold_time_attribute = - __ATTR(hold_time, S_IRUGO | S_IWUSR, bql_show_hold_time, - bql_set_hold_time); +static struct netdev_queue_attribute bql_hold_time_attribute __ro_after_init + = __ATTR(hold_time, S_IRUGO | S_IWUSR, + bql_show_hold_time, bql_set_hold_time); static ssize_t bql_show_inflight(struct netdev_queue *queue, - struct netdev_queue_attribute *attr, char *buf) { struct dql *dql = &queue->dql; @@ -1169,33 +1158,31 @@ static ssize_t bql_show_inflight(struct netdev_queue *queue, return sprintf(buf, "%u\n", dql->num_queued - dql->num_completed); } -static struct netdev_queue_attribute bql_inflight_attribute = +static struct netdev_queue_attribute bql_inflight_attribute __ro_after_init = __ATTR(inflight, S_IRUGO, bql_show_inflight, NULL); #define BQL_ATTR(NAME, FIELD) \ static ssize_t bql_show_ ## NAME(struct netdev_queue *queue, \ - struct netdev_queue_attribute *attr, \ char *buf) \ { \ return bql_show(buf, queue->dql.FIELD); \ } \ \ static ssize_t bql_set_ ## NAME(struct netdev_queue *queue, \ - struct netdev_queue_attribute *attr, \ const char *buf, size_t len) \ { \ return bql_set(buf, len, &queue->dql.FIELD); \ } \ \ -static struct netdev_queue_attribute bql_ ## NAME ## _attribute = \ - __ATTR(NAME, S_IRUGO | S_IWUSR, bql_show_ ## NAME, \ - bql_set_ ## NAME); +static struct netdev_queue_attribute bql_ ## NAME ## _attribute __ro_after_init \ + = __ATTR(NAME, S_IRUGO | S_IWUSR, \ + bql_show_ ## NAME, bql_set_ ## NAME) -BQL_ATTR(limit, limit) -BQL_ATTR(limit_max, max_limit) -BQL_ATTR(limit_min, min_limit) +BQL_ATTR(limit, limit); +BQL_ATTR(limit_max, max_limit); +BQL_ATTR(limit_min, min_limit); -static struct attribute *dql_attrs[] = { +static struct attribute *dql_attrs[] __ro_after_init = { &bql_limit_attribute.attr, &bql_limit_max_attribute.attr, &bql_limit_min_attribute.attr, @@ -1211,8 +1198,8 @@ static const struct attribute_group dql_group = { #endif /* CONFIG_BQL */ #ifdef CONFIG_XPS -static ssize_t show_xps_map(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, char *buf) +static ssize_t xps_cpus_show(struct netdev_queue *queue, + char *buf) { struct net_device *dev = queue->dev; int cpu, len, num_tc = 1, tc = 0; @@ -1258,9 +1245,8 @@ static ssize_t show_xps_map(struct netdev_queue *queue, return len < PAGE_SIZE ? len : -EINVAL; } -static ssize_t store_xps_map(struct netdev_queue *queue, - struct netdev_queue_attribute *attribute, - const char *buf, size_t len) +static ssize_t xps_cpus_store(struct netdev_queue *queue, + const char *buf, size_t len) { struct net_device *dev = queue->dev; unsigned long index; @@ -1288,11 +1274,11 @@ static ssize_t store_xps_map(struct netdev_queue *queue, return err ? : len; } -static struct netdev_queue_attribute xps_cpus_attribute = - __ATTR(xps_cpus, S_IRUGO | S_IWUSR, show_xps_map, store_xps_map); +static struct netdev_queue_attribute xps_cpus_attribute __ro_after_init + = __ATTR_RW(xps_cpus); #endif /* CONFIG_XPS */ -static struct attribute *netdev_queue_default_attrs[] = { +static struct attribute *netdev_queue_default_attrs[] __ro_after_init = { &queue_trans_timeout.attr, &queue_traffic_class.attr, #ifdef CONFIG_XPS @@ -1322,7 +1308,7 @@ static const void *netdev_queue_namespace(struct kobject *kobj) return ns; } -static struct kobj_type netdev_queue_ktype = { +static struct kobj_type netdev_queue_ktype __ro_after_init = { .sysfs_ops = &netdev_queue_sysfs_ops, .release = netdev_queue_release, .default_attrs = netdev_queue_default_attrs, @@ -1337,23 +1323,22 @@ static int netdev_queue_add_kobject(struct net_device *dev, int index) kobj->kset = dev->queues_kset; error = kobject_init_and_add(kobj, &netdev_queue_ktype, NULL, - "tx-%u", index); + "tx-%u", index); if (error) - goto exit; + return error; #ifdef CONFIG_BQL error = sysfs_create_group(kobj, &dql_group); - if (error) - goto exit; + if (error) { + kobject_put(kobj); + return error; + } #endif kobject_uevent(kobj, KOBJ_ADD); dev_hold(queue->dev); return 0; -exit: - kobject_put(kobj); - return error; } #endif /* CONFIG_SYSFS */ @@ -1395,7 +1380,7 @@ static int register_queue_kobjects(struct net_device *dev) #ifdef CONFIG_SYSFS dev->queues_kset = kset_create_and_add("queues", - NULL, &dev->dev.kobj); + NULL, &dev->dev.kobj); if (!dev->queues_kset) return -ENOMEM; real_rx = dev->real_num_rx_queues; @@ -1463,7 +1448,7 @@ static const void *net_netlink_ns(struct sock *sk) return sock_net(sk); } -struct kobj_ns_type_operations net_ns_type_operations = { +const struct kobj_ns_type_operations net_ns_type_operations = { .type = KOBJ_NS_TYPE_NET, .current_may_mount = net_current_may_mount, .grab_current_ns = net_grab_current_ns, @@ -1485,7 +1470,8 @@ static int netdev_uevent(struct device *d, struct kobj_uevent_env *env) /* pass ifindex to uevent. * ifindex is useful as it won't change (interface name may change) - * and is what RtNetlink uses natively. */ + * and is what RtNetlink uses natively. + */ retval = add_uevent_var(env, "IFINDEX=%d", dev->ifindex); exit: @@ -1513,7 +1499,7 @@ static const void *net_namespace(struct device *d) return dev_net(dev); } -static struct class net_class = { +static struct class net_class __ro_after_init = { .name = "net", .dev_release = netdev_release, .dev_groups = net_class_groups, @@ -1560,7 +1546,7 @@ EXPORT_SYMBOL(of_find_net_device_by_node); */ void netdev_unregister_kobject(struct net_device *ndev) { - struct device *dev = &(ndev->dev); + struct device *dev = &ndev->dev; if (!atomic_read(&dev_net(ndev)->count)) dev_set_uevent_suppress(dev, 1); @@ -1577,7 +1563,7 @@ void netdev_unregister_kobject(struct net_device *ndev) /* Create sysfs entries for network device. */ int netdev_register_kobject(struct net_device *ndev) { - struct device *dev = &(ndev->dev); + struct device *dev = &ndev->dev; const struct attribute_group **groups = ndev->sysfs_groups; int error = 0; @@ -1620,14 +1606,14 @@ int netdev_register_kobject(struct net_device *ndev) return error; } -int netdev_class_create_file_ns(struct class_attribute *class_attr, +int netdev_class_create_file_ns(const struct class_attribute *class_attr, const void *ns) { return class_create_file_ns(&net_class, class_attr, ns); } EXPORT_SYMBOL(netdev_class_create_file_ns); -void netdev_class_remove_file_ns(struct class_attribute *class_attr, +void netdev_class_remove_file_ns(const struct class_attribute *class_attr, const void *ns) { class_remove_file_ns(&net_class, class_attr, ns); diff --git a/net/core/net-traces.c b/net/core/net-traces.c index 92da5e4ceb4f..1132820c8e62 100644 --- a/net/core/net-traces.c +++ b/net/core/net-traces.c @@ -32,10 +32,18 @@ #include <trace/events/sock.h> #include <trace/events/udp.h> #include <trace/events/fib.h> +#include <trace/events/qdisc.h> #if IS_ENABLED(CONFIG_IPV6) #include <trace/events/fib6.h> EXPORT_TRACEPOINT_SYMBOL_GPL(fib6_table_lookup); #endif +#if IS_ENABLED(CONFIG_BRIDGE) +#include <trace/events/bridge.h> +EXPORT_TRACEPOINT_SYMBOL_GPL(br_fdb_add); +EXPORT_TRACEPOINT_SYMBOL_GPL(br_fdb_external_learn_add); +EXPORT_TRACEPOINT_SYMBOL_GPL(fdb_delete); +EXPORT_TRACEPOINT_SYMBOL_GPL(br_fdb_update); +#endif EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb); diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 8726d051f31d..6cfdc7c84c48 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -855,9 +855,10 @@ static int __init net_ns_init(void) register_pernet_subsys(&net_ns_ops); - rtnl_register(PF_UNSPEC, RTM_NEWNSID, rtnl_net_newid, NULL, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWNSID, rtnl_net_newid, NULL, + RTNL_FLAG_DOIT_UNLOCKED); rtnl_register(PF_UNSPEC, RTM_GETNSID, rtnl_net_getid, rtnl_net_dumpid, - NULL); + RTNL_FLAG_DOIT_UNLOCKED); return 0; } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index d3408a693166..912731bed7b7 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -277,7 +277,7 @@ static void zap_completion_queue(void) struct sk_buff *skb = clist; clist = clist->next; if (!skb_irq_freeable(skb)) { - refcount_inc(&skb->users); + refcount_set(&skb->users, 1); dev_kfree_skb_any(skb); /* put this one back */ } else { __kfree_skb(skb); @@ -666,7 +666,7 @@ int netpoll_setup(struct netpoll *np) int err; rtnl_lock(); - if (np->dev_name) { + if (np->dev_name[0]) { struct net *net = current->nsproxy->net_ns; ndev = __dev_get_by_name(net, np->dev_name); } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index d1ba90980be1..a78fd61da0ec 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -62,7 +62,7 @@ struct rtnl_link { rtnl_doit_func doit; rtnl_dumpit_func dumpit; - rtnl_calcit_func calcit; + unsigned int flags; }; static DEFINE_MUTEX(rtnl_mutex); @@ -127,7 +127,8 @@ bool lockdep_rtnl_is_held(void) EXPORT_SYMBOL(lockdep_rtnl_is_held); #endif /* #ifdef CONFIG_PROVE_LOCKING */ -static struct rtnl_link *rtnl_msg_handlers[RTNL_FAMILY_MAX + 1]; +static struct rtnl_link __rcu *rtnl_msg_handlers[RTNL_FAMILY_MAX + 1]; +static refcount_t rtnl_msg_handlers_ref[RTNL_FAMILY_MAX + 1]; static inline int rtm_msgindex(int msgtype) { @@ -143,58 +144,13 @@ static inline int rtm_msgindex(int msgtype) return msgindex; } -static rtnl_doit_func rtnl_get_doit(int protocol, int msgindex) -{ - struct rtnl_link *tab; - - if (protocol <= RTNL_FAMILY_MAX) - tab = rtnl_msg_handlers[protocol]; - else - tab = NULL; - - if (tab == NULL || tab[msgindex].doit == NULL) - tab = rtnl_msg_handlers[PF_UNSPEC]; - - return tab[msgindex].doit; -} - -static rtnl_dumpit_func rtnl_get_dumpit(int protocol, int msgindex) -{ - struct rtnl_link *tab; - - if (protocol <= RTNL_FAMILY_MAX) - tab = rtnl_msg_handlers[protocol]; - else - tab = NULL; - - if (tab == NULL || tab[msgindex].dumpit == NULL) - tab = rtnl_msg_handlers[PF_UNSPEC]; - - return tab[msgindex].dumpit; -} - -static rtnl_calcit_func rtnl_get_calcit(int protocol, int msgindex) -{ - struct rtnl_link *tab; - - if (protocol <= RTNL_FAMILY_MAX) - tab = rtnl_msg_handlers[protocol]; - else - tab = NULL; - - if (tab == NULL || tab[msgindex].calcit == NULL) - tab = rtnl_msg_handlers[PF_UNSPEC]; - - return tab[msgindex].calcit; -} - /** * __rtnl_register - Register a rtnetlink message type * @protocol: Protocol family or PF_UNSPEC * @msgtype: rtnetlink message type * @doit: Function pointer called for each request message * @dumpit: Function pointer called for each dump request (NLM_F_DUMP) message - * @calcit: Function pointer to calc size of dump message + * @flags: rtnl_link_flags to modifiy behaviour of doit/dumpit functions * * Registers the specified function pointers (at least one of them has * to be non-NULL) to be called whenever a request message for the @@ -208,7 +164,7 @@ static rtnl_calcit_func rtnl_get_calcit(int protocol, int msgindex) */ int __rtnl_register(int protocol, int msgtype, rtnl_doit_func doit, rtnl_dumpit_func dumpit, - rtnl_calcit_func calcit) + unsigned int flags) { struct rtnl_link *tab; int msgindex; @@ -216,23 +172,20 @@ int __rtnl_register(int protocol, int msgtype, BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX); msgindex = rtm_msgindex(msgtype); - tab = rtnl_msg_handlers[protocol]; + tab = rcu_dereference_raw(rtnl_msg_handlers[protocol]); if (tab == NULL) { tab = kcalloc(RTM_NR_MSGTYPES, sizeof(*tab), GFP_KERNEL); if (tab == NULL) return -ENOBUFS; - rtnl_msg_handlers[protocol] = tab; + rcu_assign_pointer(rtnl_msg_handlers[protocol], tab); } if (doit) tab[msgindex].doit = doit; - if (dumpit) tab[msgindex].dumpit = dumpit; - - if (calcit) - tab[msgindex].calcit = calcit; + tab[msgindex].flags |= flags; return 0; } @@ -249,9 +202,9 @@ EXPORT_SYMBOL_GPL(__rtnl_register); */ void rtnl_register(int protocol, int msgtype, rtnl_doit_func doit, rtnl_dumpit_func dumpit, - rtnl_calcit_func calcit) + unsigned int flags) { - if (__rtnl_register(protocol, msgtype, doit, dumpit, calcit) < 0) + if (__rtnl_register(protocol, msgtype, doit, dumpit, flags) < 0) panic("Unable to register rtnetlink message handler, " "protocol = %d, message type = %d\n", protocol, msgtype); @@ -267,17 +220,23 @@ EXPORT_SYMBOL_GPL(rtnl_register); */ int rtnl_unregister(int protocol, int msgtype) { + struct rtnl_link *handlers; int msgindex; BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX); msgindex = rtm_msgindex(msgtype); - if (rtnl_msg_handlers[protocol] == NULL) + rtnl_lock(); + handlers = rtnl_dereference(rtnl_msg_handlers[protocol]); + if (!handlers) { + rtnl_unlock(); return -ENOENT; + } - rtnl_msg_handlers[protocol][msgindex].doit = NULL; - rtnl_msg_handlers[protocol][msgindex].dumpit = NULL; - rtnl_msg_handlers[protocol][msgindex].calcit = NULL; + handlers[msgindex].doit = NULL; + handlers[msgindex].dumpit = NULL; + handlers[msgindex].flags = 0; + rtnl_unlock(); return 0; } @@ -292,10 +251,20 @@ EXPORT_SYMBOL_GPL(rtnl_unregister); */ void rtnl_unregister_all(int protocol) { + struct rtnl_link *handlers; + BUG_ON(protocol < 0 || protocol > RTNL_FAMILY_MAX); - kfree(rtnl_msg_handlers[protocol]); - rtnl_msg_handlers[protocol] = NULL; + rtnl_lock(); + handlers = rtnl_dereference(rtnl_msg_handlers[protocol]); + RCU_INIT_POINTER(rtnl_msg_handlers[protocol], NULL); + rtnl_unlock(); + + synchronize_net(); + + while (refcount_read(&rtnl_msg_handlers_ref[protocol]) > 1) + schedule(); + kfree(handlers); } EXPORT_SYMBOL_GPL(rtnl_unregister_all); @@ -433,16 +402,24 @@ static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev) { struct net_device *master_dev; const struct rtnl_link_ops *ops; + size_t size = 0; - master_dev = netdev_master_upper_dev_get((struct net_device *) dev); + rcu_read_lock(); + + master_dev = netdev_master_upper_dev_get_rcu((struct net_device *)dev); if (!master_dev) - return 0; + goto out; + ops = master_dev->rtnl_link_ops; if (!ops || !ops->get_slave_size) - return 0; + goto out; /* IFLA_INFO_SLAVE_DATA + nested data */ - return nla_total_size(sizeof(struct nlattr)) + + size = nla_total_size(sizeof(struct nlattr)) + ops->get_slave_size(master_dev, dev); + +out: + rcu_read_unlock(); + return size; } static size_t rtnl_link_get_size(const struct net_device *dev) @@ -1644,8 +1621,6 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) s_h = cb->args[0]; s_idx = cb->args[1]; - cb->seq = net->dev_base_seq; - /* A hack to preserve kernel<->userspace interface. * The correct header is ifinfomsg. It is consistent with rtnl_getlink. * However, before Linux v3.9 the code here assumed rtgenmsg and that's @@ -1691,8 +1666,6 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) goto out_err; } - - nl_dump_check_consistent(cb, nlmsg_hdr(skb)); cont: idx++; } @@ -1702,6 +1675,8 @@ out: out_err: cb->args[1] = idx; cb->args[0] = h; + cb->seq = net->dev_base_seq; + nl_dump_check_consistent(cb, nlmsg_hdr(skb)); return err; } @@ -2031,7 +2006,8 @@ static int do_setlink(const struct sk_buff *skb, struct sockaddr *sa; int len; - len = sizeof(sa_family_t) + dev->addr_len; + len = sizeof(sa_family_t) + max_t(size_t, dev->addr_len, + sizeof(*sa)); sa = kmalloc(len, GFP_KERNEL); if (!sa) { err = -ENOMEM; @@ -2830,11 +2806,13 @@ static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh) * traverse the list of net devices and compute the minimum * buffer size based upon the filter mask. */ - list_for_each_entry(dev, &net->dev_base_head, dev_list) { + rcu_read_lock(); + for_each_netdev_rcu(net, dev) { min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size, if_nlmsg_size(dev, ext_filter_mask)); } + rcu_read_unlock(); return nlmsg_total_size(min_ifinfo_dump_size); } @@ -2846,19 +2824,29 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb) if (s_idx == 0) s_idx = 1; + for (idx = 1; idx <= RTNL_FAMILY_MAX; idx++) { int type = cb->nlh->nlmsg_type-RTM_BASE; + struct rtnl_link *handlers; + rtnl_dumpit_func dumpit; + if (idx < s_idx || idx == PF_PACKET) continue; - if (rtnl_msg_handlers[idx] == NULL || - rtnl_msg_handlers[idx][type].dumpit == NULL) + + handlers = rtnl_dereference(rtnl_msg_handlers[idx]); + if (!handlers) continue; + + dumpit = READ_ONCE(handlers[type].dumpit); + if (!dumpit) + continue; + if (idx > s_idx) { memset(&cb->args[0], 0, sizeof(cb->args)); cb->prev_seq = 0; cb->seq = 0; } - if (rtnl_msg_handlers[idx][type].dumpit(skb, cb)) + if (dumpit(skb, cb)) break; } cb->family = idx; @@ -4161,11 +4149,13 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); + struct rtnl_link *handlers; + int err = -EOPNOTSUPP; rtnl_doit_func doit; + unsigned int flags; int kind; int family; int type; - int err; type = nlh->nlmsg_type; if (type > RTM_MAX) @@ -4183,20 +4173,40 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN)) return -EPERM; + if (family >= ARRAY_SIZE(rtnl_msg_handlers)) + family = PF_UNSPEC; + + rcu_read_lock(); + handlers = rcu_dereference(rtnl_msg_handlers[family]); + if (!handlers) { + family = PF_UNSPEC; + handlers = rcu_dereference(rtnl_msg_handlers[family]); + } + if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) { struct sock *rtnl; rtnl_dumpit_func dumpit; - rtnl_calcit_func calcit; u16 min_dump_alloc = 0; - dumpit = rtnl_get_dumpit(family, type); - if (dumpit == NULL) - return -EOPNOTSUPP; - calcit = rtnl_get_calcit(family, type); - if (calcit) - min_dump_alloc = calcit(skb, nlh); + dumpit = READ_ONCE(handlers[type].dumpit); + if (!dumpit) { + family = PF_UNSPEC; + handlers = rcu_dereference(rtnl_msg_handlers[PF_UNSPEC]); + if (!handlers) + goto err_unlock; + + dumpit = READ_ONCE(handlers[type].dumpit); + if (!dumpit) + goto err_unlock; + } + + refcount_inc(&rtnl_msg_handlers_ref[family]); + + if (type == RTM_GETLINK - RTM_BASE) + min_dump_alloc = rtnl_calcit(skb, nlh); + + rcu_read_unlock(); - __rtnl_unlock(); rtnl = net->rtnl; { struct netlink_dump_control c = { @@ -4205,22 +4215,47 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, }; err = netlink_dump_start(rtnl, skb, nlh, &c); } - rtnl_lock(); + refcount_dec(&rtnl_msg_handlers_ref[family]); return err; } - doit = rtnl_get_doit(family, type); - if (doit == NULL) - return -EOPNOTSUPP; + doit = READ_ONCE(handlers[type].doit); + if (!doit) { + family = PF_UNSPEC; + handlers = rcu_dereference(rtnl_msg_handlers[family]); + } + + flags = READ_ONCE(handlers[type].flags); + if (flags & RTNL_FLAG_DOIT_UNLOCKED) { + refcount_inc(&rtnl_msg_handlers_ref[family]); + doit = READ_ONCE(handlers[type].doit); + rcu_read_unlock(); + if (doit) + err = doit(skb, nlh, extack); + refcount_dec(&rtnl_msg_handlers_ref[family]); + return err; + } - return doit(skb, nlh, extack); + rcu_read_unlock(); + + rtnl_lock(); + handlers = rtnl_dereference(rtnl_msg_handlers[family]); + if (handlers) { + doit = READ_ONCE(handlers[type].doit); + if (doit) + err = doit(skb, nlh, extack); + } + rtnl_unlock(); + return err; + +err_unlock: + rcu_read_unlock(); + return -EOPNOTSUPP; } static void rtnetlink_rcv(struct sk_buff *skb) { - rtnl_lock(); netlink_rcv_skb(skb, &rtnetlink_rcv_msg); - rtnl_unlock(); } static int rtnetlink_bind(struct net *net, int group) @@ -4241,6 +4276,7 @@ static int rtnetlink_event(struct notifier_block *this, unsigned long event, voi switch (event) { case NETDEV_REBOOT: + case NETDEV_CHANGEADDR: case NETDEV_CHANGENAME: case NETDEV_FEAT_CHANGE: case NETDEV_BONDING_FAILOVER: @@ -4292,29 +4328,34 @@ static struct pernet_operations rtnetlink_net_ops = { void __init rtnetlink_init(void) { + int i; + + for (i = 0; i < ARRAY_SIZE(rtnl_msg_handlers_ref); i++) + refcount_set(&rtnl_msg_handlers_ref[i], 1); + if (register_pernet_subsys(&rtnetlink_net_ops)) panic("rtnetlink_init: cannot initialize rtnetlink\n"); register_netdevice_notifier(&rtnetlink_dev_notifier); rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink, - rtnl_dump_ifinfo, rtnl_calcit); - rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL, NULL); + rtnl_dump_ifinfo, 0); + rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL, 0); - rtnl_register(PF_UNSPEC, RTM_GETADDR, NULL, rtnl_dump_all, NULL); - rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all, NULL); - rtnl_register(PF_UNSPEC, RTM_GETNETCONF, NULL, rtnl_dump_all, NULL); + rtnl_register(PF_UNSPEC, RTM_GETADDR, NULL, rtnl_dump_all, 0); + rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all, 0); + rtnl_register(PF_UNSPEC, RTM_GETNETCONF, NULL, rtnl_dump_all, 0); - rtnl_register(PF_BRIDGE, RTM_NEWNEIGH, rtnl_fdb_add, NULL, NULL); - rtnl_register(PF_BRIDGE, RTM_DELNEIGH, rtnl_fdb_del, NULL, NULL); - rtnl_register(PF_BRIDGE, RTM_GETNEIGH, NULL, rtnl_fdb_dump, NULL); + rtnl_register(PF_BRIDGE, RTM_NEWNEIGH, rtnl_fdb_add, NULL, 0); + rtnl_register(PF_BRIDGE, RTM_DELNEIGH, rtnl_fdb_del, NULL, 0); + rtnl_register(PF_BRIDGE, RTM_GETNEIGH, NULL, rtnl_fdb_dump, 0); - rtnl_register(PF_BRIDGE, RTM_GETLINK, NULL, rtnl_bridge_getlink, NULL); - rtnl_register(PF_BRIDGE, RTM_DELLINK, rtnl_bridge_dellink, NULL, NULL); - rtnl_register(PF_BRIDGE, RTM_SETLINK, rtnl_bridge_setlink, NULL, NULL); + rtnl_register(PF_BRIDGE, RTM_GETLINK, NULL, rtnl_bridge_getlink, 0); + rtnl_register(PF_BRIDGE, RTM_DELLINK, rtnl_bridge_dellink, NULL, 0); + rtnl_register(PF_BRIDGE, RTM_SETLINK, rtnl_bridge_setlink, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETSTATS, rtnl_stats_get, rtnl_stats_dump, - NULL); + 0); } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f990eb8b30a9..68065d7d383f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -158,31 +158,6 @@ out: * */ -struct sk_buff *__alloc_skb_head(gfp_t gfp_mask, int node) -{ - struct sk_buff *skb; - - /* Get the HEAD */ - skb = kmem_cache_alloc_node(skbuff_head_cache, - gfp_mask & ~__GFP_DMA, node); - if (!skb) - goto out; - - /* - * Only clear those fields we need to clear, not those that we will - * actually initialise below. Hence, don't put any more fields after - * the tail pointer in struct sk_buff! - */ - memset(skb, 0, offsetof(struct sk_buff, tail)); - skb->head = NULL; - skb->truesize = sizeof(struct sk_buff); - refcount_set(&skb->users, 1); - - skb->mac_header = (typeof(skb->mac_header))~0U; -out: - return skb; -} - /** * __alloc_skb - allocate a network buffer * @size: size to allocate @@ -592,21 +567,10 @@ static void skb_release_data(struct sk_buff *skb) for (i = 0; i < shinfo->nr_frags; i++) __skb_frag_unref(&shinfo->frags[i]); - /* - * If skb buf is from userspace, we need to notify the caller - * the lower device DMA has done; - */ - if (shinfo->tx_flags & SKBTX_DEV_ZEROCOPY) { - struct ubuf_info *uarg; - - uarg = shinfo->destructor_arg; - if (uarg->callback) - uarg->callback(uarg, true); - } - if (shinfo->frag_list) kfree_skb_list(shinfo->frag_list); + skb_zcopy_clear(skb, true); skb_free_head(skb); } @@ -720,14 +684,7 @@ EXPORT_SYMBOL(kfree_skb_list); */ void skb_tx_error(struct sk_buff *skb) { - if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { - struct ubuf_info *uarg; - - uarg = skb_shinfo(skb)->destructor_arg; - if (uarg->callback) - uarg->callback(uarg, false); - skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; - } + skb_zcopy_clear(skb, true); } EXPORT_SYMBOL(skb_tx_error); @@ -762,8 +719,7 @@ void consume_stateless_skb(struct sk_buff *skb) return; trace_consume_skb(skb); - if (likely(skb->head)) - skb_release_data(skb); + skb_release_data(skb); kfree_skbmem(skb); } @@ -941,6 +897,267 @@ struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src) } EXPORT_SYMBOL_GPL(skb_morph); +static int mm_account_pinned_pages(struct mmpin *mmp, size_t size) +{ + unsigned long max_pg, num_pg, new_pg, old_pg; + struct user_struct *user; + + if (capable(CAP_IPC_LOCK) || !size) + return 0; + + num_pg = (size >> PAGE_SHIFT) + 2; /* worst case */ + max_pg = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + user = mmp->user ? : current_user(); + + do { + old_pg = atomic_long_read(&user->locked_vm); + new_pg = old_pg + num_pg; + if (new_pg > max_pg) + return -ENOBUFS; + } while (atomic_long_cmpxchg(&user->locked_vm, old_pg, new_pg) != + old_pg); + + if (!mmp->user) { + mmp->user = get_uid(user); + mmp->num_pg = num_pg; + } else { + mmp->num_pg += num_pg; + } + + return 0; +} + +static void mm_unaccount_pinned_pages(struct mmpin *mmp) +{ + if (mmp->user) { + atomic_long_sub(mmp->num_pg, &mmp->user->locked_vm); + free_uid(mmp->user); + } +} + +struct ubuf_info *sock_zerocopy_alloc(struct sock *sk, size_t size) +{ + struct ubuf_info *uarg; + struct sk_buff *skb; + + WARN_ON_ONCE(!in_task()); + + if (!sock_flag(sk, SOCK_ZEROCOPY)) + return NULL; + + skb = sock_omalloc(sk, 0, GFP_KERNEL); + if (!skb) + return NULL; + + BUILD_BUG_ON(sizeof(*uarg) > sizeof(skb->cb)); + uarg = (void *)skb->cb; + uarg->mmp.user = NULL; + + if (mm_account_pinned_pages(&uarg->mmp, size)) { + kfree_skb(skb); + return NULL; + } + + uarg->callback = sock_zerocopy_callback; + uarg->id = ((u32)atomic_inc_return(&sk->sk_zckey)) - 1; + uarg->len = 1; + uarg->bytelen = size; + uarg->zerocopy = 1; + refcount_set(&uarg->refcnt, 1); + sock_hold(sk); + + return uarg; +} +EXPORT_SYMBOL_GPL(sock_zerocopy_alloc); + +static inline struct sk_buff *skb_from_uarg(struct ubuf_info *uarg) +{ + return container_of((void *)uarg, struct sk_buff, cb); +} + +struct ubuf_info *sock_zerocopy_realloc(struct sock *sk, size_t size, + struct ubuf_info *uarg) +{ + if (uarg) { + const u32 byte_limit = 1 << 19; /* limit to a few TSO */ + u32 bytelen, next; + + /* realloc only when socket is locked (TCP, UDP cork), + * so uarg->len and sk_zckey access is serialized + */ + if (!sock_owned_by_user(sk)) { + WARN_ON_ONCE(1); + return NULL; + } + + bytelen = uarg->bytelen + size; + if (uarg->len == USHRT_MAX - 1 || bytelen > byte_limit) { + /* TCP can create new skb to attach new uarg */ + if (sk->sk_type == SOCK_STREAM) + goto new_alloc; + return NULL; + } + + next = (u32)atomic_read(&sk->sk_zckey); + if ((u32)(uarg->id + uarg->len) == next) { + if (mm_account_pinned_pages(&uarg->mmp, size)) + return NULL; + uarg->len++; + uarg->bytelen = bytelen; + atomic_set(&sk->sk_zckey, ++next); + sock_zerocopy_get(uarg); + return uarg; + } + } + +new_alloc: + return sock_zerocopy_alloc(sk, size); +} +EXPORT_SYMBOL_GPL(sock_zerocopy_realloc); + +static bool skb_zerocopy_notify_extend(struct sk_buff *skb, u32 lo, u16 len) +{ + struct sock_exterr_skb *serr = SKB_EXT_ERR(skb); + u32 old_lo, old_hi; + u64 sum_len; + + old_lo = serr->ee.ee_info; + old_hi = serr->ee.ee_data; + sum_len = old_hi - old_lo + 1ULL + len; + + if (sum_len >= (1ULL << 32)) + return false; + + if (lo != old_hi + 1) + return false; + + serr->ee.ee_data += len; + return true; +} + +void sock_zerocopy_callback(struct ubuf_info *uarg, bool success) +{ + struct sk_buff *tail, *skb = skb_from_uarg(uarg); + struct sock_exterr_skb *serr; + struct sock *sk = skb->sk; + struct sk_buff_head *q; + unsigned long flags; + u32 lo, hi; + u16 len; + + mm_unaccount_pinned_pages(&uarg->mmp); + + /* if !len, there was only 1 call, and it was aborted + * so do not queue a completion notification + */ + if (!uarg->len || sock_flag(sk, SOCK_DEAD)) + goto release; + + len = uarg->len; + lo = uarg->id; + hi = uarg->id + len - 1; + + serr = SKB_EXT_ERR(skb); + memset(serr, 0, sizeof(*serr)); + serr->ee.ee_errno = 0; + serr->ee.ee_origin = SO_EE_ORIGIN_ZEROCOPY; + serr->ee.ee_data = hi; + serr->ee.ee_info = lo; + if (!success) + serr->ee.ee_code |= SO_EE_CODE_ZEROCOPY_COPIED; + + q = &sk->sk_error_queue; + spin_lock_irqsave(&q->lock, flags); + tail = skb_peek_tail(q); + if (!tail || SKB_EXT_ERR(tail)->ee.ee_origin != SO_EE_ORIGIN_ZEROCOPY || + !skb_zerocopy_notify_extend(tail, lo, len)) { + __skb_queue_tail(q, skb); + skb = NULL; + } + spin_unlock_irqrestore(&q->lock, flags); + + sk->sk_error_report(sk); + +release: + consume_skb(skb); + sock_put(sk); +} +EXPORT_SYMBOL_GPL(sock_zerocopy_callback); + +void sock_zerocopy_put(struct ubuf_info *uarg) +{ + if (uarg && refcount_dec_and_test(&uarg->refcnt)) { + if (uarg->callback) + uarg->callback(uarg, uarg->zerocopy); + else + consume_skb(skb_from_uarg(uarg)); + } +} +EXPORT_SYMBOL_GPL(sock_zerocopy_put); + +void sock_zerocopy_put_abort(struct ubuf_info *uarg) +{ + if (uarg) { + struct sock *sk = skb_from_uarg(uarg)->sk; + + atomic_dec(&sk->sk_zckey); + uarg->len--; + + sock_zerocopy_put(uarg); + } +} +EXPORT_SYMBOL_GPL(sock_zerocopy_put_abort); + +extern int __zerocopy_sg_from_iter(struct sock *sk, struct sk_buff *skb, + struct iov_iter *from, size_t length); + +int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb, + struct msghdr *msg, int len, + struct ubuf_info *uarg) +{ + struct ubuf_info *orig_uarg = skb_zcopy(skb); + struct iov_iter orig_iter = msg->msg_iter; + int err, orig_len = skb->len; + + /* An skb can only point to one uarg. This edge case happens when + * TCP appends to an skb, but zerocopy_realloc triggered a new alloc. + */ + if (orig_uarg && uarg != orig_uarg) + return -EEXIST; + + err = __zerocopy_sg_from_iter(sk, skb, &msg->msg_iter, len); + if (err == -EFAULT || (err == -EMSGSIZE && skb->len == orig_len)) { + /* Streams do not free skb on error. Reset to prev state. */ + msg->msg_iter = orig_iter; + ___pskb_trim(skb, orig_len); + return err; + } + + skb_zcopy_set(skb, uarg); + return skb->len - orig_len; +} +EXPORT_SYMBOL_GPL(skb_zerocopy_iter_stream); + +static int skb_zerocopy_clone(struct sk_buff *nskb, struct sk_buff *orig, + gfp_t gfp_mask) +{ + if (skb_zcopy(orig)) { + if (skb_zcopy(nskb)) { + /* !gfp_mask callers are verified to !skb_zcopy(nskb) */ + if (!gfp_mask) { + WARN_ON_ONCE(1); + return -ENOMEM; + } + if (skb_uarg(nskb) == skb_uarg(orig)) + return 0; + if (skb_copy_ubufs(nskb, GFP_ATOMIC)) + return -EIO; + } + skb_zcopy_set(nskb, skb_uarg(orig)); + } + return 0; +} + /** * skb_copy_ubufs - copy userspace skb frags buffers to kernel * @skb: the skb to modify @@ -958,15 +1175,19 @@ EXPORT_SYMBOL_GPL(skb_morph); */ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) { - int i; int num_frags = skb_shinfo(skb)->nr_frags; struct page *page, *head = NULL; - struct ubuf_info *uarg = skb_shinfo(skb)->destructor_arg; + int i, new_frags; + u32 d_off; - for (i = 0; i < num_frags; i++) { - u8 *vaddr; - skb_frag_t *f = &skb_shinfo(skb)->frags[i]; + if (!num_frags) + return 0; + + if (skb_shared(skb) || skb_unclone(skb, gfp_mask)) + return -EINVAL; + new_frags = (__skb_pagelen(skb) + PAGE_SIZE - 1) >> PAGE_SHIFT; + for (i = 0; i < new_frags; i++) { page = alloc_page(gfp_mask); if (!page) { while (head) { @@ -976,28 +1197,51 @@ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) } return -ENOMEM; } - vaddr = kmap_atomic(skb_frag_page(f)); - memcpy(page_address(page), - vaddr + f->page_offset, skb_frag_size(f)); - kunmap_atomic(vaddr); set_page_private(page, (unsigned long)head); head = page; } + page = head; + d_off = 0; + for (i = 0; i < num_frags; i++) { + skb_frag_t *f = &skb_shinfo(skb)->frags[i]; + u32 p_off, p_len, copied; + struct page *p; + u8 *vaddr; + + skb_frag_foreach_page(f, f->page_offset, skb_frag_size(f), + p, p_off, p_len, copied) { + u32 copy, done = 0; + vaddr = kmap_atomic(p); + + while (done < p_len) { + if (d_off == PAGE_SIZE) { + d_off = 0; + page = (struct page *)page_private(page); + } + copy = min_t(u32, PAGE_SIZE - d_off, p_len - done); + memcpy(page_address(page) + d_off, + vaddr + p_off + done, copy); + done += copy; + d_off += copy; + } + kunmap_atomic(vaddr); + } + } + /* skb frags release userspace buffers */ for (i = 0; i < num_frags; i++) skb_frag_unref(skb, i); - uarg->callback(uarg, false); - /* skb frags point to kernel buffers */ - for (i = num_frags - 1; i >= 0; i--) { - __skb_fill_page_desc(skb, i, head, 0, - skb_shinfo(skb)->frags[i].size); + for (i = 0; i < new_frags - 1; i++) { + __skb_fill_page_desc(skb, i, head, 0, PAGE_SIZE); head = (struct page *)page_private(head); } + __skb_fill_page_desc(skb, new_frags - 1, head, 0, d_off); + skb_shinfo(skb)->nr_frags = new_frags; - skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; + skb_zcopy_clear(skb, false); return 0; } EXPORT_SYMBOL_GPL(skb_copy_ubufs); @@ -1158,7 +1402,8 @@ struct sk_buff *__pskb_copy_fclone(struct sk_buff *skb, int headroom, if (skb_shinfo(skb)->nr_frags) { int i; - if (skb_orphan_frags(skb, gfp_mask)) { + if (skb_orphan_frags(skb, gfp_mask) || + skb_zerocopy_clone(n, skb, gfp_mask)) { kfree_skb(n); n = NULL; goto out; @@ -1235,9 +1480,10 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, * be since all we did is relocate the values */ if (skb_cloned(skb)) { - /* copy this zero copy skb frags */ if (skb_orphan_frags(skb, gfp_mask)) goto nofrags; + if (skb_zcopy(skb)) + refcount_inc(&skb_uarg(skb)->refcnt); for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) skb_frag_ref(skb, i); @@ -1363,18 +1609,20 @@ struct sk_buff *skb_copy_expand(const struct sk_buff *skb, EXPORT_SYMBOL(skb_copy_expand); /** - * skb_pad - zero pad the tail of an skb + * __skb_pad - zero pad the tail of an skb * @skb: buffer to pad * @pad: space to pad + * @free_on_error: free buffer on error * * Ensure that a buffer is followed by a padding area that is zero * filled. Used by network drivers which may DMA or transfer data * beyond the buffer end onto the wire. * - * May return error in out of memory cases. The skb is freed on error. + * May return error in out of memory cases. The skb is freed on error + * if @free_on_error is true. */ -int skb_pad(struct sk_buff *skb, int pad) +int __skb_pad(struct sk_buff *skb, int pad, bool free_on_error) { int err; int ntail; @@ -1403,10 +1651,11 @@ int skb_pad(struct sk_buff *skb, int pad) return 0; free_skb: - kfree_skb(skb); + if (free_on_error) + kfree_skb(skb); return err; } -EXPORT_SYMBOL(skb_pad); +EXPORT_SYMBOL(__skb_pad); /** * pskb_put - add data to the tail of a potentially fragmented buffer @@ -1719,6 +1968,8 @@ pull_pages: if (eat) { skb_shinfo(skb)->frags[k].page_offset += eat; skb_frag_size_sub(&skb_shinfo(skb)->frags[k], eat); + if (!i) + goto end; eat = 0; } k++; @@ -1726,9 +1977,13 @@ pull_pages: } skb_shinfo(skb)->nr_frags = k; +end: skb->tail += delta; skb->data_len -= delta; + if (!skb->data_len) + skb_zcopy_clear(skb, false); + return skb_tail_pointer(skb); } EXPORT_SYMBOL(__pskb_pull_tail); @@ -1776,16 +2031,20 @@ int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len) end = start + skb_frag_size(f); if ((copy = end - offset) > 0) { + u32 p_off, p_len, copied; + struct page *p; u8 *vaddr; if (copy > len) copy = len; - vaddr = kmap_atomic(skb_frag_page(f)); - memcpy(to, - vaddr + f->page_offset + offset - start, - copy); - kunmap_atomic(vaddr); + skb_frag_foreach_page(f, + f->page_offset + offset - start, + copy, p, p_off, p_len, copied) { + vaddr = kmap_atomic(p); + memcpy(to + copied, vaddr + p_off, p_len); + kunmap_atomic(vaddr); + } if ((len -= copy) == 0) return 0; @@ -2005,6 +2264,107 @@ int skb_splice_bits(struct sk_buff *skb, struct sock *sk, unsigned int offset, } EXPORT_SYMBOL_GPL(skb_splice_bits); +/* Send skb data on a socket. Socket must be locked. */ +int skb_send_sock_locked(struct sock *sk, struct sk_buff *skb, int offset, + int len) +{ + unsigned int orig_len = len; + struct sk_buff *head = skb; + unsigned short fragidx; + int slen, ret; + +do_frag_list: + + /* Deal with head data */ + while (offset < skb_headlen(skb) && len) { + struct kvec kv; + struct msghdr msg; + + slen = min_t(int, len, skb_headlen(skb) - offset); + kv.iov_base = skb->data + offset; + kv.iov_len = slen; + memset(&msg, 0, sizeof(msg)); + + ret = kernel_sendmsg_locked(sk, &msg, &kv, 1, slen); + if (ret <= 0) + goto error; + + offset += ret; + len -= ret; + } + + /* All the data was skb head? */ + if (!len) + goto out; + + /* Make offset relative to start of frags */ + offset -= skb_headlen(skb); + + /* Find where we are in frag list */ + for (fragidx = 0; fragidx < skb_shinfo(skb)->nr_frags; fragidx++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[fragidx]; + + if (offset < frag->size) + break; + + offset -= frag->size; + } + + for (; len && fragidx < skb_shinfo(skb)->nr_frags; fragidx++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[fragidx]; + + slen = min_t(size_t, len, frag->size - offset); + + while (slen) { + ret = kernel_sendpage_locked(sk, frag->page.p, + frag->page_offset + offset, + slen, MSG_DONTWAIT); + if (ret <= 0) + goto error; + + len -= ret; + offset += ret; + slen -= ret; + } + + offset = 0; + } + + if (len) { + /* Process any frag lists */ + + if (skb == head) { + if (skb_has_frag_list(skb)) { + skb = skb_shinfo(skb)->frag_list; + goto do_frag_list; + } + } else if (skb->next) { + skb = skb->next; + goto do_frag_list; + } + } + +out: + return orig_len - len; + +error: + return orig_len == len ? ret : orig_len - len; +} +EXPORT_SYMBOL_GPL(skb_send_sock_locked); + +/* Send skb data on a socket. */ +int skb_send_sock(struct sock *sk, struct sk_buff *skb, int offset, int len) +{ + int ret = 0; + + lock_sock(sk); + ret = skb_send_sock_locked(sk, skb, offset, len); + release_sock(sk); + + return ret; +} +EXPORT_SYMBOL_GPL(skb_send_sock); + /** * skb_store_bits - store bits from kernel buffer to skb * @skb: destination buffer @@ -2044,15 +2404,20 @@ int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len) end = start + skb_frag_size(frag); if ((copy = end - offset) > 0) { + u32 p_off, p_len, copied; + struct page *p; u8 *vaddr; if (copy > len) copy = len; - vaddr = kmap_atomic(skb_frag_page(frag)); - memcpy(vaddr + frag->page_offset + offset - start, - from, copy); - kunmap_atomic(vaddr); + skb_frag_foreach_page(frag, + frag->page_offset + offset - start, + copy, p, p_off, p_len, copied) { + vaddr = kmap_atomic(p); + memcpy(vaddr + p_off, from + copied, p_len); + kunmap_atomic(vaddr); + } if ((len -= copy) == 0) return 0; @@ -2117,20 +2482,27 @@ __wsum __skb_checksum(const struct sk_buff *skb, int offset, int len, end = start + skb_frag_size(frag); if ((copy = end - offset) > 0) { + u32 p_off, p_len, copied; + struct page *p; __wsum csum2; u8 *vaddr; if (copy > len) copy = len; - vaddr = kmap_atomic(skb_frag_page(frag)); - csum2 = ops->update(vaddr + frag->page_offset + - offset - start, copy, 0); - kunmap_atomic(vaddr); - csum = ops->combine(csum, csum2, pos, copy); + + skb_frag_foreach_page(frag, + frag->page_offset + offset - start, + copy, p, p_off, p_len, copied) { + vaddr = kmap_atomic(p); + csum2 = ops->update(vaddr + p_off, p_len, 0); + kunmap_atomic(vaddr); + csum = ops->combine(csum, csum2, pos, p_len); + pos += p_len; + } + if (!(len -= copy)) return csum; offset += copy; - pos += copy; } start = end; } @@ -2203,24 +2575,31 @@ __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, end = start + skb_frag_size(&skb_shinfo(skb)->frags[i]); if ((copy = end - offset) > 0) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + u32 p_off, p_len, copied; + struct page *p; __wsum csum2; u8 *vaddr; - skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; if (copy > len) copy = len; - vaddr = kmap_atomic(skb_frag_page(frag)); - csum2 = csum_partial_copy_nocheck(vaddr + - frag->page_offset + - offset - start, to, - copy, 0); - kunmap_atomic(vaddr); - csum = csum_block_add(csum, csum2, pos); + + skb_frag_foreach_page(frag, + frag->page_offset + offset - start, + copy, p, p_off, p_len, copied) { + vaddr = kmap_atomic(p); + csum2 = csum_partial_copy_nocheck(vaddr + p_off, + to + copied, + p_len, 0); + kunmap_atomic(vaddr); + csum = csum_block_add(csum, csum2, pos); + pos += p_len; + } + if (!(len -= copy)) return csum; offset += copy; to += copy; - pos += copy; } start = end; } @@ -2360,6 +2739,7 @@ skb_zerocopy(struct sk_buff *to, struct sk_buff *from, int len, int hlen) skb_tx_error(from); return -ENOMEM; } + skb_zerocopy_clone(to, from, GFP_ATOMIC); for (i = 0; i < skb_shinfo(from)->nr_frags; i++) { if (!len) @@ -2657,6 +3037,7 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len) skb_shinfo(skb1)->tx_flags |= skb_shinfo(skb)->tx_flags & SKBTX_SHARED_FRAG; + skb_zerocopy_clone(skb1, skb, 0); if (len < pos) /* Split line is inside header. */ skb_split_inside_header(skb, skb1, len, pos); else /* Second chunk has no header, nothing to copy. */ @@ -2700,6 +3081,8 @@ int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen) if (skb_headlen(skb)) return 0; + if (skb_zcopy(tgt) || skb_zcopy(skb)) + return 0; todo = shiftlen; from = 0; @@ -3273,6 +3656,8 @@ normal: skb_shinfo(nskb)->tx_flags |= skb_shinfo(head_skb)->tx_flags & SKBTX_SHARED_FRAG; + if (skb_zerocopy_clone(nskb, head_skb, GFP_ATOMIC)) + goto err; while (pos < offset + len) { if (i >= nfrags) { @@ -4396,6 +4781,8 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, if (skb_has_frag_list(to) || skb_has_frag_list(from)) return false; + if (skb_zcopy(to) || skb_zcopy(from)) + return false; if (skb_headlen(from) != 0) { struct page *page; diff --git a/net/core/sock.c b/net/core/sock.c index ac2a404c73eb..9b7b6bbb2a23 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -307,16 +307,6 @@ static struct lock_class_key af_wlock_keys[AF_MAX]; static struct lock_class_key af_elock_keys[AF_MAX]; static struct lock_class_key af_kern_callback_keys[AF_MAX]; -/* Take into consideration the size of the struct sk_buff overhead in the - * determination of these values, since that is non-constant across - * platforms. This makes socket queueing behavior and performance - * not depend upon such differences. - */ -#define _SK_MEM_PACKETS 256 -#define _SK_MEM_OVERHEAD SKB_TRUESIZE(256) -#define SK_WMEM_MAX (_SK_MEM_OVERHEAD * _SK_MEM_PACKETS) -#define SK_RMEM_MAX (_SK_MEM_OVERHEAD * _SK_MEM_PACKETS) - /* Run time adjustable parameters. */ __u32 sysctl_wmem_max __read_mostly = SK_WMEM_MAX; EXPORT_SYMBOL(sysctl_wmem_max); @@ -1055,6 +1045,20 @@ set_rcvbuf: if (val == 1) dst_negative_advice(sk); break; + + case SO_ZEROCOPY: + if (sk->sk_family != PF_INET && sk->sk_family != PF_INET6) + ret = -ENOTSUPP; + else if (sk->sk_protocol != IPPROTO_TCP) + ret = -ENOTSUPP; + else if (sk->sk_state != TCP_CLOSE) + ret = -EBUSY; + else if (val < 0 || val > 1) + ret = -EINVAL; + else + sock_valbool_flag(sk, SOCK_ZEROCOPY, valbool); + break; + default: ret = -ENOPROTOOPT; break; @@ -1383,6 +1387,10 @@ int sock_getsockopt(struct socket *sock, int level, int optname, v.val64 = sock_gen_cookie(sk); break; + case SO_ZEROCOPY: + v.val = sock_flag(sk, SOCK_ZEROCOPY); + break; + default: /* We implement the SO_SNDLOWAT etc to not be settable * (1003.1g 7). @@ -1670,6 +1678,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) atomic_set(&newsk->sk_drops, 0); newsk->sk_send_head = NULL; newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK; + atomic_set(&newsk->sk_zckey, 0); sock_reset_flag(newsk, SOCK_DONE); @@ -1757,7 +1766,7 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst) sk->sk_route_caps |= NETIF_F_GSO_SOFTWARE; sk->sk_route_caps &= ~sk->sk_route_nocaps; if (sk_can_gso(sk)) { - if (dst->header_len) { + if (dst->header_len && !xfrm_dst_offload_ok(dst)) { sk->sk_route_caps &= ~NETIF_F_GSO_MASK; } else { sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM; @@ -1923,6 +1932,33 @@ struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force, } EXPORT_SYMBOL(sock_wmalloc); +static void sock_ofree(struct sk_buff *skb) +{ + struct sock *sk = skb->sk; + + atomic_sub(skb->truesize, &sk->sk_omem_alloc); +} + +struct sk_buff *sock_omalloc(struct sock *sk, unsigned long size, + gfp_t priority) +{ + struct sk_buff *skb; + + /* small safe race: SKB_TRUESIZE may differ from final skb->truesize */ + if (atomic_read(&sk->sk_omem_alloc) + SKB_TRUESIZE(size) > + sysctl_optmem_max) + return NULL; + + skb = alloc_skb(size, priority); + if (!skb) + return NULL; + + atomic_add(skb->truesize, &sk->sk_omem_alloc); + skb->sk = sk; + skb->destructor = sock_ofree; + return skb; +} + /* * Allocate a memory block from the socket's option memory buffer. */ @@ -2408,9 +2444,6 @@ EXPORT_SYMBOL(__sk_mem_reclaim); int sk_set_peek_off(struct sock *sk, int val) { - if (val < 0) - return -EINVAL; - sk->sk_peek_off = val; return 0; } @@ -2500,6 +2533,12 @@ int sock_no_sendmsg(struct socket *sock, struct msghdr *m, size_t len) } EXPORT_SYMBOL(sock_no_sendmsg); +int sock_no_sendmsg_locked(struct sock *sk, struct msghdr *m, size_t len) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL(sock_no_sendmsg_locked); + int sock_no_recvmsg(struct socket *sock, struct msghdr *m, size_t len, int flags) { @@ -2528,6 +2567,22 @@ ssize_t sock_no_sendpage(struct socket *sock, struct page *page, int offset, siz } EXPORT_SYMBOL(sock_no_sendpage); +ssize_t sock_no_sendpage_locked(struct sock *sk, struct page *page, + int offset, size_t size, int flags) +{ + ssize_t res; + struct msghdr msg = {.msg_flags = flags}; + struct kvec iov; + char *kaddr = kmap(page); + + iov.iov_base = kaddr + offset; + iov.iov_len = size; + res = kernel_sendmsg_locked(sk, &msg, &iov, 1, size); + kunmap(page); + return res; +} +EXPORT_SYMBOL(sock_no_sendpage_locked); + /* * Default Socket Callbacks */ @@ -2673,6 +2728,7 @@ void sock_init_data(struct socket *sock, struct sock *sk) sk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT; sk->sk_stamp = SK_DEFAULT_STAMP; + atomic_set(&sk->sk_zckey, 0); #ifdef CONFIG_NET_RX_BUSY_POLL sk->sk_napi_id = 0; diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c index 733f523707ac..bae7d78aa068 100644 --- a/net/dcb/dcbnl.c +++ b/net/dcb/dcbnl.c @@ -1938,8 +1938,8 @@ static int __init dcbnl_init(void) { INIT_LIST_HEAD(&dcb_app_list); - rtnl_register(PF_UNSPEC, RTM_GETDCB, dcb_doit, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_SETDCB, dcb_doit, NULL, NULL); + rtnl_register(PF_UNSPEC, RTM_GETDCB, dcb_doit, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_SETDCB, dcb_doit, NULL, 0); return 0; } diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 1704948e6a12..f227f002c73d 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -1471,9 +1471,12 @@ int dccp_feat_init(struct sock *sk) * singleton values (which always leads to failure). * These settings can still (later) be overridden via sockopts. */ - if (ccid_get_builtin_ccids(&tx.val, &tx.len) || - ccid_get_builtin_ccids(&rx.val, &rx.len)) + if (ccid_get_builtin_ccids(&tx.val, &tx.len)) return -ENOBUFS; + if (ccid_get_builtin_ccids(&rx.val, &rx.len)) { + kfree(tx.val); + return -ENOBUFS; + } if (!dccp_feat_prefer(sysctl_dccp_tx_ccid, tx.val, tx.len) || !dccp_feat_prefer(sysctl_dccp_rx_ccid, rx.val, rx.len)) diff --git a/net/dccp/input.c b/net/dccp/input.c index 4a05d7876850..fa6be9750bb4 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -126,7 +126,7 @@ static int dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb) static u16 dccp_reset_code_convert(const u8 code) { - const u16 error_code[] = { + static const u16 error_code[] = { [DCCP_RESET_CODE_CLOSED] = 0, /* normal termination */ [DCCP_RESET_CODE_UNSPECIFIED] = 0, /* nothing known */ [DCCP_RESET_CODE_ABORTED] = ECONNRESET, diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index f85d901f4e3f..001c08696334 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -256,7 +256,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info) sk = __inet_lookup_established(net, &dccp_hashinfo, iph->daddr, dh->dccph_dport, iph->saddr, ntohs(dh->dccph_sport), - inet_iif(skb)); + inet_iif(skb), 0); if (!sk) { __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); return; @@ -631,6 +631,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) goto drop_and_free; inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT); + reqsk_put(req); return 0; drop_and_free: @@ -803,7 +804,7 @@ static int dccp_v4_rcv(struct sk_buff *skb) lookup: sk = __inet_lookup_skb(&dccp_hashinfo, skb, __dccp_hdr_len(dh), - dh->dccph_sport, dh->dccph_dport, &refcounted); + dh->dccph_sport, dh->dccph_dport, 0, &refcounted); if (!sk) { dccp_pr_debug("failed to look up flow ID in table and " "get corresponding socket\n"); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c376af5bfdfb..5df7857fc0f3 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -16,6 +16,7 @@ #include <linux/random.h> #include <linux/slab.h> #include <linux/xfrm.h> +#include <linux/string.h> #include <net/addrconf.h> #include <net/inet_common.h> @@ -30,6 +31,7 @@ #include <net/ip6_checksum.h> #include <net/xfrm.h> #include <net/secure_seq.h> +#include <net/sock.h> #include "dccp.h" #include "ipv6.h" @@ -89,7 +91,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, sk = __inet6_lookup_established(net, &dccp_hashinfo, &hdr->daddr, dh->dccph_dport, &hdr->saddr, ntohs(dh->dccph_sport), - inet6_iif(skb)); + inet6_iif(skb), 0); if (!sk) { __ICMP6_INC_STATS(net, __in6_dev_get(skb->dev), @@ -380,6 +382,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) goto drop_and_free; inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT); + reqsk_put(req); return 0; drop_and_free: @@ -596,19 +599,13 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) --ANK (980728) */ if (np->rxopt.all) - /* - * FIXME: Add handling of IPV6_PKTOPTIONS skb. See the comments below - * (wrt ipv6_pktopions) and net/ipv6/tcp_ipv6.c for an example. - */ opt_skb = skb_clone(skb, GFP_ATOMIC); if (sk->sk_state == DCCP_OPEN) { /* Fast path */ if (dccp_rcv_established(sk, skb, dccp_hdr(skb), skb->len)) goto reset; - if (opt_skb) { - /* XXX This is where we would goto ipv6_pktoptions. */ - __kfree_skb(opt_skb); - } + if (opt_skb) + goto ipv6_pktoptions; return 0; } @@ -639,10 +636,8 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (dccp_rcv_state_process(sk, skb, dccp_hdr(skb), skb->len)) goto reset; - if (opt_skb) { - /* XXX This is where we would goto ipv6_pktoptions. */ - __kfree_skb(opt_skb); - } + if (opt_skb) + goto ipv6_pktoptions; return 0; reset: @@ -652,6 +647,35 @@ discard: __kfree_skb(opt_skb); kfree_skb(skb); return 0; + +/* Handling IPV6_PKTOPTIONS skb the similar + * way it's done for net/ipv6/tcp_ipv6.c + */ +ipv6_pktoptions: + if (!((1 << sk->sk_state) & (DCCPF_CLOSED | DCCPF_LISTEN))) { + if (np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo) + np->mcast_oif = inet6_iif(opt_skb); + if (np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) + np->mcast_hops = ipv6_hdr(opt_skb)->hop_limit; + if (np->rxopt.bits.rxflow || np->rxopt.bits.rxtclass) + np->rcv_flowinfo = ip6_flowinfo(ipv6_hdr(opt_skb)); + if (np->repflow) + np->flow_label = ip6_flowlabel(ipv6_hdr(opt_skb)); + if (ipv6_opt_accepted(sk, opt_skb, + &DCCP_SKB_CB(opt_skb)->header.h6)) { + skb_set_owner_r(opt_skb, sk); + memmove(IP6CB(opt_skb), + &DCCP_SKB_CB(opt_skb)->header.h6, + sizeof(struct inet6_skb_parm)); + opt_skb = xchg(&np->pktoptions, opt_skb); + } else { + __kfree_skb(opt_skb); + opt_skb = xchg(&np->pktoptions, NULL); + } + } + + kfree_skb(opt_skb); + return 0; } static int dccp_v6_rcv(struct sk_buff *skb) @@ -686,7 +710,7 @@ static int dccp_v6_rcv(struct sk_buff *skb) lookup: sk = __inet6_lookup_skb(&dccp_hashinfo, skb, __dccp_hdr_len(dh), dh->dccph_sport, dh->dccph_dport, - inet6_iif(skb), &refcounted); + inet6_iif(skb), 0, &refcounted); if (!sk) { dccp_pr_debug("failed to look up flow ID in table and " "get corresponding socket\n"); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 9fe25bf63296..b68168fcc06a 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -24,6 +24,7 @@ #include <net/checksum.h> #include <net/inet_sock.h> +#include <net/inet_common.h> #include <net/sock.h> #include <net/xfrm.h> @@ -170,6 +171,15 @@ const char *dccp_packet_name(const int type) EXPORT_SYMBOL_GPL(dccp_packet_name); +static void dccp_sk_destruct(struct sock *sk) +{ + struct dccp_sock *dp = dccp_sk(sk); + + ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); + dp->dccps_hc_tx_ccid = NULL; + inet_sock_destruct(sk); +} + int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) { struct dccp_sock *dp = dccp_sk(sk); @@ -179,6 +189,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) icsk->icsk_syn_retries = sysctl_dccp_request_retries; sk->sk_state = DCCP_CLOSED; sk->sk_write_space = dccp_write_space; + sk->sk_destruct = dccp_sk_destruct; icsk->icsk_sync_mss = dccp_sync_mss; dp->dccps_mss_cache = 536; dp->dccps_rate_last = jiffies; @@ -201,10 +212,7 @@ void dccp_destroy_sock(struct sock *sk) { struct dccp_sock *dp = dccp_sk(sk); - /* - * DCCP doesn't use sk_write_queue, just sk_send_head - * for retransmissions - */ + __skb_queue_purge(&sk->sk_write_queue); if (sk->sk_send_head != NULL) { kfree_skb(sk->sk_send_head); sk->sk_send_head = NULL; @@ -222,8 +230,7 @@ void dccp_destroy_sock(struct sock *sk) dp->dccps_hc_rx_ackvec = NULL; } ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); - ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); - dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL; + dp->dccps_hc_rx_ccid = NULL; /* clean up feature negotiation state */ dccp_feat_list_purge(&dp->dccps_featneg); diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index fa0110b57ca1..4d339de56862 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -1419,9 +1419,9 @@ void __init dn_dev_init(void) dn_dev_devices_on(); - rtnl_register(PF_DECnet, RTM_NEWADDR, dn_nl_newaddr, NULL, NULL); - rtnl_register(PF_DECnet, RTM_DELADDR, dn_nl_deladdr, NULL, NULL); - rtnl_register(PF_DECnet, RTM_GETADDR, NULL, dn_nl_dump_ifaddr, NULL); + rtnl_register(PF_DECnet, RTM_NEWADDR, dn_nl_newaddr, NULL, 0); + rtnl_register(PF_DECnet, RTM_DELADDR, dn_nl_deladdr, NULL, 0); + rtnl_register(PF_DECnet, RTM_GETADDR, NULL, dn_nl_dump_ifaddr, 0); proc_create("decnet_dev", S_IRUGO, init_net.proc_net, &dn_dev_seq_fops); diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c index f9f6fb3f3c5b..3d37464c8b4a 100644 --- a/net/decnet/dn_fib.c +++ b/net/decnet/dn_fib.c @@ -791,8 +791,8 @@ void __init dn_fib_init(void) register_dnaddr_notifier(&dn_fib_dnaddr_notifier); - rtnl_register(PF_DECnet, RTM_NEWROUTE, dn_fib_rtm_newroute, NULL, NULL); - rtnl_register(PF_DECnet, RTM_DELROUTE, dn_fib_rtm_delroute, NULL, NULL); + rtnl_register(PF_DECnet, RTM_NEWROUTE, dn_fib_rtm_newroute, NULL, 0); + rtnl_register(PF_DECnet, RTM_DELROUTE, dn_fib_rtm_delroute, NULL, 0); } diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 21dedf6fd0f7..22bf0b95d6ed 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -94,7 +94,7 @@ struct neigh_table dn_neigh_table = { [NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ, [NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ, [NEIGH_VAR_GC_STALETIME] = 60 * HZ, - [NEIGH_VAR_QUEUE_LEN_BYTES] = 64*1024, + [NEIGH_VAR_QUEUE_LEN_BYTES] = SK_WMEM_MAX, [NEIGH_VAR_PROXY_QLEN] = 0, [NEIGH_VAR_ANYCAST_DELAY] = 0, [NEIGH_VAR_PROXY_DELAY] = 0, diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index bcbe548f8854..0bd3afd01dd2 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1922,10 +1922,10 @@ void __init dn_route_init(void) #ifdef CONFIG_DECNET_ROUTER rtnl_register(PF_DECnet, RTM_GETROUTE, dn_cache_getroute, - dn_fib_dump, NULL); + dn_fib_dump, 0); #else rtnl_register(PF_DECnet, RTM_GETROUTE, dn_cache_getroute, - dn_cache_dump, NULL); + dn_cache_dump, 0); #endif } diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index aa8ffecc46a4..ab395e55cd78 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -115,7 +115,7 @@ static inline void dnrmg_receive_user_skb(struct sk_buff *skb) RCV_SKB_FAIL(-EINVAL); } -static struct nf_hook_ops dnrmg_ops __read_mostly = { +static const struct nf_hook_ops dnrmg_ops = { .hook = dnrmg_hook, .pf = NFPROTO_DECNET, .hooknum = NF_DN_ROUTE, diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 416ac4ef9ba9..03c58b0eb082 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -67,17 +67,17 @@ const struct dsa_device_ops *dsa_device_ops[DSA_TAG_LAST] = { [DSA_TAG_PROTO_NONE] = &none_ops, }; -int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, - struct dsa_port *dport, int port) +int dsa_cpu_dsa_setup(struct dsa_port *port) { - struct device_node *port_dn = dport->dn; + struct device_node *port_dn = port->dn; + struct dsa_switch *ds = port->ds; struct phy_device *phydev; int ret, mode; if (of_phy_is_fixed_link(port_dn)) { ret = of_phy_register_fixed_link(port_dn); if (ret) { - dev_err(dev, "failed to register fixed PHY\n"); + dev_err(ds->dev, "failed to register fixed PHY\n"); return ret; } phydev = of_phy_find_device(port_dn); @@ -90,7 +90,7 @@ int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, genphy_config_init(phydev); genphy_read_status(phydev); if (ds->ops->adjust_link) - ds->ops->adjust_link(ds, port, phydev); + ds->ops->adjust_link(ds, port->index, phydev); put_device(&phydev->mdio.dev); } @@ -186,10 +186,12 @@ struct net_device *dsa_dev_to_net_device(struct device *dev) EXPORT_SYMBOL_GPL(dsa_dev_to_net_device); static int dsa_switch_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 *unused) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct sk_buff *nskb = NULL; + struct pcpu_sw_netstats *s; + struct dsa_slave_priv *p; if (unlikely(dst == NULL)) { kfree_skb(skb); @@ -200,19 +202,23 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev, if (!skb) return 0; - nskb = dst->rcv(skb, dev, pt, orig_dev); + nskb = dst->rcv(skb, dev, pt); if (!nskb) { kfree_skb(skb); return 0; } skb = nskb; + p = netdev_priv(skb->dev); skb_push(skb, ETH_HLEN); skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, skb->dev); - skb->dev->stats.rx_packets++; - skb->dev->stats.rx_bytes += skb->len; + s = this_cpu_ptr(p->stats64); + u64_stats_update_begin(&s->syncp); + s->rx_packets++; + s->rx_bytes += skb->len; + u64_stats_update_end(&s->syncp); netif_receive_skb(skb); @@ -220,6 +226,11 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev, } #ifdef CONFIG_PM_SLEEP +static bool dsa_is_port_initialized(struct dsa_switch *ds, int p) +{ + return ds->enabled_port_mask & (1 << p) && ds->ports[p].netdev; +} + int dsa_switch_suspend(struct dsa_switch *ds) { int i, ret = 0; @@ -271,10 +282,22 @@ static struct packet_type dsa_pack_type __read_mostly = { .func = dsa_switch_rcv, }; +static struct workqueue_struct *dsa_owq; + +bool dsa_schedule_work(struct work_struct *work) +{ + return queue_work(dsa_owq, work); +} + static int __init dsa_init_module(void) { int rc; + dsa_owq = alloc_ordered_workqueue("dsa_ordered", + WQ_MEM_RECLAIM); + if (!dsa_owq) + return -ENOMEM; + rc = dsa_slave_register_notifier(); if (rc) return rc; @@ -294,6 +317,7 @@ static void __exit dsa_cleanup_module(void) dsa_slave_unregister_notifier(); dev_remove_pack(&dsa_pack_type); dsa_legacy_unregister(); + destroy_workqueue(dsa_owq); } module_exit(dsa_cleanup_module); diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 56e46090526b..873af0108e24 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -219,7 +219,7 @@ static int dsa_dsa_port_apply(struct dsa_port *port) struct dsa_switch *ds = port->ds; int err; - err = dsa_cpu_dsa_setup(ds, ds->dev, port, port->index); + err = dsa_cpu_dsa_setup(port); if (err) { dev_warn(ds->dev, "Failed to setup dsa port %d: %d\n", port->index, err); @@ -243,7 +243,7 @@ static int dsa_cpu_port_apply(struct dsa_port *port) struct dsa_switch *ds = port->ds; int err; - err = dsa_cpu_dsa_setup(ds, ds->dev, port, port->index); + err = dsa_cpu_dsa_setup(port); if (err) { dev_warn(ds->dev, "Failed to setup cpu port %d: %d\n", port->index, err); @@ -275,7 +275,7 @@ static int dsa_user_port_apply(struct dsa_port *port) if (!name) name = "eth%d"; - err = dsa_slave_create(ds, ds->dev, port->index, name); + err = dsa_slave_create(port, name); if (err) { dev_warn(ds->dev, "Failed to create slave %d: %d\n", port->index, err); @@ -509,21 +509,22 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, dst->cpu_dp->netdev = ethernet_dev; } + /* Initialize cpu_port_mask now for drv->setup() + * to have access to a correct value, just like what + * net/dsa/dsa.c::dsa_switch_setup_one does. + */ + ds->cpu_port_mask |= BIT(index); + tag_protocol = ds->ops->get_tag_protocol(ds); dst->tag_ops = dsa_resolve_tag_protocol(tag_protocol); if (IS_ERR(dst->tag_ops)) { dev_warn(ds->dev, "No tagger for this switch\n"); + ds->cpu_port_mask &= ~BIT(index); return PTR_ERR(dst->tag_ops); } dst->rcv = dst->tag_ops->rcv; - /* Initialize cpu_port_mask now for drv->setup() - * to have access to a correct value, just like what - * net/dsa/dsa.c::dsa_switch_setup_one does. - */ - ds->cpu_port_mask |= BIT(index); - return 0; } @@ -576,7 +577,7 @@ static int dsa_dst_parse(struct dsa_switch_tree *dst) return err; } - if (!dst->cpu_dp->netdev) { + if (!dst->cpu_dp) { pr_warn("Tree has no master device\n"); return -EINVAL; } diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index 55982cc39b24..9c3eeb72462d 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -43,10 +43,10 @@ struct dsa_notifier_bridge_info { /* DSA_NOTIFIER_FDB_* */ struct dsa_notifier_fdb_info { - const struct switchdev_obj_port_fdb *fdb; - struct switchdev_trans *trans; int sw_index; int port; + const unsigned char *addr; + u16 vid; }; /* DSA_NOTIFIER_MDB_* */ @@ -65,18 +65,13 @@ struct dsa_notifier_vlan_info { int port; }; -struct dsa_device_ops { - struct sk_buff *(*xmit)(struct sk_buff *skb, struct net_device *dev); - struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev); -}; - struct dsa_slave_priv { /* Copy of dp->ds->dst->tag_ops->xmit for faster access in hot path */ struct sk_buff * (*xmit)(struct sk_buff *skb, struct net_device *dev); + struct pcpu_sw_netstats *stats64; + /* DSA port data, such as switch, port index, etc. */ struct dsa_port *dp; @@ -99,16 +94,23 @@ struct dsa_slave_priv { }; /* dsa.c */ -int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, - struct dsa_port *dport, int port); +int dsa_cpu_dsa_setup(struct dsa_port *port); void dsa_cpu_dsa_destroy(struct dsa_port *dport); const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol); int dsa_cpu_port_ethtool_setup(struct dsa_port *cpu_dp); void dsa_cpu_port_ethtool_restore(struct dsa_port *cpu_dp); +bool dsa_schedule_work(struct work_struct *work); /* legacy.c */ int dsa_legacy_register(void); void dsa_legacy_unregister(void); +int dsa_legacy_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], + struct net_device *dev, + const unsigned char *addr, u16 vid, + u16 flags); +int dsa_legacy_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], + struct net_device *dev, + const unsigned char *addr, u16 vid); /* port.c */ int dsa_port_set_state(struct dsa_port *dp, u8 state, @@ -120,35 +122,25 @@ int dsa_port_vlan_filtering(struct dsa_port *dp, bool vlan_filtering, struct switchdev_trans *trans); int dsa_port_ageing_time(struct dsa_port *dp, clock_t ageing_clock, struct switchdev_trans *trans); -int dsa_port_fdb_add(struct dsa_port *dp, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans); -int dsa_port_fdb_del(struct dsa_port *dp, - const struct switchdev_obj_port_fdb *fdb); -int dsa_port_fdb_dump(struct dsa_port *dp, struct switchdev_obj_port_fdb *fdb, - switchdev_obj_dump_cb_t *cb); +int dsa_port_fdb_add(struct dsa_port *dp, const unsigned char *addr, + u16 vid); +int dsa_port_fdb_del(struct dsa_port *dp, const unsigned char *addr, + u16 vid); int dsa_port_mdb_add(struct dsa_port *dp, const struct switchdev_obj_port_mdb *mdb, struct switchdev_trans *trans); int dsa_port_mdb_del(struct dsa_port *dp, const struct switchdev_obj_port_mdb *mdb); -int dsa_port_mdb_dump(struct dsa_port *dp, struct switchdev_obj_port_mdb *mdb, - switchdev_obj_dump_cb_t *cb); int dsa_port_vlan_add(struct dsa_port *dp, const struct switchdev_obj_port_vlan *vlan, struct switchdev_trans *trans); int dsa_port_vlan_del(struct dsa_port *dp, const struct switchdev_obj_port_vlan *vlan); -int dsa_port_vlan_dump(struct dsa_port *dp, - struct switchdev_obj_port_vlan *vlan, - switchdev_obj_dump_cb_t *cb); - /* slave.c */ extern const struct dsa_device_ops notag_netdev_ops; void dsa_slave_mii_bus_init(struct dsa_switch *ds); void dsa_cpu_port_ethtool_init(struct ethtool_ops *ops); -int dsa_slave_create(struct dsa_switch *ds, struct device *parent, - int port, const char *name); +int dsa_slave_create(struct dsa_port *port, const char *name); void dsa_slave_destroy(struct net_device *slave_dev); int dsa_slave_suspend(struct net_device *slave_dev); int dsa_slave_resume(struct net_device *slave_dev); diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 1d7a3282f2a7..91e6f7981d39 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -78,25 +78,23 @@ dsa_switch_probe(struct device *parent, struct device *host_dev, int sw_addr, } /* basic switch operations **************************************************/ -static int dsa_cpu_dsa_setups(struct dsa_switch *ds, struct device *dev) +static int dsa_cpu_dsa_setups(struct dsa_switch *ds) { - struct dsa_port *dport; int ret, port; for (port = 0; port < ds->num_ports; port++) { if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))) continue; - dport = &ds->ports[port]; - ret = dsa_cpu_dsa_setup(ds, dev, dport, port); + ret = dsa_cpu_dsa_setup(&ds->ports[port]); if (ret) return ret; } return 0; } -static int dsa_switch_setup_one(struct dsa_switch *ds, struct net_device *master, - struct device *parent) +static int dsa_switch_setup_one(struct dsa_switch *ds, + struct net_device *master) { const struct dsa_switch_ops *ops = ds->ops; struct dsa_switch_tree *dst = ds->dst; @@ -176,7 +174,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct net_device *master } if (!ds->slave_mii_bus && ops->phy_read) { - ds->slave_mii_bus = devm_mdiobus_alloc(parent); + ds->slave_mii_bus = devm_mdiobus_alloc(ds->dev); if (!ds->slave_mii_bus) return -ENOMEM; dsa_slave_mii_bus_init(ds); @@ -196,14 +194,14 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct net_device *master if (!(ds->enabled_port_mask & (1 << i))) continue; - ret = dsa_slave_create(ds, parent, i, cd->port_names[i]); + ret = dsa_slave_create(&ds->ports[i], cd->port_names[i]); if (ret < 0) netdev_err(master, "[%d]: can't create dsa slave device for port %d(%s): %d\n", index, i, cd->port_names[i], ret); } /* Perform configuration of the CPU and DSA ports */ - ret = dsa_cpu_dsa_setups(ds, parent); + ret = dsa_cpu_dsa_setups(ds); if (ret < 0) netdev_err(master, "[%d] : can't configure CPU and DSA ports\n", index); @@ -252,7 +250,7 @@ dsa_switch_setup(struct dsa_switch_tree *dst, struct net_device *master, ds->ops = ops; ds->priv = priv; - ret = dsa_switch_setup_one(ds, master, parent); + ret = dsa_switch_setup_one(ds, master); if (ret) return ERR_PTR(ret); @@ -741,6 +739,28 @@ static int dsa_resume(struct device *d) } #endif +/* legacy way, bypassing the bridge *****************************************/ +int dsa_legacy_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], + struct net_device *dev, + const unsigned char *addr, u16 vid, + u16 flags) +{ + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_port *dp = p->dp; + + return dsa_port_fdb_add(dp, addr, vid); +} + +int dsa_legacy_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], + struct net_device *dev, + const unsigned char *addr, u16 vid) +{ + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_port *dp = p->dp; + + return dsa_port_fdb_del(dp, addr, vid); +} + static SIMPLE_DEV_PM_OPS(dsa_pm_ops, dsa_suspend, dsa_resume); static const struct of_device_id dsa_of_match_table[] = { diff --git a/net/dsa/port.c b/net/dsa/port.c index efc3bce3a89d..659676ba3f8b 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -146,43 +146,33 @@ int dsa_port_ageing_time(struct dsa_port *dp, clock_t ageing_clock, return dsa_port_notify(dp, DSA_NOTIFIER_AGEING_TIME, &info); } -int dsa_port_fdb_add(struct dsa_port *dp, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans) +int dsa_port_fdb_add(struct dsa_port *dp, const unsigned char *addr, + u16 vid) { struct dsa_notifier_fdb_info info = { .sw_index = dp->ds->index, .port = dp->index, - .trans = trans, - .fdb = fdb, + .addr = addr, + .vid = vid, }; return dsa_port_notify(dp, DSA_NOTIFIER_FDB_ADD, &info); } -int dsa_port_fdb_del(struct dsa_port *dp, - const struct switchdev_obj_port_fdb *fdb) +int dsa_port_fdb_del(struct dsa_port *dp, const unsigned char *addr, + u16 vid) { struct dsa_notifier_fdb_info info = { .sw_index = dp->ds->index, .port = dp->index, - .fdb = fdb, + .addr = addr, + .vid = vid, + }; return dsa_port_notify(dp, DSA_NOTIFIER_FDB_DEL, &info); } -int dsa_port_fdb_dump(struct dsa_port *dp, struct switchdev_obj_port_fdb *fdb, - switchdev_obj_dump_cb_t *cb) -{ - struct dsa_switch *ds = dp->ds; - - if (ds->ops->port_fdb_dump) - return ds->ops->port_fdb_dump(ds, dp->index, fdb, cb); - - return -EOPNOTSUPP; -} - int dsa_port_mdb_add(struct dsa_port *dp, const struct switchdev_obj_port_mdb *mdb, struct switchdev_trans *trans) @@ -209,17 +199,6 @@ int dsa_port_mdb_del(struct dsa_port *dp, return dsa_port_notify(dp, DSA_NOTIFIER_MDB_DEL, &info); } -int dsa_port_mdb_dump(struct dsa_port *dp, struct switchdev_obj_port_mdb *mdb, - switchdev_obj_dump_cb_t *cb) -{ - struct dsa_switch *ds = dp->ds; - - if (ds->ops->port_mdb_dump) - return ds->ops->port_mdb_dump(ds, dp->index, mdb, cb); - - return -EOPNOTSUPP; -} - int dsa_port_vlan_add(struct dsa_port *dp, const struct switchdev_obj_port_vlan *vlan, struct switchdev_trans *trans) @@ -245,15 +224,3 @@ int dsa_port_vlan_del(struct dsa_port *dp, return dsa_port_notify(dp, DSA_NOTIFIER_VLAN_DEL, &info); } - -int dsa_port_vlan_dump(struct dsa_port *dp, - struct switchdev_obj_port_vlan *vlan, - switchdev_obj_dump_cb_t *cb) -{ - struct dsa_switch *ds = dp->ds; - - if (ds->ops->port_vlan_dump) - return ds->ops->port_vlan_dump(ds, dp->index, vlan, cb); - - return -EOPNOTSUPP; -} diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 9507bd38cf04..2afa99506f8b 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -199,6 +199,83 @@ out: return 0; } +struct dsa_slave_dump_ctx { + struct net_device *dev; + struct sk_buff *skb; + struct netlink_callback *cb; + int idx; +}; + +static int +dsa_slave_port_fdb_do_dump(const unsigned char *addr, u16 vid, + bool is_static, void *data) +{ + struct dsa_slave_dump_ctx *dump = data; + u32 portid = NETLINK_CB(dump->cb->skb).portid; + u32 seq = dump->cb->nlh->nlmsg_seq; + struct nlmsghdr *nlh; + struct ndmsg *ndm; + + if (dump->idx < dump->cb->args[2]) + goto skip; + + nlh = nlmsg_put(dump->skb, portid, seq, RTM_NEWNEIGH, + sizeof(*ndm), NLM_F_MULTI); + if (!nlh) + return -EMSGSIZE; + + ndm = nlmsg_data(nlh); + ndm->ndm_family = AF_BRIDGE; + ndm->ndm_pad1 = 0; + ndm->ndm_pad2 = 0; + ndm->ndm_flags = NTF_SELF; + ndm->ndm_type = 0; + ndm->ndm_ifindex = dump->dev->ifindex; + ndm->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE; + + if (nla_put(dump->skb, NDA_LLADDR, ETH_ALEN, addr)) + goto nla_put_failure; + + if (vid && nla_put_u16(dump->skb, NDA_VLAN, vid)) + goto nla_put_failure; + + nlmsg_end(dump->skb, nlh); + +skip: + dump->idx++; + return 0; + +nla_put_failure: + nlmsg_cancel(dump->skb, nlh); + return -EMSGSIZE; +} + +static int +dsa_slave_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb, + struct net_device *dev, struct net_device *filter_dev, + int *idx) +{ + struct dsa_slave_dump_ctx dump = { + .dev = dev, + .skb = skb, + .cb = cb, + .idx = *idx, + }; + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_port *dp = p->dp; + struct dsa_switch *ds = dp->ds; + int err; + + if (!ds->ops->port_fdb_dump) + return -EOPNOTSUPP; + + err = ds->ops->port_fdb_dump(ds, dp->index, + dsa_slave_port_fdb_do_dump, + &dump); + *idx = dump.idx; + return err; +} + static int dsa_slave_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { struct dsa_slave_priv *p = netdev_priv(dev); @@ -250,9 +327,6 @@ static int dsa_slave_port_obj_add(struct net_device *dev, */ switch (obj->id) { - case SWITCHDEV_OBJ_ID_PORT_FDB: - err = dsa_port_fdb_add(dp, SWITCHDEV_OBJ_PORT_FDB(obj), trans); - break; case SWITCHDEV_OBJ_ID_PORT_MDB: err = dsa_port_mdb_add(dp, SWITCHDEV_OBJ_PORT_MDB(obj), trans); break; @@ -276,9 +350,6 @@ static int dsa_slave_port_obj_del(struct net_device *dev, int err; switch (obj->id) { - case SWITCHDEV_OBJ_ID_PORT_FDB: - err = dsa_port_fdb_del(dp, SWITCHDEV_OBJ_PORT_FDB(obj)); - break; case SWITCHDEV_OBJ_ID_PORT_MDB: err = dsa_port_mdb_del(dp, SWITCHDEV_OBJ_PORT_MDB(obj)); break; @@ -293,32 +364,6 @@ static int dsa_slave_port_obj_del(struct net_device *dev, return err; } -static int dsa_slave_port_obj_dump(struct net_device *dev, - struct switchdev_obj *obj, - switchdev_obj_dump_cb_t *cb) -{ - struct dsa_slave_priv *p = netdev_priv(dev); - struct dsa_port *dp = p->dp; - int err; - - switch (obj->id) { - case SWITCHDEV_OBJ_ID_PORT_FDB: - err = dsa_port_fdb_dump(dp, SWITCHDEV_OBJ_PORT_FDB(obj), cb); - break; - case SWITCHDEV_OBJ_ID_PORT_MDB: - err = dsa_port_mdb_dump(dp, SWITCHDEV_OBJ_PORT_MDB(obj), cb); - break; - case SWITCHDEV_OBJ_ID_PORT_VLAN: - err = dsa_port_vlan_dump(dp, SWITCHDEV_OBJ_PORT_VLAN(obj), cb); - break; - default: - err = -EOPNOTSUPP; - break; - } - - return err; -} - static int dsa_slave_port_attr_get(struct net_device *dev, struct switchdev_attr *attr) { @@ -330,6 +375,9 @@ static int dsa_slave_port_attr_get(struct net_device *dev, attr->u.ppid.id_len = sizeof(ds->index); memcpy(&attr->u.ppid.id, &ds->index, attr->u.ppid.id_len); break; + case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS_SUPPORT: + attr->u.brport_flags_support = 0; + break; default: return -EOPNOTSUPP; } @@ -352,10 +400,14 @@ static inline netdev_tx_t dsa_netpoll_send_skb(struct dsa_slave_priv *p, static netdev_tx_t dsa_slave_xmit(struct sk_buff *skb, struct net_device *dev) { struct dsa_slave_priv *p = netdev_priv(dev); + struct pcpu_sw_netstats *s; struct sk_buff *nskb; - dev->stats.tx_packets++; - dev->stats.tx_bytes += skb->len; + s = this_cpu_ptr(p->stats64); + u64_stats_update_begin(&s->syncp); + s->tx_packets++; + s->tx_bytes += skb->len; + u64_stats_update_end(&s->syncp); /* Transmit function may have to reallocate the original SKB, * in which case it must have freed it. Only free it here on error. @@ -594,11 +646,26 @@ static void dsa_slave_get_ethtool_stats(struct net_device *dev, { struct dsa_slave_priv *p = netdev_priv(dev); struct dsa_switch *ds = p->dp->ds; - - data[0] = dev->stats.tx_packets; - data[1] = dev->stats.tx_bytes; - data[2] = dev->stats.rx_packets; - data[3] = dev->stats.rx_bytes; + struct pcpu_sw_netstats *s; + unsigned int start; + int i; + + for_each_possible_cpu(i) { + u64 tx_packets, tx_bytes, rx_packets, rx_bytes; + + s = per_cpu_ptr(p->stats64, i); + do { + start = u64_stats_fetch_begin_irq(&s->syncp); + tx_packets = s->tx_packets; + tx_bytes = s->tx_bytes; + rx_packets = s->rx_packets; + rx_bytes = s->rx_bytes; + } while (u64_stats_fetch_retry_irq(&s->syncp, start)); + data[0] += tx_packets; + data[1] += tx_bytes; + data[2] += rx_packets; + data[3] += rx_bytes; + } if (ds->ops->get_ethtool_stats) ds->ops->get_ethtool_stats(ds, p->dp->index, data + 4); } @@ -648,17 +715,24 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e) struct dsa_switch *ds = p->dp->ds; int ret; - if (!ds->ops->set_eee) + /* Port's PHY and MAC both need to be EEE capable */ + if (!p->phy) + return -ENODEV; + + if (!ds->ops->set_mac_eee) return -EOPNOTSUPP; - ret = ds->ops->set_eee(ds, p->dp->index, p->phy, e); + ret = ds->ops->set_mac_eee(ds, p->dp->index, e); if (ret) return ret; - if (p->phy) - ret = phy_ethtool_set_eee(p->phy, e); + if (e->eee_enabled) { + ret = phy_init_eee(p->phy, 0); + if (ret) + return ret; + } - return ret; + return phy_ethtool_set_eee(p->phy, e); } static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e) @@ -667,17 +741,18 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e) struct dsa_switch *ds = p->dp->ds; int ret; - if (!ds->ops->get_eee) + /* Port's PHY and MAC both need to be EEE capable */ + if (!p->phy) + return -ENODEV; + + if (!ds->ops->get_mac_eee) return -EOPNOTSUPP; - ret = ds->ops->get_eee(ds, p->dp->index, e); + ret = ds->ops->get_mac_eee(ds, p->dp->index, e); if (ret) return ret; - if (p->phy) - ret = phy_ethtool_get_eee(p->phy, e); - - return ret; + return phy_ethtool_get_eee(p->phy, e); } #ifdef CONFIG_NET_POLL_CONTROLLER @@ -747,12 +822,12 @@ dsa_slave_mall_tc_entry_find(struct dsa_slave_priv *p, } static int dsa_slave_add_cls_matchall(struct net_device *dev, - __be16 protocol, struct tc_cls_matchall_offload *cls, bool ingress) { struct dsa_slave_priv *p = netdev_priv(dev); struct dsa_mall_tc_entry *mall_tc_entry; + __be16 protocol = cls->common.protocol; struct dsa_switch *ds = p->dp->ds; struct net *net = dev_net(dev); struct dsa_slave_priv *to_p; @@ -765,7 +840,7 @@ static int dsa_slave_add_cls_matchall(struct net_device *dev, if (!ds->ops->port_mirror_add) return err; - if (!tc_single_action(cls->exts)) + if (!tcf_exts_has_one_action(cls->exts)) return err; tcf_exts_to_list(cls->exts, &actions); @@ -836,31 +911,71 @@ static void dsa_slave_del_cls_matchall(struct net_device *dev, kfree(mall_tc_entry); } -static int dsa_slave_setup_tc(struct net_device *dev, u32 handle, - u32 chain_index, __be16 protocol, - struct tc_to_netdev *tc) +static int dsa_slave_setup_tc_cls_matchall(struct net_device *dev, + struct tc_cls_matchall_offload *cls) { - bool ingress = TC_H_MAJ(handle) == TC_H_MAJ(TC_H_INGRESS); + bool ingress; - if (chain_index) + if (is_classid_clsact_ingress(cls->common.classid)) + ingress = true; + else if (is_classid_clsact_egress(cls->common.classid)) + ingress = false; + else return -EOPNOTSUPP; - switch (tc->type) { - case TC_SETUP_MATCHALL: - switch (tc->cls_mall->command) { - case TC_CLSMATCHALL_REPLACE: - return dsa_slave_add_cls_matchall(dev, protocol, - tc->cls_mall, - ingress); - case TC_CLSMATCHALL_DESTROY: - dsa_slave_del_cls_matchall(dev, tc->cls_mall); - return 0; - } + if (cls->common.chain_index) + return -EOPNOTSUPP; + + switch (cls->command) { + case TC_CLSMATCHALL_REPLACE: + return dsa_slave_add_cls_matchall(dev, cls, ingress); + case TC_CLSMATCHALL_DESTROY: + dsa_slave_del_cls_matchall(dev, cls); + return 0; default: return -EOPNOTSUPP; } } +static int dsa_slave_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) +{ + switch (type) { + case TC_SETUP_CLSMATCHALL: + return dsa_slave_setup_tc_cls_matchall(dev, type_data); + default: + return -EOPNOTSUPP; + } +} + +static void dsa_slave_get_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct dsa_slave_priv *p = netdev_priv(dev); + struct pcpu_sw_netstats *s; + unsigned int start; + int i; + + netdev_stats_to_stats64(stats, &dev->stats); + for_each_possible_cpu(i) { + u64 tx_packets, tx_bytes, rx_packets, rx_bytes; + + s = per_cpu_ptr(p->stats64, i); + do { + start = u64_stats_fetch_begin_irq(&s->syncp); + tx_packets = s->tx_packets; + tx_bytes = s->tx_bytes; + rx_packets = s->rx_packets; + rx_bytes = s->rx_bytes; + } while (u64_stats_fetch_retry_irq(&s->syncp, start)); + + stats->tx_packets += tx_packets; + stats->tx_bytes += tx_bytes; + stats->rx_packets += rx_packets; + stats->rx_bytes += rx_bytes; + } +} + void dsa_cpu_port_ethtool_init(struct ethtool_ops *ops) { ops->get_sset_count = dsa_cpu_port_get_sset_count; @@ -921,9 +1036,9 @@ static const struct net_device_ops dsa_slave_netdev_ops = { .ndo_change_rx_flags = dsa_slave_change_rx_flags, .ndo_set_rx_mode = dsa_slave_set_rx_mode, .ndo_set_mac_address = dsa_slave_set_mac_address, - .ndo_fdb_add = switchdev_port_fdb_add, - .ndo_fdb_del = switchdev_port_fdb_del, - .ndo_fdb_dump = switchdev_port_fdb_dump, + .ndo_fdb_add = dsa_legacy_fdb_add, + .ndo_fdb_del = dsa_legacy_fdb_del, + .ndo_fdb_dump = dsa_slave_fdb_dump, .ndo_do_ioctl = dsa_slave_ioctl, .ndo_get_iflink = dsa_slave_get_iflink, #ifdef CONFIG_NET_POLL_CONTROLLER @@ -931,11 +1046,9 @@ static const struct net_device_ops dsa_slave_netdev_ops = { .ndo_netpoll_cleanup = dsa_slave_netpoll_cleanup, .ndo_poll_controller = dsa_slave_poll_controller, #endif - .ndo_bridge_getlink = switchdev_port_bridge_getlink, - .ndo_bridge_setlink = switchdev_port_bridge_setlink, - .ndo_bridge_dellink = switchdev_port_bridge_dellink, .ndo_get_phys_port_name = dsa_slave_get_phys_port_name, .ndo_setup_tc = dsa_slave_setup_tc, + .ndo_get_stats64 = dsa_slave_get_stats64, }; static const struct switchdev_ops dsa_slave_switchdev_ops = { @@ -943,7 +1056,6 @@ static const struct switchdev_ops dsa_slave_switchdev_ops = { .switchdev_port_attr_set = dsa_slave_port_attr_set, .switchdev_port_obj_add = dsa_slave_port_obj_add, .switchdev_port_obj_del = dsa_slave_port_obj_del, - .switchdev_port_obj_dump = dsa_slave_port_obj_dump, }; static struct device_type dsa_type = { @@ -1134,9 +1246,9 @@ int dsa_slave_resume(struct net_device *slave_dev) return 0; } -int dsa_slave_create(struct dsa_switch *ds, struct device *parent, - int port, const char *name) +int dsa_slave_create(struct dsa_port *port, const char *name) { + struct dsa_switch *ds = port->ds; struct dsa_switch_tree *dst = ds->dst; struct net_device *master; struct net_device *slave_dev; @@ -1147,8 +1259,12 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, cpu_dp = ds->dst->cpu_dp; master = cpu_dp->netdev; - slave_dev = alloc_netdev(sizeof(struct dsa_slave_priv), name, - NET_NAME_UNKNOWN, ether_setup); + if (!ds->num_tx_queues) + ds->num_tx_queues = 1; + + slave_dev = alloc_netdev_mqs(sizeof(struct dsa_slave_priv), name, + NET_NAME_UNKNOWN, ether_setup, + ds->num_tx_queues, 1); if (slave_dev == NULL) return -ENOMEM; @@ -1166,12 +1282,17 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, netdev_for_each_tx_queue(slave_dev, dsa_slave_set_lockdep_class_one, NULL); - SET_NETDEV_DEV(slave_dev, parent); - slave_dev->dev.of_node = ds->ports[port].dn; + SET_NETDEV_DEV(slave_dev, port->ds->dev); + slave_dev->dev.of_node = port->dn; slave_dev->vlan_features = master->vlan_features; p = netdev_priv(slave_dev); - p->dp = &ds->ports[port]; + p->stats64 = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); + if (!p->stats64) { + free_netdev(slave_dev); + return -ENOMEM; + } + p->dp = port; INIT_LIST_HEAD(&p->mall_tc_list); p->xmit = dst->tag_ops->xmit; @@ -1179,12 +1300,13 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, p->old_link = -1; p->old_duplex = -1; - ds->ports[port].netdev = slave_dev; + port->netdev = slave_dev; ret = register_netdev(slave_dev); if (ret) { netdev_err(master, "error %d registering interface %s\n", ret, slave_dev->name); - ds->ports[port].netdev = NULL; + port->netdev = NULL; + free_percpu(p->stats64); free_netdev(slave_dev); return ret; } @@ -1195,6 +1317,7 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, if (ret) { netdev_err(master, "error %d setting up slave phy\n", ret); unregister_netdev(slave_dev); + free_percpu(p->stats64); free_netdev(slave_dev); return ret; } @@ -1217,6 +1340,7 @@ void dsa_slave_destroy(struct net_device *slave_dev) of_phy_deregister_fixed_link(port_dn); } unregister_netdev(slave_dev); + free_percpu(p->stats64); free_netdev(slave_dev); } @@ -1259,19 +1383,142 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb, return NOTIFY_DONE; } +struct dsa_switchdev_event_work { + struct work_struct work; + struct switchdev_notifier_fdb_info fdb_info; + struct net_device *dev; + unsigned long event; +}; + +static void dsa_slave_switchdev_event_work(struct work_struct *work) +{ + struct dsa_switchdev_event_work *switchdev_work = + container_of(work, struct dsa_switchdev_event_work, work); + struct net_device *dev = switchdev_work->dev; + struct switchdev_notifier_fdb_info *fdb_info; + struct dsa_slave_priv *p = netdev_priv(dev); + int err; + + rtnl_lock(); + switch (switchdev_work->event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + fdb_info = &switchdev_work->fdb_info; + err = dsa_port_fdb_add(p->dp, fdb_info->addr, fdb_info->vid); + if (err) { + netdev_dbg(dev, "fdb add failed err=%d\n", err); + break; + } + call_switchdev_notifiers(SWITCHDEV_FDB_OFFLOADED, dev, + &fdb_info->info); + break; + + case SWITCHDEV_FDB_DEL_TO_DEVICE: + fdb_info = &switchdev_work->fdb_info; + err = dsa_port_fdb_del(p->dp, fdb_info->addr, fdb_info->vid); + if (err) { + netdev_dbg(dev, "fdb del failed err=%d\n", err); + dev_close(dev); + } + break; + } + rtnl_unlock(); + + kfree(switchdev_work->fdb_info.addr); + kfree(switchdev_work); + dev_put(dev); +} + +static int +dsa_slave_switchdev_fdb_work_init(struct dsa_switchdev_event_work * + switchdev_work, + const struct switchdev_notifier_fdb_info * + fdb_info) +{ + memcpy(&switchdev_work->fdb_info, fdb_info, + sizeof(switchdev_work->fdb_info)); + switchdev_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC); + if (!switchdev_work->fdb_info.addr) + return -ENOMEM; + ether_addr_copy((u8 *)switchdev_work->fdb_info.addr, + fdb_info->addr); + return 0; +} + +/* Called under rcu_read_lock() */ +static int dsa_slave_switchdev_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = switchdev_notifier_info_to_dev(ptr); + struct dsa_switchdev_event_work *switchdev_work; + + if (!dsa_slave_dev_check(dev)) + return NOTIFY_DONE; + + switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC); + if (!switchdev_work) + return NOTIFY_BAD; + + INIT_WORK(&switchdev_work->work, + dsa_slave_switchdev_event_work); + switchdev_work->dev = dev; + switchdev_work->event = event; + + switch (event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: /* fall through */ + case SWITCHDEV_FDB_DEL_TO_DEVICE: + if (dsa_slave_switchdev_fdb_work_init(switchdev_work, + ptr)) + goto err_fdb_work_init; + dev_hold(dev); + break; + default: + kfree(switchdev_work); + return NOTIFY_DONE; + } + + dsa_schedule_work(&switchdev_work->work); + return NOTIFY_OK; + +err_fdb_work_init: + kfree(switchdev_work); + return NOTIFY_BAD; +} + static struct notifier_block dsa_slave_nb __read_mostly = { - .notifier_call = dsa_slave_netdevice_event, + .notifier_call = dsa_slave_netdevice_event, +}; + +static struct notifier_block dsa_slave_switchdev_notifier = { + .notifier_call = dsa_slave_switchdev_event, }; int dsa_slave_register_notifier(void) { - return register_netdevice_notifier(&dsa_slave_nb); + int err; + + err = register_netdevice_notifier(&dsa_slave_nb); + if (err) + return err; + + err = register_switchdev_notifier(&dsa_slave_switchdev_notifier); + if (err) + goto err_switchdev_nb; + + return 0; + +err_switchdev_nb: + unregister_netdevice_notifier(&dsa_slave_nb); + return err; } void dsa_slave_unregister_notifier(void) { int err; + err = unregister_switchdev_notifier(&dsa_slave_switchdev_notifier); + if (err) + pr_err("DSA: failed to unregister switchdev notifier (%d)\n", err); + err = unregister_netdevice_notifier(&dsa_slave_nb); if (err) pr_err("DSA: failed to unregister slave notifier (%d)\n", err); diff --git a/net/dsa/switch.c b/net/dsa/switch.c index 97e2e9c8cf3f..e6c06aa349a6 100644 --- a/net/dsa/switch.c +++ b/net/dsa/switch.c @@ -83,30 +83,20 @@ static int dsa_switch_bridge_leave(struct dsa_switch *ds, static int dsa_switch_fdb_add(struct dsa_switch *ds, struct dsa_notifier_fdb_info *info) { - const struct switchdev_obj_port_fdb *fdb = info->fdb; - struct switchdev_trans *trans = info->trans; - /* Do not care yet about other switch chips of the fabric */ if (ds->index != info->sw_index) return 0; - if (switchdev_trans_ph_prepare(trans)) { - if (!ds->ops->port_fdb_prepare || !ds->ops->port_fdb_add) - return -EOPNOTSUPP; - - return ds->ops->port_fdb_prepare(ds, info->port, fdb, trans); - } - - ds->ops->port_fdb_add(ds, info->port, fdb, trans); + if (!ds->ops->port_fdb_add) + return -EOPNOTSUPP; - return 0; + return ds->ops->port_fdb_add(ds, info->port, info->addr, + info->vid); } static int dsa_switch_fdb_del(struct dsa_switch *ds, struct dsa_notifier_fdb_info *info) { - const struct switchdev_obj_port_fdb *fdb = info->fdb; - /* Do not care yet about other switch chips of the fabric */ if (ds->index != info->sw_index) return 0; @@ -114,7 +104,8 @@ static int dsa_switch_fdb_del(struct dsa_switch *ds, if (!ds->ops->port_fdb_del) return -EOPNOTSUPP; - return ds->ops->port_fdb_del(ds, info->port, fdb); + return ds->ops->port_fdb_del(ds, info->port, info->addr, + info->vid); } static int dsa_switch_mdb_add(struct dsa_switch *ds, diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index c697d9815177..dbb016434ace 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -62,6 +62,7 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev) { struct dsa_slave_priv *p = netdev_priv(dev); + u16 queue = skb_get_queue_mapping(skb); u8 *brcm_tag; if (skb_cow_head(skb, BRCM_TAG_LEN) < 0) @@ -78,7 +79,7 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev * deprecated */ brcm_tag[0] = (1 << BRCM_OPCODE_SHIFT) | - ((skb->priority << BRCM_IG_TC_SHIFT) & BRCM_IG_TC_MASK); + ((queue & BRCM_IG_TC_MASK) << BRCM_IG_TC_SHIFT); brcm_tag[1] = 0; brcm_tag[2] = 0; if (p->dp->index == 8) @@ -89,8 +90,7 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev } static struct sk_buff *brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c index 12867a4b458f..fbf9ca954773 100644 --- a/net/dsa/tag_dsa.c +++ b/net/dsa/tag_dsa.c @@ -65,8 +65,7 @@ static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *dsa_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_switch *ds; diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c index 67a9d26f9075..76367ba1b2e2 100644 --- a/net/dsa/tag_edsa.c +++ b/net/dsa/tag_edsa.c @@ -78,8 +78,7 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *edsa_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_switch *ds; diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c index fab41de8e983..010ca0a336c4 100644 --- a/net/dsa/tag_ksz.c +++ b/net/dsa/tag_ksz.c @@ -42,6 +42,10 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) padlen = (skb->len >= ETH_ZLEN) ? 0 : ETH_ZLEN - skb->len; if (skb_tailroom(skb) >= padlen + KSZ_INGRESS_TAG_LEN) { + /* Let dsa_slave_xmit() free skb */ + if (__skb_put_padto(skb, skb->len + padlen, false)) + return NULL; + nskb = skb; } else { nskb = alloc_skb(NET_IP_ALIGN + skb->len + @@ -56,12 +60,15 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) skb_set_transport_header(nskb, skb_transport_header(skb) - skb->head); skb_copy_and_csum_dev(skb, skb_put(nskb, skb->len)); - kfree_skb(skb); - } - /* skb is freed when it fails */ - if (skb_put_padto(nskb, nskb->len + padlen)) - return NULL; + /* Let skb_put_padto() free nskb, and let dsa_slave_xmit() free + * skb + */ + if (skb_put_padto(nskb, nskb->len + padlen)) + return NULL; + + consume_skb(skb); + } tag = skb_put(nskb, KSZ_INGRESS_TAG_LEN); tag[0] = 0; @@ -71,8 +78,7 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *ksz_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); diff --git a/net/dsa/tag_lan9303.c b/net/dsa/tag_lan9303.c index 247774d149f9..0b9826105e42 100644 --- a/net/dsa/tag_lan9303.c +++ b/net/dsa/tag_lan9303.c @@ -39,7 +39,6 @@ */ #define LAN9303_TAG_LEN 4 -#define LAN9303_MAX_PORTS 3 static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev) { @@ -69,7 +68,7 @@ static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, struct net_device *orig_dev) + struct packet_type *pt) { u16 *lan9303_tag; struct dsa_switch_tree *dst = dev->dsa_ptr; @@ -104,7 +103,7 @@ static struct sk_buff *lan9303_rcv(struct sk_buff *skb, struct net_device *dev, source_port = ntohs(lan9303_tag[1]) & 0x3; - if (source_port >= LAN9303_MAX_PORTS) { + if (source_port >= ds->num_ports) { dev_warn_ratelimited(&dev->dev, "Dropping packet due to invalid source port\n"); return NULL; } diff --git a/net/dsa/tag_mtk.c b/net/dsa/tag_mtk.c index 2f32b7ea3365..ec8ee5f43255 100644 --- a/net/dsa/tag_mtk.c +++ b/net/dsa/tag_mtk.c @@ -44,8 +44,7 @@ static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb, } static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_switch *ds; @@ -87,7 +86,17 @@ static struct sk_buff *mtk_tag_rcv(struct sk_buff *skb, struct net_device *dev, return skb; } +static int mtk_tag_flow_dissect(const struct sk_buff *skb, __be16 *proto, + int *offset) +{ + *offset = 4; + *proto = ((__be16 *)skb->data)[1]; + + return 0; +} + const struct dsa_device_ops mtk_netdev_ops = { - .xmit = mtk_tag_xmit, - .rcv = mtk_tag_rcv, + .xmit = mtk_tag_xmit, + .rcv = mtk_tag_rcv, + .flow_dissect = mtk_tag_flow_dissect, }; diff --git a/net/dsa/tag_qca.c b/net/dsa/tag_qca.c index 1867a3d11f28..1d4c70711c0f 100644 --- a/net/dsa/tag_qca.c +++ b/net/dsa/tag_qca.c @@ -63,8 +63,7 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *qca_tag_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c index b09e56214005..d2fd4923aa3e 100644 --- a/net/dsa/tag_trailer.c +++ b/net/dsa/tag_trailer.c @@ -40,7 +40,7 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev) skb_set_network_header(nskb, skb_network_header(skb) - skb->head); skb_set_transport_header(nskb, skb_transport_header(skb) - skb->head); skb_copy_and_csum_dev(skb, skb_put(nskb, skb->len)); - kfree_skb(skb); + consume_skb(skb); if (padlen) { skb_put_zero(nskb, padlen); @@ -56,8 +56,7 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev) } static struct sk_buff *trailer_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt, - struct net_device *orig_dev) + struct packet_type *pt) { struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index 4e7bdb213cd0..172d8309f89e 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -314,7 +314,8 @@ static void send_hsr_supervision_frame(struct hsr_port *master, hsr_sp = skb_put(skb, sizeof(struct hsr_sup_payload)); ether_addr_copy(hsr_sp->MacAddressA, master->dev->dev_addr); - skb_put_padto(skb, ETH_ZLEN + HSR_HLEN); + if (skb_put_padto(skb, ETH_ZLEN + HSR_HLEN)) + return; hsr_forward_skb(skb, master); return; diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index de2661cd0328..974765b7d92a 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -54,7 +54,7 @@ static int open_count; -static struct header_ops lowpan_header_ops = { +static const struct header_ops lowpan_header_ops = { .create = lowpan_header_create, }; diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 30d875dff6b5..f85b08baff16 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -580,19 +580,14 @@ static int __net_init lowpan_frags_init_net(struct net *net) { struct netns_ieee802154_lowpan *ieee802154_lowpan = net_ieee802154_lowpan(net); - int res; ieee802154_lowpan->frags.high_thresh = IPV6_FRAG_HIGH_THRESH; ieee802154_lowpan->frags.low_thresh = IPV6_FRAG_LOW_THRESH; ieee802154_lowpan->frags.timeout = IPV6_FRAG_TIMEOUT; - res = inet_frags_init_net(&ieee802154_lowpan->frags); - if (res) - return res; - res = lowpan_frags_ns_sysctl_register(net); - if (res) - inet_frags_uninit_net(&ieee802154_lowpan->frags); - return res; + inet_frags_init_net(&ieee802154_lowpan->frags); + + return lowpan_frags_ns_sysctl_register(net); } static void __net_exit lowpan_frags_exit_net(struct net *net) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 76c2077c3f5b..e31108e5ef79 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -944,6 +944,8 @@ const struct proto_ops inet_stream_ops = { .sendpage = inet_sendpage, .splice_read = tcp_splice_read, .read_sock = tcp_read_sock, + .sendmsg_locked = tcp_sendmsg_locked, + .sendpage_locked = tcp_sendpage_locked, .peek_len = tcp_peek_len, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_sock_common_setsockopt, @@ -1219,10 +1221,9 @@ EXPORT_SYMBOL(inet_sk_rebuild_header); struct sk_buff *inet_gso_segment(struct sk_buff *skb, netdev_features_t features) { - bool udpfrag = false, fixedid = false, gso_partial, encap; + bool fixedid = false, gso_partial, encap; struct sk_buff *segs = ERR_PTR(-EINVAL); const struct net_offload *ops; - unsigned int offset = 0; struct iphdr *iph; int proto, tot_len; int nhoff; @@ -1257,7 +1258,6 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb, segs = ERR_PTR(-EPROTONOSUPPORT); if (!skb->encapsulation || encap) { - udpfrag = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP); fixedid = !!(skb_shinfo(skb)->gso_type & SKB_GSO_TCP_FIXEDID); /* fixed ID is invalid if DF bit is not set */ @@ -1277,13 +1277,7 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb, skb = segs; do { iph = (struct iphdr *)(skb_mac_header(skb) + nhoff); - if (udpfrag) { - iph->frag_off = htons(offset >> 3); - if (skb->next) - iph->frag_off |= htons(IP_MF); - offset += skb->len - nhoff - ihl; - tot_len = skb->len - nhoff; - } else if (skb_is_gso(skb)) { + if (skb_is_gso(skb)) { if (!fixedid) { iph->id = htons(id); id += skb_shinfo(skb)->gso_segs; @@ -1602,6 +1596,9 @@ static const struct net_protocol igmp_protocol = { }; #endif +/* thinking of making this const? Don't. + * early_demux can change based on sysctl. + */ static struct net_protocol tcp_protocol = { .early_demux = tcp_v4_early_demux, .early_demux_handler = tcp_v4_early_demux, @@ -1612,6 +1609,9 @@ static struct net_protocol tcp_protocol = { .icmp_strict_tag_validation = 1, }; +/* thinking of making this const? Don't. + * early_demux can change based on sysctl. + */ static struct net_protocol udp_protocol = { .early_demux = udp_v4_early_demux, .early_demux_handler = udp_v4_early_demux, @@ -1731,6 +1731,13 @@ static __net_init int inet_init_net(struct net *net) net->ipv4.sysctl_ip_prot_sock = PROT_SOCK; #endif + /* Some igmp sysctl, whose values are always used */ + net->ipv4.sysctl_igmp_max_memberships = 20; + net->ipv4.sysctl_igmp_max_msf = 10; + /* IGMP reports for link-local multicast groups are enabled by default */ + net->ipv4.sysctl_igmp_llm_reports = 1; + net->ipv4.sysctl_igmp_qrv = 2; + return 0; } @@ -1771,6 +1778,11 @@ static const struct net_offload ipip_offload = { }, }; +static int __init ipip_offload_init(void) +{ + return inet_add_offload(&ipip_offload, IPPROTO_IPIP); +} + static int __init ipv4_offload_init(void) { /* @@ -1780,9 +1792,10 @@ static int __init ipv4_offload_init(void) pr_crit("%s: Cannot add UDP protocol offload\n", __func__); if (tcpv4_offload_init() < 0) pr_crit("%s: Cannot add TCP protocol offload\n", __func__); + if (ipip_offload_init() < 0) + pr_crit("%s: Cannot add IPIP protocol offload\n", __func__); dev_add_offload(&ip_packet_offload); - inet_add_offload(&ipip_offload, IPPROTO_IPIP); return 0; } diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 8b52179ddc6e..7c45b8896709 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -171,7 +171,7 @@ struct neigh_table arp_tbl = { [NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ, [NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ, [NEIGH_VAR_GC_STALETIME] = 60 * HZ, - [NEIGH_VAR_QUEUE_LEN_BYTES] = 64 * 1024, + [NEIGH_VAR_QUEUE_LEN_BYTES] = SK_WMEM_MAX, [NEIGH_VAR_PROXY_QLEN] = 64, [NEIGH_VAR_ANYCAST_DELAY] = 1 * HZ, [NEIGH_VAR_PROXY_DELAY] = (8 * HZ) / 10, diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index c4c6e1969ed0..2ae8f54cb321 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -1523,9 +1523,17 @@ unsigned char *cipso_v4_optptr(const struct sk_buff *skb) int taglen; for (optlen = iph->ihl*4 - sizeof(struct iphdr); optlen > 0; ) { - if (optptr[0] == IPOPT_CIPSO) + switch (optptr[0]) { + case IPOPT_CIPSO: return optptr; - taglen = optptr[1]; + case IPOPT_END: + return NULL; + case IPOPT_NOOP: + taglen = 1; + break; + default: + taglen = optptr[1]; + } optlen -= taglen; optptr += taglen; } diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 38d9af9b917c..d7adc0616599 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2491,9 +2491,9 @@ void __init devinet_init(void) rtnl_af_register(&inet_af_ops); - rtnl_register(PF_INET, RTM_NEWADDR, inet_rtm_newaddr, NULL, NULL); - rtnl_register(PF_INET, RTM_DELADDR, inet_rtm_deladdr, NULL, NULL); - rtnl_register(PF_INET, RTM_GETADDR, NULL, inet_dump_ifaddr, NULL); + rtnl_register(PF_INET, RTM_NEWADDR, inet_rtm_newaddr, NULL, 0); + rtnl_register(PF_INET, RTM_DELADDR, inet_rtm_deladdr, NULL, 0); + rtnl_register(PF_INET, RTM_GETADDR, NULL, inet_dump_ifaddr, 0); rtnl_register(PF_INET, RTM_GETNETCONF, inet_netconf_get_devconf, - inet_netconf_dump_devconf, NULL); + inet_netconf_dump_devconf, 0); } diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 0cbee0a666ff..b00e4a43b4dc 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -258,7 +258,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * esp_output_udp_encap(x, skb, esp); if (!skb_cloned(skb)) { - if (tailen <= skb_availroom(skb)) { + if (tailen <= skb_tailroom(skb)) { nfrags = 1; trailer = skb; tail = skb_tail_pointer(trailer); @@ -292,8 +292,6 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * kunmap_atomic(vaddr); - spin_unlock_bh(&x->lock); - nfrags = skb_shinfo(skb)->nr_frags; __skb_fill_page_desc(skb, nfrags, page, pfrag->offset, @@ -301,6 +299,9 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * skb_shinfo(skb)->nr_frags = ++nfrags; pfrag->offset = pfrag->offset + allocsize; + + spin_unlock_bh(&x->lock); + nfrags++; skb->len += tailen; @@ -381,7 +382,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; if (!esp->inplace) { int allocsize; @@ -392,7 +393,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * spin_lock_bh(&x->lock); if (unlikely(!skb_page_frag_refill(allocsize, pfrag, GFP_ATOMIC))) { spin_unlock_bh(&x->lock); - goto error; + goto error_free; } skb_shinfo(skb)->nr_frags = 1; @@ -409,7 +410,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; } if ((x->props.flags & XFRM_STATE_ESN)) @@ -442,8 +443,9 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * if (sg != dsg) esp_ssg_unref(x, tmp); - kfree(tmp); +error_free: + kfree(tmp); error: return err; } @@ -499,18 +501,59 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) return esp_output_tail(x, skb, &esp); } +static inline int esp_remove_trailer(struct sk_buff *skb) +{ + struct xfrm_state *x = xfrm_input_state(skb); + struct xfrm_offload *xo = xfrm_offload(skb); + struct crypto_aead *aead = x->data; + int alen, hlen, elen; + int padlen, trimlen; + __wsum csumdiff; + u8 nexthdr[2]; + int ret; + + alen = crypto_aead_authsize(aead); + hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); + elen = skb->len - hlen; + + if (xo && (xo->flags & XFRM_ESP_NO_TRAILER)) { + ret = xo->proto; + goto out; + } + + if (skb_copy_bits(skb, skb->len - alen - 2, nexthdr, 2)) + BUG(); + + ret = -EINVAL; + padlen = nexthdr[0]; + if (padlen + 2 + alen >= elen) { + net_dbg_ratelimited("ipsec esp packet is garbage padlen=%d, elen=%d\n", + padlen + 2, elen - alen); + goto out; + } + + trimlen = alen + padlen + 2; + if (skb->ip_summed == CHECKSUM_COMPLETE) { + csumdiff = skb_checksum(skb, skb->len - trimlen, trimlen, 0); + skb->csum = csum_block_sub(skb->csum, csumdiff, + skb->len - trimlen); + } + pskb_trim(skb, skb->len - trimlen); + + ret = nexthdr[1]; + +out: + return ret; +} + int esp_input_done2(struct sk_buff *skb, int err) { const struct iphdr *iph; struct xfrm_state *x = xfrm_input_state(skb); struct xfrm_offload *xo = xfrm_offload(skb); struct crypto_aead *aead = x->data; - int alen = crypto_aead_authsize(aead); int hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); - int elen = skb->len - hlen; int ihl; - u8 nexthdr[2]; - int padlen; if (!xo || (xo && !(xo->flags & CRYPTO_DONE))) kfree(ESP_SKB_CB(skb)->tmp); @@ -518,16 +561,10 @@ int esp_input_done2(struct sk_buff *skb, int err) if (unlikely(err)) goto out; - if (skb_copy_bits(skb, skb->len-alen-2, nexthdr, 2)) - BUG(); - - err = -EINVAL; - padlen = nexthdr[0]; - if (padlen + 2 + alen >= elen) + err = esp_remove_trailer(skb); + if (unlikely(err < 0)) goto out; - /* ... check padding bits here. Silly. :-) */ - iph = ip_hdr(skb); ihl = iph->ihl * 4; @@ -568,15 +605,12 @@ int esp_input_done2(struct sk_buff *skb, int err) skb->ip_summed = CHECKSUM_UNNECESSARY; } - pskb_trim(skb, skb->len - alen - padlen - 2); - __skb_pull(skb, hlen); + skb_pull_rcsum(skb, hlen); if (x->props.mode == XFRM_MODE_TUNNEL) skb_reset_transport_header(skb); else skb_set_transport_header(skb, -ihl); - err = nexthdr[1]; - /* RFC4303: Drop dummy packets without any error */ if (err == IPPROTO_NONE) err = -EINVAL; @@ -695,8 +729,10 @@ skip_cow: sg_init_table(sg, nfrags); err = skb_to_sgvec(skb, sg, 0, skb->len); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + kfree(tmp); goto out; + } skb->ip_summed = CHECKSUM_NONE; diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c index e0666016a764..f8b918c766b0 100644 --- a/net/ipv4/esp4_offload.c +++ b/net/ipv4/esp4_offload.c @@ -182,11 +182,13 @@ out: static int esp_input_tail(struct xfrm_state *x, struct sk_buff *skb) { struct crypto_aead *aead = x->data; + struct xfrm_offload *xo = xfrm_offload(skb); if (!pskb_may_pull(skb, sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead))) return -EINVAL; - skb->ip_summed = CHECKSUM_NONE; + if (!(xo->flags & CRYPTO_DONE)) + skb->ip_summed = CHECKSUM_NONE; return esp_input_done2(skb, 0); } @@ -257,7 +259,7 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features_ esp.seqno = cpu_to_be64(xo->seq.low + ((u64)xo->seq.hi << 32)); err = esp_output_tail(x, skb, &esp); - if (err < 0) + if (err) return err; secpath_reset(skb); @@ -303,3 +305,4 @@ module_init(esp4_offload_init); module_exit(esp4_offload_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Steffen Klassert <steffen.klassert@secunet.com>"); +MODULE_ALIAS_XFRM_OFFLOAD_TYPE(AF_INET, XFRM_PROTO_ESP); diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 4e678fa892dd..37819ab4cc74 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -1247,22 +1247,28 @@ static int __net_init ip_fib_net_init(struct net *net) int err; size_t size = sizeof(struct hlist_head) * FIB_TABLE_HASHSZ; - net->ipv4.fib_seq = 0; + err = fib4_notifier_init(net); + if (err) + return err; /* Avoid false sharing : Use at least a full cache line */ size = max_t(size_t, size, L1_CACHE_BYTES); net->ipv4.fib_table_hash = kzalloc(size, GFP_KERNEL); - if (!net->ipv4.fib_table_hash) - return -ENOMEM; + if (!net->ipv4.fib_table_hash) { + err = -ENOMEM; + goto err_table_hash_alloc; + } err = fib4_rules_init(net); if (err < 0) - goto fail; + goto err_rules_init; return 0; -fail: +err_rules_init: kfree(net->ipv4.fib_table_hash); +err_table_hash_alloc: + fib4_notifier_exit(net); return err; } @@ -1292,6 +1298,7 @@ static void ip_fib_net_exit(struct net *net) #endif rtnl_unlock(); kfree(net->ipv4.fib_table_hash); + fib4_notifier_exit(net); } static int __net_init fib_net_init(struct net *net) @@ -1334,13 +1341,14 @@ static struct pernet_operations fib_net_ops = { void __init ip_fib_init(void) { - rtnl_register(PF_INET, RTM_NEWROUTE, inet_rtm_newroute, NULL, NULL); - rtnl_register(PF_INET, RTM_DELROUTE, inet_rtm_delroute, NULL, NULL); - rtnl_register(PF_INET, RTM_GETROUTE, NULL, inet_dump_fib, NULL); + fib_trie_init(); register_pernet_subsys(&fib_net_ops); + register_netdevice_notifier(&fib_netdev_notifier); register_inetaddr_notifier(&fib_inetaddr_notifier); - fib_trie_init(); + rtnl_register(PF_INET, RTM_NEWROUTE, inet_rtm_newroute, NULL, 0); + rtnl_register(PF_INET, RTM_DELROUTE, inet_rtm_delroute, NULL, 0); + rtnl_register(PF_INET, RTM_GETROUTE, NULL, inet_dump_fib, 0); } diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index 769ab87ebc4b..5b2af19cfb5b 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -32,6 +32,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg, struct netlink_ext_ack *extack); int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, struct netlink_ext_ack *extack); +bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi); int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u32 tb_id, u8 type, __be32 dst, int dst_len, u8 tos, struct fib_info *fi, unsigned int); diff --git a/net/ipv4/fib_notifier.c b/net/ipv4/fib_notifier.c index e0714d975947..cfd420b0572c 100644 --- a/net/ipv4/fib_notifier.c +++ b/net/ipv4/fib_notifier.c @@ -1,86 +1,73 @@ #include <linux/rtnetlink.h> #include <linux/notifier.h> -#include <linux/rcupdate.h> +#include <linux/socket.h> #include <linux/kernel.h> +#include <linux/export.h> #include <net/net_namespace.h> +#include <net/fib_notifier.h> #include <net/netns/ipv4.h> #include <net/ip_fib.h> -static ATOMIC_NOTIFIER_HEAD(fib_chain); - -int call_fib_notifier(struct notifier_block *nb, struct net *net, - enum fib_event_type event_type, - struct fib_notifier_info *info) +int call_fib4_notifier(struct notifier_block *nb, struct net *net, + enum fib_event_type event_type, + struct fib_notifier_info *info) { - info->net = net; - return nb->notifier_call(nb, event_type, info); + info->family = AF_INET; + return call_fib_notifier(nb, net, event_type, info); } -int call_fib_notifiers(struct net *net, enum fib_event_type event_type, - struct fib_notifier_info *info) +int call_fib4_notifiers(struct net *net, enum fib_event_type event_type, + struct fib_notifier_info *info) { + ASSERT_RTNL(); + + info->family = AF_INET; net->ipv4.fib_seq++; - info->net = net; - return atomic_notifier_call_chain(&fib_chain, event_type, info); + return call_fib_notifiers(net, event_type, info); } -static unsigned int fib_seq_sum(void) +static unsigned int fib4_seq_read(struct net *net) { - unsigned int fib_seq = 0; - struct net *net; - - rtnl_lock(); - for_each_net(net) - fib_seq += net->ipv4.fib_seq; - rtnl_unlock(); + ASSERT_RTNL(); - return fib_seq; + return net->ipv4.fib_seq + fib4_rules_seq_read(net); } -static bool fib_dump_is_consistent(struct notifier_block *nb, - void (*cb)(struct notifier_block *nb), - unsigned int fib_seq) +static int fib4_dump(struct net *net, struct notifier_block *nb) { - atomic_notifier_chain_register(&fib_chain, nb); - if (fib_seq == fib_seq_sum()) - return true; - atomic_notifier_chain_unregister(&fib_chain, nb); - if (cb) - cb(nb); - return false; + int err; + + err = fib4_rules_dump(net, nb); + if (err) + return err; + + fib_notify(net, nb); + + return 0; } -#define FIB_DUMP_MAX_RETRIES 5 -int register_fib_notifier(struct notifier_block *nb, - void (*cb)(struct notifier_block *nb)) -{ - int retries = 0; +static const struct fib_notifier_ops fib4_notifier_ops_template = { + .family = AF_INET, + .fib_seq_read = fib4_seq_read, + .fib_dump = fib4_dump, + .owner = THIS_MODULE, +}; - do { - unsigned int fib_seq = fib_seq_sum(); - struct net *net; +int __net_init fib4_notifier_init(struct net *net) +{ + struct fib_notifier_ops *ops; - /* Mutex semantics guarantee that every change done to - * FIB tries before we read the change sequence counter - * is now visible to us. - */ - rcu_read_lock(); - for_each_net_rcu(net) { - fib_rules_notify(net, nb); - fib_notify(net, nb); - } - rcu_read_unlock(); + net->ipv4.fib_seq = 0; - if (fib_dump_is_consistent(nb, cb, fib_seq)) - return 0; - } while (++retries < FIB_DUMP_MAX_RETRIES); + ops = fib_notifier_ops_register(&fib4_notifier_ops_template, net); + if (IS_ERR(ops)) + return PTR_ERR(ops); + net->ipv4.notifier_ops = ops; - return -EBUSY; + return 0; } -EXPORT_SYMBOL(register_fib_notifier); -int unregister_fib_notifier(struct notifier_block *nb) +void __net_exit fib4_notifier_exit(struct net *net) { - return atomic_notifier_chain_unregister(&fib_chain, nb); + fib_notifier_ops_unregister(net->ipv4.notifier_ops); } -EXPORT_SYMBOL(unregister_fib_notifier); diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 778ecf977eb2..35d646a62ad4 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -68,6 +68,16 @@ bool fib4_rule_default(const struct fib_rule *rule) } EXPORT_SYMBOL_GPL(fib4_rule_default); +int fib4_rules_dump(struct net *net, struct notifier_block *nb) +{ + return fib_rules_dump(net, nb, AF_INET); +} + +unsigned int fib4_rules_seq_read(struct net *net) +{ + return fib_rules_seq_read(net, AF_INET); +} + int __fib_lookup(struct net *net, struct flowi4 *flp, struct fib_result *res, unsigned int flags) { @@ -185,38 +195,6 @@ static struct fib_table *fib_empty_table(struct net *net) return NULL; } -static int call_fib_rule_notifier(struct notifier_block *nb, struct net *net, - enum fib_event_type event_type, - struct fib_rule *rule) -{ - struct fib_rule_notifier_info info = { - .rule = rule, - }; - - return call_fib_notifier(nb, net, event_type, &info.info); -} - -static int call_fib_rule_notifiers(struct net *net, - enum fib_event_type event_type, - struct fib_rule *rule) -{ - struct fib_rule_notifier_info info = { - .rule = rule, - }; - - return call_fib_notifiers(net, event_type, &info.info); -} - -/* Called with rcu_read_lock() */ -void fib_rules_notify(struct net *net, struct notifier_block *nb) -{ - struct fib_rules_ops *ops = net->ipv4.rules_ops; - struct fib_rule *rule; - - list_for_each_entry_rcu(rule, &ops->rules_list, list) - call_fib_rule_notifier(nb, net, FIB_EVENT_RULE_ADD, rule); -} - static const struct nla_policy fib4_rule_policy[FRA_MAX+1] = { FRA_GENERIC_POLICY, [FRA_FLOW] = { .type = NLA_U32 }, @@ -273,7 +251,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, rule4->tos = frh->tos; net->ipv4.fib_has_custom_rules = true; - call_fib_rule_notifiers(net, FIB_EVENT_RULE_ADD, rule); err = 0; errout: @@ -295,7 +272,6 @@ static int fib4_rule_delete(struct fib_rule *rule) net->ipv4.fib_num_tclassid_users--; #endif net->ipv4.fib_has_custom_rules = true; - call_fib_rule_notifiers(net, FIB_EVENT_RULE_DEL, rule); errout: return err; } diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 222100103808..57a5d48acee8 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -44,6 +44,7 @@ #include <net/netlink.h> #include <net/nexthop.h> #include <net/lwtunnel.h> +#include <net/fib_notifier.h> #include "fib_lookup.h" @@ -219,7 +220,7 @@ static void free_fib_info_rcu(struct rcu_head *head) } endfor_nexthops(fi); m = fi->fib_metrics; - if (m != &dst_default_metrics && atomic_dec_and_test(&m->refcnt)) + if (m != &dst_default_metrics && refcount_dec_and_test(&m->refcnt)) kfree(m); kfree(fi); } @@ -695,6 +696,40 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, return 0; } +bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi) +{ + struct nlattr *nla; + int remaining; + + if (!cfg->fc_mx) + return true; + + nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) { + int type = nla_type(nla); + u32 val; + + if (!type) + continue; + if (type > RTAX_MAX) + return false; + + if (type == RTAX_CC_ALGO) { + char tmp[TCP_CA_NAME_MAX]; + bool ecn_ca = false; + + nla_strlcpy(tmp, nla, sizeof(tmp)); + val = tcp_ca_get_key_by_name(tmp, &ecn_ca); + } else { + val = nla_get_u32(nla); + } + + if (fi->fib_metrics->metrics[type - 1] != val) + return false; + } + + return true; +} + /* * Picture @@ -1083,15 +1118,17 @@ struct fib_info *fib_create_info(struct fib_config *cfg, fi = kzalloc(sizeof(*fi)+nhs*sizeof(struct fib_nh), GFP_KERNEL); if (!fi) goto failure; - fib_info_cnt++; if (cfg->fc_mx) { fi->fib_metrics = kzalloc(sizeof(*fi->fib_metrics), GFP_KERNEL); - if (!fi->fib_metrics) - goto failure; - atomic_set(&fi->fib_metrics->refcnt, 1); - } else + if (unlikely(!fi->fib_metrics)) { + kfree(fi); + return ERR_PTR(err); + } + refcount_set(&fi->fib_metrics->refcnt, 1); + } else { fi->fib_metrics = (struct dst_metrics *)&dst_default_metrics; - + } + fib_info_cnt++; fi->fib_net = net; fi->fib_protocol = cfg->fc_protocol; fi->fib_scope = cfg->fc_scope; @@ -1342,6 +1379,8 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event, IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev)) rtm->rtm_flags |= RTNH_F_DEAD; } + if (fi->fib_nh->nh_flags & RTNH_F_OFFLOAD) + rtm->rtm_flags |= RTNH_F_OFFLOAD; #ifdef CONFIG_IP_ROUTE_CLASSID if (fi->fib_nh[0].nh_tclassid && nla_put_u32(skb, RTA_FLOW, fi->fib_nh[0].nh_tclassid)) @@ -1449,14 +1488,14 @@ static int call_fib_nh_notifiers(struct fib_nh *fib_nh, if (IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) && fib_nh->nh_flags & RTNH_F_LINKDOWN) break; - return call_fib_notifiers(dev_net(fib_nh->nh_dev), event_type, - &info.info); + return call_fib4_notifiers(dev_net(fib_nh->nh_dev), event_type, + &info.info); case FIB_EVENT_NH_DEL: - if ((IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) && + if ((in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) && fib_nh->nh_flags & RTNH_F_LINKDOWN) || (fib_nh->nh_flags & RTNH_F_DEAD)) - return call_fib_notifiers(dev_net(fib_nh->nh_dev), - event_type, &info.info); + return call_fib4_notifiers(dev_net(fib_nh->nh_dev), + event_type, &info.info); default: break; } diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 64668c69dda6..c636650a6a70 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -81,6 +81,7 @@ #include <net/tcp.h> #include <net/sock.h> #include <net/ip_fib.h> +#include <net/fib_notifier.h> #include <trace/events/fib.h> #include "fib_lookup.h" @@ -97,7 +98,7 @@ static int call_fib_entry_notifier(struct notifier_block *nb, struct net *net, .type = type, .tb_id = tb_id, }; - return call_fib_notifier(nb, net, event_type, &info.info); + return call_fib4_notifier(nb, net, event_type, &info.info); } static int call_fib_entry_notifiers(struct net *net, @@ -113,7 +114,7 @@ static int call_fib_entry_notifiers(struct net *net, .type = type, .tb_id = tb_id, }; - return call_fib_notifiers(net, event_type, &info.info); + return call_fib4_notifiers(net, event_type, &info.info); } #define MAX_STAT_DEPTH 32 @@ -1562,7 +1563,8 @@ int fib_table_delete(struct net *net, struct fib_table *tb, fi->fib_prefsrc == cfg->fc_prefsrc) && (!cfg->fc_protocol || fi->fib_protocol == cfg->fc_protocol) && - fib_nh_match(cfg, fi, extack) == 0) { + fib_nh_match(cfg, fi, extack) == 0 && + fib_metrics_match(cfg, fi)) { fa_to_delete = fa; break; } diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index 8e0257d01200..1540db65241a 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -450,6 +450,7 @@ out_unlock: out: NAPI_GRO_CB(skb)->flush |= flush; skb_gro_remcsum_cleanup(skb, &grc); + skb->remcsum_offload = 0; return pp; } diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c index d5cac99170b1..416bb304a281 100644 --- a/net/ipv4/gre_offload.c +++ b/net/ipv4/gre_offload.c @@ -24,7 +24,7 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, __be16 protocol = skb->protocol; u16 mac_len = skb->mac_len; int gre_offset, outer_hlen; - bool need_csum, ufo, gso_partial; + bool need_csum, gso_partial; if (!skb->encapsulation) goto out; @@ -47,20 +47,8 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM); skb->encap_hdr_csum = need_csum; - ufo = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP); - features &= skb->dev->hw_enc_features; - /* The only checksum offload we care about from here on out is the - * outer one so strip the existing checksum feature flags based - * on the fact that we will be computing our checksum in software. - */ - if (ufo) { - features &= ~NETIF_F_CSUM_MASK; - if (!need_csum) - features |= NETIF_F_HW_CSUM; - } - /* segment inner packet. */ segs = skb_mac_gso_segment(skb, features); if (IS_ERR_OR_NULL(segs)) { diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index c2be26b98b5f..681e33998e03 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -412,7 +412,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) int type = icmp_param->data.icmph.type; int code = icmp_param->data.icmph.code; - if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb)) + if (ip_options_echo(net, &icmp_param->replyopts.opt.opt, skb)) return; /* Needed by both icmp_global_allow and icmp_xmit_lock */ @@ -694,7 +694,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) iph->tos; mark = IP4_REPLY_MARK(net, skb_in->mark); - if (ip_options_echo(&icmp_param.replyopts.opt.opt, skb_in)) + if (ip_options_echo(net, &icmp_param.replyopts.opt.opt, skb_in)) goto out_unlock; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 28f14afd0dd3..ab183af0b5b6 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1007,10 +1007,18 @@ int igmp_rcv(struct sk_buff *skb) { /* This basically follows the spec line by line -- see RFC1112 */ struct igmphdr *ih; - struct in_device *in_dev = __in_dev_get_rcu(skb->dev); + struct net_device *dev = skb->dev; + struct in_device *in_dev; int len = skb->len; bool dropped = true; + if (netif_is_l3_master(dev)) { + dev = dev_get_by_index_rcu(dev_net(dev), IPCB(skb)->iif); + if (!dev) + goto drop; + } + + in_dev = __in_dev_get_rcu(dev); if (!in_dev) goto drop; @@ -2549,7 +2557,8 @@ done: /* * check if a multicast source filter allows delivery for a given <src,dst,intf> */ -int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr, int dif) +int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr, + int dif, int sdif) { struct inet_sock *inet = inet_sk(sk); struct ip_mc_socklist *pmc; @@ -2564,7 +2573,8 @@ int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr, int dif) rcu_read_lock(); for_each_pmc_rcu(inet, pmc) { if (pmc->multi.imr_multiaddr.s_addr == loc_addr && - pmc->multi.imr_ifindex == dif) + (pmc->multi.imr_ifindex == dif || + (sdif && pmc->multi.imr_ifindex == sdif))) break; } ret = inet->mc_all; @@ -2974,12 +2984,6 @@ static int __net_init igmp_net_init(struct net *net) goto out_sock; } - /* Sysctl initialization */ - net->ipv4.sysctl_igmp_max_memberships = 20; - net->ipv4.sysctl_igmp_max_msf = 10; - /* IGMP reports for link-local multicast groups are enabled by default */ - net->ipv4.sysctl_igmp_llm_reports = 1; - net->ipv4.sysctl_igmp_qrv = 2; return 0; out_sock: diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 3828b3a805cd..c9c35b61a027 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -93,8 +93,17 @@ void inet_diag_msg_common_fill(struct inet_diag_msg *r, struct sock *sk) } EXPORT_SYMBOL_GPL(inet_diag_msg_common_fill); -static size_t inet_sk_attr_size(void) +static size_t inet_sk_attr_size(struct sock *sk, + const struct inet_diag_req_v2 *req, + bool net_admin) { + const struct inet_diag_handler *handler; + size_t aux = 0; + + handler = inet_diag_table[req->sdiag_protocol]; + if (handler && handler->idiag_get_aux_size) + aux = handler->idiag_get_aux_size(sk, net_admin); + return nla_total_size(sizeof(struct tcp_info)) + nla_total_size(1) /* INET_DIAG_SHUTDOWN */ + nla_total_size(1) /* INET_DIAG_TOS */ @@ -105,6 +114,7 @@ static size_t inet_sk_attr_size(void) + nla_total_size(SK_MEMINFO_VARS * sizeof(u32)) + nla_total_size(TCP_CA_NAME_MAX) + nla_total_size(sizeof(struct tcpvegas_info)) + + aux + 64; } @@ -260,6 +270,10 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, handler->idiag_get_info(sk, r, info); + if (ext & (1 << (INET_DIAG_INFO - 1)) && handler->idiag_get_aux) + if (handler->idiag_get_aux(sk, net_admin, skb) < 0) + goto errout; + if (sk->sk_state < TCP_TIME_WAIT) { union tcp_cc_info info; size_t sz = 0; @@ -274,6 +288,17 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, goto errout; } + if (ext & (1 << (INET_DIAG_CLASS_ID - 1))) { + u32 classid = 0; + +#ifdef CONFIG_SOCK_CGROUP_DATA + classid = sock_cgroup_classid(&sk->sk_cgrp_data); +#endif + + if (nla_put_u32(skb, INET_DIAG_CLASS_ID, classid)) + goto errout; + } + out: nlmsg_end(skb, nlh); return 0; @@ -438,6 +463,7 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, const struct nlmsghdr *nlh, const struct inet_diag_req_v2 *req) { + bool net_admin = netlink_net_capable(in_skb, CAP_NET_ADMIN); struct net *net = sock_net(in_skb->sk); struct sk_buff *rep; struct sock *sk; @@ -447,7 +473,7 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, if (IS_ERR(sk)) return PTR_ERR(sk); - rep = nlmsg_new(inet_sk_attr_size(), GFP_KERNEL); + rep = nlmsg_new(inet_sk_attr_size(sk, req, net_admin), GFP_KERNEL); if (!rep) { err = -ENOMEM; goto out; @@ -456,8 +482,7 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, err = sk_diag_fill(sk, rep, req, sk_user_ns(NETLINK_CB(in_skb).sk), NETLINK_CB(in_skb).portid, - nlh->nlmsg_seq, 0, nlh, - netlink_net_capable(in_skb, CAP_NET_ADMIN)); + nlh->nlmsg_seq, 0, nlh, net_admin); if (err < 0) { WARN_ON(err == -EMSGSIZE); nlmsg_free(rep); diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 96e95e83cc61..af74d0433453 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -234,10 +234,8 @@ evict_again: cond_resched(); if (read_seqretry(&f->rnd_seqlock, seq) || - percpu_counter_sum(&nf->mem)) + sum_frag_mem_limit(nf)) goto evict_again; - - percpu_counter_destroy(&nf->mem); } EXPORT_SYMBOL(inet_frags_exit_net); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 2e3389d614d1..597bb4cfe805 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -170,7 +170,7 @@ EXPORT_SYMBOL_GPL(__inet_inherit_port); static inline int compute_score(struct sock *sk, struct net *net, const unsigned short hnum, const __be32 daddr, - const int dif, bool exact_dif) + const int dif, const int sdif, bool exact_dif) { int score = -1; struct inet_sock *inet = inet_sk(sk); @@ -185,9 +185,13 @@ static inline int compute_score(struct sock *sk, struct net *net, score += 4; } if (sk->sk_bound_dev_if || exact_dif) { - if (sk->sk_bound_dev_if != dif) + bool dev_match = (sk->sk_bound_dev_if == dif || + sk->sk_bound_dev_if == sdif); + + if (exact_dif && !dev_match) return -1; - score += 4; + if (sk->sk_bound_dev_if && dev_match) + score += 4; } if (sk->sk_incoming_cpu == raw_smp_processor_id()) score++; @@ -208,7 +212,7 @@ struct sock *__inet_lookup_listener(struct net *net, struct sk_buff *skb, int doff, const __be32 saddr, __be16 sport, const __be32 daddr, const unsigned short hnum, - const int dif) + const int dif, const int sdif) { unsigned int hash = inet_lhashfn(net, hnum); struct inet_listen_hashbucket *ilb = &hashinfo->listening_hash[hash]; @@ -218,7 +222,8 @@ struct sock *__inet_lookup_listener(struct net *net, u32 phash = 0; sk_for_each_rcu(sk, &ilb->head) { - score = compute_score(sk, net, hnum, daddr, dif, exact_dif); + score = compute_score(sk, net, hnum, daddr, + dif, sdif, exact_dif); if (score > hiscore) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -268,7 +273,7 @@ struct sock *__inet_lookup_established(struct net *net, struct inet_hashinfo *hashinfo, const __be32 saddr, const __be16 sport, const __be32 daddr, const u16 hnum, - const int dif) + const int dif, const int sdif) { INET_ADDR_COOKIE(acookie, saddr, daddr); const __portpair ports = INET_COMBINED_PORTS(sport, hnum); @@ -286,11 +291,12 @@ begin: if (sk->sk_hash != hash) continue; if (likely(INET_MATCH(sk, net, acookie, - saddr, daddr, ports, dif))) { + saddr, daddr, ports, dif, sdif))) { if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt))) goto out; if (unlikely(!INET_MATCH(sk, net, acookie, - saddr, daddr, ports, dif))) { + saddr, daddr, ports, + dif, sdif))) { sock_gen_put(sk); goto begin; } @@ -321,9 +327,10 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, __be32 daddr = inet->inet_rcv_saddr; __be32 saddr = inet->inet_daddr; int dif = sk->sk_bound_dev_if; + struct net *net = sock_net(sk); + int sdif = l3mdev_master_ifindex_by_index(net, dif); INET_ADDR_COOKIE(acookie, saddr, daddr); const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); - struct net *net = sock_net(sk); unsigned int hash = inet_ehashfn(net, daddr, lport, saddr, inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); @@ -339,7 +346,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, continue; if (likely(INET_MATCH(sk2, net, acookie, - saddr, daddr, ports, dif))) { + saddr, daddr, ports, dif, sdif))) { if (sk2->sk_state == TCP_TIME_WAIT) { tw = inet_twsk(sk2); if (twsk_unique(sk, sk2, twp)) diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index c5a117cc6619..e7eb590c86ce 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -33,7 +33,7 @@ * also be removed if the pool is overloaded i.e. if the total amount of * entries is greater-or-equal than the threshold. * - * Node pool is organised as an AVL tree. + * Node pool is organised as an RB tree. * Such an implementation has been chosen not just for fun. It's a way to * prevent easy and efficient DoS attacks by creating hash collisions. A huge * amount of long living nodes in a single hash slot would significantly delay @@ -45,7 +45,7 @@ * AND reference count being 0. * 3. Global variable peer_total is modified under the pool lock. * 4. struct inet_peer fields modification: - * avl_left, avl_right, avl_parent, avl_height: pool lock + * rb_node: pool lock * refcnt: atomically against modifications on other CPU; * usually under some other lock to prevent node disappearing * daddr: unchangeable @@ -53,30 +53,15 @@ static struct kmem_cache *peer_cachep __read_mostly; -static LIST_HEAD(gc_list); -static const int gc_delay = 60 * HZ; -static struct delayed_work gc_work; -static DEFINE_SPINLOCK(gc_lock); - -#define node_height(x) x->avl_height - -#define peer_avl_empty ((struct inet_peer *)&peer_fake_node) -#define peer_avl_empty_rcu ((struct inet_peer __rcu __force *)&peer_fake_node) -static const struct inet_peer peer_fake_node = { - .avl_left = peer_avl_empty_rcu, - .avl_right = peer_avl_empty_rcu, - .avl_height = 0 -}; - void inet_peer_base_init(struct inet_peer_base *bp) { - bp->root = peer_avl_empty_rcu; + bp->rb_root = RB_ROOT; seqlock_init(&bp->lock); bp->total = 0; } EXPORT_SYMBOL_GPL(inet_peer_base_init); -#define PEER_MAXDEPTH 40 /* sufficient for about 2^27 nodes */ +#define PEER_MAX_GC 32 /* Exported for sysctl_net_ipv4. */ int inet_peer_threshold __read_mostly = 65536 + 128; /* start to throw entries more @@ -84,53 +69,6 @@ int inet_peer_threshold __read_mostly = 65536 + 128; /* start to throw entries m int inet_peer_minttl __read_mostly = 120 * HZ; /* TTL under high load: 120 sec */ int inet_peer_maxttl __read_mostly = 10 * 60 * HZ; /* usual time to live: 10 min */ -static void inetpeer_gc_worker(struct work_struct *work) -{ - struct inet_peer *p, *n, *c; - struct list_head list; - - spin_lock_bh(&gc_lock); - list_replace_init(&gc_list, &list); - spin_unlock_bh(&gc_lock); - - if (list_empty(&list)) - return; - - list_for_each_entry_safe(p, n, &list, gc_list) { - - if (need_resched()) - cond_resched(); - - c = rcu_dereference_protected(p->avl_left, 1); - if (c != peer_avl_empty) { - list_add_tail(&c->gc_list, &list); - p->avl_left = peer_avl_empty_rcu; - } - - c = rcu_dereference_protected(p->avl_right, 1); - if (c != peer_avl_empty) { - list_add_tail(&c->gc_list, &list); - p->avl_right = peer_avl_empty_rcu; - } - - n = list_entry(p->gc_list.next, struct inet_peer, gc_list); - - if (refcount_read(&p->refcnt) == 1) { - list_del(&p->gc_list); - kmem_cache_free(peer_cachep, p); - } - } - - if (list_empty(&list)) - return; - - spin_lock_bh(&gc_lock); - list_splice(&list, &gc_list); - spin_unlock_bh(&gc_lock); - - schedule_delayed_work(&gc_work, gc_delay); -} - /* Called from ip_output.c:ip_init */ void __init inet_initpeers(void) { @@ -153,225 +91,65 @@ void __init inet_initpeers(void) sizeof(struct inet_peer), 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); - - INIT_DEFERRABLE_WORK(&gc_work, inetpeer_gc_worker); } -#define rcu_deref_locked(X, BASE) \ - rcu_dereference_protected(X, lockdep_is_held(&(BASE)->lock.lock)) - -/* - * Called with local BH disabled and the pool lock held. - */ -#define lookup(_daddr, _stack, _base) \ -({ \ - struct inet_peer *u; \ - struct inet_peer __rcu **v; \ - \ - stackptr = _stack; \ - *stackptr++ = &_base->root; \ - for (u = rcu_deref_locked(_base->root, _base); \ - u != peer_avl_empty;) { \ - int cmp = inetpeer_addr_cmp(_daddr, &u->daddr); \ - if (cmp == 0) \ - break; \ - if (cmp == -1) \ - v = &u->avl_left; \ - else \ - v = &u->avl_right; \ - *stackptr++ = v; \ - u = rcu_deref_locked(*v, _base); \ - } \ - u; \ -}) - -/* - * Called with rcu_read_lock() - * Because we hold no lock against a writer, its quite possible we fall - * in an endless loop. - * But every pointer we follow is guaranteed to be valid thanks to RCU. - * We exit from this function if number of links exceeds PEER_MAXDEPTH - */ -static struct inet_peer *lookup_rcu(const struct inetpeer_addr *daddr, - struct inet_peer_base *base) +/* Called with rcu_read_lock() or base->lock held */ +static struct inet_peer *lookup(const struct inetpeer_addr *daddr, + struct inet_peer_base *base, + unsigned int seq, + struct inet_peer *gc_stack[], + unsigned int *gc_cnt, + struct rb_node **parent_p, + struct rb_node ***pp_p) { - struct inet_peer *u = rcu_dereference(base->root); - int count = 0; + struct rb_node **pp, *parent, *next; + struct inet_peer *p; + + pp = &base->rb_root.rb_node; + parent = NULL; + while (1) { + int cmp; - while (u != peer_avl_empty) { - int cmp = inetpeer_addr_cmp(daddr, &u->daddr); + next = rcu_dereference_raw(*pp); + if (!next) + break; + parent = next; + p = rb_entry(parent, struct inet_peer, rb_node); + cmp = inetpeer_addr_cmp(daddr, &p->daddr); if (cmp == 0) { - /* Before taking a reference, check if this entry was - * deleted (refcnt=0) - */ - if (!refcount_inc_not_zero(&u->refcnt)) { - u = NULL; - } - return u; + if (!refcount_inc_not_zero(&p->refcnt)) + break; + return p; + } + if (gc_stack) { + if (*gc_cnt < PEER_MAX_GC) + gc_stack[(*gc_cnt)++] = p; + } else if (unlikely(read_seqretry(&base->lock, seq))) { + break; } if (cmp == -1) - u = rcu_dereference(u->avl_left); + pp = &(*pp)->rb_left; else - u = rcu_dereference(u->avl_right); - if (unlikely(++count == PEER_MAXDEPTH)) - break; + pp = &(*pp)->rb_right; } + *parent_p = parent; + *pp_p = pp; return NULL; } -/* Called with local BH disabled and the pool lock held. */ -#define lookup_rightempty(start, base) \ -({ \ - struct inet_peer *u; \ - struct inet_peer __rcu **v; \ - *stackptr++ = &start->avl_left; \ - v = &start->avl_left; \ - for (u = rcu_deref_locked(*v, base); \ - u->avl_right != peer_avl_empty_rcu;) { \ - v = &u->avl_right; \ - *stackptr++ = v; \ - u = rcu_deref_locked(*v, base); \ - } \ - u; \ -}) - -/* Called with local BH disabled and the pool lock held. - * Variable names are the proof of operation correctness. - * Look into mm/map_avl.c for more detail description of the ideas. - */ -static void peer_avl_rebalance(struct inet_peer __rcu **stack[], - struct inet_peer __rcu ***stackend, - struct inet_peer_base *base) -{ - struct inet_peer __rcu **nodep; - struct inet_peer *node, *l, *r; - int lh, rh; - - while (stackend > stack) { - nodep = *--stackend; - node = rcu_deref_locked(*nodep, base); - l = rcu_deref_locked(node->avl_left, base); - r = rcu_deref_locked(node->avl_right, base); - lh = node_height(l); - rh = node_height(r); - if (lh > rh + 1) { /* l: RH+2 */ - struct inet_peer *ll, *lr, *lrl, *lrr; - int lrh; - ll = rcu_deref_locked(l->avl_left, base); - lr = rcu_deref_locked(l->avl_right, base); - lrh = node_height(lr); - if (lrh <= node_height(ll)) { /* ll: RH+1 */ - RCU_INIT_POINTER(node->avl_left, lr); /* lr: RH or RH+1 */ - RCU_INIT_POINTER(node->avl_right, r); /* r: RH */ - node->avl_height = lrh + 1; /* RH+1 or RH+2 */ - RCU_INIT_POINTER(l->avl_left, ll); /* ll: RH+1 */ - RCU_INIT_POINTER(l->avl_right, node); /* node: RH+1 or RH+2 */ - l->avl_height = node->avl_height + 1; - RCU_INIT_POINTER(*nodep, l); - } else { /* ll: RH, lr: RH+1 */ - lrl = rcu_deref_locked(lr->avl_left, base);/* lrl: RH or RH-1 */ - lrr = rcu_deref_locked(lr->avl_right, base);/* lrr: RH or RH-1 */ - RCU_INIT_POINTER(node->avl_left, lrr); /* lrr: RH or RH-1 */ - RCU_INIT_POINTER(node->avl_right, r); /* r: RH */ - node->avl_height = rh + 1; /* node: RH+1 */ - RCU_INIT_POINTER(l->avl_left, ll); /* ll: RH */ - RCU_INIT_POINTER(l->avl_right, lrl); /* lrl: RH or RH-1 */ - l->avl_height = rh + 1; /* l: RH+1 */ - RCU_INIT_POINTER(lr->avl_left, l); /* l: RH+1 */ - RCU_INIT_POINTER(lr->avl_right, node); /* node: RH+1 */ - lr->avl_height = rh + 2; - RCU_INIT_POINTER(*nodep, lr); - } - } else if (rh > lh + 1) { /* r: LH+2 */ - struct inet_peer *rr, *rl, *rlr, *rll; - int rlh; - rr = rcu_deref_locked(r->avl_right, base); - rl = rcu_deref_locked(r->avl_left, base); - rlh = node_height(rl); - if (rlh <= node_height(rr)) { /* rr: LH+1 */ - RCU_INIT_POINTER(node->avl_right, rl); /* rl: LH or LH+1 */ - RCU_INIT_POINTER(node->avl_left, l); /* l: LH */ - node->avl_height = rlh + 1; /* LH+1 or LH+2 */ - RCU_INIT_POINTER(r->avl_right, rr); /* rr: LH+1 */ - RCU_INIT_POINTER(r->avl_left, node); /* node: LH+1 or LH+2 */ - r->avl_height = node->avl_height + 1; - RCU_INIT_POINTER(*nodep, r); - } else { /* rr: RH, rl: RH+1 */ - rlr = rcu_deref_locked(rl->avl_right, base);/* rlr: LH or LH-1 */ - rll = rcu_deref_locked(rl->avl_left, base);/* rll: LH or LH-1 */ - RCU_INIT_POINTER(node->avl_right, rll); /* rll: LH or LH-1 */ - RCU_INIT_POINTER(node->avl_left, l); /* l: LH */ - node->avl_height = lh + 1; /* node: LH+1 */ - RCU_INIT_POINTER(r->avl_right, rr); /* rr: LH */ - RCU_INIT_POINTER(r->avl_left, rlr); /* rlr: LH or LH-1 */ - r->avl_height = lh + 1; /* r: LH+1 */ - RCU_INIT_POINTER(rl->avl_right, r); /* r: LH+1 */ - RCU_INIT_POINTER(rl->avl_left, node); /* node: LH+1 */ - rl->avl_height = lh + 2; - RCU_INIT_POINTER(*nodep, rl); - } - } else { - node->avl_height = (lh > rh ? lh : rh) + 1; - } - } -} - -/* Called with local BH disabled and the pool lock held. */ -#define link_to_pool(n, base) \ -do { \ - n->avl_height = 1; \ - n->avl_left = peer_avl_empty_rcu; \ - n->avl_right = peer_avl_empty_rcu; \ - /* lockless readers can catch us now */ \ - rcu_assign_pointer(**--stackptr, n); \ - peer_avl_rebalance(stack, stackptr, base); \ -} while (0) - static void inetpeer_free_rcu(struct rcu_head *head) { kmem_cache_free(peer_cachep, container_of(head, struct inet_peer, rcu)); } -static void unlink_from_pool(struct inet_peer *p, struct inet_peer_base *base, - struct inet_peer __rcu **stack[PEER_MAXDEPTH]) -{ - struct inet_peer __rcu ***stackptr, ***delp; - - if (lookup(&p->daddr, stack, base) != p) - BUG(); - delp = stackptr - 1; /* *delp[0] == p */ - if (p->avl_left == peer_avl_empty_rcu) { - *delp[0] = p->avl_right; - --stackptr; - } else { - /* look for a node to insert instead of p */ - struct inet_peer *t; - t = lookup_rightempty(p, base); - BUG_ON(rcu_deref_locked(*stackptr[-1], base) != t); - **--stackptr = t->avl_left; - /* t is removed, t->daddr > x->daddr for any - * x in p->avl_left subtree. - * Put t in the old place of p. */ - RCU_INIT_POINTER(*delp[0], t); - t->avl_left = p->avl_left; - t->avl_right = p->avl_right; - t->avl_height = p->avl_height; - BUG_ON(delp[1] != &p->avl_left); - delp[1] = &t->avl_left; /* was &p->avl_left */ - } - peer_avl_rebalance(stack, stackptr, base); - base->total--; - call_rcu(&p->rcu, inetpeer_free_rcu); -} - /* perform garbage collect on all items stacked during a lookup */ -static int inet_peer_gc(struct inet_peer_base *base, - struct inet_peer __rcu **stack[PEER_MAXDEPTH], - struct inet_peer __rcu ***stackptr) +static void inet_peer_gc(struct inet_peer_base *base, + struct inet_peer *gc_stack[], + unsigned int gc_cnt) { - struct inet_peer *p, *gchead = NULL; + struct inet_peer *p; __u32 delta, ttl; - int cnt = 0; + int i; if (base->total >= inet_peer_threshold) ttl = 0; /* be aggressive */ @@ -379,43 +157,38 @@ static int inet_peer_gc(struct inet_peer_base *base, ttl = inet_peer_maxttl - (inet_peer_maxttl - inet_peer_minttl) / HZ * base->total / inet_peer_threshold * HZ; - stackptr--; /* last stack slot is peer_avl_empty */ - while (stackptr > stack) { - stackptr--; - p = rcu_deref_locked(**stackptr, base); - if (refcount_read(&p->refcnt) == 1) { - smp_rmb(); - delta = (__u32)jiffies - p->dtime; - if (delta >= ttl && refcount_dec_if_one(&p->refcnt)) { - p->gc_next = gchead; - gchead = p; - } - } + for (i = 0; i < gc_cnt; i++) { + p = gc_stack[i]; + delta = (__u32)jiffies - p->dtime; + if (delta < ttl || !refcount_dec_if_one(&p->refcnt)) + gc_stack[i] = NULL; } - while ((p = gchead) != NULL) { - gchead = p->gc_next; - cnt++; - unlink_from_pool(p, base, stack); + for (i = 0; i < gc_cnt; i++) { + p = gc_stack[i]; + if (p) { + rb_erase(&p->rb_node, &base->rb_root); + base->total--; + call_rcu(&p->rcu, inetpeer_free_rcu); + } } - return cnt; } struct inet_peer *inet_getpeer(struct inet_peer_base *base, const struct inetpeer_addr *daddr, int create) { - struct inet_peer __rcu **stack[PEER_MAXDEPTH], ***stackptr; - struct inet_peer *p; - unsigned int sequence; - int invalidated, gccnt = 0; + struct inet_peer *p, *gc_stack[PEER_MAX_GC]; + struct rb_node **pp, *parent; + unsigned int gc_cnt, seq; + int invalidated; /* Attempt a lockless lookup first. * Because of a concurrent writer, we might not find an existing entry. */ rcu_read_lock(); - sequence = read_seqbegin(&base->lock); - p = lookup_rcu(daddr, base); - invalidated = read_seqretry(&base->lock, sequence); + seq = read_seqbegin(&base->lock); + p = lookup(daddr, base, seq, NULL, &gc_cnt, &parent, &pp); + invalidated = read_seqretry(&base->lock, seq); rcu_read_unlock(); if (p) @@ -428,36 +201,31 @@ struct inet_peer *inet_getpeer(struct inet_peer_base *base, /* retry an exact lookup, taking the lock before. * At least, nodes should be hot in our cache. */ + parent = NULL; write_seqlock_bh(&base->lock); -relookup: - p = lookup(daddr, stack, base); - if (p != peer_avl_empty) { - refcount_inc(&p->refcnt); - write_sequnlock_bh(&base->lock); - return p; - } - if (!gccnt) { - gccnt = inet_peer_gc(base, stack, stackptr); - if (gccnt && create) - goto relookup; - } - p = create ? kmem_cache_alloc(peer_cachep, GFP_ATOMIC) : NULL; - if (p) { - p->daddr = *daddr; - refcount_set(&p->refcnt, 2); - atomic_set(&p->rid, 0); - p->metrics[RTAX_LOCK-1] = INETPEER_METRICS_NEW; - p->rate_tokens = 0; - /* 60*HZ is arbitrary, but chosen enough high so that the first - * calculation of tokens is at its maximum. - */ - p->rate_last = jiffies - 60*HZ; - INIT_LIST_HEAD(&p->gc_list); - /* Link the node. */ - link_to_pool(p, base); - base->total++; + gc_cnt = 0; + p = lookup(daddr, base, seq, gc_stack, &gc_cnt, &parent, &pp); + if (!p && create) { + p = kmem_cache_alloc(peer_cachep, GFP_ATOMIC); + if (p) { + p->daddr = *daddr; + refcount_set(&p->refcnt, 2); + atomic_set(&p->rid, 0); + p->metrics[RTAX_LOCK-1] = INETPEER_METRICS_NEW; + p->rate_tokens = 0; + /* 60*HZ is arbitrary, but chosen enough high so that the first + * calculation of tokens is at its maximum. + */ + p->rate_last = jiffies - 60*HZ; + + rb_link_node(&p->rb_node, parent, pp); + rb_insert_color(&p->rb_node, &base->rb_root); + base->total++; + } } + if (gc_cnt) + inet_peer_gc(base, gc_stack, gc_cnt); write_sequnlock_bh(&base->lock); return p; @@ -467,8 +235,9 @@ EXPORT_SYMBOL_GPL(inet_getpeer); void inet_putpeer(struct inet_peer *p) { p->dtime = (__u32)jiffies; - smp_mb__before_atomic(); - refcount_dec(&p->refcnt); + + if (refcount_dec_and_test(&p->refcnt)) + call_rcu(&p->rcu, inetpeer_free_rcu); } EXPORT_SYMBOL_GPL(inet_putpeer); @@ -513,30 +282,16 @@ bool inet_peer_xrlim_allow(struct inet_peer *peer, int timeout) } EXPORT_SYMBOL(inet_peer_xrlim_allow); -static void inetpeer_inval_rcu(struct rcu_head *head) -{ - struct inet_peer *p = container_of(head, struct inet_peer, gc_rcu); - - spin_lock_bh(&gc_lock); - list_add_tail(&p->gc_list, &gc_list); - spin_unlock_bh(&gc_lock); - - schedule_delayed_work(&gc_work, gc_delay); -} - void inetpeer_invalidate_tree(struct inet_peer_base *base) { - struct inet_peer *root; - - write_seqlock_bh(&base->lock); + struct inet_peer *p, *n; - root = rcu_deref_locked(base->root, base); - if (root != peer_avl_empty) { - base->root = peer_avl_empty_rcu; - base->total = 0; - call_rcu(&root->gc_rcu, inetpeer_inval_rcu); + rbtree_postorder_for_each_entry_safe(p, n, &base->rb_root, rb_node) { + inet_putpeer(p); + cond_resched(); } - write_sequnlock_bh(&base->lock); + base->rb_root = RB_ROOT; + base->total = 0; } EXPORT_SYMBOL(inetpeer_invalidate_tree); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 9a8cfac503dc..46408c220d9d 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -844,8 +844,6 @@ static void __init ip4_frags_ctl_register(void) static int __net_init ipv4_frags_init_net(struct net *net) { - int res; - /* Fragment cache limits. * * The fragment memory accounting code, (tries to) account for @@ -871,13 +869,9 @@ static int __net_init ipv4_frags_init_net(struct net *net) net->ipv4.frags.max_dist = 64; - res = inet_frags_init_net(&net->ipv4.frags); - if (res) - return res; - res = ip4_frags_ns_ctl_register(net); - if (res) - inet_frags_uninit_net(&net->ipv4.frags); - return res; + inet_frags_init_net(&net->ipv4.frags); + + return ip4_frags_ns_ctl_register(net); } static void __net_exit ipv4_frags_exit_net(struct net *net) diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 7a7829e839c2..0162fb955b33 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -48,6 +48,7 @@ #include <net/rtnetlink.h> #include <net/gre.h> #include <net/dst_metadata.h> +#include <net/erspan.h> /* Problems & solutions @@ -112,9 +113,12 @@ MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN"); static struct rtnl_link_ops ipgre_link_ops __read_mostly; static int ipgre_tunnel_init(struct net_device *dev); +static void erspan_build_header(struct sk_buff *skb, + __be32 id, u32 index, bool truncate); static unsigned int ipgre_net_id __read_mostly; static unsigned int gre_tap_net_id __read_mostly; +static unsigned int erspan_net_id __read_mostly; static void ipgre_err(struct sk_buff *skb, u32 info, const struct tnl_ptk_info *tpi) @@ -246,6 +250,81 @@ static void gre_err(struct sk_buff *skb, u32 info) ipgre_err(skb, info, &tpi); } +static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, + int gre_hdr_len) +{ + struct net *net = dev_net(skb->dev); + struct metadata_dst *tun_dst = NULL; + struct ip_tunnel_net *itn; + struct ip_tunnel *tunnel; + struct erspanhdr *ershdr; + const struct iphdr *iph; + __be32 session_id; + __be32 index; + int len; + + itn = net_generic(net, erspan_net_id); + len = gre_hdr_len + sizeof(*ershdr); + + if (unlikely(!pskb_may_pull(skb, len))) + return -ENOMEM; + + iph = ip_hdr(skb); + ershdr = (struct erspanhdr *)(skb->data + gre_hdr_len); + + /* The original GRE header does not have key field, + * Use ERSPAN 10-bit session ID as key. + */ + session_id = cpu_to_be32(ntohs(ershdr->session_id)); + tpi->key = session_id; + index = ershdr->md.index; + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, + tpi->flags | TUNNEL_KEY, + iph->saddr, iph->daddr, tpi->key); + + if (tunnel) { + if (__iptunnel_pull_header(skb, + gre_hdr_len + sizeof(*ershdr), + htons(ETH_P_TEB), + false, false) < 0) + goto drop; + + if (tunnel->collect_md) { + struct ip_tunnel_info *info; + struct erspan_metadata *md; + __be64 tun_id; + __be16 flags; + + tpi->flags |= TUNNEL_KEY; + flags = tpi->flags; + tun_id = key32_to_tunnel_id(tpi->key); + + tun_dst = ip_tun_rx_dst(skb, flags, + tun_id, sizeof(*md)); + if (!tun_dst) + return PACKET_REJECT; + + md = ip_tunnel_info_opts(&tun_dst->u.tun_info); + if (!md) + return PACKET_REJECT; + + md->index = index; + info = &tun_dst->u.tun_info; + info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + info->options_len = sizeof(*md); + } else { + tunnel->index = ntohl(index); + } + + skb_reset_mac_header(skb); + ip_tunnel_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error); + return PACKET_RCVD; + } +drop: + kfree_skb(skb); + return PACKET_RCVD; +} + static int __ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct ip_tunnel_net *itn, int hdr_len, bool raw_proto) { @@ -328,6 +407,11 @@ static int gre_rcv(struct sk_buff *skb) if (hdr_len < 0) goto drop; + if (unlikely(tpi.proto == htons(ETH_P_ERSPAN))) { + if (erspan_rcv(skb, &tpi, hdr_len) == PACKET_RCVD) + return 0; + } + if (ipgre_rcv(skb, &tpi, hdr_len) == PACKET_RCVD) return 0; @@ -376,39 +460,33 @@ static struct rtable *gre_get_rt(struct sk_buff *skb, return ip_route_output_key(net, fl); } -static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, - __be16 proto) +static struct rtable *prepare_fb_xmit(struct sk_buff *skb, + struct net_device *dev, + struct flowi4 *fl, + int tunnel_hlen) { struct ip_tunnel_info *tun_info; const struct ip_tunnel_key *key; struct rtable *rt = NULL; - struct flowi4 fl; int min_headroom; - int tunnel_hlen; - __be16 df, flags; bool use_cache; int err; tun_info = skb_tunnel_info(skb); - if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) || - ip_tunnel_info_af(tun_info) != AF_INET)) - goto err_free_skb; - key = &tun_info->key; use_cache = ip_tunnel_dst_cache_usable(skb, tun_info); + if (use_cache) - rt = dst_cache_get_ip4(&tun_info->dst_cache, &fl.saddr); + rt = dst_cache_get_ip4(&tun_info->dst_cache, &fl->saddr); if (!rt) { - rt = gre_get_rt(skb, dev, &fl, key); + rt = gre_get_rt(skb, dev, fl, key); if (IS_ERR(rt)) - goto err_free_skb; + goto err_free_skb; if (use_cache) dst_cache_set_ip4(&tun_info->dst_cache, &rt->dst, - fl.saddr); + fl->saddr); } - tunnel_hlen = gre_calc_hlen(key->tun_flags); - min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len + tunnel_hlen + sizeof(struct iphdr); if (skb_headroom(skb) < min_headroom || skb_header_cloned(skb)) { @@ -420,6 +498,37 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, if (unlikely(err)) goto err_free_rt; } + return rt; + +err_free_rt: + ip_rt_put(rt); +err_free_skb: + kfree_skb(skb); + dev->stats.tx_dropped++; + return NULL; +} + +static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, + __be16 proto) +{ + struct ip_tunnel_info *tun_info; + const struct ip_tunnel_key *key; + struct rtable *rt = NULL; + struct flowi4 fl; + int tunnel_hlen; + __be16 df, flags; + + tun_info = skb_tunnel_info(skb); + if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) || + ip_tunnel_info_af(tun_info) != AF_INET)) + goto err_free_skb; + + key = &tun_info->key; + tunnel_hlen = gre_calc_hlen(key->tun_flags); + + rt = prepare_fb_xmit(skb, dev, &fl, tunnel_hlen); + if (!rt) + return; /* Push Tunnel header. */ if (gre_handle_offloads(skb, !!(tun_info->key.tun_flags & TUNNEL_CSUM))) @@ -442,6 +551,64 @@ err_free_skb: dev->stats.tx_dropped++; } +static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev, + __be16 proto) +{ + struct ip_tunnel *tunnel = netdev_priv(dev); + struct ip_tunnel_info *tun_info; + const struct ip_tunnel_key *key; + struct erspan_metadata *md; + struct rtable *rt = NULL; + bool truncate = false; + struct flowi4 fl; + int tunnel_hlen; + __be16 df; + + tun_info = skb_tunnel_info(skb); + if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) || + ip_tunnel_info_af(tun_info) != AF_INET)) + goto err_free_skb; + + key = &tun_info->key; + + /* ERSPAN has fixed 8 byte GRE header */ + tunnel_hlen = 8 + sizeof(struct erspanhdr); + + rt = prepare_fb_xmit(skb, dev, &fl, tunnel_hlen); + if (!rt) + return; + + if (gre_handle_offloads(skb, false)) + goto err_free_rt; + + if (skb->len > dev->mtu) { + pskb_trim(skb, dev->mtu); + truncate = true; + } + + md = ip_tunnel_info_opts(tun_info); + if (!md) + goto err_free_rt; + + erspan_build_header(skb, tunnel_id_to_key32(key->tun_id), + ntohl(md->index), truncate); + + gre_build_header(skb, 8, TUNNEL_SEQ, + htons(ETH_P_ERSPAN), 0, htonl(tunnel->o_seqno++)); + + df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; + + iptunnel_xmit(skb->sk, rt, skb, fl.saddr, key->u.ipv4.dst, IPPROTO_GRE, + key->tos, key->ttl, df, false); + return; + +err_free_rt: + ip_rt_put(rt); +err_free_skb: + kfree_skb(skb); + dev->stats.tx_dropped++; +} + static int gre_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb) { struct ip_tunnel_info *info = skb_tunnel_info(skb); @@ -503,6 +670,86 @@ free_skb: return NETDEV_TX_OK; } +static inline u8 tos_to_cos(u8 tos) +{ + u8 dscp, cos; + + dscp = tos >> 2; + cos = dscp >> 3; + return cos; +} + +static void erspan_build_header(struct sk_buff *skb, + __be32 id, u32 index, bool truncate) +{ + struct iphdr *iphdr = ip_hdr(skb); + struct ethhdr *eth = eth_hdr(skb); + enum erspan_encap_type enc_type; + struct erspanhdr *ershdr; + struct qtag_prefix { + __be16 eth_type; + __be16 tci; + } *qp; + u16 vlan_tci = 0; + + enc_type = ERSPAN_ENCAP_NOVLAN; + + /* If mirrored packet has vlan tag, extract tci and + * perserve vlan header in the mirrored frame. + */ + if (eth->h_proto == htons(ETH_P_8021Q)) { + qp = (struct qtag_prefix *)(skb->data + 2 * ETH_ALEN); + vlan_tci = ntohs(qp->tci); + enc_type = ERSPAN_ENCAP_INFRAME; + } + + skb_push(skb, sizeof(*ershdr)); + ershdr = (struct erspanhdr *)skb->data; + memset(ershdr, 0, sizeof(*ershdr)); + + ershdr->ver_vlan = htons((vlan_tci & VLAN_MASK) | + (ERSPAN_VERSION << VER_OFFSET)); + ershdr->session_id = htons((u16)(ntohl(id) & ID_MASK) | + ((tos_to_cos(iphdr->tos) << COS_OFFSET) & COS_MASK) | + (enc_type << EN_OFFSET & EN_MASK) | + ((truncate << T_OFFSET) & T_MASK)); + ershdr->md.index = htonl(index & INDEX_MASK); +} + +static netdev_tx_t erspan_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + struct ip_tunnel *tunnel = netdev_priv(dev); + bool truncate = false; + + if (tunnel->collect_md) { + erspan_fb_xmit(skb, dev, skb->protocol); + return NETDEV_TX_OK; + } + + if (gre_handle_offloads(skb, false)) + goto free_skb; + + if (skb_cow_head(skb, dev->needed_headroom)) + goto free_skb; + + if (skb->len > dev->mtu) { + pskb_trim(skb, dev->mtu); + truncate = true; + } + + /* Push ERSPAN header */ + erspan_build_header(skb, tunnel->parms.o_key, tunnel->index, truncate); + tunnel->parms.o_flags &= ~TUNNEL_KEY; + __gre_xmit(skb, dev, &tunnel->parms.iph, htons(ETH_P_ERSPAN)); + return NETDEV_TX_OK; + +free_skb: + kfree_skb(skb); + dev->stats.tx_dropped++; + return NETDEV_TX_OK; +} + static netdev_tx_t gre_tap_xmit(struct sk_buff *skb, struct net_device *dev) { @@ -828,6 +1075,42 @@ out: return ipgre_tunnel_validate(tb, data, extack); } +static int erspan_validate(struct nlattr *tb[], struct nlattr *data[], + struct netlink_ext_ack *extack) +{ + __be16 flags = 0; + int ret; + + if (!data) + return 0; + + ret = ipgre_tap_validate(tb, data, extack); + if (ret) + return ret; + + /* ERSPAN should only have GRE sequence and key flag */ + if (data[IFLA_GRE_OFLAGS]) + flags |= nla_get_be16(data[IFLA_GRE_OFLAGS]); + if (data[IFLA_GRE_IFLAGS]) + flags |= nla_get_be16(data[IFLA_GRE_IFLAGS]); + if (!data[IFLA_GRE_COLLECT_METADATA] && + flags != (GRE_SEQ | GRE_KEY)) + return -EINVAL; + + /* ERSPAN Session ID only has 10-bit. Since we reuse + * 32-bit key field as ID, check it's range. + */ + if (data[IFLA_GRE_IKEY] && + (ntohl(nla_get_be32(data[IFLA_GRE_IKEY])) & ~ID_MASK)) + return -EINVAL; + + if (data[IFLA_GRE_OKEY] && + (ntohl(nla_get_be32(data[IFLA_GRE_OKEY])) & ~ID_MASK)) + return -EINVAL; + + return 0; +} + static int ipgre_netlink_parms(struct net_device *dev, struct nlattr *data[], struct nlattr *tb[], @@ -892,6 +1175,13 @@ static int ipgre_netlink_parms(struct net_device *dev, if (data[IFLA_GRE_FWMARK]) *fwmark = nla_get_u32(data[IFLA_GRE_FWMARK]); + if (data[IFLA_GRE_ERSPAN_INDEX]) { + t->index = nla_get_u32(data[IFLA_GRE_ERSPAN_INDEX]); + + if (t->index & ~INDEX_MASK) + return -EINVAL; + } + return 0; } @@ -949,6 +1239,36 @@ static const struct net_device_ops gre_tap_netdev_ops = { .ndo_fill_metadata_dst = gre_fill_metadata_dst, }; +static int erspan_tunnel_init(struct net_device *dev) +{ + struct ip_tunnel *tunnel = netdev_priv(dev); + int t_hlen; + + tunnel->tun_hlen = 8; + tunnel->parms.iph.protocol = IPPROTO_GRE; + t_hlen = tunnel->hlen + sizeof(struct iphdr) + sizeof(struct erspanhdr); + + dev->needed_headroom = LL_MAX_HEADER + t_hlen + 4; + dev->mtu = ETH_DATA_LEN - t_hlen - 4; + dev->features |= GRE_FEATURES; + dev->hw_features |= GRE_FEATURES; + dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + + return ip_tunnel_init(dev); +} + +static const struct net_device_ops erspan_netdev_ops = { + .ndo_init = erspan_tunnel_init, + .ndo_uninit = ip_tunnel_uninit, + .ndo_start_xmit = erspan_xmit, + .ndo_set_mac_address = eth_mac_addr, + .ndo_validate_addr = eth_validate_addr, + .ndo_change_mtu = ip_tunnel_change_mtu, + .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_iflink = ip_tunnel_get_iflink, + .ndo_fill_metadata_dst = gre_fill_metadata_dst, +}; + static void ipgre_tap_setup(struct net_device *dev) { ether_setup(dev); @@ -1041,6 +1361,8 @@ static size_t ipgre_get_size(const struct net_device *dev) nla_total_size(1) + /* IFLA_GRE_FWMARK */ nla_total_size(4) + + /* IFLA_GRE_ERSPAN_INDEX */ + nla_total_size(4) + 0; } @@ -1083,12 +1405,25 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) goto nla_put_failure; } + if (t->index) + if (nla_put_u32(skb, IFLA_GRE_ERSPAN_INDEX, t->index)) + goto nla_put_failure; + return 0; nla_put_failure: return -EMSGSIZE; } +static void erspan_setup(struct net_device *dev) +{ + ether_setup(dev); + dev->netdev_ops = &erspan_netdev_ops; + dev->priv_flags &= ~IFF_TX_SKB_SHARING; + dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + ip_tunnel_setup(dev, erspan_net_id); +} + static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = { [IFLA_GRE_LINK] = { .type = NLA_U32 }, [IFLA_GRE_IFLAGS] = { .type = NLA_U16 }, @@ -1107,6 +1442,7 @@ static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = { [IFLA_GRE_COLLECT_METADATA] = { .type = NLA_FLAG }, [IFLA_GRE_IGNORE_DF] = { .type = NLA_U8 }, [IFLA_GRE_FWMARK] = { .type = NLA_U32 }, + [IFLA_GRE_ERSPAN_INDEX] = { .type = NLA_U32 }, }; static struct rtnl_link_ops ipgre_link_ops __read_mostly = { @@ -1139,6 +1475,21 @@ static struct rtnl_link_ops ipgre_tap_ops __read_mostly = { .get_link_net = ip_tunnel_get_link_net, }; +static struct rtnl_link_ops erspan_link_ops __read_mostly = { + .kind = "erspan", + .maxtype = IFLA_GRE_MAX, + .policy = ipgre_policy, + .priv_size = sizeof(struct ip_tunnel), + .setup = erspan_setup, + .validate = erspan_validate, + .newlink = ipgre_newlink, + .changelink = ipgre_changelink, + .dellink = ip_tunnel_dellink, + .get_size = ipgre_get_size, + .fill_info = ipgre_fill_info, + .get_link_net = ip_tunnel_get_link_net, +}; + struct net_device *gretap_fb_dev_create(struct net *net, const char *name, u8 name_assign_type) { @@ -1202,6 +1553,26 @@ static struct pernet_operations ipgre_tap_net_ops = { .size = sizeof(struct ip_tunnel_net), }; +static int __net_init erspan_init_net(struct net *net) +{ + return ip_tunnel_init_net(net, erspan_net_id, + &erspan_link_ops, "erspan0"); +} + +static void __net_exit erspan_exit_net(struct net *net) +{ + struct ip_tunnel_net *itn = net_generic(net, erspan_net_id); + + ip_tunnel_delete_net(itn, &erspan_link_ops); +} + +static struct pernet_operations erspan_net_ops = { + .init = erspan_init_net, + .exit = erspan_exit_net, + .id = &erspan_net_id, + .size = sizeof(struct ip_tunnel_net), +}; + static int __init ipgre_init(void) { int err; @@ -1214,7 +1585,11 @@ static int __init ipgre_init(void) err = register_pernet_device(&ipgre_tap_net_ops); if (err < 0) - goto pnet_tap_faied; + goto pnet_tap_failed; + + err = register_pernet_device(&erspan_net_ops); + if (err < 0) + goto pnet_erspan_failed; err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO); if (err < 0) { @@ -1230,15 +1605,23 @@ static int __init ipgre_init(void) if (err < 0) goto tap_ops_failed; + err = rtnl_link_register(&erspan_link_ops); + if (err < 0) + goto erspan_link_failed; + return 0; +erspan_link_failed: + rtnl_link_unregister(&ipgre_tap_ops); tap_ops_failed: rtnl_link_unregister(&ipgre_link_ops); rtnl_link_failed: gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO); add_proto_failed: + unregister_pernet_device(&erspan_net_ops); +pnet_erspan_failed: unregister_pernet_device(&ipgre_tap_net_ops); -pnet_tap_faied: +pnet_tap_failed: unregister_pernet_device(&ipgre_net_ops); return err; } @@ -1247,9 +1630,11 @@ static void __exit ipgre_fini(void) { rtnl_link_unregister(&ipgre_tap_ops); rtnl_link_unregister(&ipgre_link_ops); + rtnl_link_unregister(&erspan_link_ops); gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO); unregister_pernet_device(&ipgre_tap_net_ops); unregister_pernet_device(&ipgre_net_ops); + unregister_pernet_device(&erspan_net_ops); } module_init(ipgre_init); @@ -1257,5 +1642,7 @@ module_exit(ipgre_fini); MODULE_LICENSE("GPL"); MODULE_ALIAS_RTNL_LINK("gre"); MODULE_ALIAS_RTNL_LINK("gretap"); +MODULE_ALIAS_RTNL_LINK("erspan"); MODULE_ALIAS_NETDEV("gre0"); MODULE_ALIAS_NETDEV("gretap0"); +MODULE_ALIAS_NETDEV("erspan0"); diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 93157f2f4758..525ae88d1e58 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -86,8 +86,8 @@ void ip_options_build(struct sk_buff *skb, struct ip_options *opt, * NOTE: dopt cannot point to skb. */ -int __ip_options_echo(struct ip_options *dopt, struct sk_buff *skb, - const struct ip_options *sopt) +int __ip_options_echo(struct net *net, struct ip_options *dopt, + struct sk_buff *skb, const struct ip_options *sopt) { unsigned char *sptr, *dptr; int soffset, doffset; @@ -140,7 +140,7 @@ int __ip_options_echo(struct ip_options *dopt, struct sk_buff *skb, __be32 addr; memcpy(&addr, dptr+soffset-1, 4); - if (inet_addr_type(dev_net(skb_dst(skb)->dev), addr) != RTN_UNICAST) { + if (inet_addr_type(net, addr) != RTN_UNICAST) { dopt->ts_needtime = 1; soffset += 8; } @@ -174,9 +174,6 @@ int __ip_options_echo(struct ip_options *dopt, struct sk_buff *skb, doffset -= 4; } if (doffset > 3) { - __be32 daddr = fib_compute_spec_dst(skb); - - memcpy(&start[doffset-1], &daddr, 4); dopt->faddr = faddr; dptr[0] = start[0]; dptr[1] = doffset+3; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 7eb252dcecee..e8e675be60ec 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -599,6 +599,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, hlen = iph->ihl * 4; mtu = mtu - hlen; /* Size of data space */ IPCB(skb)->flags |= IPSKB_FRAG_COMPLETE; + ll_rs = LL_RESERVED_SPACE(rt->dst.dev); /* When frag_list is given, use it. First, check its validity: * some transformers could create wrong frag_list or break existing @@ -614,14 +615,15 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, if (first_len - hlen > mtu || ((first_len - hlen) & 7) || ip_is_fragment(iph) || - skb_cloned(skb)) + skb_cloned(skb) || + skb_headroom(skb) < ll_rs) goto slow_path; skb_walk_frags(skb, frag) { /* Correct geometry. */ if (frag->len > mtu || ((frag->len & 7) && frag->next) || - skb_headroom(frag) < hlen) + skb_headroom(frag) < hlen + ll_rs) goto slow_path_clean; /* Partially cloned skb? */ @@ -711,8 +713,6 @@ slow_path: left = skb->len - hlen; /* Space per frame */ ptr = hlen; /* Where to start from */ - ll_rs = LL_RESERVED_SPACE(rt->dst.dev); - /* * Fragment the datagram. */ @@ -853,61 +853,6 @@ csum_page(struct page *page, int offset, int copy) return csum; } -static inline int ip_ufo_append_data(struct sock *sk, - struct sk_buff_head *queue, - int getfrag(void *from, char *to, int offset, int len, - int odd, struct sk_buff *skb), - void *from, int length, int hh_len, int fragheaderlen, - int transhdrlen, int maxfraglen, unsigned int flags) -{ - struct sk_buff *skb; - int err; - - /* There is support for UDP fragmentation offload by network - * device, so create one single skb packet containing complete - * udp datagram - */ - skb = skb_peek_tail(queue); - if (!skb) { - skb = sock_alloc_send_skb(sk, - hh_len + fragheaderlen + transhdrlen + 20, - (flags & MSG_DONTWAIT), &err); - - if (!skb) - return err; - - /* reserve space for Hardware header */ - skb_reserve(skb, hh_len); - - /* create space for UDP/IP header */ - skb_put(skb, fragheaderlen + transhdrlen); - - /* initialize network header pointer */ - skb_reset_network_header(skb); - - /* initialize protocol header pointer */ - skb->transport_header = skb->network_header + fragheaderlen; - - skb->csum = 0; - - if (flags & MSG_CONFIRM) - skb_set_dst_pending_confirm(skb, 1); - - __skb_queue_tail(queue, skb); - } else if (skb_is_gso(skb)) { - goto append; - } - - skb->ip_summed = CHECKSUM_PARTIAL; - /* specify the length of each IP datagram fragment */ - skb_shinfo(skb)->gso_size = maxfraglen - fragheaderlen; - skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - -append: - return skb_append_datato_frags(sk, skb, getfrag, from, - (length - transhdrlen)); -} - static int __ip_append_data(struct sock *sk, struct flowi4 *fl4, struct sk_buff_head *queue, @@ -965,18 +910,6 @@ static int __ip_append_data(struct sock *sk, csummode = CHECKSUM_PARTIAL; cork->length += length; - if ((((length + (skb ? skb->len : fragheaderlen)) > mtu) || - (skb && skb_is_gso(skb))) && - (sk->sk_protocol == IPPROTO_UDP) && - (rt->dst.dev->features & NETIF_F_UFO) && !dst_xfrm(&rt->dst) && - (sk->sk_type == SOCK_DGRAM) && !sk->sk_no_check_tx) { - err = ip_ufo_append_data(sk, queue, getfrag, from, length, - hh_len, fragheaderlen, transhdrlen, - maxfraglen, flags); - if (err) - goto error; - return 0; - } /* So, what's going on in the loop below? * @@ -1287,27 +1220,14 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, if (!skb) return -EINVAL; - if ((size + skb->len > mtu) && - (sk->sk_protocol == IPPROTO_UDP) && - (rt->dst.dev->features & NETIF_F_UFO)) { - if (skb->ip_summed != CHECKSUM_PARTIAL) - return -EOPNOTSUPP; - - skb_shinfo(skb)->gso_size = mtu - fragheaderlen; - skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - } cork->length += size; while (size > 0) { - if (skb_is_gso(skb)) { - len = size; - } else { + /* Check if the remaining data fits into current packet. */ + len = mtu - skb->len; + if (len < size) + len = maxfraglen - skb->len; - /* Check if the remaining data fits into current packet. */ - len = mtu - skb->len; - if (len < size) - len = maxfraglen - skb->len; - } if (len <= 0) { struct sk_buff *skb_prev; int alloclen; @@ -1601,7 +1521,7 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, int err; int oif; - if (__ip_options_echo(&replyopts.opt.opt, skb, sopt)) + if (__ip_options_echo(net, &replyopts.opt.opt, skb, sopt)) return; ipc.addr = daddr; diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index ecc4b4a2413e..e558e4f9597b 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -80,7 +80,8 @@ static void ip_cmsg_recv_opts(struct msghdr *msg, struct sk_buff *skb) } -static void ip_cmsg_recv_retopts(struct msghdr *msg, struct sk_buff *skb) +static void ip_cmsg_recv_retopts(struct net *net, struct msghdr *msg, + struct sk_buff *skb) { unsigned char optbuf[sizeof(struct ip_options) + 40]; struct ip_options *opt = (struct ip_options *)optbuf; @@ -88,7 +89,7 @@ static void ip_cmsg_recv_retopts(struct msghdr *msg, struct sk_buff *skb) if (IPCB(skb)->opt.optlen == 0) return; - if (ip_options_echo(opt, skb)) { + if (ip_options_echo(net, opt, skb)) { msg->msg_flags |= MSG_CTRUNC; return; } @@ -204,7 +205,7 @@ void ip_cmsg_recv_offset(struct msghdr *msg, struct sock *sk, } if (flags & IP_CMSG_RETOPTS) { - ip_cmsg_recv_retopts(msg, skb); + ip_cmsg_recv_retopts(sock_net(sk), msg, skb); flags &= ~IP_CMSG_RETOPTS; if (!flags) @@ -1206,6 +1207,7 @@ e_inval: void ipv4_pktinfo_prepare(const struct sock *sk, struct sk_buff *skb) { struct in_pktinfo *pktinfo = PKTINFO_SKB_CB(skb); + bool l3slave = ipv4_l3mdev_skb(IPCB(skb)->flags); bool prepare = (inet_sk(sk)->cmsg_flags & IP_CMSG_PKTINFO) || ipv6_sk_rxinfo(sk); @@ -1219,7 +1221,7 @@ void ipv4_pktinfo_prepare(const struct sock *sk, struct sk_buff *skb) * (e.g., process binds socket to eth0 for Tx which is * redirected to loopback in the rtable/dst). */ - if (pktinfo->ipi_ifindex == LOOPBACK_IFINDEX) + if (pktinfo->ipi_ifindex == LOOPBACK_IFINDEX || l3slave) pktinfo->ipi_ifindex = inet_iif(skb); pktinfo->ipi_spec_dst.s_addr = fib_compute_spec_dst(skb); @@ -1227,14 +1229,7 @@ void ipv4_pktinfo_prepare(const struct sock *sk, struct sk_buff *skb) pktinfo->ipi_ifindex = 0; pktinfo->ipi_spec_dst.s_addr = 0; } - /* We need to keep the dst for __ip_options_echo() - * We could restrict the test to opt.ts_needtime || opt.srr, - * but the following is good enough as IP options are not often used. - */ - if (unlikely(IPCB(skb)->opt.optlen)) - skb_dst_force(skb); - else - skb_dst_drop(skb); + skb_dst_drop(skb); } int ip_setsockopt(struct sock *sk, int level, diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 0192c255e508..5ed63d250950 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -584,33 +584,6 @@ static struct rtnl_link_ops vti_link_ops __read_mostly = { .get_link_net = ip_tunnel_get_link_net, }; -static bool is_vti_tunnel(const struct net_device *dev) -{ - return dev->netdev_ops == &vti_netdev_ops; -} - -static int vti_device_event(struct notifier_block *unused, - unsigned long event, void *ptr) -{ - struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct ip_tunnel *tunnel = netdev_priv(dev); - - if (!is_vti_tunnel(dev)) - return NOTIFY_DONE; - - switch (event) { - case NETDEV_DOWN: - if (!net_eq(tunnel->net, dev_net(dev))) - xfrm_garbage_collect(tunnel->net); - break; - } - return NOTIFY_DONE; -} - -static struct notifier_block vti_notifier_block __read_mostly = { - .notifier_call = vti_device_event, -}; - static int __init vti_init(void) { const char *msg; @@ -618,8 +591,6 @@ static int __init vti_init(void) pr_info("IPv4 over IPsec tunneling driver\n"); - register_netdevice_notifier(&vti_notifier_block); - msg = "tunnel device"; err = register_pernet_device(&vti_net_ops); if (err < 0) @@ -652,7 +623,6 @@ xfrm_proto_ah_failed: xfrm_proto_esp_failed: unregister_pernet_device(&vti_net_ops); pernet_dev_failed: - unregister_netdevice_notifier(&vti_notifier_block); pr_err("vti init: failed to register %s\n", msg); return err; } @@ -664,7 +634,6 @@ static void __exit vti_fini(void) xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH); xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP); unregister_pernet_device(&vti_net_ops); - unregister_netdevice_notifier(&vti_notifier_block); } module_init(vti_init); diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 06863ea3fc5b..c9b3e6e069ae 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -3114,14 +3114,14 @@ int __init ip_mr_init(void) } #endif rtnl_register(RTNL_FAMILY_IPMR, RTM_GETROUTE, - ipmr_rtm_getroute, ipmr_rtm_dumproute, NULL); + ipmr_rtm_getroute, ipmr_rtm_dumproute, 0); rtnl_register(RTNL_FAMILY_IPMR, RTM_NEWROUTE, - ipmr_rtm_route, NULL, NULL); + ipmr_rtm_route, NULL, 0); rtnl_register(RTNL_FAMILY_IPMR, RTM_DELROUTE, - ipmr_rtm_route, NULL, NULL); + ipmr_rtm_route, NULL, 0); rtnl_register(RTNL_FAMILY_IPMR, RTM_GETLINK, - NULL, ipmr_rtm_dumplink, NULL); + NULL, ipmr_rtm_dumplink, 0); return 0; #ifdef CONFIG_IP_PIMSM_V2 diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 0bc3c3d73e61..e04457198f93 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -268,14 +268,14 @@ unsigned int arpt_do_table(struct sk_buff *skb, acpar.targinfo = t->data; verdict = t->u.kernel.target->target(skb, &acpar); - /* Target might have changed stuff. */ - arp = arp_hdr(skb); - - if (verdict == XT_CONTINUE) + if (verdict == XT_CONTINUE) { + /* Target might have changed stuff. */ + arp = arp_hdr(skb); e = arpt_next_entry(e); - else + } else { /* Verdict */ break; + } } while (!acpar.hotdrop); xt_write_recseq_end(addend); local_bh_enable(); @@ -1117,7 +1117,6 @@ compat_copy_entry_from_user(struct compat_arpt_entry *e, void **dstptr, struct xt_table_info *newinfo, unsigned char *base) { struct xt_entry_target *t; - struct xt_target *target; struct arpt_entry *de; unsigned int origsize; int h; @@ -1132,7 +1131,6 @@ compat_copy_entry_from_user(struct compat_arpt_entry *e, void **dstptr, de->target_offset = e->target_offset - (origsize - *size); t = compat_arpt_get_target(e); - target = t->u.kernel.target; xt_compat_target_from_user(t, dstptr, size); de->next_offset = e->next_offset - (origsize - *size); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 2a55a40211cb..576cba2b57e9 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -35,12 +35,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); MODULE_DESCRIPTION("IPv4 packet filter"); -#ifdef CONFIG_NETFILTER_DEBUG -#define IP_NF_ASSERT(x) WARN_ON(!(x)) -#else -#define IP_NF_ASSERT(x) -#endif - void *ipt_alloc_initial_table(const struct xt_table *info) { return xt_alloc_initial_table(ipt, IPT); @@ -151,7 +145,7 @@ static const char *const comments[] = { [NF_IP_TRACE_COMMENT_POLICY] = "policy", }; -static struct nf_loginfo trace_loginfo = { +static const struct nf_loginfo trace_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { @@ -263,7 +257,7 @@ ipt_do_table(struct sk_buff *skb, acpar.hotdrop = false; acpar.state = state; - IP_NF_ASSERT(table->valid_hooks & (1 << hook)); + WARN_ON(!(table->valid_hooks & (1 << hook))); local_bh_disable(); addend = xt_write_recseq_begin(); private = table->private; @@ -293,7 +287,7 @@ ipt_do_table(struct sk_buff *skb, const struct xt_entry_match *ematch; struct xt_counters *counter; - IP_NF_ASSERT(e); + WARN_ON(!e); if (!ip_packet_match(ip, indev, outdev, &e->ip, acpar.fragoff)) { no_match: @@ -312,7 +306,7 @@ ipt_do_table(struct sk_buff *skb, ADD_COUNTER(*counter, skb->len, 1); t = ipt_get_target(e); - IP_NF_ASSERT(t->u.kernel.target); + WARN_ON(!t->u.kernel.target); #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) /* The packet is traced: log it */ @@ -352,13 +346,14 @@ ipt_do_table(struct sk_buff *skb, acpar.targinfo = t->data; verdict = t->u.kernel.target->target(skb, &acpar); - /* Target might have changed stuff. */ - ip = ip_hdr(skb); - if (verdict == XT_CONTINUE) + if (verdict == XT_CONTINUE) { + /* Target might have changed stuff. */ + ip = ip_hdr(skb); e = ipt_next_entry(e); - else + } else { /* Verdict */ break; + } } while (!acpar.hotdrop); xt_write_recseq_end(addend); @@ -1355,7 +1350,6 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, struct xt_table_info *newinfo, unsigned char *base) { struct xt_entry_target *t; - struct xt_target *target; struct ipt_entry *de; unsigned int origsize; int h; @@ -1374,7 +1368,6 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, de->target_offset = e->target_offset - (origsize - *size); t = compat_ipt_get_target(e); - target = t->u.kernel.target; xt_compat_target_from_user(t, dstptr, size); de->next_offset = e->next_offset - (origsize - *size); diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 7d72decb80f9..17b4ca562944 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -117,7 +117,8 @@ clusterip_config_entry_put(struct net *net, struct clusterip_config *c) * functions are also incrementing the refcount on their own, * so it's safe to remove the entry even if it's in use. */ #ifdef CONFIG_PROC_FS - proc_remove(c->pde); + if (cn->procdir) + proc_remove(c->pde); #endif return; } @@ -624,7 +625,7 @@ arp_mangle(void *priv, return NF_ACCEPT; } -static struct nf_hook_ops cip_arp_ops __read_mostly = { +static const struct nf_hook_ops cip_arp_ops = { .hook = arp_mangle, .pf = NFPROTO_ARP, .hooknum = NF_ARP_OUT, @@ -815,6 +816,7 @@ static void clusterip_net_exit(struct net *net) #ifdef CONFIG_PROC_FS struct clusterip_net *cn = net_generic(net, clusterip_net_id); proc_remove(cn->procdir); + cn->procdir = NULL; #endif nf_unregister_net_hook(net, &cip_arp_ops); } diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c index f1528f7175a8..811689e523c3 100644 --- a/net/ipv4/netfilter/ipt_SYNPROXY.c +++ b/net/ipv4/netfilter/ipt_SYNPROXY.c @@ -416,7 +416,7 @@ static unsigned int ipv4_synproxy_hook(void *priv, return NF_ACCEPT; } -static struct nf_hook_ops ipv4_synproxy_ops[] __read_mostly = { +static const struct nf_hook_ops ipv4_synproxy_ops[] = { { .hook = ipv4_synproxy_hook, .pf = NFPROTO_IPV4, diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 138a24bc76ad..a1a07b338ccf 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -67,7 +67,7 @@ static unsigned int iptable_nat_ipv4_local_fn(void *priv, return nf_nat_ipv4_local_fn(priv, skb, state, iptable_nat_do_chain); } -static struct nf_hook_ops nf_nat_ipv4_ops[] __read_mostly = { +static const struct nf_hook_ops nf_nat_ipv4_ops[] = { /* Before packet filtering, change destination */ { .hook = iptable_nat_ipv4_in, diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 2e14ed11a35c..fe374da4bc13 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -63,13 +63,6 @@ static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -static void ipv4_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "src=%pI4 dst=%pI4 ", - &tuple->src.u3.ip, &tuple->dst.u3.ip); -} - static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, unsigned int *dataoff, u_int8_t *protonum) { @@ -174,7 +167,7 @@ static unsigned int ipv4_conntrack_local(void *priv, /* Connection tracking may drop packets, but never alters them, so make it the first hook. */ -static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = { +static const struct nf_hook_ops ipv4_conntrack_ops[] = { { .hook = ipv4_conntrack_in, .pf = NFPROTO_IPV4, @@ -303,11 +296,6 @@ static int ipv4_nlattr_to_tuple(struct nlattr *tb[], return 0; } - -static int ipv4_nlattr_tuple_size(void) -{ - return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1); -} #endif static struct nf_sockopt_ops so_getorigdst = { @@ -358,16 +346,15 @@ static void ipv4_hooks_unregister(struct net *net) struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = { .l3proto = PF_INET, - .name = "ipv4", .pkt_to_tuple = ipv4_pkt_to_tuple, .invert_tuple = ipv4_invert_tuple, - .print_tuple = ipv4_print_tuple, .get_l4proto = ipv4_get_l4proto, #if IS_ENABLED(CONFIG_NF_CT_NETLINK) .tuple_to_nlattr = ipv4_tuple_to_nlattr, - .nlattr_tuple_size = ipv4_nlattr_tuple_size, .nlattr_to_tuple = ipv4_nlattr_to_tuple, .nla_policy = ipv4_nla_policy, + .nla_size = NLA_ALIGN(NLA_HDRLEN + sizeof(u32)) + /* CTA_IP_V4_SRC */ + NLA_ALIGN(NLA_HDRLEN + sizeof(u32)), /* CTA_IP_V4_DST */ #endif .net_ns_get = ipv4_hooks_register, .net_ns_put = ipv4_hooks_unregister, @@ -398,24 +385,12 @@ static struct nf_conntrack_l4proto *builtin_l4proto4[] = { static int ipv4_net_init(struct net *net) { - int ret = 0; - - ret = nf_ct_l4proto_pernet_register(net, builtin_l4proto4, - ARRAY_SIZE(builtin_l4proto4)); - if (ret < 0) - return ret; - ret = nf_ct_l3proto_pernet_register(net, &nf_conntrack_l3proto_ipv4); - if (ret < 0) { - pr_err("nf_conntrack_ipv4: pernet registration failed\n"); - nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4, - ARRAY_SIZE(builtin_l4proto4)); - } - return ret; + return nf_ct_l4proto_pernet_register(net, builtin_l4proto4, + ARRAY_SIZE(builtin_l4proto4)); } static void ipv4_net_exit(struct net *net) { - nf_ct_l3proto_pernet_unregister(net, &nf_conntrack_l3proto_ipv4); nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4, ARRAY_SIZE(builtin_l4proto4)); } @@ -433,6 +408,11 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) need_conntrack(); +#if IS_ENABLED(CONFIG_NF_CT_NETLINK) + if (WARN_ON(nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1) != + nf_conntrack_l3proto_ipv4.nla_size)) + return -EINVAL; +#endif ret = nf_register_sockopt(&so_getorigdst); if (ret < 0) { pr_err("Unable to register netfilter socket option\n"); diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 73c591d8a9a8..a046c298413a 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -71,16 +71,6 @@ static bool icmp_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void icmp_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "type=%u code=%u id=%u ", - tuple->dst.u.icmp.type, - tuple->dst.u.icmp.code, - ntohs(tuple->src.u.icmp.id)); -} - static unsigned int *icmp_get_timeouts(struct net *net) { return &icmp_pernet(net)->timeout; @@ -92,7 +82,6 @@ static int icmp_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeout) { /* Do not immediately delete the connection after the first @@ -137,7 +126,7 @@ icmp_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, enum ip_conntrack_info ctinfo; struct nf_conntrack_zone tmp; - NF_CT_ASSERT(!skb_nfct(skb)); + WARN_ON(skb_nfct(skb)); zone = nf_ct_zone_tmpl(tmpl, skb, &tmp); /* Are they talking about one of our connections? */ @@ -362,10 +351,8 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_icmp __read_mostly = { .l3proto = PF_INET, .l4proto = IPPROTO_ICMP, - .name = "icmp", .pkt_to_tuple = icmp_pkt_to_tuple, .invert_tuple = icmp_invert_tuple, - .print_tuple = icmp_print_tuple, .packet = icmp_packet, .get_timeouts = icmp_get_timeouts, .new = icmp_new, diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index 346bf7ccac08..37fe1616ca0b 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -90,7 +90,7 @@ static unsigned int ipv4_conntrack_defrag(void *priv, return NF_ACCEPT; } -static struct nf_hook_ops ipv4_defrag_ops[] = { +static const struct nf_hook_ops ipv4_defrag_ops[] = { { .hook = ipv4_conntrack_defrag, .pf = NFPROTO_IPV4, diff --git a/net/ipv4/netfilter/nf_log_arp.c b/net/ipv4/netfilter/nf_log_arp.c index 2f3895ddc275..df5c2a2061a4 100644 --- a/net/ipv4/netfilter/nf_log_arp.c +++ b/net/ipv4/netfilter/nf_log_arp.c @@ -25,7 +25,7 @@ #include <linux/netfilter/xt_LOG.h> #include <net/netfilter/nf_log.h> -static struct nf_loginfo default_loginfo = { +static const struct nf_loginfo default_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { diff --git a/net/ipv4/netfilter/nf_log_ipv4.c b/net/ipv4/netfilter/nf_log_ipv4.c index c83a9963269b..4388de0e5380 100644 --- a/net/ipv4/netfilter/nf_log_ipv4.c +++ b/net/ipv4/netfilter/nf_log_ipv4.c @@ -24,7 +24,7 @@ #include <linux/netfilter/xt_LOG.h> #include <net/netfilter/nf_log.h> -static struct nf_loginfo default_loginfo = { +static const struct nf_loginfo default_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c index 574f7ebba0b6..ac8342dcb55e 100644 --- a/net/ipv4/netfilter/nf_nat_h323.c +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -252,16 +252,16 @@ static int nat_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, if (set_h245_addr(skb, protoff, data, dataoff, taddr, &ct->tuplehash[!dir].tuple.dst.u3, htons((port & htons(1)) ? nated_port + 1 : - nated_port)) == 0) { - /* Save ports */ - info->rtp_port[i][dir] = rtp_port; - info->rtp_port[i][!dir] = htons(nated_port); - } else { + nated_port))) { nf_ct_unexpect_related(rtp_exp); nf_ct_unexpect_related(rtcp_exp); return -1; } + /* Save ports */ + info->rtp_port[i][dir] = rtp_port; + info->rtp_port[i][!dir] = htons(nated_port); + /* Success */ pr_debug("nf_nat_h323: expect RTP %pI4:%hu->%pI4:%hu\n", &rtp_exp->tuple.src.u3.ip, @@ -370,15 +370,15 @@ static int nat_h245(struct sk_buff *skb, struct nf_conn *ct, /* Modify signal */ if (set_h225_addr(skb, protoff, data, dataoff, taddr, &ct->tuplehash[!dir].tuple.dst.u3, - htons(nated_port)) == 0) { - /* Save ports */ - info->sig_port[dir] = port; - info->sig_port[!dir] = htons(nated_port); - } else { + htons(nated_port))) { nf_ct_unexpect_related(exp); return -1; } + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + pr_debug("nf_nat_q931: expect H.245 %pI4:%hu->%pI4:%hu\n", &exp->tuple.src.u3.ip, ntohs(exp->tuple.src.u.tcp.port), @@ -462,24 +462,27 @@ static int nat_q931(struct sk_buff *skb, struct nf_conn *ct, /* Modify signal */ if (set_h225_addr(skb, protoff, data, 0, &taddr[idx], &ct->tuplehash[!dir].tuple.dst.u3, - htons(nated_port)) == 0) { - /* Save ports */ - info->sig_port[dir] = port; - info->sig_port[!dir] = htons(nated_port); - - /* Fix for Gnomemeeting */ - if (idx > 0 && - get_h225_addr(ct, *data, &taddr[0], &addr, &port) && - (ntohl(addr.ip) & 0xff000000) == 0x7f000000) { - set_h225_addr(skb, protoff, data, 0, &taddr[0], - &ct->tuplehash[!dir].tuple.dst.u3, - info->sig_port[!dir]); - } - } else { + htons(nated_port))) { nf_ct_unexpect_related(exp); return -1; } + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + + /* Fix for Gnomemeeting */ + if (idx > 0 && + get_h225_addr(ct, *data, &taddr[0], &addr, &port) && + (ntohl(addr.ip) & 0xff000000) == 0x7f000000) { + if (set_h225_addr(skb, protoff, data, 0, &taddr[0], + &ct->tuplehash[!dir].tuple.dst.u3, + info->sig_port[!dir])) { + nf_ct_unexpect_related(exp); + return -1; + } + } + /* Success */ pr_debug("nf_nat_ras: expect Q.931 %pI4:%hu->%pI4:%hu\n", &exp->tuple.src.u3.ip, @@ -550,9 +553,9 @@ static int nat_callforwarding(struct sk_buff *skb, struct nf_conn *ct, } /* Modify signal */ - if (!set_h225_addr(skb, protoff, data, dataoff, taddr, - &ct->tuplehash[!dir].tuple.dst.u3, - htons(nated_port)) == 0) { + if (set_h225_addr(skb, protoff, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port))) { nf_ct_unexpect_related(exp); return -1; } diff --git a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c index feedd759ca80..a0f37b208268 100644 --- a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c @@ -190,7 +190,7 @@ int nf_nat_icmp_reply_translation(struct sk_buff *skb, struct nf_conntrack_tuple target; unsigned long statusbit; - NF_CT_ASSERT(ctinfo == IP_CT_RELATED || ctinfo == IP_CT_RELATED_REPLY); + WARN_ON(ctinfo != IP_CT_RELATED && ctinfo != IP_CT_RELATED_REPLY); if (!skb_make_writable(skb, hdrlen + sizeof(*inside))) return 0; @@ -306,8 +306,8 @@ nf_nat_ipv4_fn(void *priv, struct sk_buff *skb, default: /* ESTABLISHED */ - NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED || - ctinfo == IP_CT_ESTABLISHED_REPLY); + WARN_ON(ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED_REPLY); if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out)) goto oif_changed; } diff --git a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c index f39037fca923..0c366aad89cb 100644 --- a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c +++ b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c @@ -34,12 +34,12 @@ nf_nat_masquerade_ipv4(struct sk_buff *skb, unsigned int hooknum, const struct rtable *rt; __be32 newsrc, nh; - NF_CT_ASSERT(hooknum == NF_INET_POST_ROUTING); + WARN_ON(hooknum != NF_INET_POST_ROUTING); ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || - ctinfo == IP_CT_RELATED_REPLY)); + WARN_ON(!(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || + ctinfo == IP_CT_RELATED_REPLY))); /* Source address is 0.0.0.0 - locally generated packet that is * probably not supposed to be masqueraded. @@ -96,7 +96,7 @@ static int masq_device_event(struct notifier_block *this, * conntracks which were associated with that device, * and forget them. */ - NF_CT_ASSERT(dev->ifindex != 0); + WARN_ON(dev->ifindex == 0); nf_ct_iterate_cleanup_net(net, device_cmp, (void *)(long)dev->ifindex, 0, 0); diff --git a/net/ipv4/netfilter/nf_tables_arp.c b/net/ipv4/netfilter/nf_tables_arp.c index 805c8ddfe860..4bbc273b45e8 100644 --- a/net/ipv4/netfilter/nf_tables_arp.c +++ b/net/ipv4/netfilter/nf_tables_arp.c @@ -72,8 +72,7 @@ static const struct nf_chain_type filter_arp = { .family = NFPROTO_ARP, .owner = THIS_MODULE, .hook_mask = (1 << NF_ARP_IN) | - (1 << NF_ARP_OUT) | - (1 << NF_ARP_FORWARD), + (1 << NF_ARP_OUT), }; static int __init nf_tables_arp_init(void) diff --git a/net/ipv4/netfilter/nft_fib_ipv4.c b/net/ipv4/netfilter/nft_fib_ipv4.c index de3681df2ce7..e50976e3c213 100644 --- a/net/ipv4/netfilter/nft_fib_ipv4.c +++ b/net/ipv4/netfilter/nft_fib_ipv4.c @@ -32,9 +32,10 @@ void nft_fib4_eval_type(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_fib *priv = nft_expr_priv(expr); + int noff = skb_network_offset(pkt->skb); u32 *dst = ®s->data[priv->dreg]; const struct net_device *dev = NULL; - const struct iphdr *iph; + struct iphdr *iph, _iph; __be32 addr; if (priv->flags & NFTA_FIB_F_IIF) @@ -42,7 +43,12 @@ void nft_fib4_eval_type(const struct nft_expr *expr, struct nft_regs *regs, else if (priv->flags & NFTA_FIB_F_OIF) dev = nft_out(pkt); - iph = ip_hdr(pkt->skb); + iph = skb_header_pointer(pkt->skb, noff, sizeof(_iph), &_iph); + if (!iph) { + regs->verdict.code = NFT_BREAK; + return; + } + if (priv->flags & NFTA_FIB_F_DADDR) addr = iph->daddr; else @@ -61,8 +67,9 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_fib *priv = nft_expr_priv(expr); + int noff = skb_network_offset(pkt->skb); u32 *dest = ®s->data[priv->dreg]; - const struct iphdr *iph; + struct iphdr *iph, _iph; struct fib_result res; struct flowi4 fl4 = { .flowi4_scope = RT_SCOPE_UNIVERSE, @@ -95,7 +102,12 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, return; } - iph = ip_hdr(pkt->skb); + iph = skb_header_pointer(pkt->skb, noff, sizeof(_iph), &_iph); + if (!iph) { + regs->verdict.code = NFT_BREAK; + return; + } + if (ipv4_is_zeronet(iph->saddr)) { if (ipv4_is_lbcast(iph->daddr) || ipv4_is_local_multicast(iph->daddr)) { diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 43eb6567b3a0..127153f1ed8a 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -206,12 +206,7 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("DelayedACKLost", LINUX_MIB_DELAYEDACKLOST), SNMP_MIB_ITEM("ListenOverflows", LINUX_MIB_LISTENOVERFLOWS), SNMP_MIB_ITEM("ListenDrops", LINUX_MIB_LISTENDROPS), - SNMP_MIB_ITEM("TCPPrequeued", LINUX_MIB_TCPPREQUEUED), - SNMP_MIB_ITEM("TCPDirectCopyFromBacklog", LINUX_MIB_TCPDIRECTCOPYFROMBACKLOG), - SNMP_MIB_ITEM("TCPDirectCopyFromPrequeue", LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE), - SNMP_MIB_ITEM("TCPPrequeueDropped", LINUX_MIB_TCPPREQUEUEDROPPED), SNMP_MIB_ITEM("TCPHPHits", LINUX_MIB_TCPHPHITS), - SNMP_MIB_ITEM("TCPHPHitsToUser", LINUX_MIB_TCPHPHITSTOUSER), SNMP_MIB_ITEM("TCPPureAcks", LINUX_MIB_TCPPUREACKS), SNMP_MIB_ITEM("TCPHPAcks", LINUX_MIB_TCPHPACKS), SNMP_MIB_ITEM("TCPRenoRecovery", LINUX_MIB_TCPRENORECOVERY), @@ -230,14 +225,12 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("TCPSackFailures", LINUX_MIB_TCPSACKFAILURES), SNMP_MIB_ITEM("TCPLossFailures", LINUX_MIB_TCPLOSSFAILURES), SNMP_MIB_ITEM("TCPFastRetrans", LINUX_MIB_TCPFASTRETRANS), - SNMP_MIB_ITEM("TCPForwardRetrans", LINUX_MIB_TCPFORWARDRETRANS), SNMP_MIB_ITEM("TCPSlowStartRetrans", LINUX_MIB_TCPSLOWSTARTRETRANS), SNMP_MIB_ITEM("TCPTimeouts", LINUX_MIB_TCPTIMEOUTS), SNMP_MIB_ITEM("TCPLossProbes", LINUX_MIB_TCPLOSSPROBES), SNMP_MIB_ITEM("TCPLossProbeRecovery", LINUX_MIB_TCPLOSSPROBERECOVERY), SNMP_MIB_ITEM("TCPRenoRecoveryFail", LINUX_MIB_TCPRENORECOVERYFAIL), SNMP_MIB_ITEM("TCPSackRecoveryFail", LINUX_MIB_TCPSACKRECOVERYFAIL), - SNMP_MIB_ITEM("TCPSchedulerFailed", LINUX_MIB_TCPSCHEDULERFAILED), SNMP_MIB_ITEM("TCPRcvCollapsed", LINUX_MIB_TCPRCVCOLLAPSED), SNMP_MIB_ITEM("TCPDSACKOldSent", LINUX_MIB_TCPDSACKOLDSENT), SNMP_MIB_ITEM("TCPDSACKOfoSent", LINUX_MIB_TCPDSACKOFOSENT), diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index b0bb5d0a30bd..33b70bfd1122 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -122,7 +122,8 @@ void raw_unhash_sk(struct sock *sk) EXPORT_SYMBOL_GPL(raw_unhash_sk); struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, - unsigned short num, __be32 raddr, __be32 laddr, int dif) + unsigned short num, __be32 raddr, __be32 laddr, + int dif, int sdif) { sk_for_each_from(sk) { struct inet_sock *inet = inet_sk(sk); @@ -130,7 +131,8 @@ struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, if (net_eq(sock_net(sk), net) && inet->inet_num == num && !(inet->inet_daddr && inet->inet_daddr != raddr) && !(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) && - !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)) + !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif && + sk->sk_bound_dev_if != sdif)) goto found; /* gotcha */ } sk = NULL; @@ -171,6 +173,7 @@ static int icmp_filter(const struct sock *sk, const struct sk_buff *skb) */ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash) { + int sdif = inet_sdif(skb); struct sock *sk; struct hlist_head *head; int delivered = 0; @@ -184,13 +187,13 @@ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash) net = dev_net(skb->dev); sk = __raw_v4_lookup(net, __sk_head(head), iph->protocol, iph->saddr, iph->daddr, - skb->dev->ifindex); + skb->dev->ifindex, sdif); while (sk) { delivered = 1; if ((iph->protocol != IPPROTO_ICMP || !icmp_filter(sk, skb)) && ip_mc_sf_allow(sk, iph->daddr, iph->saddr, - skb->dev->ifindex)) { + skb->dev->ifindex, sdif)) { struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ @@ -199,7 +202,7 @@ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash) } sk = __raw_v4_lookup(net, sk_next(sk), iph->protocol, iph->saddr, iph->daddr, - skb->dev->ifindex); + skb->dev->ifindex, sdif); } out: read_unlock(&raw_v4_hashinfo.lock); @@ -297,12 +300,15 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info) read_lock(&raw_v4_hashinfo.lock); raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]); if (raw_sk) { + int dif = skb->dev->ifindex; + int sdif = inet_sdif(skb); + iph = (const struct iphdr *)skb->data; net = dev_net(skb->dev); while ((raw_sk = __raw_v4_lookup(net, raw_sk, protocol, iph->daddr, iph->saddr, - skb->dev->ifindex)) != NULL) { + dif, sdif)) != NULL) { raw_err(raw_sk, skb, info); raw_sk = sk_next(raw_sk); iph = (const struct iphdr *)skb->data; diff --git a/net/ipv4/raw_diag.c b/net/ipv4/raw_diag.c index e1a51ca68d23..c200065ef9a5 100644 --- a/net/ipv4/raw_diag.c +++ b/net/ipv4/raw_diag.c @@ -46,13 +46,13 @@ static struct sock *raw_lookup(struct net *net, struct sock *from, sk = __raw_v4_lookup(net, from, r->sdiag_raw_protocol, r->id.idiag_dst[0], r->id.idiag_src[0], - r->id.idiag_if); + r->id.idiag_if, 0); #if IS_ENABLED(CONFIG_IPV6) else sk = __raw_v6_lookup(net, from, r->sdiag_raw_protocol, (const struct in6_addr *)r->id.idiag_src, (const struct in6_addr *)r->id.idiag_dst, - r->id.idiag_if); + r->id.idiag_if, 0); #endif return sk; } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 0383e66f59bc..94d4cd2d5ea4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1267,7 +1267,7 @@ static unsigned int ipv4_mtu(const struct dst_entry *dst) if (mtu) return mtu; - mtu = dst->dev->mtu; + mtu = READ_ONCE(dst->dev->mtu); if (unlikely(dst_metric_locked(dst, RTAX_MTU))) { if (rt->rt_uses_gateway && mtu > 576) @@ -1398,7 +1398,7 @@ static void ipv4_dst_destroy(struct dst_entry *dst) struct dst_metrics *p = (struct dst_metrics *)DST_METRICS_PTR(dst); struct rtable *rt = (struct rtable *) dst; - if (p != &dst_default_metrics && atomic_dec_and_test(&p->refcnt)) + if (p != &dst_default_metrics && refcount_dec_and_test(&p->refcnt)) kfree(p); if (!list_empty(&rt->rt_uncached)) { @@ -1456,7 +1456,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr, dst_init_metrics(&rt->dst, fi->fib_metrics->metrics, true); if (fi->fib_metrics != &dst_default_metrics) { rt->dst._metrics |= DST_METRICS_REFCOUNTED; - atomic_inc(&fi->fib_metrics->refcnt); + refcount_inc(&fi->fib_metrics->refcnt); } #ifdef CONFIG_IP_ROUTE_CLASSID rt->dst.tclassid = nh->nh_tclassid; @@ -2236,7 +2236,7 @@ add: if (!rth) return ERR_PTR(-ENOBUFS); - rth->rt_iif = orig_oif ? : 0; + rth->rt_iif = orig_oif; if (res->table) rth->rt_table_id = res->table->tb_id; @@ -2439,6 +2439,12 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, /* L3 master device is the loopback for that domain */ dev_out = l3mdev_master_dev_rcu(FIB_RES_DEV(*res)) ? : net->loopback_dev; + + /* make sure orig_oif points to fib result device even + * though packet rx/tx happens over loopback or l3mdev + */ + orig_oif = FIB_RES_OIF(*res); + fl4->flowi4_oif = dev_out->ifindex; flags |= RTCF_LOCAL; goto make_route; @@ -2750,26 +2756,34 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, err = 0; if (IS_ERR(rt)) err = PTR_ERR(rt); + else + skb_dst_set(skb, &rt->dst); } if (err) goto errout_free; - skb_dst_set(skb, &rt->dst); if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; if (rtm->rtm_flags & RTM_F_LOOKUP_TABLE) table_id = rt->rt_table_id; - if (rtm->rtm_flags & RTM_F_FIB_MATCH) + if (rtm->rtm_flags & RTM_F_FIB_MATCH) { + if (!res.fi) { + err = fib_props[res.type].error; + if (!err) + err = -EHOSTUNREACH; + goto errout_free; + } err = fib_dump_info(skb, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, RTM_NEWROUTE, table_id, rt->rt_type, res.prefix, res.prefixlen, fl4.flowi4_tos, res.fi, 0); - else + } else { err = rt_fill_info(net, dst, src, table_id, &fl4, skb, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq); + } if (err < 0) goto errout_free; @@ -3067,7 +3081,8 @@ int __init ip_rt_init(void) xfrm_init(); xfrm4_init(); #endif - rtnl_register(PF_INET, RTM_GETROUTE, inet_rtm_getroute, NULL, NULL); + rtnl_register(PF_INET, RTM_GETROUTE, inet_rtm_getroute, NULL, + RTNL_FLAG_DOIT_UNLOCKED); #ifdef CONFIG_SYSCTL register_pernet_subsys(&sysctl_route_ops); diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 0905cf04c2a4..b1bb1b3a1082 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -335,6 +335,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) treq->rcv_isn = ntohl(th->seq) - 1; treq->snt_isn = cookie; treq->ts_off = 0; + treq->txhash = net_tx_rndhash(); req->mss = mss; ireq->ir_num = ntohs(th->dest); ireq->ir_rmt_port = th->source; @@ -354,7 +355,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) /* We throwed the options of the initial SYN away, so we hope * the ACK carries the same options again (see RFC1122 4.2.3.8) */ - ireq->opt = tcp_v4_save_options(skb); + ireq->opt = tcp_v4_save_options(sock_net(sk), skb); if (security_inet_conn_request(sk, skb, req)) { reqsk_free(req); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 9bf809726066..0d3c038d7b04 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -45,6 +45,9 @@ static int tcp_syn_retries_max = MAX_TCP_SYNCNT; static int ip_ping_group_range_min[] = { 0, 0 }; static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; +/* obsolete */ +static int sysctl_tcp_low_latency __read_mostly; + /* Update system visible IP port range */ static void set_local_port_range(struct net *net, int range[2]) { diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 71ce33decd97..5091402720ab 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -269,6 +269,7 @@ #include <linux/err.h> #include <linux/time.h> #include <linux/slab.h> +#include <linux/errqueue.h> #include <net/icmp.h> #include <net/inet_common.h> @@ -388,6 +389,19 @@ static int retrans_to_secs(u8 retrans, int timeout, int rto_max) return period; } +static u64 tcp_compute_delivery_rate(const struct tcp_sock *tp) +{ + u32 rate = READ_ONCE(tp->rate_delivered); + u32 intv = READ_ONCE(tp->rate_interval_us); + u64 rate64 = 0; + + if (rate && intv) { + rate64 = (u64)rate * tp->mss_cache * USEC_PER_SEC; + do_div(rate64, intv); + } + return rate64; +} + /* Address-family independent initialization for a tcp_sock. * * NOTE: A lot of things set to zero explicitly by call to @@ -400,7 +414,6 @@ void tcp_init_sock(struct sock *sk) tp->out_of_order_queue = RB_ROOT; tcp_init_xmit_timers(sk); - tcp_prequeue_init(tp); INIT_LIST_HEAD(&tp->tsq_node); icsk->icsk_rto = TCP_TIMEOUT_INIT; @@ -1034,23 +1047,29 @@ out_err: } EXPORT_SYMBOL_GPL(do_tcp_sendpages); -int tcp_sendpage(struct sock *sk, struct page *page, int offset, - size_t size, int flags) +int tcp_sendpage_locked(struct sock *sk, struct page *page, int offset, + size_t size, int flags) { - ssize_t res; - if (!(sk->sk_route_caps & NETIF_F_SG) || !sk_check_csum_caps(sk)) - return sock_no_sendpage(sk->sk_socket, page, offset, size, - flags); - - lock_sock(sk); + return sock_no_sendpage_locked(sk, page, offset, size, flags); tcp_rate_check_app_limited(sk); /* is sending application-limited? */ - res = do_tcp_sendpages(sk, page, offset, size, flags); + return do_tcp_sendpages(sk, page, offset, size, flags); +} +EXPORT_SYMBOL_GPL(tcp_sendpage_locked); + +int tcp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags) +{ + int ret; + + lock_sock(sk); + ret = tcp_sendpage_locked(sk, page, offset, size, flags); release_sock(sk); - return res; + + return ret; } EXPORT_SYMBOL(tcp_sendpage); @@ -1144,9 +1163,10 @@ static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, return err; } -int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) +int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) { struct tcp_sock *tp = tcp_sk(sk); + struct ubuf_info *uarg = NULL; struct sk_buff *skb; struct sockcm_cookie sockc; int flags, err, copied = 0; @@ -1155,9 +1175,25 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) bool sg; long timeo; - lock_sock(sk); - flags = msg->msg_flags; + + if (flags & MSG_ZEROCOPY && size) { + if (sk->sk_state != TCP_ESTABLISHED) { + err = -EINVAL; + goto out_err; + } + + skb = tcp_send_head(sk) ? tcp_write_queue_tail(sk) : NULL; + uarg = sock_zerocopy_realloc(sk, size, skb_zcopy(skb)); + if (!uarg) { + err = -ENOBUFS; + goto out_err; + } + + if (!(sk_check_csum_caps(sk) && sk->sk_route_caps & NETIF_F_SG)) + uarg->zerocopy = 0; + } + if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect)) { err = tcp_sendmsg_fastopen(sk, msg, &copied_syn, size); if (err == -EINPROGRESS && copied_syn > 0) @@ -1281,7 +1317,7 @@ new_segment: err = skb_add_data_nocache(sk, skb, &msg->msg_iter, copy); if (err) goto do_fault; - } else { + } else if (!uarg || !uarg->zerocopy) { bool merge = true; int i = skb_shinfo(skb)->nr_frags; struct page_frag *pfrag = sk_page_frag(sk); @@ -1319,6 +1355,13 @@ new_segment: page_ref_inc(pfrag->page); } pfrag->offset += copy; + } else { + err = skb_zerocopy_iter_stream(sk, skb, msg, copy, uarg); + if (err == -EMSGSIZE || err == -EEXIST) + goto new_segment; + if (err < 0) + goto do_error; + copy = err; } if (!copied) @@ -1365,7 +1408,7 @@ out: tcp_push(sk, flags, mss_now, tp->nonagle, size_goal); } out_nopush: - release_sock(sk); + sock_zerocopy_put(uarg); return copied + copied_syn; do_fault: @@ -1382,6 +1425,7 @@ do_error: if (copied + copied_syn) goto out; out_err: + sock_zerocopy_put_abort(uarg); err = sk_stream_error(sk, flags, err); /* make sure we wake any epoll edge trigger waiter */ if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && @@ -1389,9 +1433,20 @@ out_err: sk->sk_write_space(sk); tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED); } - release_sock(sk); return err; } +EXPORT_SYMBOL_GPL(tcp_sendmsg_locked); + +int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) +{ + int ret; + + lock_sock(sk); + ret = tcp_sendmsg_locked(sk, msg, size); + release_sock(sk); + + return ret; +} EXPORT_SYMBOL(tcp_sendmsg); /* @@ -1525,20 +1580,6 @@ static void tcp_cleanup_rbuf(struct sock *sk, int copied) tcp_send_ack(sk); } -static void tcp_prequeue_process(struct sock *sk) -{ - struct sk_buff *skb; - struct tcp_sock *tp = tcp_sk(sk); - - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPPREQUEUED); - - while ((skb = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) - sk_backlog_rcv(sk, skb); - - /* Clear memory counter. */ - tp->ucopy.memory = 0; -} - static struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off) { struct sk_buff *skb; @@ -1652,6 +1693,61 @@ int tcp_peek_len(struct socket *sock) } EXPORT_SYMBOL(tcp_peek_len); +static void tcp_update_recv_tstamps(struct sk_buff *skb, + struct scm_timestamping *tss) +{ + if (skb->tstamp) + tss->ts[0] = ktime_to_timespec(skb->tstamp); + else + tss->ts[0] = (struct timespec) {0}; + + if (skb_hwtstamps(skb)->hwtstamp) + tss->ts[2] = ktime_to_timespec(skb_hwtstamps(skb)->hwtstamp); + else + tss->ts[2] = (struct timespec) {0}; +} + +/* Similar to __sock_recv_timestamp, but does not require an skb */ +void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, + struct scm_timestamping *tss) +{ + struct timeval tv; + bool has_timestamping = false; + + if (tss->ts[0].tv_sec || tss->ts[0].tv_nsec) { + if (sock_flag(sk, SOCK_RCVTSTAMP)) { + if (sock_flag(sk, SOCK_RCVTSTAMPNS)) { + put_cmsg(msg, SOL_SOCKET, SCM_TIMESTAMPNS, + sizeof(tss->ts[0]), &tss->ts[0]); + } else { + tv.tv_sec = tss->ts[0].tv_sec; + tv.tv_usec = tss->ts[0].tv_nsec / 1000; + + put_cmsg(msg, SOL_SOCKET, SCM_TIMESTAMP, + sizeof(tv), &tv); + } + } + + if (sk->sk_tsflags & SOF_TIMESTAMPING_SOFTWARE) + has_timestamping = true; + else + tss->ts[0] = (struct timespec) {0}; + } + + if (tss->ts[2].tv_sec || tss->ts[2].tv_nsec) { + if (sk->sk_tsflags & SOF_TIMESTAMPING_RAW_HARDWARE) + has_timestamping = true; + else + tss->ts[2] = (struct timespec) {0}; + } + + if (has_timestamping) { + tss->ts[1] = (struct timespec) {0}; + put_cmsg(msg, SOL_SOCKET, SCM_TIMESTAMPING, + sizeof(*tss), tss); + } +} + /* * This routine copies from a sock struct into the user buffer. * @@ -1671,9 +1767,10 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int err; int target; /* Read at least this many bytes */ long timeo; - struct task_struct *user_recv = NULL; struct sk_buff *skb, *last; u32 urg_hole = 0; + struct scm_timestamping tss; + bool has_tss = false; if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len, addr_len); @@ -1806,51 +1903,6 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, tcp_cleanup_rbuf(sk, copied); - if (!sysctl_tcp_low_latency && tp->ucopy.task == user_recv) { - /* Install new reader */ - if (!user_recv && !(flags & (MSG_TRUNC | MSG_PEEK))) { - user_recv = current; - tp->ucopy.task = user_recv; - tp->ucopy.msg = msg; - } - - tp->ucopy.len = len; - - WARN_ON(tp->copied_seq != tp->rcv_nxt && - !(flags & (MSG_PEEK | MSG_TRUNC))); - - /* Ugly... If prequeue is not empty, we have to - * process it before releasing socket, otherwise - * order will be broken at second iteration. - * More elegant solution is required!!! - * - * Look: we have the following (pseudo)queues: - * - * 1. packets in flight - * 2. backlog - * 3. prequeue - * 4. receive_queue - * - * Each queue can be processed only if the next ones - * are empty. At this point we have empty receive_queue. - * But prequeue _can_ be not empty after 2nd iteration, - * when we jumped to start of loop because backlog - * processing added something to receive_queue. - * We cannot release_sock(), because backlog contains - * packets arrived _after_ prequeued ones. - * - * Shortly, algorithm is clear --- to process all - * the queues in order. We could make it more directly, - * requeueing packets from backlog to prequeue, if - * is not empty. It is more elegant, but eats cycles, - * unfortunately. - */ - if (!skb_queue_empty(&tp->ucopy.prequeue)) - goto do_prequeue; - - /* __ Set realtime policy in scheduler __ */ - } - if (copied >= target) { /* Do not sleep, just process backlog. */ release_sock(sk); @@ -1859,31 +1911,6 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, sk_wait_data(sk, &timeo, last); } - if (user_recv) { - int chunk; - - /* __ Restore normal policy in scheduler __ */ - - chunk = len - tp->ucopy.len; - if (chunk != 0) { - NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMBACKLOG, chunk); - len -= chunk; - copied += chunk; - } - - if (tp->rcv_nxt == tp->copied_seq && - !skb_queue_empty(&tp->ucopy.prequeue)) { -do_prequeue: - tcp_prequeue_process(sk); - - chunk = len - tp->ucopy.len; - if (chunk != 0) { - NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk); - len -= chunk; - copied += chunk; - } - } - } if ((flags & MSG_PEEK) && (peek_seq - copied - urg_hole != tp->copied_seq)) { net_dbg_ratelimited("TCP(%s:%d): Application bug, race in MSG_PEEK\n", @@ -1941,6 +1968,10 @@ skip_copy: if (used + offset < skb->len) continue; + if (TCP_SKB_CB(skb)->has_rxtstamp) { + tcp_update_recv_tstamps(skb, &tss); + has_tss = true; + } if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) goto found_fin_ok; if (!(flags & MSG_PEEK)) @@ -1955,29 +1986,13 @@ skip_copy: break; } while (len > 0); - if (user_recv) { - if (!skb_queue_empty(&tp->ucopy.prequeue)) { - int chunk; - - tp->ucopy.len = copied > 0 ? len : 0; - - tcp_prequeue_process(sk); - - if (copied > 0 && (chunk = len - tp->ucopy.len) != 0) { - NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPDIRECTCOPYFROMPREQUEUE, chunk); - len -= chunk; - copied += chunk; - } - } - - tp->ucopy.task = NULL; - tp->ucopy.len = 0; - } - /* According to UNIX98, msg_name/msg_namelen are ignored * on connected socket. I was just happy when found this 8) --ANK */ + if (has_tss) + tcp_recv_timestamp(msg, sk, &tss); + /* Clean up data we have read: This will do ACK frames. */ tcp_cleanup_rbuf(sk, copied); @@ -2481,7 +2496,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, name[val] = 0; lock_sock(sk); - err = tcp_set_congestion_control(sk, name, true); + err = tcp_set_congestion_control(sk, name, true, true); release_sock(sk); return err; } @@ -2823,7 +2838,7 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info) { const struct tcp_sock *tp = tcp_sk(sk); /* iff sk_type == SOCK_STREAM */ const struct inet_connection_sock *icsk = inet_csk(sk); - u32 now, intv; + u32 now; u64 rate64; bool slow; u32 rate; @@ -2922,13 +2937,9 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info) info->tcpi_data_segs_out = tp->data_segs_out; info->tcpi_delivery_rate_app_limited = tp->rate_app_limited ? 1 : 0; - rate = READ_ONCE(tp->rate_delivered); - intv = READ_ONCE(tp->rate_interval_us); - if (rate && intv) { - rate64 = (u64)rate * tp->mss_cache * USEC_PER_SEC; - do_div(rate64, intv); + rate64 = tcp_compute_delivery_rate(tp); + if (rate64) info->tcpi_delivery_rate = rate64; - } unlock_sock_fast(sk, slow); } EXPORT_SYMBOL_GPL(tcp_get_info); @@ -2938,8 +2949,12 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) const struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *stats; struct tcp_info info; + u64 rate64; + u32 rate; - stats = alloc_skb(5 * nla_total_size_64bit(sizeof(u64)), GFP_ATOMIC); + stats = alloc_skb(7 * nla_total_size_64bit(sizeof(u64)) + + 3 * nla_total_size(sizeof(u32)) + + 2 * nla_total_size(sizeof(u8)), GFP_ATOMIC); if (!stats) return NULL; @@ -2954,6 +2969,20 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) tp->data_segs_out, TCP_NLA_PAD); nla_put_u64_64bit(stats, TCP_NLA_TOTAL_RETRANS, tp->total_retrans, TCP_NLA_PAD); + + rate = READ_ONCE(sk->sk_pacing_rate); + rate64 = rate != ~0U ? rate : ~0ULL; + nla_put_u64_64bit(stats, TCP_NLA_PACING_RATE, rate64, TCP_NLA_PAD); + + rate64 = tcp_compute_delivery_rate(tp); + nla_put_u64_64bit(stats, TCP_NLA_DELIVERY_RATE, rate64, TCP_NLA_PAD); + + nla_put_u32(stats, TCP_NLA_SND_CWND, tp->snd_cwnd); + nla_put_u32(stats, TCP_NLA_REORDERING, tp->reordering); + nla_put_u32(stats, TCP_NLA_MIN_RTT, tcp_min_rtt(tp)); + + nla_put_u8(stats, TCP_NLA_RECUR_RETRANS, inet_csk(sk)->icsk_retransmits); + nla_put_u8(stats, TCP_NLA_DELIVERY_RATE_APP_LMT, !!tp->rate_app_limited); return stats; } diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index dbcc9352a48f..69ee877574d0 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -112,7 +112,8 @@ struct bbr { cwnd_gain:10, /* current gain for setting cwnd */ full_bw_cnt:3, /* number of rounds without large bw gains */ cycle_idx:3, /* current index in pacing_gain cycle array */ - unused_b:6; + has_seen_rtt:1, /* have we seen an RTT sample yet? */ + unused_b:5; u32 prior_cwnd; /* prior cwnd upon entering loss recovery */ u32 full_bw; /* recent bw, to estimate if pipe is full */ }; @@ -211,6 +212,35 @@ static u64 bbr_rate_bytes_per_sec(struct sock *sk, u64 rate, int gain) return rate >> BW_SCALE; } +/* Convert a BBR bw and gain factor to a pacing rate in bytes per second. */ +static u32 bbr_bw_to_pacing_rate(struct sock *sk, u32 bw, int gain) +{ + u64 rate = bw; + + rate = bbr_rate_bytes_per_sec(sk, rate, gain); + rate = min_t(u64, rate, sk->sk_max_pacing_rate); + return rate; +} + +/* Initialize pacing rate to: high_gain * init_cwnd / RTT. */ +static void bbr_init_pacing_rate_from_rtt(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct bbr *bbr = inet_csk_ca(sk); + u64 bw; + u32 rtt_us; + + if (tp->srtt_us) { /* any RTT sample yet? */ + rtt_us = max(tp->srtt_us >> 3, 1U); + bbr->has_seen_rtt = 1; + } else { /* no RTT sample yet */ + rtt_us = USEC_PER_MSEC; /* use nominal default RTT */ + } + bw = (u64)tp->snd_cwnd * BW_UNIT; + do_div(bw, rtt_us); + sk->sk_pacing_rate = bbr_bw_to_pacing_rate(sk, bw, bbr_high_gain); +} + /* Pace using current bw estimate and a gain factor. In order to help drive the * network toward lower queues while maintaining high utilization and low * latency, the average pacing rate aims to be slightly (~1%) lower than the @@ -220,12 +250,13 @@ static u64 bbr_rate_bytes_per_sec(struct sock *sk, u64 rate, int gain) */ static void bbr_set_pacing_rate(struct sock *sk, u32 bw, int gain) { + struct tcp_sock *tp = tcp_sk(sk); struct bbr *bbr = inet_csk_ca(sk); - u64 rate = bw; + u32 rate = bbr_bw_to_pacing_rate(sk, bw, gain); - rate = bbr_rate_bytes_per_sec(sk, rate, gain); - rate = min_t(u64, rate, sk->sk_max_pacing_rate); - if (bbr->mode != BBR_STARTUP || rate > sk->sk_pacing_rate) + if (unlikely(!bbr->has_seen_rtt && tp->srtt_us)) + bbr_init_pacing_rate_from_rtt(sk); + if (bbr_full_bw_reached(sk) || rate > sk->sk_pacing_rate) sk->sk_pacing_rate = rate; } @@ -798,7 +829,6 @@ static void bbr_init(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct bbr *bbr = inet_csk_ca(sk); - u64 bw; bbr->prior_cwnd = 0; bbr->tso_segs_goal = 0; /* default segs per skb until first ACK */ @@ -814,11 +844,8 @@ static void bbr_init(struct sock *sk) minmax_reset(&bbr->bw, bbr->rtt_cnt, 0); /* init max bw to 0 */ - /* Initialize pacing rate to: high_gain * init_cwnd / RTT. */ - bw = (u64)tp->snd_cwnd * BW_UNIT; - do_div(bw, (tp->srtt_us >> 3) ? : USEC_PER_MSEC); - sk->sk_pacing_rate = 0; /* force an update of sk_pacing_rate */ - bbr_set_pacing_rate(sk, bw, bbr_high_gain); + bbr->has_seen_rtt = 0; + bbr_init_pacing_rate_from_rtt(sk); bbr->restore_cwnd = 0; bbr->round_start = 0; diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c index 609965f0e298..fc3614377413 100644 --- a/net/ipv4/tcp_bic.c +++ b/net/ipv4/tcp_bic.c @@ -49,7 +49,6 @@ MODULE_PARM_DESC(smooth_part, "log(B/(B*Smin))/log(B/(B-1))+B, # of RTT from Wma struct bictcp { u32 cnt; /* increase cwnd by 1 after ACKs */ u32 last_max_cwnd; /* last maximum snd_cwnd */ - u32 loss_cwnd; /* congestion window at last loss */ u32 last_cwnd; /* the last snd_cwnd */ u32 last_time; /* time when updated last_cwnd */ u32 epoch_start; /* beginning of an epoch */ @@ -72,7 +71,6 @@ static void bictcp_init(struct sock *sk) struct bictcp *ca = inet_csk_ca(sk); bictcp_reset(ca); - ca->loss_cwnd = 0; if (initial_ssthresh) tcp_sk(sk)->snd_ssthresh = initial_ssthresh; @@ -172,22 +170,12 @@ static u32 bictcp_recalc_ssthresh(struct sock *sk) else ca->last_max_cwnd = tp->snd_cwnd; - ca->loss_cwnd = tp->snd_cwnd; - if (tp->snd_cwnd <= low_window) return max(tp->snd_cwnd >> 1U, 2U); else return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U); } -static u32 bictcp_undo_cwnd(struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - const struct bictcp *ca = inet_csk_ca(sk); - - return max(tp->snd_cwnd, ca->loss_cwnd); -} - static void bictcp_state(struct sock *sk, u8 new_state) { if (new_state == TCP_CA_Loss) @@ -214,7 +202,7 @@ static struct tcp_congestion_ops bictcp __read_mostly = { .ssthresh = bictcp_recalc_ssthresh, .cong_avoid = bictcp_cong_avoid, .set_state = bictcp_state, - .undo_cwnd = bictcp_undo_cwnd, + .undo_cwnd = tcp_reno_undo_cwnd, .pkts_acked = bictcp_acked, .owner = THIS_MODULE, .name = "bic", diff --git a/net/ipv4/tcp_cdg.c b/net/ipv4/tcp_cdg.c index 50a0f3e51d5b..66ac69f7bd19 100644 --- a/net/ipv4/tcp_cdg.c +++ b/net/ipv4/tcp_cdg.c @@ -85,7 +85,6 @@ struct cdg { u8 state; u8 delack; u32 rtt_seq; - u32 undo_cwnd; u32 shadow_wnd; u16 backoff_cnt; u16 sample_cnt; @@ -330,8 +329,6 @@ static u32 tcp_cdg_ssthresh(struct sock *sk) struct cdg *ca = inet_csk_ca(sk); struct tcp_sock *tp = tcp_sk(sk); - ca->undo_cwnd = tp->snd_cwnd; - if (ca->state == CDG_BACKOFF) return max(2U, (tp->snd_cwnd * min(1024U, backoff_beta)) >> 10); @@ -344,13 +341,6 @@ static u32 tcp_cdg_ssthresh(struct sock *sk) return max(2U, tp->snd_cwnd >> 1); } -static u32 tcp_cdg_undo_cwnd(struct sock *sk) -{ - struct cdg *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->undo_cwnd); -} - static void tcp_cdg_cwnd_event(struct sock *sk, const enum tcp_ca_event ev) { struct cdg *ca = inet_csk_ca(sk); @@ -403,7 +393,7 @@ struct tcp_congestion_ops tcp_cdg __read_mostly = { .cong_avoid = tcp_cdg_cong_avoid, .cwnd_event = tcp_cdg_cwnd_event, .pkts_acked = tcp_cdg_acked, - .undo_cwnd = tcp_cdg_undo_cwnd, + .undo_cwnd = tcp_reno_undo_cwnd, .ssthresh = tcp_cdg_ssthresh, .release = tcp_cdg_release, .init = tcp_cdg_init, diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index fde983f6376b..2f26124fd160 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -189,8 +189,8 @@ void tcp_init_congestion_control(struct sock *sk) INET_ECN_dontxmit(sk); } -void tcp_reinit_congestion_control(struct sock *sk, - const struct tcp_congestion_ops *ca) +static void tcp_reinit_congestion_control(struct sock *sk, + const struct tcp_congestion_ops *ca) { struct inet_connection_sock *icsk = inet_csk(sk); @@ -338,7 +338,7 @@ out: * tcp_reinit_congestion_control (if the current congestion control was * already initialized. */ -int tcp_set_congestion_control(struct sock *sk, const char *name, bool load) +int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, bool reinit) { struct inet_connection_sock *icsk = inet_csk(sk); const struct tcp_congestion_ops *ca; @@ -360,9 +360,18 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load) if (!ca) { err = -ENOENT; } else if (!load) { - icsk->icsk_ca_ops = ca; - if (!try_module_get(ca->owner)) + const struct tcp_congestion_ops *old_ca = icsk->icsk_ca_ops; + + if (try_module_get(ca->owner)) { + if (reinit) { + tcp_reinit_congestion_control(sk, ca); + } else { + icsk->icsk_ca_ops = ca; + module_put(old_ca->owner); + } + } else { err = -EBUSY; + } } else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN))) { err = -EPERM; @@ -456,7 +465,7 @@ u32 tcp_reno_undo_cwnd(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - return max(tp->snd_cwnd, tp->snd_ssthresh << 1); + return max(tp->snd_cwnd, tp->prior_cwnd); } EXPORT_SYMBOL_GPL(tcp_reno_undo_cwnd); diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index 57ae5b5ae643..78bfadfcf342 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -83,7 +83,6 @@ MODULE_PARM_DESC(hystart_ack_delta, "spacing between ack's indicating train (mse struct bictcp { u32 cnt; /* increase cwnd by 1 after ACKs */ u32 last_max_cwnd; /* last maximum snd_cwnd */ - u32 loss_cwnd; /* congestion window at last loss */ u32 last_cwnd; /* the last snd_cwnd */ u32 last_time; /* time when updated last_cwnd */ u32 bic_origin_point;/* origin point of bic function */ @@ -142,7 +141,6 @@ static void bictcp_init(struct sock *sk) struct bictcp *ca = inet_csk_ca(sk); bictcp_reset(ca); - ca->loss_cwnd = 0; if (hystart) bictcp_hystart_reset(sk); @@ -366,18 +364,9 @@ static u32 bictcp_recalc_ssthresh(struct sock *sk) else ca->last_max_cwnd = tp->snd_cwnd; - ca->loss_cwnd = tp->snd_cwnd; - return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U); } -static u32 bictcp_undo_cwnd(struct sock *sk) -{ - struct bictcp *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); -} - static void bictcp_state(struct sock *sk, u8 new_state) { if (new_state == TCP_CA_Loss) { @@ -470,7 +459,7 @@ static struct tcp_congestion_ops cubictcp __read_mostly = { .ssthresh = bictcp_recalc_ssthresh, .cong_avoid = bictcp_cong_avoid, .set_state = bictcp_state, - .undo_cwnd = bictcp_undo_cwnd, + .undo_cwnd = tcp_reno_undo_cwnd, .cwnd_event = bictcp_cwnd_event, .pkts_acked = bictcp_acked, .owner = THIS_MODULE, diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index a748c74aa8b7..abbf0edcf6c2 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -16,6 +16,7 @@ #include <linux/tcp.h> +#include <net/netlink.h> #include <net/tcp.h> static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, @@ -36,6 +37,100 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, tcp_get_info(sk, info); } +#ifdef CONFIG_TCP_MD5SIG +static void tcp_diag_md5sig_fill(struct tcp_diag_md5sig *info, + const struct tcp_md5sig_key *key) +{ + info->tcpm_family = key->family; + info->tcpm_prefixlen = key->prefixlen; + info->tcpm_keylen = key->keylen; + memcpy(info->tcpm_key, key->key, key->keylen); + + if (key->family == AF_INET) + info->tcpm_addr[0] = key->addr.a4.s_addr; + #if IS_ENABLED(CONFIG_IPV6) + else if (key->family == AF_INET6) + memcpy(&info->tcpm_addr, &key->addr.a6, + sizeof(info->tcpm_addr)); + #endif +} + +static int tcp_diag_put_md5sig(struct sk_buff *skb, + const struct tcp_md5sig_info *md5sig) +{ + const struct tcp_md5sig_key *key; + struct tcp_diag_md5sig *info; + struct nlattr *attr; + int md5sig_count = 0; + + hlist_for_each_entry_rcu(key, &md5sig->head, node) + md5sig_count++; + if (md5sig_count == 0) + return 0; + + attr = nla_reserve(skb, INET_DIAG_MD5SIG, + md5sig_count * sizeof(struct tcp_diag_md5sig)); + if (!attr) + return -EMSGSIZE; + + info = nla_data(attr); + memset(info, 0, md5sig_count * sizeof(struct tcp_diag_md5sig)); + hlist_for_each_entry_rcu(key, &md5sig->head, node) { + tcp_diag_md5sig_fill(info++, key); + if (--md5sig_count == 0) + break; + } + + return 0; +} +#endif + +static int tcp_diag_get_aux(struct sock *sk, bool net_admin, + struct sk_buff *skb) +{ +#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); + if (md5sig) + err = tcp_diag_put_md5sig(skb, md5sig); + rcu_read_unlock(); + if (err < 0) + return err; + } +#endif + + return 0; +} + +static size_t tcp_diag_get_aux_size(struct sock *sk, bool net_admin) +{ + size_t size = 0; + +#ifdef CONFIG_TCP_MD5SIG + if (net_admin && sk_fullsock(sk)) { + const struct tcp_md5sig_info *md5sig; + const struct tcp_md5sig_key *key; + size_t md5sig_count = 0; + + rcu_read_lock(); + md5sig = rcu_dereference(tcp_sk(sk)->md5sig_info); + if (md5sig) { + hlist_for_each_entry_rcu(key, &md5sig->head, node) + md5sig_count++; + } + rcu_read_unlock(); + size += nla_total_size(md5sig_count * + sizeof(struct tcp_diag_md5sig)); + } +#endif + + return size; +} + static void tcp_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, const struct inet_diag_req_v2 *r, struct nlattr *bc) { @@ -68,13 +163,15 @@ static int tcp_diag_destroy(struct sk_buff *in_skb, #endif static const struct inet_diag_handler tcp_diag_handler = { - .dump = tcp_diag_dump, - .dump_one = tcp_diag_dump_one, - .idiag_get_info = tcp_diag_get_info, - .idiag_type = IPPROTO_TCP, - .idiag_info_size = sizeof(struct tcp_info), + .dump = tcp_diag_dump, + .dump_one = tcp_diag_dump_one, + .idiag_get_info = tcp_diag_get_info, + .idiag_get_aux = tcp_diag_get_aux, + .idiag_get_aux_size = tcp_diag_get_aux_size, + .idiag_type = IPPROTO_TCP, + .idiag_info_size = sizeof(struct tcp_info), #ifdef CONFIG_INET_DIAG_DESTROY - .destroy = tcp_diag_destroy, + .destroy = tcp_diag_destroy, #endif }; diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index ce9c7fef200f..e3c33220c418 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -171,7 +171,6 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb) static struct sock *tcp_fastopen_create_child(struct sock *sk, struct sk_buff *skb, - struct dst_entry *dst, struct request_sock *req) { struct tcp_sock *tp; @@ -278,8 +277,7 @@ static bool tcp_fastopen_queue_check(struct sock *sk) */ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, - struct tcp_fastopen_cookie *foc, - struct dst_entry *dst) + struct tcp_fastopen_cookie *foc) { struct tcp_fastopen_cookie valid_foc = { .len = -1 }; bool syn_data = TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1; @@ -312,7 +310,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, * data in SYN_RECV state. */ fastopen: - child = tcp_fastopen_create_child(sk, skb, dst, req); + child = tcp_fastopen_create_child(sk, skb, req); if (child) { foc->len = -1; NET_INC_STATS(sock_net(sk), diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c index 6d9879e93648..d1c33c91eadc 100644 --- a/net/ipv4/tcp_highspeed.c +++ b/net/ipv4/tcp_highspeed.c @@ -94,7 +94,6 @@ static const struct hstcp_aimd_val { struct hstcp { u32 ai; - u32 loss_cwnd; }; static void hstcp_init(struct sock *sk) @@ -153,22 +152,14 @@ static u32 hstcp_ssthresh(struct sock *sk) const struct tcp_sock *tp = tcp_sk(sk); struct hstcp *ca = inet_csk_ca(sk); - ca->loss_cwnd = tp->snd_cwnd; /* Do multiplicative decrease */ return max(tp->snd_cwnd - ((tp->snd_cwnd * hstcp_aimd_vals[ca->ai].md) >> 8), 2U); } -static u32 hstcp_cwnd_undo(struct sock *sk) -{ - const struct hstcp *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); -} - static struct tcp_congestion_ops tcp_highspeed __read_mostly = { .init = hstcp_init, .ssthresh = hstcp_ssthresh, - .undo_cwnd = hstcp_cwnd_undo, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = hstcp_cong_avoid, .owner = THIS_MODULE, diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c index 3eb78cde6ff0..082d479462fa 100644 --- a/net/ipv4/tcp_htcp.c +++ b/net/ipv4/tcp_htcp.c @@ -66,7 +66,6 @@ static inline void htcp_reset(struct htcp *ca) static u32 htcp_cwnd_undo(struct sock *sk) { - const struct tcp_sock *tp = tcp_sk(sk); struct htcp *ca = inet_csk_ca(sk); if (ca->undo_last_cong) { @@ -76,7 +75,7 @@ static u32 htcp_cwnd_undo(struct sock *sk) ca->undo_last_cong = 0; } - return max(tp->snd_cwnd, (tp->snd_ssthresh << 7) / ca->beta); + return tcp_reno_undo_cwnd(sk); } static inline void measure_rtt(struct sock *sk, u32 srtt) diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 60352ff4f5a8..7c843578f233 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -48,7 +48,6 @@ struct illinois { u32 end_seq; /* right edge of current RTT */ u32 alpha; /* Additive increase */ u32 beta; /* Muliplicative decrease */ - u32 loss_cwnd; /* cwnd on loss */ u16 acked; /* # packets acked by current ACK */ u8 rtt_above; /* average rtt has gone above threshold */ u8 rtt_low; /* # of rtts measurements below threshold */ @@ -297,18 +296,10 @@ static u32 tcp_illinois_ssthresh(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct illinois *ca = inet_csk_ca(sk); - ca->loss_cwnd = tp->snd_cwnd; /* Multiplicative decrease */ return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->beta) >> BETA_SHIFT), 2U); } -static u32 tcp_illinois_cwnd_undo(struct sock *sk) -{ - const struct illinois *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); -} - /* Extract info for Tcp socket info provided via netlink. */ static size_t tcp_illinois_info(struct sock *sk, u32 ext, int *attr, union tcp_cc_info *info) @@ -336,7 +327,7 @@ static size_t tcp_illinois_info(struct sock *sk, u32 ext, int *attr, static struct tcp_congestion_ops tcp_illinois __read_mostly = { .init = tcp_illinois_init, .ssthresh = tcp_illinois_ssthresh, - .undo_cwnd = tcp_illinois_cwnd_undo, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_illinois_cong_avoid, .set_state = tcp_illinois_state, .get_info = tcp_illinois_info, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2920e0cb09f8..c5d7656beeee 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -107,6 +107,7 @@ int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; #define FLAG_ORIG_SACK_ACKED 0x200 /* Never retransmitted data are (s)acked */ #define FLAG_SND_UNA_ADVANCED 0x400 /* Snd_una was changed (!= FLAG_DATA_ACKED) */ #define FLAG_DSACKING_ACK 0x800 /* SACK blocks contained D-SACK info */ +#define FLAG_SET_XMIT_TIMER 0x1000 /* Set TLP or RTO timer */ #define FLAG_SACK_RENEGING 0x2000 /* snd_una advanced to a sacked seq */ #define FLAG_UPDATE_TS_RECENT 0x4000 /* tcp_replace_ts_recent() */ #define FLAG_NO_CHALLENGE_ACK 0x8000 /* do not call tcp_send_challenge_ack() */ @@ -1951,6 +1952,7 @@ void tcp_enter_loss(struct sock *sk) !after(tp->high_seq, tp->snd_una) || (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(sk); + tp->prior_cwnd = tp->snd_cwnd; tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); tcp_ca_event(sk, CA_EVENT_LOSS); tcp_init_undo(tp); @@ -2520,8 +2522,8 @@ static inline void tcp_end_cwnd_reduction(struct sock *sk) return; /* Reset cwnd to ssthresh in CWR or Recovery (unless it's undone) */ - if (inet_csk(sk)->icsk_ca_state == TCP_CA_CWR || - (tp->undo_marker && tp->snd_ssthresh < TCP_INFINITE_SSTHRESH)) { + if (tp->snd_ssthresh < TCP_INFINITE_SSTHRESH && + (inet_csk(sk)->icsk_ca_state == TCP_CA_CWR || tp->undo_marker)) { tp->snd_cwnd = tp->snd_ssthresh; tp->snd_cwnd_stamp = tcp_jiffies32; } @@ -3004,21 +3006,24 @@ void tcp_rearm_rto(struct sock *sk) /* Offset the time elapsed after installing regular RTO */ if (icsk->icsk_pending == ICSK_TIME_REO_TIMEOUT || icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) { - struct sk_buff *skb = tcp_write_queue_head(sk); - u64 rto_time_stamp = skb->skb_mstamp + - jiffies_to_usecs(rto); - s64 delta_us = rto_time_stamp - tp->tcp_mstamp; + s64 delta_us = tcp_rto_delta_us(sk); /* delta_us may not be positive if the socket is locked * when the retrans timer fires and is rescheduled. */ - if (delta_us > 0) - rto = usecs_to_jiffies(delta_us); + rto = usecs_to_jiffies(max_t(int, delta_us, 1)); } inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, rto, TCP_RTO_MAX); } } +/* Try to schedule a loss probe; if that doesn't work, then schedule an RTO. */ +static void tcp_set_xmit_timer(struct sock *sk) +{ + if (!tcp_schedule_loss_probe(sk)) + tcp_rearm_rto(sk); +} + /* If we get here, the whole TSO packet has not been acked. */ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb) { @@ -3180,7 +3185,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, ca_rtt_us, sack->rate); if (flag & FLAG_ACKED) { - tcp_rearm_rto(sk); + flag |= FLAG_SET_XMIT_TIMER; /* set TLP or RTO timer */ if (unlikely(icsk->icsk_mtup.probe_size && !after(tp->mtu_probe.probe_seq_end, tp->snd_una))) { tcp_mtup_probe_success(sk); @@ -3208,7 +3213,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, * after when the head was last (re)transmitted. Otherwise the * timeout may continue to extend in loss recovery. */ - tcp_rearm_rto(sk); + flag |= FLAG_SET_XMIT_TIMER; /* set TLP or RTO timer */ } if (icsk->icsk_ca_ops->pkts_acked) { @@ -3580,9 +3585,6 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) if (after(ack, tp->snd_nxt)) goto invalid_ack; - if (icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) - tcp_rearm_rto(sk); - if (after(ack, prior_snd_una)) { flag |= FLAG_SND_UNA_ADVANCED; icsk->icsk_retransmits = 0; @@ -3647,18 +3649,20 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una, &acked, &sack_state); + if (tp->tlp_high_seq) + tcp_process_tlp_ack(sk, ack, flag); + /* If needed, reset TLP/RTO timer; RACK may later override this. */ + if (flag & FLAG_SET_XMIT_TIMER) + tcp_set_xmit_timer(sk); + if (tcp_ack_is_dubious(sk, flag)) { is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP)); tcp_fastretrans_alert(sk, acked, is_dupack, &flag, &rexmit); } - if (tp->tlp_high_seq) - tcp_process_tlp_ack(sk, ack, flag); if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) sk_dst_confirm(sk); - if (icsk->icsk_pending == ICSK_TIME_RETRANS) - tcp_schedule_loss_probe(sk); delivered = tp->delivered - delivered; /* freshly ACKed or SACKed */ lost = tp->lost - lost; /* freshly marked lost */ tcp_rate_gen(sk, delivered, lost, sack_state.rate); @@ -4264,9 +4268,15 @@ static void tcp_sack_remove(struct tcp_sock *tp) tp->rx_opt.num_sacks = num_sacks; } +enum tcp_queue { + OOO_QUEUE, + RCV_QUEUE, +}; + /** * tcp_try_coalesce - try to merge skb to prior one * @sk: socket + * @dest: destination queue * @to: prior buffer * @from: buffer to add in queue * @fragstolen: pointer to boolean @@ -4278,6 +4288,7 @@ static void tcp_sack_remove(struct tcp_sock *tp) * Returns true if caller should free @from instead of queueing it */ static bool tcp_try_coalesce(struct sock *sk, + enum tcp_queue dest, struct sk_buff *to, struct sk_buff *from, bool *fragstolen) @@ -4299,6 +4310,15 @@ static bool tcp_try_coalesce(struct sock *sk, TCP_SKB_CB(to)->end_seq = TCP_SKB_CB(from)->end_seq; TCP_SKB_CB(to)->ack_seq = TCP_SKB_CB(from)->ack_seq; TCP_SKB_CB(to)->tcp_flags |= TCP_SKB_CB(from)->tcp_flags; + + if (TCP_SKB_CB(from)->has_rxtstamp) { + TCP_SKB_CB(to)->has_rxtstamp = true; + if (dest == OOO_QUEUE) + TCP_SKB_CB(to)->swtstamp = TCP_SKB_CB(from)->swtstamp; + else + to->tstamp = from->tstamp; + } + return true; } @@ -4333,6 +4353,9 @@ static void tcp_ofo_queue(struct sock *sk) } p = rb_next(p); rb_erase(&skb->rbnode, &tp->out_of_order_queue); + /* Replace tstamp which was stomped by rbnode */ + if (TCP_SKB_CB(skb)->has_rxtstamp) + skb->tstamp = TCP_SKB_CB(skb)->swtstamp; if (unlikely(!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt))) { SOCK_DEBUG(sk, "ofo packet was already received\n"); @@ -4344,7 +4367,8 @@ static void tcp_ofo_queue(struct sock *sk) TCP_SKB_CB(skb)->end_seq); tail = skb_peek_tail(&sk->sk_receive_queue); - eaten = tail && tcp_try_coalesce(sk, tail, skb, &fragstolen); + eaten = tail && tcp_try_coalesce(sk, RCV_QUEUE, + tail, skb, &fragstolen); tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); fin = TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN; if (!eaten) @@ -4398,6 +4422,10 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) return; } + /* Stash tstamp to avoid being stomped on by rbnode */ + if (TCP_SKB_CB(skb)->has_rxtstamp) + TCP_SKB_CB(skb)->swtstamp = skb->tstamp; + /* Disable header prediction. */ tp->pred_flags = 0; inet_csk_schedule_ack(sk); @@ -4425,7 +4453,8 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) /* In the typical case, we are adding an skb to the end of the list. * Use of ooo_last_skb avoids the O(Log(N)) rbtree lookup. */ - if (tcp_try_coalesce(sk, tp->ooo_last_skb, skb, &fragstolen)) { + if (tcp_try_coalesce(sk, OOO_QUEUE, tp->ooo_last_skb, + skb, &fragstolen)) { coalesce_done: tcp_grow_window(sk, skb); kfree_skb_partial(skb, fragstolen); @@ -4475,7 +4504,8 @@ coalesce_done: __kfree_skb(skb1); goto merge_right; } - } else if (tcp_try_coalesce(sk, skb1, skb, &fragstolen)) { + } else if (tcp_try_coalesce(sk, OOO_QUEUE, skb1, + skb, &fragstolen)) { goto coalesce_done; } p = &parent->rb_right; @@ -4526,7 +4556,8 @@ static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, int __skb_pull(skb, hdrlen); eaten = (tail && - tcp_try_coalesce(sk, tail, skb, fragstolen)) ? 1 : 0; + tcp_try_coalesce(sk, RCV_QUEUE, tail, + skb, fragstolen)) ? 1 : 0; tcp_rcv_nxt_update(tcp_sk(sk), TCP_SKB_CB(skb)->end_seq); if (!eaten) { __skb_queue_tail(&sk->sk_receive_queue, skb); @@ -4588,8 +4619,8 @@ err: static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); - bool fragstolen = false; - int eaten = -1; + bool fragstolen; + int eaten; if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) { __kfree_skb(skb); @@ -4611,32 +4642,13 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) goto out_of_window; /* Ok. In sequence. In window. */ - if (tp->ucopy.task == current && - tp->copied_seq == tp->rcv_nxt && tp->ucopy.len && - sock_owned_by_user(sk) && !tp->urg_data) { - int chunk = min_t(unsigned int, skb->len, - tp->ucopy.len); - - __set_current_state(TASK_RUNNING); - - if (!skb_copy_datagram_msg(skb, 0, tp->ucopy.msg, chunk)) { - tp->ucopy.len -= chunk; - tp->copied_seq += chunk; - eaten = (chunk == skb->len); - tcp_rcv_space_adjust(sk); - } - } - - if (eaten <= 0) { queue_and_out: - if (eaten < 0) { - if (skb_queue_len(&sk->sk_receive_queue) == 0) - sk_forced_mem_schedule(sk, skb->truesize); - else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) - goto drop; - } - eaten = tcp_queue_rcv(sk, skb, 0, &fragstolen); - } + if (skb_queue_len(&sk->sk_receive_queue) == 0) + sk_forced_mem_schedule(sk, skb->truesize); + else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) + goto drop; + + eaten = tcp_queue_rcv(sk, skb, 0, &fragstolen); tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); if (skb->len) tcp_event_data_recv(sk, skb); @@ -5186,26 +5198,6 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t } } -static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen) -{ - struct tcp_sock *tp = tcp_sk(sk); - int chunk = skb->len - hlen; - int err; - - if (skb_csum_unnecessary(skb)) - err = skb_copy_datagram_msg(skb, hlen, tp->ucopy.msg, chunk); - else - err = skb_copy_and_csum_datagram_msg(skb, hlen, tp->ucopy.msg); - - if (!err) { - tp->ucopy.len -= chunk; - tp->copied_seq += chunk; - tcp_rcv_space_adjust(sk); - } - - return err; -} - /* Accept RST for rcv_nxt - 1 after a FIN. * When tcp connections are abruptly terminated from Mac OSX (via ^C), a * FIN is sent followed by a RST packet. The RST is sent with the same @@ -5358,8 +5350,9 @@ discard: * tcp_data_queue when everything is OK. */ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb, - const struct tcphdr *th, unsigned int len) + const struct tcphdr *th) { + unsigned int len = skb->len; struct tcp_sock *tp = tcp_sk(sk); tcp_mstamp_refresh(tp); @@ -5445,56 +5438,28 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb, int eaten = 0; bool fragstolen = false; - if (tp->ucopy.task == current && - tp->copied_seq == tp->rcv_nxt && - len - tcp_header_len <= tp->ucopy.len && - sock_owned_by_user(sk)) { - __set_current_state(TASK_RUNNING); - - if (!tcp_copy_to_iovec(sk, skb, tcp_header_len)) { - /* Predicted packet is in window by definition. - * seq == rcv_nxt and rcv_wup <= rcv_nxt. - * Hence, check seq<=rcv_wup reduces to: - */ - if (tcp_header_len == - (sizeof(struct tcphdr) + - TCPOLEN_TSTAMP_ALIGNED) && - tp->rcv_nxt == tp->rcv_wup) - tcp_store_ts_recent(tp); - - tcp_rcv_rtt_measure_ts(sk, skb); - - __skb_pull(skb, tcp_header_len); - tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); - NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPHPHITSTOUSER); - eaten = 1; - } - } - if (!eaten) { - if (tcp_checksum_complete(skb)) - goto csum_error; + if (tcp_checksum_complete(skb)) + goto csum_error; - if ((int)skb->truesize > sk->sk_forward_alloc) - goto step5; + if ((int)skb->truesize > sk->sk_forward_alloc) + goto step5; - /* Predicted packet is in window by definition. - * seq == rcv_nxt and rcv_wup <= rcv_nxt. - * Hence, check seq<=rcv_wup reduces to: - */ - if (tcp_header_len == - (sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) && - tp->rcv_nxt == tp->rcv_wup) - tcp_store_ts_recent(tp); + /* Predicted packet is in window by definition. + * seq == rcv_nxt and rcv_wup <= rcv_nxt. + * Hence, check seq<=rcv_wup reduces to: + */ + if (tcp_header_len == + (sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) && + tp->rcv_nxt == tp->rcv_wup) + tcp_store_ts_recent(tp); - tcp_rcv_rtt_measure_ts(sk, skb); + tcp_rcv_rtt_measure_ts(sk, skb); - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPHPHITS); + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPHPHITS); - /* Bulk data transfer: receiver */ - eaten = tcp_queue_rcv(sk, skb, tcp_header_len, - &fragstolen); - } + /* Bulk data transfer: receiver */ + eaten = tcp_queue_rcv(sk, skb, tcp_header_len, + &fragstolen); tcp_event_data_recv(sk, skb); @@ -5588,7 +5553,6 @@ void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) __tcp_fast_path_on(tp, tp->snd_wnd); else tp->pred_flags = 0; - } static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, @@ -6303,9 +6267,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); struct sock *fastopen_sk = NULL; - struct dst_entry *dst = NULL; struct request_sock *req; bool want_cookie = false; + struct dst_entry *dst; struct flowi fl; /* TW buckets are converted to open requests without @@ -6355,6 +6319,10 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, if (tmp_opt.tstamp_ok) tcp_rsk(req)->ts_off = af_ops->init_ts_off(net, skb); + dst = af_ops->route_req(sk, &fl, req); + if (!dst) + goto drop_and_free; + if (!want_cookie && !isn) { /* Kill the following clause, if you dislike this way. */ if (!net->ipv4.sysctl_tcp_syncookies && @@ -6375,11 +6343,6 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, isn = af_ops->init_seq(skb); } - if (!dst) { - dst = af_ops->route_req(sk, &fl, req); - if (!dst) - goto drop_and_free; - } tcp_ecn_create_request(req, skb, sk, dst); @@ -6395,7 +6358,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, tcp_openreq_init_rwin(req, sk, dst); if (!want_cookie) { tcp_reqsk_record_syn(sk, req, skb); - fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc, dst); + fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc); } if (fastopen_sk) { af_ops->send_synack(fastopen_sk, dst, &fl, req, diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a20e7f03d5f7..a63486afa7a7 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -85,8 +85,6 @@ #include <crypto/hash.h> #include <linux/scatterlist.h> -int sysctl_tcp_low_latency __read_mostly; - #ifdef CONFIG_TCP_MD5SIG static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, __be32 daddr, __be32 saddr, const struct tcphdr *th); @@ -385,7 +383,7 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) sk = __inet_lookup_established(net, &tcp_hashinfo, iph->daddr, th->dest, iph->saddr, ntohs(th->source), - inet_iif(icmp_skb)); + inet_iif(icmp_skb), 0); if (!sk) { __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); return; @@ -661,7 +659,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) sk1 = __inet_lookup_listener(net, &tcp_hashinfo, NULL, 0, ip_hdr(skb)->saddr, th->source, ip_hdr(skb)->daddr, - ntohs(th->source), inet_iif(skb)); + ntohs(th->source), inet_iif(skb), + tcp_v4_sdif(skb)); /* don't send rst if it can't find key */ if (!sk1) goto out; @@ -1269,7 +1268,7 @@ static void tcp_v4_init_req(struct request_sock *req, sk_rcv_saddr_set(req_to_sk(req), ip_hdr(skb)->daddr); sk_daddr_set(req_to_sk(req), ip_hdr(skb)->saddr); - ireq->opt = tcp_v4_save_options(skb); + ireq->opt = tcp_v4_save_options(sock_net(sk_listener), skb); } static struct dst_entry *tcp_v4_route_req(const struct sock *sk, @@ -1458,7 +1457,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) sk->sk_rx_dst = NULL; } } - tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len); + tcp_rcv_established(sk, skb, tcp_hdr(skb)); return 0; } @@ -1525,7 +1524,7 @@ void tcp_v4_early_demux(struct sk_buff *skb) sk = __inet_lookup_established(dev_net(skb->dev), &tcp_hashinfo, iph->saddr, th->source, iph->daddr, ntohs(th->dest), - skb->skb_iif); + skb->skb_iif, inet_sdif(skb)); if (sk) { skb->sk = sk; skb->destructor = sock_edemux; @@ -1541,61 +1540,6 @@ void tcp_v4_early_demux(struct sk_buff *skb) } } -/* Packet is added to VJ-style prequeue for processing in process - * context, if a reader task is waiting. Apparently, this exciting - * idea (VJ's mail "Re: query about TCP header on tcp-ip" of 07 Sep 93) - * failed somewhere. Latency? Burstiness? Well, at least now we will - * see, why it failed. 8)8) --ANK - * - */ -bool tcp_prequeue(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - if (sysctl_tcp_low_latency || !tp->ucopy.task) - return false; - - if (skb->len <= tcp_hdrlen(skb) && - skb_queue_len(&tp->ucopy.prequeue) == 0) - return false; - - /* Before escaping RCU protected region, we need to take care of skb - * dst. Prequeue is only enabled for established sockets. - * For such sockets, we might need the skb dst only to set sk->sk_rx_dst - * Instead of doing full sk_rx_dst validity here, let's perform - * an optimistic check. - */ - if (likely(sk->sk_rx_dst)) - skb_dst_drop(skb); - else - skb_dst_force_safe(skb); - - __skb_queue_tail(&tp->ucopy.prequeue, skb); - tp->ucopy.memory += skb->truesize; - if (skb_queue_len(&tp->ucopy.prequeue) >= 32 || - tp->ucopy.memory + atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) { - struct sk_buff *skb1; - - BUG_ON(sock_owned_by_user(sk)); - __NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPPREQUEUEDROPPED, - skb_queue_len(&tp->ucopy.prequeue)); - - while ((skb1 = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) - sk_backlog_rcv(sk, skb1); - - tp->ucopy.memory = 0; - } else if (skb_queue_len(&tp->ucopy.prequeue) == 1) { - wake_up_interruptible_sync_poll(sk_sleep(sk), - POLLIN | POLLRDNORM | POLLRDBAND); - if (!inet_csk_ack_scheduled(sk)) - inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, - (3 * tcp_rto_min(sk)) / 4, - TCP_RTO_MAX); - } - return true; -} -EXPORT_SYMBOL(tcp_prequeue); - bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) { u32 limit = sk->sk_rcvbuf + sk->sk_sndbuf; @@ -1645,6 +1589,7 @@ EXPORT_SYMBOL(tcp_filter); int tcp_v4_rcv(struct sk_buff *skb) { struct net *net = dev_net(skb->dev); + int sdif = inet_sdif(skb); const struct iphdr *iph; const struct tcphdr *th; bool refcounted; @@ -1692,10 +1637,12 @@ int tcp_v4_rcv(struct sk_buff *skb) TCP_SKB_CB(skb)->tcp_tw_isn = 0; TCP_SKB_CB(skb)->ip_dsfield = ipv4_get_dsfield(iph); TCP_SKB_CB(skb)->sacked = 0; + TCP_SKB_CB(skb)->has_rxtstamp = + skb->tstamp || skb_hwtstamps(skb)->hwtstamp; lookup: sk = __inet_lookup_skb(&tcp_hashinfo, skb, __tcp_hdrlen(th), th->source, - th->dest, &refcounted); + th->dest, sdif, &refcounted); if (!sk) goto no_tcp_socket; @@ -1722,6 +1669,8 @@ process: */ sock_hold(sk); refcounted = true; + if (tcp_filter(sk, skb)) + goto discard_and_relse; nsk = tcp_check_req(sk, skb, req, false); if (!nsk) { reqsk_put(req); @@ -1729,8 +1678,6 @@ process: } if (nsk == sk) { reqsk_put(req); - } else if (tcp_filter(sk, skb)) { - goto discard_and_relse; } else if (tcp_child_process(sk, nsk, skb)) { tcp_v4_send_reset(nsk, skb); goto discard_and_relse; @@ -1770,8 +1717,7 @@ process: tcp_segs_in(tcp_sk(sk), skb); ret = 0; if (!sock_owned_by_user(sk)) { - if (!tcp_prequeue(sk, skb)) - ret = tcp_v4_do_rcv(sk, skb); + ret = tcp_v4_do_rcv(sk, skb); } else if (tcp_add_backlog(sk, skb)) { goto discard_and_relse; } @@ -1824,7 +1770,8 @@ do_time_wait: __tcp_hdrlen(th), iph->saddr, th->source, iph->daddr, th->dest, - inet_iif(skb)); + inet_iif(skb), + sdif); if (sk2) { inet_twsk_deschedule_put(inet_twsk(sk)); sk = sk2; @@ -1936,9 +1883,6 @@ void tcp_v4_destroy_sock(struct sock *sk) } #endif - /* Clean prequeue, it must be empty really */ - __skb_queue_purge(&tp->ucopy.prequeue); - /* Clean up a referenced TCP bind bucket. */ if (inet_csk(sk)->icsk_bind_hash) inet_put_port(sk); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0ff83c1637d8..188a6f31356d 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -445,7 +445,6 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->snd_sml = newtp->snd_una = newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1; - tcp_prequeue_init(newtp); INIT_LIST_HEAD(&newtp->tsq_node); tcp_init_wl(newtp, treq->rcv_isn); diff --git a/net/ipv4/tcp_nv.c b/net/ipv4/tcp_nv.c index 6d650ed3cb59..1ff73982e28c 100644 --- a/net/ipv4/tcp_nv.c +++ b/net/ipv4/tcp_nv.c @@ -86,7 +86,6 @@ struct tcpnv { * < 0 => less than 1 packet/RTT */ u8 available8; u16 available16; - u32 loss_cwnd; /* cwnd at last loss */ u8 nv_allow_cwnd_growth:1, /* whether cwnd can grow */ nv_reset:1, /* whether to reset values */ nv_catchup:1; /* whether we are growing because @@ -121,7 +120,6 @@ static inline void tcpnv_reset(struct tcpnv *ca, struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); ca->nv_reset = 0; - ca->loss_cwnd = 0; ca->nv_no_cong_cnt = 0; ca->nv_rtt_cnt = 0; ca->nv_last_rtt = 0; @@ -177,19 +175,10 @@ static void tcpnv_cong_avoid(struct sock *sk, u32 ack, u32 acked) static u32 tcpnv_recalc_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - struct tcpnv *ca = inet_csk_ca(sk); - ca->loss_cwnd = tp->snd_cwnd; return max((tp->snd_cwnd * nv_loss_dec_factor) >> 10, 2U); } -static u32 tcpnv_undo_cwnd(struct sock *sk) -{ - struct tcpnv *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); -} - static void tcpnv_state(struct sock *sk, u8 new_state) { struct tcpnv *ca = inet_csk_ca(sk); @@ -446,7 +435,7 @@ static struct tcp_congestion_ops tcpnv __read_mostly = { .ssthresh = tcpnv_recalc_ssthresh, .cong_avoid = tcpnv_cong_avoid, .set_state = tcpnv_state, - .undo_cwnd = tcpnv_undo_cwnd, + .undo_cwnd = tcp_reno_undo_cwnd, .pkts_acked = tcpnv_acked, .get_info = tcpnv_get_info, diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 4e985dea1dd2..5b6690d05abb 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2202,9 +2202,10 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb, static void tcp_chrono_set(struct tcp_sock *tp, const enum tcp_chrono new) { const u32 now = tcp_jiffies32; + enum tcp_chrono old = tp->chrono_type; - if (tp->chrono_type > TCP_CHRONO_UNSPEC) - tp->chrono_stat[tp->chrono_type - 1] += now - tp->chrono_start; + if (old > TCP_CHRONO_UNSPEC) + tp->chrono_stat[old - 1] += now - tp->chrono_start; tp->chrono_start = now; tp->chrono_type = new; } @@ -2376,24 +2377,14 @@ bool tcp_schedule_loss_probe(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); - u32 timeout, tlp_time_stamp, rto_time_stamp; - u32 rtt = usecs_to_jiffies(tp->srtt_us >> 3); + u32 timeout, rto_delta_us; - /* No consecutive loss probes. */ - if (WARN_ON(icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)) { - tcp_rearm_rto(sk); - return false; - } /* Don't do any loss probe on a Fast Open connection before 3WHS * finishes. */ if (tp->fastopen_rsk) return false; - /* TLP is only scheduled when next timer event is RTO. */ - if (icsk->icsk_pending != ICSK_TIME_RETRANS) - return false; - /* Schedule a loss probe in 2*RTT for SACK capable connections * in Open state, that are either limited by cwnd or application. */ @@ -2406,25 +2397,25 @@ bool tcp_schedule_loss_probe(struct sock *sk) tcp_send_head(sk)) return false; - /* Probe timeout is at least 1.5*rtt + TCP_DELACK_MAX to account + /* Probe timeout is 2*rtt. Add minimum RTO to account * for delayed ack when there's one outstanding packet. If no RTT * sample is available then probe after TCP_TIMEOUT_INIT. */ - timeout = rtt << 1 ? : TCP_TIMEOUT_INIT; - if (tp->packets_out == 1) - timeout = max_t(u32, timeout, - (rtt + (rtt >> 1) + TCP_DELACK_MAX)); - timeout = max_t(u32, timeout, msecs_to_jiffies(10)); - - /* If RTO is shorter, just schedule TLP in its place. */ - tlp_time_stamp = tcp_jiffies32 + timeout; - rto_time_stamp = (u32)inet_csk(sk)->icsk_timeout; - if ((s32)(tlp_time_stamp - rto_time_stamp) > 0) { - s32 delta = rto_time_stamp - tcp_jiffies32; - if (delta > 0) - timeout = delta; + if (tp->srtt_us) { + timeout = usecs_to_jiffies(tp->srtt_us >> 2); + if (tp->packets_out == 1) + timeout += TCP_RTO_MIN; + else + timeout += TCP_TIMEOUT_MIN; + } else { + timeout = TCP_TIMEOUT_INIT; } + /* If the RTO formula yields an earlier time, then use that time. */ + rto_delta_us = tcp_rto_delta_us(sk); /* How far in future is RTO? */ + if (rto_delta_us > 0) + timeout = min_t(u32, timeout, usecs_to_jiffies(rto_delta_us)); + inet_csk_reset_xmit_timer(sk, ICSK_TIME_LOSS_PROBE, timeout, TCP_RTO_MAX); return true; @@ -3448,6 +3439,10 @@ int tcp_connect(struct sock *sk) int err; tcp_call_bpf(sk, BPF_SOCK_OPS_TCP_CONNECT_CB); + + if (inet_csk(sk)->icsk_af_ops->rebuild_header(sk)) + return -EHOSTUNREACH; /* Routing failure or similar. */ + tcp_connect_init(sk); if (unlikely(tp->repair)) { diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c index f6c50af24a64..697f4c67b2e3 100644 --- a/net/ipv4/tcp_probe.c +++ b/net/ipv4/tcp_probe.c @@ -105,8 +105,9 @@ static inline int tcp_probe_avail(void) * Note: arguments must match tcp_rcv_established()! */ static void jtcp_rcv_established(struct sock *sk, struct sk_buff *skb, - const struct tcphdr *th, unsigned int len) + const struct tcphdr *th) { + unsigned int len = skb->len; const struct tcp_sock *tp = tcp_sk(sk); const struct inet_sock *inet = inet_sk(sk); @@ -145,7 +146,7 @@ static void jtcp_rcv_established(struct sock *sk, struct sk_buff *skb, BUG(); } - p->length = skb->len; + p->length = len; p->snd_nxt = tp->snd_nxt; p->snd_una = tp->snd_una; p->snd_cwnd = tp->snd_cwnd; diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c index fe9a493d0208..449cd914d58e 100644 --- a/net/ipv4/tcp_recovery.c +++ b/net/ipv4/tcp_recovery.c @@ -113,7 +113,7 @@ void tcp_rack_mark_lost(struct sock *sk) tp->rack.advanced = 0; tcp_rack_detect_loss(sk, &timeout); if (timeout) { - timeout = usecs_to_jiffies(timeout + TCP_REO_TIMEOUT_MIN); + timeout = usecs_to_jiffies(timeout) + TCP_TIMEOUT_MIN; inet_csk_reset_xmit_timer(sk, ICSK_TIME_REO_TIMEOUT, timeout, inet_csk(sk)->icsk_rto); } diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c index f2123075ce6e..addc122f8818 100644 --- a/net/ipv4/tcp_scalable.c +++ b/net/ipv4/tcp_scalable.c @@ -15,10 +15,6 @@ #define TCP_SCALABLE_AI_CNT 50U #define TCP_SCALABLE_MD_SCALE 3 -struct scalable { - u32 loss_cwnd; -}; - static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); @@ -36,23 +32,13 @@ static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked) static u32 tcp_scalable_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - struct scalable *ca = inet_csk_ca(sk); - - ca->loss_cwnd = tp->snd_cwnd; return max(tp->snd_cwnd - (tp->snd_cwnd>>TCP_SCALABLE_MD_SCALE), 2U); } -static u32 tcp_scalable_cwnd_undo(struct sock *sk) -{ - const struct scalable *ca = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); -} - static struct tcp_congestion_ops tcp_scalable __read_mostly = { .ssthresh = tcp_scalable_ssthresh, - .undo_cwnd = tcp_scalable_cwnd_undo, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_scalable_cong_avoid, .owner = THIS_MODULE, diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index c0feeeef962a..655dd8d7f064 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -239,7 +239,6 @@ static int tcp_write_timeout(struct sock *sk) /* Called with BH disabled */ void tcp_delack_timer_handler(struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); sk_mem_reclaim_partial(sk); @@ -254,17 +253,6 @@ void tcp_delack_timer_handler(struct sock *sk) } icsk->icsk_ack.pending &= ~ICSK_ACK_TIMER; - if (!skb_queue_empty(&tp->ucopy.prequeue)) { - struct sk_buff *skb; - - __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSCHEDULERFAILED); - - while ((skb = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) - sk_backlog_rcv(sk, skb); - - tp->ucopy.memory = 0; - } - if (inet_csk_ack_scheduled(sk)) { if (!icsk->icsk_ack.pingpong) { /* Delayed ACK missed: inflate ATO. */ @@ -652,7 +640,8 @@ static void tcp_keepalive_timer (unsigned long data) goto death; } - if (!sock_flag(sk, SOCK_KEEPOPEN) || sk->sk_state == TCP_CLOSE) + if (!sock_flag(sk, SOCK_KEEPOPEN) || + ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_SYN_SENT))) goto out; elapsed = keepalive_time_when(tp); diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp_ulp.c index 2417f55374c5..6bb9e14c710a 100644 --- a/net/ipv4/tcp_ulp.c +++ b/net/ipv4/tcp_ulp.c @@ -122,14 +122,14 @@ int tcp_set_ulp(struct sock *sk, const char *name) ulp_ops = __tcp_ulp_find_autoload(name); if (!ulp_ops) - err = -ENOENT; - else - err = ulp_ops->init(sk); + return -ENOENT; - if (err) - goto out; + err = ulp_ops->init(sk); + if (err) { + module_put(ulp_ops->owner); + return err; + } icsk->icsk_ulp_ops = ulp_ops; - out: - return err; + return 0; } diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c index 76005d4b8dfc..6fcf482d611b 100644 --- a/net/ipv4/tcp_veno.c +++ b/net/ipv4/tcp_veno.c @@ -30,7 +30,6 @@ struct veno { u32 basertt; /* the min of all Veno rtt measurements seen (in usec) */ u32 inc; /* decide whether to increase cwnd */ u32 diff; /* calculate the diff rate */ - u32 loss_cwnd; /* cwnd when loss occured */ }; /* There are several situations when we must "re-start" Veno: @@ -194,7 +193,6 @@ static u32 tcp_veno_ssthresh(struct sock *sk) const struct tcp_sock *tp = tcp_sk(sk); struct veno *veno = inet_csk_ca(sk); - veno->loss_cwnd = tp->snd_cwnd; if (veno->diff < beta) /* in "non-congestive state", cut cwnd by 1/5 */ return max(tp->snd_cwnd * 4 / 5, 2U); @@ -203,17 +201,10 @@ static u32 tcp_veno_ssthresh(struct sock *sk) return max(tp->snd_cwnd >> 1U, 2U); } -static u32 tcp_veno_cwnd_undo(struct sock *sk) -{ - const struct veno *veno = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, veno->loss_cwnd); -} - static struct tcp_congestion_ops tcp_veno __read_mostly = { .init = tcp_veno_init, .ssthresh = tcp_veno_ssthresh, - .undo_cwnd = tcp_veno_cwnd_undo, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_veno_cong_avoid, .pkts_acked = tcp_veno_pkts_acked, .set_state = tcp_veno_state, diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c index e6ff99c4bd3b..96e829b2e2fc 100644 --- a/net/ipv4/tcp_yeah.c +++ b/net/ipv4/tcp_yeah.c @@ -37,7 +37,6 @@ struct yeah { u32 fast_count; u32 pkts_acked; - u32 loss_cwnd; }; static void tcp_yeah_init(struct sock *sk) @@ -220,22 +219,14 @@ static u32 tcp_yeah_ssthresh(struct sock *sk) yeah->fast_count = 0; yeah->reno_count = max(yeah->reno_count>>1, 2U); - yeah->loss_cwnd = tp->snd_cwnd; return max_t(int, tp->snd_cwnd - reduction, 2); } -static u32 tcp_yeah_cwnd_undo(struct sock *sk) -{ - const struct yeah *yeah = inet_csk_ca(sk); - - return max(tcp_sk(sk)->snd_cwnd, yeah->loss_cwnd); -} - static struct tcp_congestion_ops tcp_yeah __read_mostly = { .init = tcp_yeah_init, .ssthresh = tcp_yeah_ssthresh, - .undo_cwnd = tcp_yeah_cwnd_undo, + .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_yeah_cong_avoid, .set_state = tcp_vegas_state, .cwnd_event = tcp_vegas_cwnd_event, diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 25294d43e147..db1c9e78c83c 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -380,8 +380,8 @@ int udp_v4_get_port(struct sock *sk, unsigned short snum) static int compute_score(struct sock *sk, struct net *net, __be32 saddr, __be16 sport, - __be32 daddr, unsigned short hnum, int dif, - bool exact_dif) + __be32 daddr, unsigned short hnum, + int dif, int sdif, bool exact_dif) { int score; struct inet_sock *inet; @@ -413,10 +413,15 @@ static int compute_score(struct sock *sk, struct net *net, } if (sk->sk_bound_dev_if || exact_dif) { - if (sk->sk_bound_dev_if != dif) + bool dev_match = (sk->sk_bound_dev_if == dif || + sk->sk_bound_dev_if == sdif); + + if (exact_dif && !dev_match) return -1; - score += 4; + if (sk->sk_bound_dev_if && dev_match) + score += 4; } + if (sk->sk_incoming_cpu == raw_smp_processor_id()) score++; return score; @@ -436,10 +441,11 @@ static u32 udp_ehashfn(const struct net *net, const __be32 laddr, /* called with rcu_read_lock() */ static struct sock *udp4_lib_lookup2(struct net *net, - __be32 saddr, __be16 sport, - __be32 daddr, unsigned int hnum, int dif, bool exact_dif, - struct udp_hslot *hslot2, - struct sk_buff *skb) + __be32 saddr, __be16 sport, + __be32 daddr, unsigned int hnum, + int dif, int sdif, bool exact_dif, + struct udp_hslot *hslot2, + struct sk_buff *skb) { struct sock *sk, *result; int score, badness, matches = 0, reuseport = 0; @@ -449,7 +455,7 @@ static struct sock *udp4_lib_lookup2(struct net *net, badness = 0; udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { score = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, exact_dif); + daddr, hnum, dif, sdif, exact_dif); if (score > badness) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -477,8 +483,8 @@ static struct sock *udp4_lib_lookup2(struct net *net, * harder than this. -DaveM */ struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, - __be16 sport, __be32 daddr, __be16 dport, - int dif, struct udp_table *udptable, struct sk_buff *skb) + __be16 sport, __be32 daddr, __be16 dport, int dif, + int sdif, struct udp_table *udptable, struct sk_buff *skb) { struct sock *sk, *result; unsigned short hnum = ntohs(dport); @@ -496,7 +502,7 @@ struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, goto begin; result = udp4_lib_lookup2(net, saddr, sport, - daddr, hnum, dif, + daddr, hnum, dif, sdif, exact_dif, hslot2, skb); if (!result) { unsigned int old_slot2 = slot2; @@ -511,7 +517,7 @@ struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, goto begin; result = udp4_lib_lookup2(net, saddr, sport, - daddr, hnum, dif, + daddr, hnum, dif, sdif, exact_dif, hslot2, skb); } return result; @@ -521,7 +527,7 @@ begin: badness = 0; sk_for_each_rcu(sk, &hslot->head) { score = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, exact_dif); + daddr, hnum, dif, sdif, exact_dif); if (score > badness) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -554,7 +560,7 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, return __udp4_lib_lookup(dev_net(skb->dev), iph->saddr, sport, iph->daddr, dport, inet_iif(skb), - udptable, skb); + inet_sdif(skb), udptable, skb); } struct sock *udp4_lib_lookup_skb(struct sk_buff *skb, @@ -576,7 +582,7 @@ struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, struct sock *sk; sk = __udp4_lib_lookup(net, saddr, sport, daddr, dport, - dif, &udp_table, NULL); + dif, 0, &udp_table, NULL); if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; return sk; @@ -587,7 +593,7 @@ EXPORT_SYMBOL_GPL(udp4_lib_lookup); static inline bool __udp_is_mcast_sock(struct net *net, struct sock *sk, __be16 loc_port, __be32 loc_addr, __be16 rmt_port, __be32 rmt_addr, - int dif, unsigned short hnum) + int dif, int sdif, unsigned short hnum) { struct inet_sock *inet = inet_sk(sk); @@ -597,9 +603,10 @@ static inline bool __udp_is_mcast_sock(struct net *net, struct sock *sk, (inet->inet_dport != rmt_port && inet->inet_dport) || (inet->inet_rcv_saddr && inet->inet_rcv_saddr != loc_addr) || ipv6_only_sock(sk) || - (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)) + (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif && + sk->sk_bound_dev_if != sdif)) return false; - if (!ip_mc_sf_allow(sk, loc_addr, rmt_addr, dif)) + if (!ip_mc_sf_allow(sk, loc_addr, rmt_addr, dif, sdif)) return false; return true; } @@ -628,8 +635,8 @@ void __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable) struct net *net = dev_net(skb->dev); sk = __udp4_lib_lookup(net, iph->daddr, uh->dest, - iph->saddr, uh->source, skb->dev->ifindex, udptable, - NULL); + iph->saddr, uh->source, skb->dev->ifindex, 0, + udptable, NULL); if (!sk) { __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); return; /* No socket for error */ @@ -802,7 +809,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4) if (is_udplite) /* UDP-Lite */ csum = udplite_csum(skb); - else if (sk->sk_no_check_tx) { /* UDP csum disabled */ + else if (sk->sk_no_check_tx) { /* UDP csum off */ skb->ip_summed = CHECKSUM_NONE; goto send; @@ -1163,34 +1170,36 @@ out: return ret; } -#if BITS_PER_LONG == 64 +#define UDP_SKB_IS_STATELESS 0x80000000 + static void udp_set_dev_scratch(struct sk_buff *skb) { - struct udp_dev_scratch *scratch; + struct udp_dev_scratch *scratch = udp_skb_scratch(skb); BUILD_BUG_ON(sizeof(struct udp_dev_scratch) > sizeof(long)); - scratch = (struct udp_dev_scratch *)&skb->dev_scratch; - scratch->truesize = skb->truesize; + scratch->_tsize_state = skb->truesize; +#if BITS_PER_LONG == 64 scratch->len = skb->len; scratch->csum_unnecessary = !!skb_csum_unnecessary(skb); scratch->is_linear = !skb_is_nonlinear(skb); +#endif + /* all head states execept sp (dst, sk, nf) are always cleared by + * udp_rcv() and we need to preserve secpath, if present, to eventually + * process IP_CMSG_PASSSEC at recvmsg() time + */ + if (likely(!skb_sec_path(skb))) + scratch->_tsize_state |= UDP_SKB_IS_STATELESS; } static int udp_skb_truesize(struct sk_buff *skb) { - return ((struct udp_dev_scratch *)&skb->dev_scratch)->truesize; -} -#else -static void udp_set_dev_scratch(struct sk_buff *skb) -{ - skb->dev_scratch = skb->truesize; + return udp_skb_scratch(skb)->_tsize_state & ~UDP_SKB_IS_STATELESS; } -static int udp_skb_truesize(struct sk_buff *skb) +static bool udp_skb_has_head_state(struct sk_buff *skb) { - return skb->dev_scratch; + return !(udp_skb_scratch(skb)->_tsize_state & UDP_SKB_IS_STATELESS); } -#endif /* fully reclaim rmem/fwd memory allocated for skb */ static void udp_rmem_release(struct sock *sk, int size, int partial, @@ -1388,6 +1397,11 @@ void skb_consume_udp(struct sock *sk, struct sk_buff *skb, int len) unlock_sock_fast(sk, slow); } + /* In the more common cases we cleared the head states previously, + * see __udp_queue_rcv_skb(). + */ + if (unlikely(udp_skb_has_head_state(skb))) + skb_release_head_state(skb); consume_stateless_skb(skb); } EXPORT_SYMBOL_GPL(skb_consume_udp); @@ -1571,7 +1585,8 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, return ip_recv_error(sk, msg, len, addr_len); try_again: - peeking = off = sk_peek_offset(sk, flags); + peeking = flags & MSG_PEEK; + off = sk_peek_offset(sk, flags); skb = __skb_recv_udp(sk, flags, noblock, &peeked, &off, &err); if (!skb) return err; @@ -1779,9 +1794,6 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) sk_mark_napi_id_once(sk, skb); } - /* clear all pending head states while they are hot in the cache */ - skb_release_head_state(skb); - rc = __udp_enqueue_schedule_skb(sk, skb); if (rc < 0) { int is_udplite = IS_UDPLITE(sk); @@ -1802,8 +1814,7 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) static struct static_key udp_encap_needed __read_mostly; void udp_encap_enable(void) { - if (!static_key_enabled(&udp_encap_needed)) - static_key_slow_inc(&udp_encap_needed); + static_key_enable(&udp_encap_needed); } EXPORT_SYMBOL(udp_encap_enable); @@ -1921,15 +1932,18 @@ drop: /* For TCP sockets, sk_rx_dst is protected by socket lock * For UDP, we use xchg() to guard against concurrent changes. */ -static void udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) +bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) { struct dst_entry *old; if (dst_hold_safe(dst)) { old = xchg(&sk->sk_rx_dst, dst); dst_release(old); + return old != dst; } + return false; } +EXPORT_SYMBOL(udp_sk_rx_dst_set); /* * Multicasts and broadcasts go to each listener. @@ -1948,6 +1962,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb, unsigned int hash2 = 0, hash2_any = 0, use_hash2 = (hslot->count > 10); unsigned int offset = offsetof(typeof(*sk), sk_node); int dif = skb->dev->ifindex; + int sdif = inet_sdif(skb); struct hlist_node *node; struct sk_buff *nskb; @@ -1962,7 +1977,7 @@ start_lookup: sk_for_each_entry_offset_rcu(sk, node, &hslot->head, offset) { if (!__udp_is_mcast_sock(net, sk, uh->dest, daddr, - uh->source, saddr, dif, hnum)) + uh->source, saddr, dif, sdif, hnum)) continue; if (!first) { @@ -2152,7 +2167,7 @@ drop: static struct sock *__udp4_lib_mcast_demux_lookup(struct net *net, __be16 loc_port, __be32 loc_addr, __be16 rmt_port, __be32 rmt_addr, - int dif) + int dif, int sdif) { struct sock *sk, *result; unsigned short hnum = ntohs(loc_port); @@ -2166,7 +2181,7 @@ static struct sock *__udp4_lib_mcast_demux_lookup(struct net *net, result = NULL; sk_for_each_rcu(sk, &hslot->head) { if (__udp_is_mcast_sock(net, sk, loc_port, loc_addr, - rmt_port, rmt_addr, dif, hnum)) { + rmt_port, rmt_addr, dif, sdif, hnum)) { if (result) return NULL; result = sk; @@ -2183,7 +2198,7 @@ static struct sock *__udp4_lib_mcast_demux_lookup(struct net *net, static struct sock *__udp4_lib_demux_lookup(struct net *net, __be16 loc_port, __be32 loc_addr, __be16 rmt_port, __be32 rmt_addr, - int dif) + int dif, int sdif) { unsigned short hnum = ntohs(loc_port); unsigned int hash2 = udp4_portaddr_hash(net, loc_addr, hnum); @@ -2195,7 +2210,7 @@ static struct sock *__udp4_lib_demux_lookup(struct net *net, udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { if (INET_MATCH(sk, net, acookie, rmt_addr, - loc_addr, ports, dif)) + loc_addr, ports, dif, sdif)) return sk; /* Only check first socket in chain */ break; @@ -2211,6 +2226,7 @@ void udp_v4_early_demux(struct sk_buff *skb) struct sock *sk = NULL; struct dst_entry *dst; int dif = skb->dev->ifindex; + int sdif = inet_sdif(skb); int ours; /* validate the packet */ @@ -2236,10 +2252,11 @@ void udp_v4_early_demux(struct sk_buff *skb) } sk = __udp4_lib_mcast_demux_lookup(net, uh->dest, iph->daddr, - uh->source, iph->saddr, dif); + uh->source, iph->saddr, + dif, sdif); } else if (skb->pkt_type == PACKET_HOST) { sk = __udp4_lib_demux_lookup(net, uh->dest, iph->daddr, - uh->source, iph->saddr, dif); + uh->source, iph->saddr, dif, sdif); } if (!sk || !refcount_inc_not_zero(&sk->sk_refcnt)) diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c index 4515836d2a3a..d0390d844ac8 100644 --- a/net/ipv4/udp_diag.c +++ b/net/ipv4/udp_diag.c @@ -45,7 +45,7 @@ static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb, sk = __udp4_lib_lookup(net, req->id.idiag_src[0], req->id.idiag_sport, req->id.idiag_dst[0], req->id.idiag_dport, - req->id.idiag_if, tbl, NULL); + req->id.idiag_if, 0, tbl, NULL); #if IS_ENABLED(CONFIG_IPV6) else if (req->sdiag_family == AF_INET6) sk = __udp6_lib_lookup(net, @@ -53,7 +53,7 @@ static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb, req->id.idiag_sport, (struct in6_addr *)req->id.idiag_dst, req->id.idiag_dport, - req->id.idiag_if, tbl, NULL); + req->id.idiag_if, 0, tbl, NULL); #endif if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; @@ -182,7 +182,7 @@ static int __udp_diag_destroy(struct sk_buff *in_skb, sk = __udp4_lib_lookup(net, req->id.idiag_dst[0], req->id.idiag_dport, req->id.idiag_src[0], req->id.idiag_sport, - req->id.idiag_if, tbl, NULL); + req->id.idiag_if, 0, tbl, NULL); #if IS_ENABLED(CONFIG_IPV6) else if (req->sdiag_family == AF_INET6) { if (ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_dst) && @@ -190,7 +190,7 @@ static int __udp_diag_destroy(struct sk_buff *in_skb, sk = __udp4_lib_lookup(net, req->id.idiag_dst[3], req->id.idiag_dport, req->id.idiag_src[3], req->id.idiag_sport, - req->id.idiag_if, tbl, NULL); + req->id.idiag_if, 0, tbl, NULL); else sk = __udp6_lib_lookup(net, @@ -198,7 +198,7 @@ static int __udp_diag_destroy(struct sk_buff *in_skb, req->id.idiag_dport, (struct in6_addr *)req->id.idiag_src, req->id.idiag_sport, - req->id.idiag_if, tbl, NULL); + req->id.idiag_if, 0, tbl, NULL); } #endif else { diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 781250151d40..97658bfc1b58 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -21,7 +21,7 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, __be16 new_protocol, bool is_ipv6) { int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb); - bool remcsum, need_csum, offload_csum, ufo, gso_partial; + bool remcsum, need_csum, offload_csum, gso_partial; struct sk_buff *segs = ERR_PTR(-EINVAL); struct udphdr *uh = udp_hdr(skb); u16 mac_offset = skb->mac_header; @@ -61,8 +61,6 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, remcsum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_TUNNEL_REMCSUM); skb->remcsum_offload = remcsum; - ufo = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP); - need_ipsec = skb_dst(skb) && dst_xfrm(skb_dst(skb)); /* Try to offload checksum if possible */ offload_csum = !!(need_csum && @@ -77,7 +75,7 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, * outer one so strip the existing checksum feature flags and * instead set the flag based on our outer checksum offload value. */ - if (remcsum || ufo) { + if (remcsum) { features &= ~NETIF_F_CSUM_MASK; if (!need_csum || offload_csum) features |= NETIF_F_HW_CSUM; @@ -189,66 +187,16 @@ out_unlock: } EXPORT_SYMBOL(skb_udp_tunnel_segment); -static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, - netdev_features_t features) +static struct sk_buff *udp4_tunnel_segment(struct sk_buff *skb, + netdev_features_t features) { struct sk_buff *segs = ERR_PTR(-EINVAL); - unsigned int mss; - __wsum csum; - struct udphdr *uh; - struct iphdr *iph; if (skb->encapsulation && (skb_shinfo(skb)->gso_type & - (SKB_GSO_UDP_TUNNEL|SKB_GSO_UDP_TUNNEL_CSUM))) { + (SKB_GSO_UDP_TUNNEL|SKB_GSO_UDP_TUNNEL_CSUM))) segs = skb_udp_tunnel_segment(skb, features, false); - goto out; - } - if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto out; - - mss = skb_shinfo(skb)->gso_size; - if (unlikely(skb->len <= mss)) - goto out; - - if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { - /* Packet is from an untrusted source, reset gso_segs. */ - - skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss); - - segs = NULL; - goto out; - } - - /* Do software UFO. Complete and fill in the UDP checksum as - * HW cannot do checksum of UDP packets sent as multiple - * IP fragments. - */ - - uh = udp_hdr(skb); - iph = ip_hdr(skb); - - uh->check = 0; - csum = skb_checksum(skb, 0, skb->len, 0); - uh->check = udp_v4_check(skb->len, iph->saddr, iph->daddr, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - - skb->ip_summed = CHECKSUM_NONE; - - /* If there is no outer header we can fake a checksum offload - * due to the fact that we have already done the checksum in - * software prior to segmenting the frame. - */ - if (!skb->encap_hdr_csum) - features |= NETIF_F_HW_CSUM; - - /* Fragment the skb. IP headers of the fragments are updated in - * inet_gso_segment() - */ - segs = skb_segment(skb, features); -out: return segs; } @@ -382,7 +330,7 @@ static int udp4_gro_complete(struct sk_buff *skb, int nhoff) static const struct net_offload udpv4_offload = { .callbacks = { - .gso_segment = udp4_ufo_fragment, + .gso_segment = udp4_tunnel_segment, .gro_receive = udp4_gro_receive, .gro_complete = udp4_gro_complete, }, diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 58bd39fb14b4..6539ff15e9a3 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -82,7 +82,8 @@ void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, struct sock *sk = sock->sk; struct udp_tunnel_info ti; - if (!dev->netdev_ops->ndo_udp_tunnel_add) + if (!dev->netdev_ops->ndo_udp_tunnel_add || + !(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) return; ti.type = type; @@ -93,6 +94,24 @@ void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, } EXPORT_SYMBOL_GPL(udp_tunnel_push_rx_port); +void udp_tunnel_drop_rx_port(struct net_device *dev, struct socket *sock, + unsigned short type) +{ + struct sock *sk = sock->sk; + struct udp_tunnel_info ti; + + if (!dev->netdev_ops->ndo_udp_tunnel_del || + !(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) + return; + + ti.type = type; + ti.sa_family = sk->sk_family; + ti.port = inet_sk(sk)->inet_sport; + + dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); +} +EXPORT_SYMBOL_GPL(udp_tunnel_drop_rx_port); + /* Notify netdevs that UDP port started listening */ void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) { @@ -109,6 +128,8 @@ void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) for_each_netdev_rcu(net, dev) { if (!dev->netdev_ops->ndo_udp_tunnel_add) continue; + if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) + continue; dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); } rcu_read_unlock(); @@ -131,6 +152,8 @@ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) for_each_netdev_rcu(net, dev) { if (!dev->netdev_ops->ndo_udp_tunnel_del) continue; + if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) + continue; dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); } rcu_read_unlock(); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 71b4ecc195c7..d7bf0b041885 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -20,7 +20,8 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, int tos, int oif, const xfrm_address_t *saddr, - const xfrm_address_t *daddr) + const xfrm_address_t *daddr, + u32 mark) { struct rtable *rt; @@ -28,6 +29,7 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, fl4->daddr = daddr->a4; fl4->flowi4_tos = tos; fl4->flowi4_oif = l3mdev_master_ifindex_by_index(net, oif); + fl4->flowi4_mark = mark; if (saddr) fl4->saddr = saddr->a4; @@ -42,20 +44,22 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, static struct dst_entry *xfrm4_dst_lookup(struct net *net, int tos, int oif, const xfrm_address_t *saddr, - const xfrm_address_t *daddr) + const xfrm_address_t *daddr, + u32 mark) { struct flowi4 fl4; - return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr); + return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr, mark); } static int xfrm4_get_saddr(struct net *net, int oif, - xfrm_address_t *saddr, xfrm_address_t *daddr) + xfrm_address_t *saddr, xfrm_address_t *daddr, + u32 mark) { struct dst_entry *dst; struct flowi4 fl4; - dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr); + dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr, mark); if (IS_ERR(dst)) return -EHOSTUNREACH; @@ -213,14 +217,6 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse) fl4->flowi4_tos = iph->tos; } -static inline int xfrm4_garbage_collect(struct dst_ops *ops) -{ - struct net *net = container_of(ops, struct net, xfrm.xfrm4_dst_ops); - - xfrm_garbage_collect_deferred(net); - return (dst_entries_get_slow(ops) > ops->gc_thresh * 2); -} - static void xfrm4_update_pmtu(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb, u32 mtu) { @@ -259,14 +255,13 @@ static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev, static struct dst_ops xfrm4_dst_ops_template = { .family = AF_INET, - .gc = xfrm4_garbage_collect, .update_pmtu = xfrm4_update_pmtu, .redirect = xfrm4_redirect, .cow_metrics = dst_cow_metrics_generic, .destroy = xfrm4_dst_destroy, .ifdown = xfrm4_dst_ifdown, .local_out = __ip_local_out, - .gc_thresh = INT_MAX, + .gc_thresh = 32768, }; static const struct xfrm_policy_afinfo xfrm4_policy_afinfo = { diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 48c452959d2c..ea71e4b0ab7a 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -308,22 +308,12 @@ config IPV6_SEG6_LWTUNNEL depends on IPV6 select LWTUNNEL select DST_CACHE + select IPV6_MULTIPLE_TABLES ---help--- Support for encapsulation of packets within an outer IPv6 header and a Segment Routing Header using the lightweight - tunnels mechanism. - - If unsure, say N. - -config IPV6_SEG6_INLINE - bool "IPv6: direct Segment Routing Header insertion " - depends on IPV6_SEG6_LWTUNNEL - ---help--- - Support for direct insertion of the Segment Routing Header, - also known as inline mode. Be aware that direct insertion of - extension headers (as opposed to encapsulation) may break - multiple mechanisms such as PMTUD or IPSec AH. Use this feature - only if you know exactly what you are doing. + tunnels mechanism. Also enable support for advanced local + processing of SRv6 packets based on their active segment. If unsure, say N. diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 217e9ff0e24b..10e342363793 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -9,7 +9,7 @@ ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ raw.o icmp.o mcast.o reassembly.o tcp_ipv6.o ping.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o \ - udp_offload.o seg6.o + udp_offload.o seg6.o fib6_notifier.o ipv6-offload := ip6_offload.o tcpv6_offload.o exthdrs_offload.o @@ -23,7 +23,7 @@ ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o ipv6-$(CONFIG_PROC_FS) += proc.o ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o ipv6-$(CONFIG_NETLABEL) += calipso.o -ipv6-$(CONFIG_IPV6_SEG6_LWTUNNEL) += seg6_iptunnel.o +ipv6-$(CONFIG_IPV6_SEG6_LWTUNNEL) += seg6_iptunnel.o seg6_local.o ipv6-$(CONFIG_IPV6_SEG6_HMAC) += seg6_hmac.o ipv6-objs += $(ipv6-y) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 3c46e9513a31..c2e2a78787ec 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3030,9 +3030,6 @@ static void sit_add_v4_addrs(struct inet6_dev *idev) static void init_loopback(struct net_device *dev) { struct inet6_dev *idev; - struct net_device *sp_dev; - struct inet6_ifaddr *sp_ifa; - struct rt6_info *sp_rt; /* ::1 */ @@ -3045,45 +3042,6 @@ static void init_loopback(struct net_device *dev) } add_addr(idev, &in6addr_loopback, 128, IFA_HOST); - - /* Add routes to other interface's IPv6 addresses */ - for_each_netdev(dev_net(dev), sp_dev) { - if (!strcmp(sp_dev->name, dev->name)) - continue; - - idev = __in6_dev_get(sp_dev); - if (!idev) - continue; - - read_lock_bh(&idev->lock); - list_for_each_entry(sp_ifa, &idev->addr_list, if_list) { - - if (sp_ifa->flags & (IFA_F_DADFAILED | IFA_F_TENTATIVE)) - continue; - - if (sp_ifa->rt) { - /* This dst has been added to garbage list when - * lo device down, release this obsolete dst and - * reallocate a new router for ifa. - */ - if (!atomic_read(&sp_ifa->rt->rt6i_ref)) { - ip6_rt_put(sp_ifa->rt); - sp_ifa->rt = NULL; - } else { - continue; - } - } - - sp_rt = addrconf_dst_alloc(idev, &sp_ifa->addr, false); - - /* Failure cases are ignored */ - if (!IS_ERR(sp_rt)) { - sp_ifa->rt = sp_rt; - ip6_ins_rt(sp_rt); - } - } - read_unlock_bh(&idev->lock); - } } void addrconf_add_linklocal(struct inet6_dev *idev, @@ -3321,11 +3279,11 @@ static void addrconf_gre_config(struct net_device *dev) static int fixup_permanent_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp) { - /* rt6i_ref == 0 means the host route was removed from the + /* !rt6i_node means the host route was removed from the * FIB, for example, if 'lo' device is taken down. In that * case regenerate the host route. */ - if (!ifp->rt || !atomic_read(&ifp->rt->rt6i_ref)) { + if (!ifp->rt || !ifp->rt->rt6i_node) { struct rt6_info *rt, *prev; rt = addrconf_dst_alloc(idev, &ifp->addr, false); @@ -5556,7 +5514,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp) * our DAD process, so we don't need * to do it again */ - if (!(ifp->rt->rt6i_node)) + if (!rcu_access_pointer(ifp->rt->rt6i_node)) ip6_ins_rt(ifp->rt); if (ifp->idev->cnf.forwarding) addrconf_join_anycast(ifp); @@ -6605,21 +6563,21 @@ int __init addrconf_init(void) rtnl_af_register(&inet6_ops); err = __rtnl_register(PF_INET6, RTM_GETLINK, NULL, inet6_dump_ifinfo, - NULL); + 0); if (err < 0) goto errout; /* Only the first call to __rtnl_register can fail */ - __rtnl_register(PF_INET6, RTM_NEWADDR, inet6_rtm_newaddr, NULL, NULL); - __rtnl_register(PF_INET6, RTM_DELADDR, inet6_rtm_deladdr, NULL, NULL); + __rtnl_register(PF_INET6, RTM_NEWADDR, inet6_rtm_newaddr, NULL, 0); + __rtnl_register(PF_INET6, RTM_DELADDR, inet6_rtm_deladdr, NULL, 0); __rtnl_register(PF_INET6, RTM_GETADDR, inet6_rtm_getaddr, - inet6_dump_ifaddr, NULL); + inet6_dump_ifaddr, 0); __rtnl_register(PF_INET6, RTM_GETMULTICAST, NULL, - inet6_dump_ifmcaddr, NULL); + inet6_dump_ifmcaddr, 0); __rtnl_register(PF_INET6, RTM_GETANYCAST, NULL, - inet6_dump_ifacaddr, NULL); + inet6_dump_ifacaddr, 0); __rtnl_register(PF_INET6, RTM_GETNETCONF, inet6_netconf_get_devconf, - inet6_netconf_dump_devconf, NULL); + inet6_netconf_dump_devconf, 0); ipv6_addr_label_rtnl_register(); diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index 7a428f65c7ec..b055bc79f56d 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -405,6 +405,18 @@ static const struct nla_policy ifal_policy[IFAL_MAX+1] = { [IFAL_LABEL] = { .len = sizeof(u32), }, }; +static bool addrlbl_ifindex_exists(struct net *net, int ifindex) +{ + + struct net_device *dev; + + rcu_read_lock(); + dev = dev_get_by_index_rcu(net, ifindex); + rcu_read_unlock(); + + return dev != NULL; +} + static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { @@ -439,7 +451,7 @@ static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh, switch (nlh->nlmsg_type) { case RTM_NEWADDRLABEL: if (ifal->ifal_index && - !__dev_get_by_index(net, ifal->ifal_index)) + !addrlbl_ifindex_exists(net, ifal->ifal_index)) return -EINVAL; err = ip6addrlbl_add(net, pfx, ifal->ifal_prefixlen, @@ -548,7 +560,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr *nlh, return -EINVAL; if (ifal->ifal_index && - !__dev_get_by_index(net, ifal->ifal_index)) + !addrlbl_ifindex_exists(net, ifal->ifal_index)) return -EINVAL; if (!tb[IFAL_ADDRESS]) @@ -593,10 +605,10 @@ out: void __init ipv6_addr_label_rtnl_register(void) { __rtnl_register(PF_INET6, RTM_NEWADDRLABEL, ip6addrlbl_newdel, - NULL, NULL); + NULL, RTNL_FLAG_DOIT_UNLOCKED); __rtnl_register(PF_INET6, RTM_DELADDRLABEL, ip6addrlbl_newdel, - NULL, NULL); + NULL, RTNL_FLAG_DOIT_UNLOCKED); __rtnl_register(PF_INET6, RTM_GETADDRLABEL, ip6addrlbl_get, - ip6addrlbl_dump, NULL); + ip6addrlbl_dump, RTNL_FLAG_DOIT_UNLOCKED); } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index a88b5b5b7955..fe5262fd6aa5 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -210,7 +210,8 @@ lookup_protocol: np->mcast_hops = IPV6_DEFAULT_MCASTHOPS; np->mc_loop = 1; np->pmtudisc = IPV6_PMTUDISC_WANT; - np->autoflowlabel = ip6_default_np_autolabel(sock_net(sk)); + np->autoflowlabel = ip6_default_np_autolabel(net); + np->repflow = net->ipv6.sysctl.flowlabel_reflect; sk->sk_ipv6only = net->ipv6.sysctl.bindv6only; /* Init the ipv4 part of the socket since we can have sockets @@ -554,6 +555,8 @@ const struct proto_ops inet6_stream_ops = { .recvmsg = inet_recvmsg, /* ok */ .mmap = sock_no_mmap, .sendpage = inet_sendpage, + .sendmsg_locked = tcp_sendmsg_locked, + .sendpage_locked = tcp_sendpage_locked, .splice_read = tcp_splice_read, .read_sock = tcp_read_sock, .peek_len = tcp_peek_len, diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9ed35473dcb5..89910e2c10f4 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -226,7 +226,7 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info int tailen = esp->tailen; if (!skb_cloned(skb)) { - if (tailen <= skb_availroom(skb)) { + if (tailen <= skb_tailroom(skb)) { nfrags = 1; trailer = skb; tail = skb_tail_pointer(trailer); @@ -260,8 +260,6 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info kunmap_atomic(vaddr); - spin_unlock_bh(&x->lock); - nfrags = skb_shinfo(skb)->nr_frags; __skb_fill_page_desc(skb, nfrags, page, pfrag->offset, @@ -269,6 +267,9 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info skb_shinfo(skb)->nr_frags = ++nfrags; pfrag->offset = pfrag->offset + allocsize; + + spin_unlock_bh(&x->lock); + nfrags++; skb->len += tailen; @@ -345,7 +346,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; if (!esp->inplace) { int allocsize; @@ -356,7 +357,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info spin_lock_bh(&x->lock); if (unlikely(!skb_page_frag_refill(allocsize, pfrag, GFP_ATOMIC))) { spin_unlock_bh(&x->lock); - goto error; + goto error_free; } skb_shinfo(skb)->nr_frags = 1; @@ -373,7 +374,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; } if ((x->props.flags & XFRM_STATE_ESN)) @@ -406,8 +407,9 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info if (sg != dsg) esp_ssg_unref(x, tmp); - kfree(tmp); +error_free: + kfree(tmp); error: return err; } @@ -461,28 +463,30 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb) return esp6_output_tail(x, skb, &esp); } -int esp6_input_done2(struct sk_buff *skb, int err) +static inline int esp_remove_trailer(struct sk_buff *skb) { struct xfrm_state *x = xfrm_input_state(skb); struct xfrm_offload *xo = xfrm_offload(skb); struct crypto_aead *aead = x->data; - int alen = crypto_aead_authsize(aead); - int hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); - int elen = skb->len - hlen; - int hdr_len = skb_network_header_len(skb); - int padlen; + int alen, hlen, elen; + int padlen, trimlen; + __wsum csumdiff; u8 nexthdr[2]; + int ret; - if (!xo || (xo && !(xo->flags & CRYPTO_DONE))) - kfree(ESP_SKB_CB(skb)->tmp); + alen = crypto_aead_authsize(aead); + hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); + elen = skb->len - hlen; - if (unlikely(err)) + if (xo && (xo->flags & XFRM_ESP_NO_TRAILER)) { + ret = xo->proto; goto out; + } if (skb_copy_bits(skb, skb->len - alen - 2, nexthdr, 2)) BUG(); - err = -EINVAL; + ret = -EINVAL; padlen = nexthdr[0]; if (padlen + 2 + alen >= elen) { net_dbg_ratelimited("ipsec esp packet is garbage padlen=%d, elen=%d\n", @@ -490,17 +494,46 @@ int esp6_input_done2(struct sk_buff *skb, int err) goto out; } - /* ... check padding bits here. Silly. :-) */ + trimlen = alen + padlen + 2; + if (skb->ip_summed == CHECKSUM_COMPLETE) { + csumdiff = skb_checksum(skb, skb->len - trimlen, trimlen, 0); + skb->csum = csum_block_sub(skb->csum, csumdiff, + skb->len - trimlen); + } + pskb_trim(skb, skb->len - trimlen); - pskb_trim(skb, skb->len - alen - padlen - 2); - __skb_pull(skb, hlen); + ret = nexthdr[1]; + +out: + return ret; +} + +int esp6_input_done2(struct sk_buff *skb, int err) +{ + struct xfrm_state *x = xfrm_input_state(skb); + struct xfrm_offload *xo = xfrm_offload(skb); + struct crypto_aead *aead = x->data; + int hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); + int hdr_len = skb_network_header_len(skb); + + if (!xo || (xo && !(xo->flags & CRYPTO_DONE))) + kfree(ESP_SKB_CB(skb)->tmp); + + if (unlikely(err)) + goto out; + + err = esp_remove_trailer(skb); + if (unlikely(err < 0)) + goto out; + + skb_postpull_rcsum(skb, skb_network_header(skb), + skb_network_header_len(skb)); + skb_pull_rcsum(skb, hlen); if (x->props.mode == XFRM_MODE_TUNNEL) skb_reset_transport_header(skb); else skb_set_transport_header(skb, -hdr_len); - err = nexthdr[1]; - /* RFC4303: Drop dummy packets without any error */ if (err == IPPROTO_NONE) err = -EINVAL; diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c index f02f131f6435..333a478aa161 100644 --- a/net/ipv6/esp6_offload.c +++ b/net/ipv6/esp6_offload.c @@ -209,11 +209,13 @@ out: static int esp6_input_tail(struct xfrm_state *x, struct sk_buff *skb) { struct crypto_aead *aead = x->data; + struct xfrm_offload *xo = xfrm_offload(skb); if (!pskb_may_pull(skb, sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead))) return -EINVAL; - skb->ip_summed = CHECKSUM_NONE; + if (!(xo->flags & CRYPTO_DONE)) + skb->ip_summed = CHECKSUM_NONE; return esp6_input_done2(skb, 0); } @@ -286,7 +288,7 @@ static int esp6_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features esp.seqno = cpu_to_be64(xo->seq.low + ((u64)xo->seq.hi << 32)); err = esp6_output_tail(x, skb, &esp); - if (err < 0) + if (err) return err; secpath_reset(skb); @@ -332,3 +334,4 @@ module_init(esp6_offload_init); module_exit(esp6_offload_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Steffen Klassert <steffen.klassert@secunet.com>"); +MODULE_ALIAS_XFRM_OFFLOAD_TYPE(AF_INET6, XFRM_PROTO_ESP); diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 4996d734f1d2..95516138e861 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -756,6 +756,7 @@ static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff) if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr))) goto drop; + IP6CB(skb)->flags |= IP6SKB_JUMBOGRAM; return true; drop: @@ -881,7 +882,7 @@ static void ipv6_push_rthdr4(struct sk_buff *skb, u8 *proto, (hops - 1) * sizeof(struct in6_addr)); sr_phdr->segments[0] = **addr_p; - *addr_p = &sr_ihdr->segments[hops - 1]; + *addr_p = &sr_ihdr->segments[sr_ihdr->segments_left]; #ifdef CONFIG_IPV6_SEG6_HMAC if (sr_has_hmac(sr_phdr)) { @@ -1173,7 +1174,7 @@ struct in6_addr *fl6_update_dst(struct flowi6 *fl6, { struct ipv6_sr_hdr *srh = (struct ipv6_sr_hdr *)opt->srcrt; - fl6->daddr = srh->segments[srh->first_segment]; + fl6->daddr = srh->segments[srh->segments_left]; break; } default: diff --git a/net/ipv6/fib6_notifier.c b/net/ipv6/fib6_notifier.c new file mode 100644 index 000000000000..05f82baaa99e --- /dev/null +++ b/net/ipv6/fib6_notifier.c @@ -0,0 +1,63 @@ +#include <linux/notifier.h> +#include <linux/socket.h> +#include <linux/kernel.h> +#include <linux/export.h> +#include <net/net_namespace.h> +#include <net/fib_notifier.h> +#include <net/netns/ipv6.h> +#include <net/ip6_fib.h> + +int call_fib6_notifier(struct notifier_block *nb, struct net *net, + enum fib_event_type event_type, + struct fib_notifier_info *info) +{ + info->family = AF_INET6; + return call_fib_notifier(nb, net, event_type, info); +} + +int call_fib6_notifiers(struct net *net, enum fib_event_type event_type, + struct fib_notifier_info *info) +{ + info->family = AF_INET6; + return call_fib_notifiers(net, event_type, info); +} + +static unsigned int fib6_seq_read(struct net *net) +{ + return fib6_tables_seq_read(net) + fib6_rules_seq_read(net); +} + +static int fib6_dump(struct net *net, struct notifier_block *nb) +{ + int err; + + err = fib6_rules_dump(net, nb); + if (err) + return err; + + return fib6_tables_dump(net, nb); +} + +static const struct fib_notifier_ops fib6_notifier_ops_template = { + .family = AF_INET6, + .fib_seq_read = fib6_seq_read, + .fib_dump = fib6_dump, + .owner = THIS_MODULE, +}; + +int __net_init fib6_notifier_init(struct net *net) +{ + struct fib_notifier_ops *ops; + + ops = fib_notifier_ops_register(&fib6_notifier_ops_template, net); + if (IS_ERR(ops)) + return PTR_ERR(ops); + net->ipv6.notifier_ops = ops; + + return 0; +} + +void __net_exit fib6_notifier_exit(struct net *net) +{ + fib_notifier_ops_unregister(net->ipv6.notifier_ops); +} diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index ec849d88a662..b240f24a6e52 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -14,6 +14,7 @@ */ #include <linux/netdevice.h> +#include <linux/notifier.h> #include <linux/export.h> #include <net/fib_rules.h> @@ -29,22 +30,65 @@ struct fib6_rule { u8 tclass; }; -struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi6 *fl6, - int flags, pol_lookup_t lookup) +static bool fib6_rule_matchall(const struct fib_rule *rule) +{ + struct fib6_rule *r = container_of(rule, struct fib6_rule, common); + + if (r->dst.plen || r->src.plen || r->tclass) + return false; + return fib_rule_matchall(rule); +} + +bool fib6_rule_default(const struct fib_rule *rule) { - struct fib_lookup_arg arg = { - .lookup_ptr = lookup, - .flags = FIB_LOOKUP_NOREF, - }; + if (!fib6_rule_matchall(rule) || rule->action != FR_ACT_TO_TBL || + rule->l3mdev) + return false; + if (rule->table != RT6_TABLE_LOCAL && rule->table != RT6_TABLE_MAIN) + return false; + return true; +} +EXPORT_SYMBOL_GPL(fib6_rule_default); - /* update flow if oif or iif point to device enslaved to l3mdev */ - l3mdev_update_flow(net, flowi6_to_flowi(fl6)); +int fib6_rules_dump(struct net *net, struct notifier_block *nb) +{ + return fib_rules_dump(net, nb, AF_INET6); +} - fib_rules_lookup(net->ipv6.fib6_rules_ops, - flowi6_to_flowi(fl6), flags, &arg); +unsigned int fib6_rules_seq_read(struct net *net) +{ + return fib_rules_seq_read(net, AF_INET6); +} - if (arg.result) - return arg.result; +struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi6 *fl6, + int flags, pol_lookup_t lookup) +{ + if (net->ipv6.fib6_has_custom_rules) { + struct fib_lookup_arg arg = { + .lookup_ptr = lookup, + .flags = FIB_LOOKUP_NOREF, + }; + + /* update flow if oif or iif point to device enslaved to l3mdev */ + l3mdev_update_flow(net, flowi6_to_flowi(fl6)); + + fib_rules_lookup(net->ipv6.fib6_rules_ops, + flowi6_to_flowi(fl6), flags, &arg); + + if (arg.result) + return arg.result; + } else { + struct rt6_info *rt; + + rt = lookup(net, net->ipv6.fib6_local_tbl, fl6, flags); + if (rt != net->ipv6.ip6_null_entry && rt->dst.error != -EAGAIN) + return &rt->dst; + ip6_rt_put(rt); + rt = lookup(net, net->ipv6.fib6_main_tbl, fl6, flags); + if (rt->dst.error != -EAGAIN) + return &rt->dst; + ip6_rt_put(rt); + } dst_hold(&net->ipv6.ip6_null_entry->dst); return &net->ipv6.ip6_null_entry->dst; @@ -214,6 +258,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, rule6->dst.plen = frh->dst_len; rule6->tclass = frh->tos; + net->ipv6.fib6_has_custom_rules = true; err = 0; errout: return err; diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 8d7b113958b1..5acb54405b10 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -399,6 +399,24 @@ relookup_failed: return ERR_PTR(err); } +static int icmp6_iif(const struct sk_buff *skb) +{ + int iif = skb->dev->ifindex; + + /* for local traffic to local address, skb dev is the loopback + * device. Check if there is a dst attached to the skb and if so + * get the real device index. + */ + if (unlikely(iif == LOOPBACK_IFINDEX)) { + const struct rt6_info *rt6 = skb_rt6_info(skb); + + if (rt6) + iif = rt6->rt6i_idev->dev->ifindex; + } + + return iif; +} + /* * Send an ICMP message in response to a packet in error */ @@ -459,9 +477,9 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, * Source addr check */ - if (__ipv6_addr_needs_scope_id(addr_type)) - iif = skb->dev->ifindex; - else { + if (__ipv6_addr_needs_scope_id(addr_type)) { + iif = icmp6_iif(skb); + } else { dst = skb_dst(skb); iif = l3mdev_master_ifindex(dst ? dst->dev : skb->dev); } @@ -508,6 +526,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, fl6.fl6_icmp_type = type; fl6.fl6_icmp_code = code; fl6.flowi6_uid = sock_net_uid(net, NULL); + fl6.mp_hash = rt6_multipath_hash(&fl6, skb); security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); sk = icmpv6_xmit_lock(net); @@ -682,7 +701,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl6.daddr = ipv6_hdr(skb)->saddr; if (saddr) fl6.saddr = *saddr; - fl6.flowi6_oif = skb->dev->ifindex; + fl6.flowi6_oif = icmp6_iif(skb); fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY; fl6.flowi6_mark = mark; fl6.flowi6_uid = sock_net_uid(net, NULL); diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c index 77f7f8c7d93d..5bd419c1abc8 100644 --- a/net/ipv6/ila/ila_xlat.c +++ b/net/ipv6/ila/ila_xlat.c @@ -208,7 +208,7 @@ ila_nf_input(void *priv, return NF_ACCEPT; } -static struct nf_hook_ops ila_nf_hook_ops[] __read_mostly = { +static const struct nf_hook_ops ila_nf_hook_ops[] = { { .hook = ila_nf_input, .pf = NFPROTO_IPV6, diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index b13b8f93079d..b01858f5deb1 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -56,7 +56,7 @@ struct sock *__inet6_lookup_established(struct net *net, const __be16 sport, const struct in6_addr *daddr, const u16 hnum, - const int dif) + const int dif, const int sdif) { struct sock *sk; const struct hlist_nulls_node *node; @@ -73,12 +73,12 @@ begin: sk_nulls_for_each_rcu(sk, node, &head->chain) { if (sk->sk_hash != hash) continue; - if (!INET6_MATCH(sk, net, saddr, daddr, ports, dif)) + if (!INET6_MATCH(sk, net, saddr, daddr, ports, dif, sdif)) continue; if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt))) goto out; - if (unlikely(!INET6_MATCH(sk, net, saddr, daddr, ports, dif))) { + if (unlikely(!INET6_MATCH(sk, net, saddr, daddr, ports, dif, sdif))) { sock_gen_put(sk); goto begin; } @@ -96,7 +96,7 @@ EXPORT_SYMBOL(__inet6_lookup_established); static inline int compute_score(struct sock *sk, struct net *net, const unsigned short hnum, const struct in6_addr *daddr, - const int dif, bool exact_dif) + const int dif, const int sdif, bool exact_dif) { int score = -1; @@ -110,9 +110,13 @@ static inline int compute_score(struct sock *sk, struct net *net, score++; } if (sk->sk_bound_dev_if || exact_dif) { - if (sk->sk_bound_dev_if != dif) + bool dev_match = (sk->sk_bound_dev_if == dif || + sk->sk_bound_dev_if == sdif); + + if (exact_dif && !dev_match) return -1; - score++; + if (sk->sk_bound_dev_if && dev_match) + score++; } if (sk->sk_incoming_cpu == raw_smp_processor_id()) score++; @@ -126,7 +130,7 @@ struct sock *inet6_lookup_listener(struct net *net, struct sk_buff *skb, int doff, const struct in6_addr *saddr, const __be16 sport, const struct in6_addr *daddr, - const unsigned short hnum, const int dif) + const unsigned short hnum, const int dif, const int sdif) { unsigned int hash = inet_lhashfn(net, hnum); struct inet_listen_hashbucket *ilb = &hashinfo->listening_hash[hash]; @@ -136,7 +140,7 @@ struct sock *inet6_lookup_listener(struct net *net, u32 phash = 0; sk_for_each(sk, &ilb->head) { - score = compute_score(sk, net, hnum, daddr, dif, exact_dif); + score = compute_score(sk, net, hnum, daddr, dif, sdif, exact_dif); if (score > hiscore) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -171,7 +175,7 @@ struct sock *inet6_lookup(struct net *net, struct inet_hashinfo *hashinfo, bool refcounted; sk = __inet6_lookup(net, hashinfo, skb, doff, saddr, sport, daddr, - ntohs(dport), dif, &refcounted); + ntohs(dport), dif, 0, &refcounted); if (sk && !refcounted && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; return sk; @@ -187,8 +191,9 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, const struct in6_addr *daddr = &sk->sk_v6_rcv_saddr; const struct in6_addr *saddr = &sk->sk_v6_daddr; const int dif = sk->sk_bound_dev_if; - const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); struct net *net = sock_net(sk); + const int sdif = l3mdev_master_ifindex_by_index(net, dif); + const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); const unsigned int hash = inet6_ehashfn(net, daddr, lport, saddr, inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); @@ -203,7 +208,8 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, if (sk2->sk_hash != hash) continue; - if (likely(INET6_MATCH(sk2, net, saddr, daddr, ports, dif))) { + if (likely(INET6_MATCH(sk2, net, saddr, daddr, ports, + dif, sdif))) { if (sk2->sk_state == TCP_TIME_WAIT) { tw = inet_twsk(sk2); if (twsk_unique(sk, sk2, twp)) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index ebb299cf72b7..a3b5c163325f 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -33,6 +33,7 @@ #include <net/ndisc.h> #include <net/addrconf.h> #include <net/lwtunnel.h> +#include <net/fib_notifier.h> #include <net/ip6_fib.h> #include <net/ip6_route.h> @@ -148,12 +149,24 @@ static struct fib6_node *node_alloc(void) return fn; } -static void node_free(struct fib6_node *fn) +static void node_free_immediate(struct fib6_node *fn) +{ + kmem_cache_free(fib6_node_kmem, fn); +} + +static void node_free_rcu(struct rcu_head *head) { + struct fib6_node *fn = container_of(head, struct fib6_node, rcu); + kmem_cache_free(fib6_node_kmem, fn); } -static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) +static void node_free(struct fib6_node *fn) +{ + call_rcu(&fn->rcu, node_free_rcu); +} + +void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) { int cpu; @@ -176,15 +189,7 @@ static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) free_percpu(non_pcpu_rt->rt6i_pcpu); non_pcpu_rt->rt6i_pcpu = NULL; } - -static void rt6_release(struct rt6_info *rt) -{ - if (atomic_dec_and_test(&rt->rt6i_ref)) { - rt6_free_pcpu(rt); - dst_dev_put(&rt->dst); - dst_release(&rt->dst); - } -} +EXPORT_SYMBOL_GPL(rt6_free_pcpu); static void fib6_link_table(struct net *net, struct fib6_table *tb) { @@ -302,6 +307,109 @@ static void __net_init fib6_tables_init(struct net *net) #endif +unsigned int fib6_tables_seq_read(struct net *net) +{ + unsigned int h, fib_seq = 0; + + rcu_read_lock(); + for (h = 0; h < FIB6_TABLE_HASHSZ; h++) { + struct hlist_head *head = &net->ipv6.fib_table_hash[h]; + struct fib6_table *tb; + + hlist_for_each_entry_rcu(tb, head, tb6_hlist) { + read_lock_bh(&tb->tb6_lock); + fib_seq += tb->fib_seq; + read_unlock_bh(&tb->tb6_lock); + } + } + rcu_read_unlock(); + + return fib_seq; +} + +static int call_fib6_entry_notifier(struct notifier_block *nb, struct net *net, + enum fib_event_type event_type, + struct rt6_info *rt) +{ + struct fib6_entry_notifier_info info = { + .rt = rt, + }; + + return call_fib6_notifier(nb, net, event_type, &info.info); +} + +static int call_fib6_entry_notifiers(struct net *net, + enum fib_event_type event_type, + struct rt6_info *rt) +{ + struct fib6_entry_notifier_info info = { + .rt = rt, + }; + + rt->rt6i_table->fib_seq++; + return call_fib6_notifiers(net, event_type, &info.info); +} + +struct fib6_dump_arg { + struct net *net; + struct notifier_block *nb; +}; + +static void fib6_rt_dump(struct rt6_info *rt, struct fib6_dump_arg *arg) +{ + if (rt == arg->net->ipv6.ip6_null_entry) + return; + call_fib6_entry_notifier(arg->nb, arg->net, FIB_EVENT_ENTRY_ADD, rt); +} + +static int fib6_node_dump(struct fib6_walker *w) +{ + struct rt6_info *rt; + + for (rt = w->leaf; rt; rt = rt->dst.rt6_next) + fib6_rt_dump(rt, w->args); + w->leaf = NULL; + return 0; +} + +static void fib6_table_dump(struct net *net, struct fib6_table *tb, + struct fib6_walker *w) +{ + w->root = &tb->tb6_root; + read_lock_bh(&tb->tb6_lock); + fib6_walk(net, w); + read_unlock_bh(&tb->tb6_lock); +} + +/* Called with rcu_read_lock() */ +int fib6_tables_dump(struct net *net, struct notifier_block *nb) +{ + struct fib6_dump_arg arg; + struct fib6_walker *w; + unsigned int h; + + w = kzalloc(sizeof(*w), GFP_ATOMIC); + if (!w) + return -ENOMEM; + + w->func = fib6_node_dump; + arg.net = net; + arg.nb = nb; + w->args = &arg; + + for (h = 0; h < FIB6_TABLE_HASHSZ; h++) { + struct hlist_head *head = &net->ipv6.fib_table_hash[h]; + struct fib6_table *tb; + + hlist_for_each_entry_rcu(tb, head, tb6_hlist) + fib6_table_dump(net, tb, w); + } + + kfree(w); + + return 0; +} + static int fib6_dump_node(struct fib6_walker *w) { int res; @@ -601,9 +709,9 @@ insert_above: if (!in || !ln) { if (in) - node_free(in); + node_free_immediate(in); if (ln) - node_free(ln); + node_free_immediate(ln); return ERR_PTR(-ENOMEM); } @@ -733,8 +841,6 @@ static void fib6_purge_rt(struct rt6_info *rt, struct fib6_node *fn, } fn = fn->parent; } - /* No more references are possible at this point. */ - BUG_ON(atomic_read(&rt->rt6i_ref) != 1); } } @@ -877,8 +983,10 @@ add: rt->dst.rt6_next = iter; *ins = rt; - rt->rt6i_node = fn; + rcu_assign_pointer(rt->rt6i_node, fn); atomic_inc(&rt->rt6i_ref); + call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_ADD, + rt); if (!info->skip_notify) inet6_rt_notify(RTM_NEWROUTE, rt, info, nlflags); info->nl_net->ipv6.rt6_stats->fib_rt_entries++; @@ -903,9 +1011,11 @@ add: return err; *ins = rt; - rt->rt6i_node = fn; + rcu_assign_pointer(rt->rt6i_node, fn); rt->dst.rt6_next = iter->dst.rt6_next; atomic_inc(&rt->rt6i_ref); + call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_REPLACE, + rt); if (!info->skip_notify) inet6_rt_notify(RTM_NEWROUTE, rt, info, NLM_F_REPLACE); if (!(fn->fn_flags & RTN_RTINFO)) { @@ -913,7 +1023,10 @@ add: fn->fn_flags |= RTN_RTINFO; } nsiblings = iter->rt6i_nsiblings; + iter->rt6i_node = NULL; fib6_purge_rt(iter, fn, info->nl_net); + if (fn->rr_ptr == iter) + fn->rr_ptr = NULL; rt6_release(iter); if (nsiblings) { @@ -925,7 +1038,10 @@ add: break; if (rt6_qualify_for_ecmp(iter)) { *ins = iter->dst.rt6_next; + iter->rt6i_node = NULL; fib6_purge_rt(iter, fn, info->nl_net); + if (fn->rr_ptr == iter) + fn->rr_ptr = NULL; rt6_release(iter); nsiblings--; } else { @@ -1014,7 +1130,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, /* Create subtree root node */ sfn = node_alloc(); if (!sfn) - goto st_failure; + goto failure; sfn->leaf = info->nl_net->ipv6.ip6_null_entry; atomic_inc(&info->nl_net->ipv6.ip6_null_entry->rt6i_ref); @@ -1031,12 +1147,12 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (IS_ERR(sn)) { /* If it is failed, discard just allocated - root, and then (in st_failure) stale node + root, and then (in failure) stale node in main tree. */ - node_free(sfn); + node_free_immediate(sfn); err = PTR_ERR(sn); - goto st_failure; + goto failure; } /* Now link new subtree to main tree */ @@ -1051,7 +1167,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (IS_ERR(sn)) { err = PTR_ERR(sn); - goto st_failure; + goto failure; } } @@ -1092,18 +1208,17 @@ out: atomic_inc(&pn->leaf->rt6i_ref); } #endif - /* Always release dst as dst->__refcnt is guaranteed - * to be taken before entering this function - */ - dst_release_immediate(&rt->dst); + goto failure; } return err; -#ifdef CONFIG_IPV6_SUBTREES - /* Subtree creation failed, probably main tree node - is orphan. If it is, shoot it. +failure: + /* fn->leaf could be NULL if fn is an intermediate node and we + * failed to add the new route to it in both subtree creation + * failure and fib6_add_rt2node() failure case. + * In both cases, fib6_repair_tree() should be called to fix + * fn->leaf. */ -st_failure: if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT))) fib6_repair_tree(info->nl_net, fn); /* Always release dst as dst->__refcnt is guaranteed @@ -1111,7 +1226,6 @@ st_failure: */ dst_release_immediate(&rt->dst); return err; -#endif } /* @@ -1459,6 +1573,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, fib6_purge_rt(rt, fn, net); + call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, rt); if (!info->skip_notify) inet6_rt_notify(RTM_DELROUTE, rt, info, 0); rt6_release(rt); @@ -1466,8 +1581,9 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, int fib6_del(struct rt6_info *rt, struct nl_info *info) { + struct fib6_node *fn = rcu_dereference_protected(rt->rt6i_node, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); struct net *net = info->nl_net; - struct fib6_node *fn = rt->rt6i_node; struct rt6_info **rtp; #if RT6_DEBUG >= 2 @@ -1656,7 +1772,9 @@ static int fib6_clean_node(struct fib6_walker *w) if (res) { #if RT6_DEBUG >= 2 pr_debug("%s: del failed: rt=%p@%p err=%d\n", - __func__, rt, rt->rt6i_node, res); + __func__, rt, + rcu_access_pointer(rt->rt6i_node), + res); #endif continue; } @@ -1778,8 +1896,10 @@ static int fib6_age(struct rt6_info *rt, void *arg) } gc_args->more++; } else if (rt->rt6i_flags & RTF_CACHE) { + if (time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) + rt->dst.obsolete = DST_OBSOLETE_KILL; if (atomic_read(&rt->dst.__refcnt) == 1 && - time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) { + rt->dst.obsolete == DST_OBSOLETE_KILL) { RT6_TRACE("aging clone %p\n", rt); return -1; } else if (rt->rt6i_flags & RTF_GATEWAY) { @@ -1839,6 +1959,11 @@ static void fib6_gc_timer_cb(unsigned long arg) static int __net_init fib6_net_init(struct net *net) { size_t size = sizeof(struct hlist_head) * FIB6_TABLE_HASHSZ; + int err; + + err = fib6_notifier_init(net); + if (err) + return err; spin_lock_init(&net->ipv6.fib6_gc_lock); rwlock_init(&net->ipv6.fib6_walker_lock); @@ -1891,6 +2016,7 @@ out_fib_table_hash: out_rt6_stats: kfree(net->ipv6.rt6_stats); out_timer: + fib6_notifier_exit(net); return -ENOMEM; } @@ -1907,6 +2033,7 @@ static void fib6_net_exit(struct net *net) kfree(net->ipv6.fib6_main_tbl); kfree(net->ipv6.fib_table_hash); kfree(net->ipv6.rt6_stats); + fib6_notifier_exit(net); } static struct pernet_operations fib6_net_ops = { @@ -1930,7 +2057,7 @@ int __init fib6_init(void) goto out_kmem_cache_create; ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib, - NULL); + 0); if (ret) goto out_unregister_subsys; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 1422d6c08377..43ca864327c7 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -673,8 +673,6 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, *prevhdr = NEXTHDR_FRAGMENT; tmp_hdr = kmemdup(skb_network_header(skb), hlen, GFP_ATOMIC); if (!tmp_hdr) { - IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), - IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; } @@ -789,8 +787,6 @@ slow_path: frag = alloc_skb(len + hlen + sizeof(struct frag_hdr) + hroom + troom, GFP_ATOMIC); if (!frag) { - IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), - IPSTATS_MIB_FRAGFAILS); err = -ENOMEM; goto fail; } @@ -1114,69 +1110,6 @@ struct dst_entry *ip6_sk_dst_lookup_flow(struct sock *sk, struct flowi6 *fl6, } EXPORT_SYMBOL_GPL(ip6_sk_dst_lookup_flow); -static inline int ip6_ufo_append_data(struct sock *sk, - struct sk_buff_head *queue, - int getfrag(void *from, char *to, int offset, int len, - int odd, struct sk_buff *skb), - void *from, int length, int hh_len, int fragheaderlen, - int exthdrlen, int transhdrlen, int mtu, - unsigned int flags, const struct flowi6 *fl6) - -{ - struct sk_buff *skb; - int err; - - /* There is support for UDP large send offload by network - * device, so create one single skb packet containing complete - * udp datagram - */ - skb = skb_peek_tail(queue); - if (!skb) { - skb = sock_alloc_send_skb(sk, - hh_len + fragheaderlen + transhdrlen + 20, - (flags & MSG_DONTWAIT), &err); - if (!skb) - return err; - - /* reserve space for Hardware header */ - skb_reserve(skb, hh_len); - - /* create space for UDP/IP header */ - skb_put(skb, fragheaderlen + transhdrlen); - - /* initialize network header pointer */ - skb_set_network_header(skb, exthdrlen); - - /* initialize protocol header pointer */ - skb->transport_header = skb->network_header + fragheaderlen; - - skb->protocol = htons(ETH_P_IPV6); - skb->csum = 0; - - if (flags & MSG_CONFIRM) - skb_set_dst_pending_confirm(skb, 1); - - __skb_queue_tail(queue, skb); - } else if (skb_is_gso(skb)) { - goto append; - } - - skb->ip_summed = CHECKSUM_PARTIAL; - /* Specify the length of each IPv6 datagram fragment. - * It has to be a multiple of 8. - */ - skb_shinfo(skb)->gso_size = (mtu - fragheaderlen - - sizeof(struct frag_hdr)) & ~7; - skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - skb_shinfo(skb)->ip6_frag_id = ipv6_select_ident(sock_net(sk), - &fl6->daddr, - &fl6->saddr); - -append: - return skb_append_datato_frags(sk, skb, getfrag, from, - (length - transhdrlen)); -} - static inline struct ipv6_opt_hdr *ip6_opt_dup(struct ipv6_opt_hdr *src, gfp_t gfp) { @@ -1385,19 +1318,6 @@ emsgsize: */ cork->length += length; - if ((((length + (skb ? skb->len : headersize)) > mtu) || - (skb && skb_is_gso(skb))) && - (sk->sk_protocol == IPPROTO_UDP) && - (rt->dst.dev->features & NETIF_F_UFO) && !dst_xfrm(&rt->dst) && - (sk->sk_type == SOCK_DGRAM) && !udp_get_no_check6_tx(sk)) { - err = ip6_ufo_append_data(sk, queue, getfrag, from, length, - hh_len, fragheaderlen, exthdrlen, - transhdrlen, mtu, flags, fl6); - if (err) - goto error; - return 0; - } - if (!skb) goto alloc_new_skb; diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 486c2305f53c..79444a4bfd6d 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -1145,33 +1145,6 @@ static struct xfrm6_protocol vti_ipcomp6_protocol __read_mostly = { .priority = 100, }; -static bool is_vti6_tunnel(const struct net_device *dev) -{ - return dev->netdev_ops == &vti6_netdev_ops; -} - -static int vti6_device_event(struct notifier_block *unused, - unsigned long event, void *ptr) -{ - struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct ip6_tnl *t = netdev_priv(dev); - - if (!is_vti6_tunnel(dev)) - return NOTIFY_DONE; - - switch (event) { - case NETDEV_DOWN: - if (!net_eq(t->net, dev_net(dev))) - xfrm_garbage_collect(t->net); - break; - } - return NOTIFY_DONE; -} - -static struct notifier_block vti6_notifier_block __read_mostly = { - .notifier_call = vti6_device_event, -}; - /** * vti6_tunnel_init - register protocol and reserve needed resources * @@ -1182,8 +1155,6 @@ static int __init vti6_tunnel_init(void) const char *msg; int err; - register_netdevice_notifier(&vti6_notifier_block); - msg = "tunnel device"; err = register_pernet_device(&vti6_net_ops); if (err < 0) @@ -1216,7 +1187,6 @@ xfrm_proto_ah_failed: xfrm_proto_esp_failed: unregister_pernet_device(&vti6_net_ops); pernet_dev_failed: - unregister_netdevice_notifier(&vti6_notifier_block); pr_err("vti6 init: failed to register %s\n", msg); return err; } @@ -1231,7 +1201,6 @@ static void __exit vti6_tunnel_cleanup(void) xfrm6_protocol_deregister(&vti_ah6_protocol, IPPROTO_AH); xfrm6_protocol_deregister(&vti_esp6_protocol, IPPROTO_ESP); unregister_pernet_device(&vti6_net_ops); - unregister_netdevice_notifier(&vti6_notifier_block); } module_init(vti6_tunnel_init); diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 7454850f2098..f5500f5444e9 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1427,7 +1427,7 @@ int __init ip6_mr_init(void) } #endif rtnl_register(RTNL_FAMILY_IP6MR, RTM_GETROUTE, NULL, - ip6mr_rtm_dumproute, NULL); + ip6mr_rtm_dumproute, 0); return 0; #ifdef CONFIG_IPV6_PIMSM_V2 add_proto_fail: diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 02d795fe3d7f..a5e466d4e093 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -242,7 +242,6 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, pktopt = xchg(&np->pktoptions, NULL); kfree_skb(pktopt); - sk->sk_destruct = inet_sock_destruct; /* * ... and add it to the refcnt debug socks count * in the new family. -acme diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 0327c1f2e6fc..266a530414d7 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -127,7 +127,7 @@ struct neigh_table nd_tbl = { [NEIGH_VAR_BASE_REACHABLE_TIME] = ND_REACHABLE_TIME, [NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ, [NEIGH_VAR_GC_STALETIME] = 60 * HZ, - [NEIGH_VAR_QUEUE_LEN_BYTES] = 64 * 1024, + [NEIGH_VAR_QUEUE_LEN_BYTES] = SK_WMEM_MAX, [NEIGH_VAR_PROXY_QLEN] = 64, [NEIGH_VAR_ANYCAST_DELAY] = 1 * HZ, [NEIGH_VAR_PROXY_DELAY] = (8 * HZ) / 10, @@ -1779,6 +1779,7 @@ static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, static struct notifier_block ndisc_netdev_notifier = { .notifier_call = ndisc_netdev_event, + .priority = ADDRCONF_NOTIFY_PRIORITY - 5, }; #ifdef CONFIG_SYSCTL diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 1f90644056ac..54b1e75eded1 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -39,12 +39,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); MODULE_DESCRIPTION("IPv6 packet filter"); -#ifdef CONFIG_NETFILTER_DEBUG -#define IP_NF_ASSERT(x) WARN_ON(!(x)) -#else -#define IP_NF_ASSERT(x) -#endif - void *ip6t_alloc_initial_table(const struct xt_table *info) { return xt_alloc_initial_table(ip6t, IP6T); @@ -176,7 +170,7 @@ static const char *const comments[] = { [NF_IP6_TRACE_COMMENT_POLICY] = "policy", }; -static struct nf_loginfo trace_loginfo = { +static const struct nf_loginfo trace_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { @@ -284,7 +278,7 @@ ip6t_do_table(struct sk_buff *skb, acpar.hotdrop = false; acpar.state = state; - IP_NF_ASSERT(table->valid_hooks & (1 << hook)); + WARN_ON(!(table->valid_hooks & (1 << hook))); local_bh_disable(); addend = xt_write_recseq_begin(); @@ -315,7 +309,7 @@ ip6t_do_table(struct sk_buff *skb, const struct xt_entry_match *ematch; struct xt_counters *counter; - IP_NF_ASSERT(e); + WARN_ON(!e); acpar.thoff = 0; if (!ip6_packet_match(skb, indev, outdev, &e->ipv6, &acpar.thoff, &acpar.fragoff, &acpar.hotdrop)) { @@ -335,7 +329,7 @@ ip6t_do_table(struct sk_buff *skb, ADD_COUNTER(*counter, skb->len, 1); t = ip6t_get_target_c(e); - IP_NF_ASSERT(t->u.kernel.target); + WARN_ON(!t->u.kernel.target); #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) /* The packet is traced: log it */ diff --git a/net/ipv6/netfilter/ip6t_SYNPROXY.c b/net/ipv6/netfilter/ip6t_SYNPROXY.c index ce203dd729e0..a5cd43d75393 100644 --- a/net/ipv6/netfilter/ip6t_SYNPROXY.c +++ b/net/ipv6/netfilter/ip6t_SYNPROXY.c @@ -438,7 +438,7 @@ static unsigned int ipv6_synproxy_hook(void *priv, return NF_ACCEPT; } -static struct nf_hook_ops ipv6_synproxy_ops[] __read_mostly = { +static const struct nf_hook_ops ipv6_synproxy_ops[] = { { .hook = ipv6_synproxy_hook, .pf = NFPROTO_IPV6, diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index 7d2bd940291f..991512576c8c 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -69,7 +69,7 @@ static unsigned int ip6table_nat_local_fn(void *priv, return nf_nat_ipv6_local_fn(priv, skb, state, ip6table_nat_do_chain); } -static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = { +static const struct nf_hook_ops nf_nat_ipv6_ops[] = { /* Before packet filtering, change destination */ { .hook = ip6table_nat_in, diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index 4e3402486833..fe01dc953c56 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -67,13 +67,6 @@ static bool ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -static void ipv6_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "src=%pI6 dst=%pI6 ", - tuple->src.u3.ip6, tuple->dst.u3.ip6); -} - static int ipv6_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, unsigned int *dataoff, u_int8_t *protonum) { @@ -191,7 +184,7 @@ static unsigned int ipv6_conntrack_local(void *priv, return nf_conntrack_in(state->net, PF_INET6, state->hook, skb); } -static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = { +static const struct nf_hook_ops ipv6_conntrack_ops[] = { { .hook = ipv6_conntrack_in, .pf = NFPROTO_IPV6, @@ -308,11 +301,6 @@ static int ipv6_nlattr_to_tuple(struct nlattr *tb[], return 0; } - -static int ipv6_nlattr_tuple_size(void) -{ - return nla_policy_len(ipv6_nla_policy, CTA_IP_MAX + 1); -} #endif static int ipv6_hooks_register(struct net *net) @@ -353,16 +341,15 @@ static void ipv6_hooks_unregister(struct net *net) struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 __read_mostly = { .l3proto = PF_INET6, - .name = "ipv6", .pkt_to_tuple = ipv6_pkt_to_tuple, .invert_tuple = ipv6_invert_tuple, - .print_tuple = ipv6_print_tuple, .get_l4proto = ipv6_get_l4proto, #if IS_ENABLED(CONFIG_NF_CT_NETLINK) .tuple_to_nlattr = ipv6_tuple_to_nlattr, - .nlattr_tuple_size = ipv6_nlattr_tuple_size, .nlattr_to_tuple = ipv6_nlattr_to_tuple, .nla_policy = ipv6_nla_policy, + .nla_size = NLA_ALIGN(NLA_HDRLEN + sizeof(u32[4])) + + NLA_ALIGN(NLA_HDRLEN + sizeof(u32[4])), #endif .net_ns_get = ipv6_hooks_register, .net_ns_put = ipv6_hooks_unregister, @@ -398,25 +385,12 @@ static struct nf_conntrack_l4proto *builtin_l4proto6[] = { static int ipv6_net_init(struct net *net) { - int ret = 0; - - ret = nf_ct_l4proto_pernet_register(net, builtin_l4proto6, - ARRAY_SIZE(builtin_l4proto6)); - if (ret < 0) - return ret; - - ret = nf_ct_l3proto_pernet_register(net, &nf_conntrack_l3proto_ipv6); - if (ret < 0) { - pr_err("nf_conntrack_ipv6: pernet registration failed.\n"); - nf_ct_l4proto_pernet_unregister(net, builtin_l4proto6, - ARRAY_SIZE(builtin_l4proto6)); - } - return ret; + return nf_ct_l4proto_pernet_register(net, builtin_l4proto6, + ARRAY_SIZE(builtin_l4proto6)); } static void ipv6_net_exit(struct net *net) { - nf_ct_l3proto_pernet_unregister(net, &nf_conntrack_l3proto_ipv6); nf_ct_l4proto_pernet_unregister(net, builtin_l4proto6, ARRAY_SIZE(builtin_l4proto6)); } @@ -434,6 +408,12 @@ static int __init nf_conntrack_l3proto_ipv6_init(void) need_conntrack(); +#if IS_ENABLED(CONFIG_NF_CT_NETLINK) + if (WARN_ON(nla_policy_len(ipv6_nla_policy, CTA_IP_MAX + 1) != + nf_conntrack_l3proto_ipv6.nla_size)) + return -EINVAL; +#endif + ret = nf_register_sockopt(&so_getorigdst6); if (ret < 0) { pr_err("Unable to register netfilter socket option\n"); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index d5f028e33f65..a9e1fd1a8536 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -84,16 +84,6 @@ static bool icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void icmpv6_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "type=%u code=%u id=%u ", - tuple->dst.u.icmp.type, - tuple->dst.u.icmp.code, - ntohs(tuple->src.u.icmp.id)); -} - static unsigned int *icmpv6_get_timeouts(struct net *net) { return &icmpv6_pernet(net)->timeout; @@ -105,7 +95,6 @@ static int icmpv6_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeout) { /* Do not immediately delete the connection after the first @@ -131,11 +120,6 @@ static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb, pr_debug("icmpv6: can't create new conn with type %u\n", type + 128); nf_ct_dump_tuple_ipv6(&ct->tuplehash[0].tuple); - if (LOG_INVALID(nf_ct_net(ct), IPPROTO_ICMPV6)) - nf_log_packet(nf_ct_net(ct), PF_INET6, 0, skb, NULL, - NULL, NULL, - "nf_ct_icmpv6: invalid new with type %d ", - type + 128); return false; } return true; @@ -144,8 +128,7 @@ static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb, static int icmpv6_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, - unsigned int icmp6off, - unsigned int hooknum) + unsigned int icmp6off) { struct nf_conntrack_tuple intuple, origtuple; const struct nf_conntrack_tuple_hash *h; @@ -153,7 +136,7 @@ icmpv6_error_message(struct net *net, struct nf_conn *tmpl, enum ip_conntrack_info ctinfo; struct nf_conntrack_zone tmp; - NF_CT_ASSERT(!skb_nfct(skb)); + WARN_ON(skb_nfct(skb)); /* Are they talking about one of our connections? */ if (!nf_ct_get_tuplepr(skb, @@ -229,7 +212,7 @@ icmpv6_error(struct net *net, struct nf_conn *tmpl, if (icmp6h->icmp6_type >= 128) return NF_ACCEPT; - return icmpv6_error_message(net, tmpl, skb, dataoff, hooknum); + return icmpv6_error_message(net, tmpl, skb, dataoff); } #if IS_ENABLED(CONFIG_NF_CT_NETLINK) @@ -367,10 +350,8 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6 __read_mostly = { .l3proto = PF_INET6, .l4proto = IPPROTO_ICMPV6, - .name = "icmpv6", .pkt_to_tuple = icmpv6_pkt_to_tuple, .invert_tuple = icmpv6_invert_tuple, - .print_tuple = icmpv6_print_tuple, .packet = icmpv6_packet, .get_timeouts = icmpv6_get_timeouts, .new = icmpv6_new, diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 986d4ca38832..b263bf3a19f7 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -622,18 +622,12 @@ EXPORT_SYMBOL_GPL(nf_ct_frag6_gather); static int nf_ct_net_init(struct net *net) { - int res; - net->nf_frag.frags.high_thresh = IPV6_FRAG_HIGH_THRESH; net->nf_frag.frags.low_thresh = IPV6_FRAG_LOW_THRESH; net->nf_frag.frags.timeout = IPV6_FRAG_TIMEOUT; - res = inet_frags_init_net(&net->nf_frag.frags); - if (res) - return res; - res = nf_ct_frag6_sysctl_register(net); - if (res) - inet_frags_uninit_net(&net->nf_frag.frags); - return res; + inet_frags_init_net(&net->nf_frag.frags); + + return nf_ct_frag6_sysctl_register(net); } static void nf_ct_net_exit(struct net *net) diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index ada60d1a991b..b326da59257f 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -74,7 +74,7 @@ static unsigned int ipv6_defrag(void *priv, return err == 0 ? NF_ACCEPT : NF_DROP; } -static struct nf_hook_ops ipv6_defrag_ops[] = { +static const struct nf_hook_ops ipv6_defrag_ops[] = { { .hook = ipv6_defrag, .pf = NFPROTO_IPV6, diff --git a/net/ipv6/netfilter/nf_log_ipv6.c b/net/ipv6/netfilter/nf_log_ipv6.c index 97c724224da7..b397a8fe88b9 100644 --- a/net/ipv6/netfilter/nf_log_ipv6.c +++ b/net/ipv6/netfilter/nf_log_ipv6.c @@ -25,7 +25,7 @@ #include <linux/netfilter/xt_LOG.h> #include <net/netfilter/nf_log.h> -static struct nf_loginfo default_loginfo = { +static const struct nf_loginfo default_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { diff --git a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c index b2b4f031b3a1..46d6dba50698 100644 --- a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c @@ -196,7 +196,7 @@ int nf_nat_icmpv6_reply_translation(struct sk_buff *skb, struct nf_conntrack_tuple target; unsigned long statusbit; - NF_CT_ASSERT(ctinfo == IP_CT_RELATED || ctinfo == IP_CT_RELATED_REPLY); + WARN_ON(ctinfo != IP_CT_RELATED && ctinfo != IP_CT_RELATED_REPLY); if (!skb_make_writable(skb, hdrlen + sizeof(*inside))) return 0; @@ -319,8 +319,8 @@ nf_nat_ipv6_fn(void *priv, struct sk_buff *skb, default: /* ESTABLISHED */ - NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED || - ctinfo == IP_CT_ESTABLISHED_REPLY); + WARN_ON(ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED_REPLY); if (nf_nat_oif_changed(state->hook, ctinfo, nat, state->out)) goto oif_changed; } diff --git a/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c b/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c index d7b679037bae..98f61fcb9108 100644 --- a/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c +++ b/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c @@ -36,8 +36,8 @@ nf_nat_masquerade_ipv6(struct sk_buff *skb, const struct nf_nat_range *range, struct nf_nat_range newrange; ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || - ctinfo == IP_CT_RELATED_REPLY)); + WARN_ON(!(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || + ctinfo == IP_CT_RELATED_REPLY))); if (ipv6_dev_get_saddr(nf_ct_net(ct), out, &ipv6_hdr(skb)->daddr, 0, &src) < 0) diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index 43f91d9b086c..54b5899543ef 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -25,9 +25,9 @@ static int get_ifindex(const struct net_device *dev) static int nft_fib6_flowi_init(struct flowi6 *fl6, const struct nft_fib *priv, const struct nft_pktinfo *pkt, - const struct net_device *dev) + const struct net_device *dev, + struct ipv6hdr *iph) { - const struct ipv6hdr *iph = ipv6_hdr(pkt->skb); int lookup_flags = 0; if (priv->flags & NFTA_FIB_F_DADDR) { @@ -55,7 +55,8 @@ static int nft_fib6_flowi_init(struct flowi6 *fl6, const struct nft_fib *priv, } static u32 __nft_fib6_eval_type(const struct nft_fib *priv, - const struct nft_pktinfo *pkt) + const struct nft_pktinfo *pkt, + struct ipv6hdr *iph) { const struct net_device *dev = NULL; const struct nf_ipv6_ops *v6ops; @@ -77,7 +78,7 @@ static u32 __nft_fib6_eval_type(const struct nft_fib *priv, else if (priv->flags & NFTA_FIB_F_OIF) dev = nft_out(pkt); - nft_fib6_flowi_init(&fl6, priv, pkt, dev); + nft_fib6_flowi_init(&fl6, priv, pkt, dev, iph); v6ops = nf_get_ipv6_ops(); if (dev && v6ops && v6ops->chk_addr(nft_net(pkt), &fl6.daddr, dev, true)) @@ -131,9 +132,17 @@ void nft_fib6_eval_type(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_fib *priv = nft_expr_priv(expr); + int noff = skb_network_offset(pkt->skb); u32 *dest = ®s->data[priv->dreg]; + struct ipv6hdr *iph, _iph; - *dest = __nft_fib6_eval_type(priv, pkt); + iph = skb_header_pointer(pkt->skb, noff, sizeof(_iph), &_iph); + if (!iph) { + regs->verdict.code = NFT_BREAK; + return; + } + + *dest = __nft_fib6_eval_type(priv, pkt, iph); } EXPORT_SYMBOL_GPL(nft_fib6_eval_type); @@ -141,8 +150,10 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_fib *priv = nft_expr_priv(expr); + int noff = skb_network_offset(pkt->skb); const struct net_device *oif = NULL; u32 *dest = ®s->data[priv->dreg]; + struct ipv6hdr *iph, _iph; struct flowi6 fl6 = { .flowi6_iif = LOOPBACK_IFINDEX, .flowi6_proto = pkt->tprot, @@ -155,7 +166,13 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs, else if (priv->flags & NFTA_FIB_F_OIF) oif = nft_out(pkt); - lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif); + iph = skb_header_pointer(pkt->skb, noff, sizeof(_iph), &_iph); + if (!iph) { + regs->verdict.code = NFT_BREAK; + return; + } + + lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif, iph); if (nft_hook(pkt) == NF_INET_PRE_ROUTING && nft_fib_is_loopback(pkt->skb, nft_in(pkt))) { diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index e9065b8d3af8..a338bbc33cf3 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -78,7 +78,7 @@ EXPORT_SYMBOL(ipv6_select_ident); int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) { - u16 offset = sizeof(struct ipv6hdr); + unsigned int offset = sizeof(struct ipv6hdr); unsigned int packet_len = skb_tail_pointer(skb) - skb_network_header(skb); int found_rhdr = 0; @@ -112,6 +112,8 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) exthdr = (struct ipv6_opt_hdr *)(skb_network_header(skb) + offset); offset += ipv6_optlen(exthdr); + if (offset > IPV6_MAXPLEN) + return -EINVAL; *nexthdr = &exthdr->nexthdr; } diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 60be012fe708..e4462b0ff801 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -72,7 +72,7 @@ EXPORT_SYMBOL_GPL(raw_v6_hashinfo); struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, unsigned short num, const struct in6_addr *loc_addr, - const struct in6_addr *rmt_addr, int dif) + const struct in6_addr *rmt_addr, int dif, int sdif) { bool is_multicast = ipv6_addr_is_multicast(loc_addr); @@ -86,7 +86,9 @@ struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, !ipv6_addr_equal(&sk->sk_v6_daddr, rmt_addr)) continue; - if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif) + if (sk->sk_bound_dev_if && + sk->sk_bound_dev_if != dif && + sk->sk_bound_dev_if != sdif) continue; if (!ipv6_addr_any(&sk->sk_v6_rcv_saddr)) { @@ -178,7 +180,8 @@ static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) goto out; net = dev_net(skb->dev); - sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, inet6_iif(skb)); + sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, + inet6_iif(skb), inet6_sdif(skb)); while (sk) { int filtered; @@ -222,7 +225,7 @@ static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) } } sk = __raw_v6_lookup(net, sk_next(sk), nexthdr, daddr, saddr, - inet6_iif(skb)); + inet6_iif(skb), inet6_sdif(skb)); } out: read_unlock(&raw_v6_hashinfo.lock); @@ -378,7 +381,7 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr, net = dev_net(skb->dev); while ((sk = __raw_v6_lookup(net, sk, nexthdr, saddr, daddr, - inet6_iif(skb)))) { + inet6_iif(skb), inet6_iif(skb)))) { rawv6_err(sk, skb, NULL, type, code, inner_offset, info); sk = sk_next(sk); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index e1da5b888cc4..846012eae526 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -714,19 +714,13 @@ static void ip6_frags_sysctl_unregister(void) static int __net_init ipv6_frags_init_net(struct net *net) { - int res; - net->ipv6.frags.high_thresh = IPV6_FRAG_HIGH_THRESH; net->ipv6.frags.low_thresh = IPV6_FRAG_LOW_THRESH; net->ipv6.frags.timeout = IPV6_FRAG_TIMEOUT; - res = inet_frags_init_net(&net->ipv6.frags); - if (res) - return res; - res = ip6_frags_ns_sysctl_register(net); - if (res) - inet_frags_uninit_net(&net->ipv6.frags); - return res; + inet_frags_init_net(&net->ipv6.frags); + + return ip6_frags_ns_sysctl_register(net); } static void __net_exit ipv6_frags_exit_net(struct net *net) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 4d30c96a819d..26cc9f483b6d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -417,14 +417,11 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, struct net_device *loopback_dev = dev_net(dev)->loopback_dev; - if (dev != loopback_dev) { - if (idev && idev->dev == dev) { - struct inet6_dev *loopback_idev = - in6_dev_get(loopback_dev); - if (loopback_idev) { - rt->rt6i_idev = loopback_idev; - in6_dev_put(idev); - } + if (idev && idev->dev != loopback_dev) { + struct inet6_dev *loopback_idev = in6_dev_get(loopback_dev); + if (loopback_idev) { + rt->rt6i_idev = loopback_idev; + in6_dev_put(idev); } } } @@ -443,21 +440,12 @@ static bool rt6_check_expired(const struct rt6_info *rt) if (time_after(jiffies, rt->dst.expires)) return true; } else if (rt->dst.from) { - return rt6_check_expired((struct rt6_info *) rt->dst.from); + return rt->dst.obsolete != DST_OBSOLETE_FORCE_CHK || + rt6_check_expired((struct rt6_info *)rt->dst.from); } return false; } -/* Multipath route selection: - * Hash based function using packet header and flowlabel. - * Adapted from fib_info_hashfn() - */ -static int rt6_info_hash_nhsfn(unsigned int candidate_count, - const struct flowi6 *fl6) -{ - return get_hash_from_flowi6(fl6) % candidate_count; -} - static struct rt6_info *rt6_multipath_select(struct rt6_info *match, struct flowi6 *fl6, int oif, int strict) @@ -465,7 +453,13 @@ static struct rt6_info *rt6_multipath_select(struct rt6_info *match, struct rt6_info *sibling, *next_sibling; int route_choosen; - route_choosen = rt6_info_hash_nhsfn(match->rt6i_nsiblings + 1, fl6); + /* We might have already computed the hash for ICMPv6 errors. In such + * case it will always be non-zero. Otherwise now is the time to do it. + */ + if (!fl6->mp_hash) + fl6->mp_hash = rt6_multipath_hash(fl6, NULL); + + route_choosen = fl6->mp_hash % (match->rt6i_nsiblings + 1); /* Don't change the route, if route_choosen == 0 * (siblings does not include ourself) */ @@ -961,10 +955,34 @@ int ip6_ins_rt(struct rt6_info *rt) return __ip6_ins_rt(rt, &info, &mxc, NULL); } +/* called with rcu_lock held */ +static struct net_device *ip6_rt_get_dev_rcu(struct rt6_info *rt) +{ + struct net_device *dev = rt->dst.dev; + + if (rt->rt6i_flags & RTF_LOCAL) { + /* for copies of local routes, dst->dev needs to be the + * device if it is a master device, the master device if + * device is enslaved, and the loopback as the default + */ + if (netif_is_l3_slave(dev) && + !rt6_need_strict(&rt->rt6i_dst.addr)) + dev = l3mdev_master_dev_rcu(dev); + else if (!netif_is_l3_master(dev)) + dev = dev_net(dev)->loopback_dev; + /* last case is netif_is_l3_master(dev) is true in which + * case we want dev returned to be dev + */ + } + + return dev; +} + static struct rt6_info *ip6_rt_cache_alloc(struct rt6_info *ort, const struct in6_addr *daddr, const struct in6_addr *saddr) { + struct net_device *dev; struct rt6_info *rt; /* @@ -974,8 +992,10 @@ static struct rt6_info *ip6_rt_cache_alloc(struct rt6_info *ort, if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU)) ort = (struct rt6_info *)ort->dst.from; - rt = __ip6_dst_alloc(dev_net(ort->dst.dev), ort->dst.dev, 0); - + rcu_read_lock(); + dev = ip6_rt_get_dev_rcu(ort); + rt = __ip6_dst_alloc(dev_net(dev), dev, 0); + rcu_read_unlock(); if (!rt) return NULL; @@ -1003,11 +1023,13 @@ static struct rt6_info *ip6_rt_cache_alloc(struct rt6_info *ort, static struct rt6_info *ip6_rt_pcpu_alloc(struct rt6_info *rt) { + struct net_device *dev; struct rt6_info *pcpu_rt; - pcpu_rt = __ip6_dst_alloc(dev_net(rt->dst.dev), - rt->dst.dev, rt->dst.flags); - + rcu_read_lock(); + dev = ip6_rt_get_dev_rcu(rt); + pcpu_rt = __ip6_dst_alloc(dev_net(dev), dev, rt->dst.flags); + rcu_read_unlock(); if (!pcpu_rt) return NULL; ip6_rt_copy_init(pcpu_rt, rt); @@ -1189,6 +1211,54 @@ struct dst_entry *ip6_route_input_lookup(struct net *net, } EXPORT_SYMBOL_GPL(ip6_route_input_lookup); +static void ip6_multipath_l3_keys(const struct sk_buff *skb, + struct flow_keys *keys) +{ + const struct ipv6hdr *outer_iph = ipv6_hdr(skb); + const struct ipv6hdr *key_iph = outer_iph; + const struct ipv6hdr *inner_iph; + const struct icmp6hdr *icmph; + struct ipv6hdr _inner_iph; + + if (likely(outer_iph->nexthdr != IPPROTO_ICMPV6)) + goto out; + + icmph = icmp6_hdr(skb); + if (icmph->icmp6_type != ICMPV6_DEST_UNREACH && + icmph->icmp6_type != ICMPV6_PKT_TOOBIG && + icmph->icmp6_type != ICMPV6_TIME_EXCEED && + icmph->icmp6_type != ICMPV6_PARAMPROB) + goto out; + + inner_iph = skb_header_pointer(skb, + skb_transport_offset(skb) + sizeof(*icmph), + sizeof(_inner_iph), &_inner_iph); + if (!inner_iph) + goto out; + + key_iph = inner_iph; +out: + memset(keys, 0, sizeof(*keys)); + keys->control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; + keys->addrs.v6addrs.src = key_iph->saddr; + keys->addrs.v6addrs.dst = key_iph->daddr; + keys->tags.flow_label = ip6_flowinfo(key_iph); + keys->basic.ip_proto = key_iph->nexthdr; +} + +/* if skb is set it will be used and fl6 can be NULL */ +u32 rt6_multipath_hash(const struct flowi6 *fl6, const struct sk_buff *skb) +{ + struct flow_keys hash_keys; + + if (skb) { + ip6_multipath_l3_keys(skb, &hash_keys); + return flow_hash_from_keys(&hash_keys); + } + + return get_hash_from_flowi6(fl6); +} + void ip6_route_input(struct sk_buff *skb) { const struct ipv6hdr *iph = ipv6_hdr(skb); @@ -1207,6 +1277,8 @@ void ip6_route_input(struct sk_buff *skb) tun_info = skb_tunnel_info(skb); if (tun_info && !(tun_info->mode & IP_TUNNEL_INFO_TX)) fl6.flowi6_tun_key.tun_id = tun_info->key.tun_id; + if (unlikely(fl6.flowi6_proto == IPPROTO_ICMPV6)) + fl6.mp_hash = rt6_multipath_hash(&fl6, skb); skb_dst_drop(skb); skb_dst_set(skb, ip6_route_input_lookup(net, skb->dev, &fl6, flags)); } @@ -1292,7 +1364,9 @@ static void rt6_dst_from_metrics_check(struct rt6_info *rt) static struct dst_entry *rt6_check(struct rt6_info *rt, u32 cookie) { - if (!rt->rt6i_node || (rt->rt6i_node->fn_sernum != cookie)) + u32 rt_cookie = 0; + + if (!rt6_get_cookie_safe(rt, &rt_cookie) || rt_cookie != cookie) return NULL; if (rt6_check_expired(rt)) @@ -1360,8 +1434,14 @@ static void ip6_link_failure(struct sk_buff *skb) if (rt->rt6i_flags & RTF_CACHE) { if (dst_hold_safe(&rt->dst)) ip6_del_rt(rt); - } else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) { - rt->rt6i_node->fn_sernum = -1; + } else { + struct fib6_node *fn; + + rcu_read_lock(); + fn = rcu_dereference(rt->rt6i_node); + if (fn && (rt->rt6i_flags & RTF_DEFAULT)) + fn->fn_sernum = -1; + rcu_read_unlock(); } } } @@ -1378,7 +1458,8 @@ static void rt6_do_update_pmtu(struct rt6_info *rt, u32 mtu) static bool rt6_cache_allowed_for_pmtu(const struct rt6_info *rt) { return !(rt->rt6i_flags & RTF_CACHE) && - (rt->rt6i_flags & RTF_PCPU || rt->rt6i_node); + (rt->rt6i_flags & RTF_PCPU || + rcu_access_pointer(rt->rt6i_node)); } static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk, @@ -2351,6 +2432,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu if (on_link) nrt->rt6i_flags &= ~RTF_GATEWAY; + nrt->rt6i_protocol = RTPROT_REDIRECT; nrt->rt6i_gateway = *(struct in6_addr *)neigh->primary_key; if (ip6_ins_rt(nrt)) @@ -2461,6 +2543,7 @@ static struct rt6_info *rt6_add_route_info(struct net *net, .fc_dst_len = prefixlen, .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO | RTF_UP | RTF_PREF(pref), + .fc_protocol = RTPROT_RA, .fc_nlinfo.portid = 0, .fc_nlinfo.nlh = NULL, .fc_nlinfo.nl_net = net, @@ -2513,6 +2596,7 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr, .fc_ifindex = dev->ifindex, .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES | RTF_PREF(pref), + .fc_protocol = RTPROT_RA, .fc_nlinfo.portid = 0, .fc_nlinfo.nlh = NULL, .fc_nlinfo.nl_net = dev_net(dev), @@ -2688,15 +2772,9 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, { u32 tb_id; struct net *net = dev_net(idev->dev); - struct net_device *dev = net->loopback_dev; + struct net_device *dev = idev->dev; struct rt6_info *rt; - /* use L3 Master device as loopback for host routes if device - * is enslaved and address is not link local or multicast - */ - if (!rt6_need_strict(addr)) - dev = l3mdev_master_dev_rcu(idev->dev) ? : dev; - rt = ip6_dst_alloc(net, dev, DST_NOCOUNT); if (!rt) return ERR_PTR(-ENOMEM); @@ -3327,6 +3405,9 @@ static int rt6_nexthop_info(struct sk_buff *skb, struct rt6_info *rt, goto nla_put_failure; } + if (rt->rt6i_nh_flags & RTNH_F_OFFLOAD) + *flags |= RTNH_F_OFFLOAD; + /* not needed for multipath encoding b/c it has a rtnexthop struct */ if (!skip_oif && rt->dst.dev && nla_put_u32(skb, RTA_OIF, rt->dst.dev->ifindex)) @@ -3424,14 +3505,6 @@ static int rt6_fill_node(struct net *net, rtm->rtm_flags = 0; rtm->rtm_scope = RT_SCOPE_UNIVERSE; rtm->rtm_protocol = rt->rt6i_protocol; - if (rt->rt6i_flags & RTF_DYNAMIC) - rtm->rtm_protocol = RTPROT_REDIRECT; - else if (rt->rt6i_flags & RTF_ADDRCONF) { - if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ROUTEINFO)) - rtm->rtm_protocol = RTPROT_RA; - else - rtm->rtm_protocol = RTPROT_KERNEL; - } if (rt->rt6i_flags & RTF_CACHE) rtm->rtm_flags |= RTM_F_CLONED; @@ -3613,8 +3686,11 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, struct net_device *dev; int flags = 0; - dev = __dev_get_by_index(net, iif); + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, iif); if (!dev) { + rcu_read_unlock(); err = -ENODEV; goto errout; } @@ -3626,15 +3702,19 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, if (!fibmatch) dst = ip6_route_input_lookup(net, dev, &fl6, flags); + else + dst = ip6_route_lookup(net, &fl6, 0); + + rcu_read_unlock(); } else { fl6.flowi6_oif = oif; if (!fibmatch) dst = ip6_route_output(net, NULL, &fl6); + else + dst = ip6_route_lookup(net, &fl6, 0); } - if (fibmatch) - dst = ip6_route_lookup(net, &fl6, 0); rt = container_of(dst, struct rt6_info, dst); if (rt->dst.error) { @@ -3729,10 +3809,10 @@ static int ip6_route_dev_notify(struct notifier_block *this, /* NETDEV_UNREGISTER could be fired for multiple times by * netdev_wait_allrefs(). Make sure we only call this once. */ - in6_dev_put(net->ipv6.ip6_null_entry->rt6i_idev); + in6_dev_put_clear(&net->ipv6.ip6_null_entry->rt6i_idev); #ifdef CONFIG_IPV6_MULTIPLE_TABLES - in6_dev_put(net->ipv6.ip6_prohibit_entry->rt6i_idev); - in6_dev_put(net->ipv6.ip6_blk_hole_entry->rt6i_idev); + in6_dev_put_clear(&net->ipv6.ip6_prohibit_entry->rt6i_idev); + in6_dev_put_clear(&net->ipv6.ip6_blk_hole_entry->rt6i_idev); #endif } @@ -3926,6 +4006,7 @@ static int __net_init ip6_route_net_init(struct net *net) ip6_template_metrics, true); #ifdef CONFIG_IPV6_MULTIPLE_TABLES + net->ipv6.fib6_has_custom_rules = false; net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template, sizeof(*net->ipv6.ip6_prohibit_entry), GFP_KERNEL); @@ -4101,9 +4182,10 @@ int __init ip6_route_init(void) goto fib6_rules_init; ret = -ENOBUFS; - if (__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL, NULL) || - __rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL, NULL) || - __rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL, NULL)) + if (__rtnl_register(PF_INET6, RTM_NEWROUTE, inet6_rtm_newroute, NULL, 0) || + __rtnl_register(PF_INET6, RTM_DELROUTE, inet6_rtm_delroute, NULL, 0) || + __rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL, + RTNL_FLAG_DOIT_UNLOCKED)) goto out_register_late_subsys; ret = register_netdevice_notifier(&ip6_route_dev_notifier); diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 15fba55e3da8..c81407770956 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -40,7 +40,7 @@ bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len) if (((srh->hdrlen + 1) << 3) != len) return false; - if (srh->segments_left != srh->first_segment) + if (srh->segments_left > srh->first_segment) return false; tlv_offset = sizeof(*srh) + ((srh->first_segment + 1) << 4); @@ -456,6 +456,10 @@ int __init seg6_init(void) err = seg6_iptunnel_init(); if (err) goto out_unregister_pernet; + + err = seg6_local_init(); + if (err) + goto out_unregister_pernet; #endif #ifdef CONFIG_IPV6_SEG6_HMAC @@ -471,6 +475,7 @@ out: #ifdef CONFIG_IPV6_SEG6_HMAC out_unregister_iptun: #ifdef CONFIG_IPV6_SEG6_LWTUNNEL + seg6_local_exit(); seg6_iptunnel_exit(); #endif #endif diff --git a/net/ipv6/seg6_hmac.c b/net/ipv6/seg6_hmac.c index f950cb53d5e3..33fb35cbfac1 100644 --- a/net/ipv6/seg6_hmac.c +++ b/net/ipv6/seg6_hmac.c @@ -12,6 +12,7 @@ */ #include <linux/errno.h> +#include <linux/kernel.h> #include <linux/types.h> #include <linux/socket.h> #include <linux/sockios.h> @@ -110,7 +111,7 @@ static struct seg6_hmac_algo *__hmac_get_algo(u8 alg_id) struct seg6_hmac_algo *algo; int i, alg_count; - alg_count = sizeof(hmac_algos) / sizeof(struct seg6_hmac_algo); + alg_count = ARRAY_SIZE(hmac_algos); for (i = 0; i < alg_count; i++) { algo = &hmac_algos[i]; if (algo->alg_id == alg_id) @@ -360,7 +361,7 @@ static int seg6_hmac_init_algo(void) struct shash_desc *shash; int i, alg_count, cpu; - alg_count = sizeof(hmac_algos) / sizeof(struct seg6_hmac_algo); + alg_count = ARRAY_SIZE(hmac_algos); for (i = 0; i < alg_count; i++) { struct crypto_shash **p_tfm; @@ -421,7 +422,7 @@ void seg6_hmac_exit(void) struct seg6_hmac_algo *algo = NULL; int i, alg_count, cpu; - alg_count = sizeof(hmac_algos) / sizeof(struct seg6_hmac_algo); + alg_count = ARRAY_SIZE(hmac_algos); for (i = 0; i < alg_count; i++) { algo = &hmac_algos[i]; for_each_possible_cpu(cpu) { diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 264d772d3c7d..bd6cc688bd19 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -91,7 +91,7 @@ static void set_tun_src(struct net *net, struct net_device *dev, } /* encapsulate an IPv6 packet within an outer IPv6 header with a given SRH */ -static int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) +int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) { struct net *net = dev_net(skb_dst(skb)->dev); struct ipv6hdr *hdr, *inner_hdr; @@ -116,15 +116,22 @@ static int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) * hlim will be decremented in ip6_forward() afterwards and * decapsulation will overwrite inner hlim with outer hlim */ - ip6_flow_hdr(hdr, ip6_tclass(ip6_flowinfo(inner_hdr)), - ip6_flowlabel(inner_hdr)); - hdr->hop_limit = inner_hdr->hop_limit; + + if (skb->protocol == htons(ETH_P_IPV6)) { + ip6_flow_hdr(hdr, ip6_tclass(ip6_flowinfo(inner_hdr)), + ip6_flowlabel(inner_hdr)); + hdr->hop_limit = inner_hdr->hop_limit; + } else { + ip6_flow_hdr(hdr, 0, 0); + hdr->hop_limit = ip6_dst_hoplimit(skb_dst(skb)); + } + hdr->nexthdr = NEXTHDR_ROUTING; isrh = (void *)hdr + sizeof(*hdr); memcpy(isrh, osrh, hdrlen); - isrh->nexthdr = NEXTHDR_IPV6; + isrh->nexthdr = proto; hdr->daddr = isrh->segments[isrh->first_segment]; set_tun_src(net, skb->dev, &hdr->daddr, &hdr->saddr); @@ -141,10 +148,10 @@ static int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) return 0; } +EXPORT_SYMBOL_GPL(seg6_do_srh_encap); /* insert an SRH within an IPv6 packet, just after the IPv6 header */ -#ifdef CONFIG_IPV6_SEG6_INLINE -static int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) +int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) { struct ipv6hdr *hdr, *oldhdr; struct ipv6_sr_hdr *isrh; @@ -193,13 +200,13 @@ static int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) return 0; } -#endif +EXPORT_SYMBOL_GPL(seg6_do_srh_inline); static int seg6_do_srh(struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct seg6_iptunnel_encap *tinfo; - int err = 0; + int proto, err = 0; tinfo = seg6_encap_lwtunnel(dst->lwtstate); @@ -209,19 +216,47 @@ static int seg6_do_srh(struct sk_buff *skb) } switch (tinfo->mode) { -#ifdef CONFIG_IPV6_SEG6_INLINE case SEG6_IPTUN_MODE_INLINE: + if (skb->protocol != htons(ETH_P_IPV6)) + return -EINVAL; + err = seg6_do_srh_inline(skb, tinfo->srh); + if (err) + return err; + skb_reset_inner_headers(skb); break; -#endif case SEG6_IPTUN_MODE_ENCAP: - err = seg6_do_srh_encap(skb, tinfo->srh); + if (skb->protocol == htons(ETH_P_IPV6)) + proto = IPPROTO_IPV6; + else if (skb->protocol == htons(ETH_P_IP)) + proto = IPPROTO_IPIP; + else + return -EINVAL; + + err = seg6_do_srh_encap(skb, tinfo->srh, proto); + if (err) + return err; + + skb->protocol = htons(ETH_P_IPV6); break; - } + case SEG6_IPTUN_MODE_L2ENCAP: + if (!skb_mac_header_was_set(skb)) + return -EINVAL; - if (err) - return err; + if (pskb_expand_head(skb, skb->mac_len, 0, GFP_ATOMIC) < 0) + return -ENOMEM; + + skb_mac_header_rebuild(skb); + skb_push(skb, skb->mac_len); + + err = seg6_do_srh_encap(skb, tinfo->srh, NEXTHDR_NONE); + if (err) + return err; + + skb->protocol = htons(ETH_P_IPV6); + break; + } ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); skb_set_transport_header(skb, sizeof(struct ipv6hdr)); @@ -336,6 +371,9 @@ static int seg6_build_state(struct nlattr *nla, struct seg6_lwt *slwt; int err; + if (family != AF_INET && family != AF_INET6) + return -EINVAL; + err = nla_parse_nested(tb, SEG6_IPTUNNEL_MAX, nla, seg6_iptunnel_policy, extack); @@ -357,12 +395,15 @@ static int seg6_build_state(struct nlattr *nla, return -EINVAL; switch (tuninfo->mode) { -#ifdef CONFIG_IPV6_SEG6_INLINE case SEG6_IPTUN_MODE_INLINE: + if (family != AF_INET6) + return -EINVAL; + break; -#endif case SEG6_IPTUN_MODE_ENCAP: break; + case SEG6_IPTUN_MODE_L2ENCAP: + break; default: return -EINVAL; } @@ -386,8 +427,11 @@ static int seg6_build_state(struct nlattr *nla, memcpy(&slwt->tuninfo, tuninfo, tuninfo_len); newts->type = LWTUNNEL_ENCAP_SEG6; - newts->flags |= LWTUNNEL_STATE_OUTPUT_REDIRECT | - LWTUNNEL_STATE_INPUT_REDIRECT; + newts->flags |= LWTUNNEL_STATE_INPUT_REDIRECT; + + if (tuninfo->mode != SEG6_IPTUN_MODE_L2ENCAP) + newts->flags |= LWTUNNEL_STATE_OUTPUT_REDIRECT; + newts->headroom = seg6_lwt_headroom(tuninfo); *ts = newts; diff --git a/net/ipv6/seg6_local.c b/net/ipv6/seg6_local.c new file mode 100644 index 000000000000..7ff54db73a48 --- /dev/null +++ b/net/ipv6/seg6_local.c @@ -0,0 +1,938 @@ +/* + * SR-IPv6 implementation + * + * Author: + * David Lebrun <david.lebrun@uclouvain.be> + * + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/types.h> +#include <linux/skbuff.h> +#include <linux/net.h> +#include <linux/module.h> +#include <net/ip.h> +#include <net/lwtunnel.h> +#include <net/netevent.h> +#include <net/netns/generic.h> +#include <net/ip6_fib.h> +#include <net/route.h> +#include <net/seg6.h> +#include <linux/seg6.h> +#include <linux/seg6_local.h> +#include <net/addrconf.h> +#include <net/ip6_route.h> +#include <net/dst_cache.h> +#ifdef CONFIG_IPV6_SEG6_HMAC +#include <net/seg6_hmac.h> +#endif +#include <linux/etherdevice.h> + +struct seg6_local_lwt; + +struct seg6_action_desc { + int action; + unsigned long attrs; + int (*input)(struct sk_buff *skb, struct seg6_local_lwt *slwt); + int static_headroom; +}; + +struct seg6_local_lwt { + int action; + struct ipv6_sr_hdr *srh; + int table; + struct in_addr nh4; + struct in6_addr nh6; + int iif; + int oif; + + int headroom; + struct seg6_action_desc *desc; +}; + +static struct seg6_local_lwt *seg6_local_lwtunnel(struct lwtunnel_state *lwt) +{ + return (struct seg6_local_lwt *)lwt->data; +} + +static struct ipv6_sr_hdr *get_srh(struct sk_buff *skb) +{ + struct ipv6_sr_hdr *srh; + int len, srhoff = 0; + + if (ipv6_find_hdr(skb, &srhoff, IPPROTO_ROUTING, NULL, NULL) < 0) + return NULL; + + if (!pskb_may_pull(skb, srhoff + sizeof(*srh))) + return NULL; + + srh = (struct ipv6_sr_hdr *)(skb->data + srhoff); + + /* make sure it's a Segment Routing header (Routing Type 4) */ + if (srh->type != IPV6_SRCRT_TYPE_4) + return NULL; + + len = (srh->hdrlen + 1) << 3; + + if (!pskb_may_pull(skb, srhoff + len)) + return NULL; + + if (!seg6_validate_srh(srh, len)) + return NULL; + + return srh; +} + +static struct ipv6_sr_hdr *get_and_validate_srh(struct sk_buff *skb) +{ + struct ipv6_sr_hdr *srh; + + srh = get_srh(skb); + if (!srh) + return NULL; + + if (srh->segments_left == 0) + return NULL; + +#ifdef CONFIG_IPV6_SEG6_HMAC + if (!seg6_hmac_validate_skb(skb)) + return NULL; +#endif + + return srh; +} + +static bool decap_and_validate(struct sk_buff *skb, int proto) +{ + struct ipv6_sr_hdr *srh; + unsigned int off = 0; + + srh = get_srh(skb); + if (srh && srh->segments_left > 0) + return false; + +#ifdef CONFIG_IPV6_SEG6_HMAC + if (srh && !seg6_hmac_validate_skb(skb)) + return false; +#endif + + if (ipv6_find_hdr(skb, &off, proto, NULL, NULL) < 0) + return false; + + if (!pskb_pull(skb, off)) + return false; + + skb_postpull_rcsum(skb, skb_network_header(skb), off); + + skb_reset_network_header(skb); + skb_reset_transport_header(skb); + skb->encapsulation = 0; + + return true; +} + +static void advance_nextseg(struct ipv6_sr_hdr *srh, struct in6_addr *daddr) +{ + struct in6_addr *addr; + + srh->segments_left--; + addr = srh->segments + srh->segments_left; + *daddr = *addr; +} + +static void lookup_nexthop(struct sk_buff *skb, struct in6_addr *nhaddr, + u32 tbl_id) +{ + struct net *net = dev_net(skb->dev); + struct ipv6hdr *hdr = ipv6_hdr(skb); + int flags = RT6_LOOKUP_F_HAS_SADDR; + struct dst_entry *dst = NULL; + struct rt6_info *rt; + struct flowi6 fl6; + + fl6.flowi6_iif = skb->dev->ifindex; + fl6.daddr = nhaddr ? *nhaddr : hdr->daddr; + fl6.saddr = hdr->saddr; + fl6.flowlabel = ip6_flowinfo(hdr); + fl6.flowi6_mark = skb->mark; + fl6.flowi6_proto = hdr->nexthdr; + + if (nhaddr) + fl6.flowi6_flags = FLOWI_FLAG_KNOWN_NH; + + if (!tbl_id) { + dst = ip6_route_input_lookup(net, skb->dev, &fl6, flags); + } else { + struct fib6_table *table; + + table = fib6_get_table(net, tbl_id); + if (!table) + goto out; + + rt = ip6_pol_route(net, table, 0, &fl6, flags); + dst = &rt->dst; + } + + if (dst && dst->dev->flags & IFF_LOOPBACK && !dst->error) { + dst_release(dst); + dst = NULL; + } + +out: + if (!dst) { + rt = net->ipv6.ip6_blk_hole_entry; + dst = &rt->dst; + dst_hold(dst); + } + + skb_dst_drop(skb); + skb_dst_set(skb, dst); +} + +/* regular endpoint function */ +static int input_action_end(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + + srh = get_and_validate_srh(skb); + if (!srh) + goto drop; + + advance_nextseg(srh, &ipv6_hdr(skb)->daddr); + + lookup_nexthop(skb, NULL, 0); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +/* regular endpoint, and forward to specified nexthop */ +static int input_action_end_x(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + + srh = get_and_validate_srh(skb); + if (!srh) + goto drop; + + advance_nextseg(srh, &ipv6_hdr(skb)->daddr); + + lookup_nexthop(skb, &slwt->nh6, 0); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +static int input_action_end_t(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + + srh = get_and_validate_srh(skb); + if (!srh) + goto drop; + + advance_nextseg(srh, &ipv6_hdr(skb)->daddr); + + lookup_nexthop(skb, NULL, slwt->table); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +/* decapsulate and forward inner L2 frame on specified interface */ +static int input_action_end_dx2(struct sk_buff *skb, + struct seg6_local_lwt *slwt) +{ + struct net *net = dev_net(skb->dev); + struct net_device *odev; + struct ethhdr *eth; + + if (!decap_and_validate(skb, NEXTHDR_NONE)) + goto drop; + + if (!pskb_may_pull(skb, ETH_HLEN)) + goto drop; + + skb_reset_mac_header(skb); + eth = (struct ethhdr *)skb->data; + + /* To determine the frame's protocol, we assume it is 802.3. This avoids + * a call to eth_type_trans(), which is not really relevant for our + * use case. + */ + if (!eth_proto_is_802_3(eth->h_proto)) + goto drop; + + odev = dev_get_by_index_rcu(net, slwt->oif); + if (!odev) + goto drop; + + /* As we accept Ethernet frames, make sure the egress device is of + * the correct type. + */ + if (odev->type != ARPHRD_ETHER) + goto drop; + + if (!(odev->flags & IFF_UP) || !netif_carrier_ok(odev)) + goto drop; + + skb_orphan(skb); + + if (skb_warn_if_lro(skb)) + goto drop; + + skb_forward_csum(skb); + + if (skb->len - ETH_HLEN > odev->mtu) + goto drop; + + skb->dev = odev; + skb->protocol = eth->h_proto; + + return dev_queue_xmit(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +/* decapsulate and forward to specified nexthop */ +static int input_action_end_dx6(struct sk_buff *skb, + struct seg6_local_lwt *slwt) +{ + struct in6_addr *nhaddr = NULL; + + /* this function accepts IPv6 encapsulated packets, with either + * an SRH with SL=0, or no SRH. + */ + + if (!decap_and_validate(skb, IPPROTO_IPV6)) + goto drop; + + if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) + goto drop; + + /* The inner packet is not associated to any local interface, + * so we do not call netif_rx(). + * + * If slwt->nh6 is set to ::, then lookup the nexthop for the + * inner packet's DA. Otherwise, use the specified nexthop. + */ + + if (!ipv6_addr_any(&slwt->nh6)) + nhaddr = &slwt->nh6; + + lookup_nexthop(skb, nhaddr, 0); + + return dst_input(skb); +drop: + kfree_skb(skb); + return -EINVAL; +} + +static int input_action_end_dx4(struct sk_buff *skb, + struct seg6_local_lwt *slwt) +{ + struct iphdr *iph; + __be32 nhaddr; + int err; + + if (!decap_and_validate(skb, IPPROTO_IPIP)) + goto drop; + + if (!pskb_may_pull(skb, sizeof(struct iphdr))) + goto drop; + + skb->protocol = htons(ETH_P_IP); + + iph = ip_hdr(skb); + + nhaddr = slwt->nh4.s_addr ?: iph->daddr; + + skb_dst_drop(skb); + + err = ip_route_input(skb, nhaddr, iph->saddr, 0, skb->dev); + if (err) + goto drop; + + return dst_input(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +static int input_action_end_dt6(struct sk_buff *skb, + struct seg6_local_lwt *slwt) +{ + if (!decap_and_validate(skb, IPPROTO_IPV6)) + goto drop; + + if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) + goto drop; + + lookup_nexthop(skb, NULL, slwt->table); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return -EINVAL; +} + +/* push an SRH on top of the current one */ +static int input_action_end_b6(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + int err = -EINVAL; + + srh = get_and_validate_srh(skb); + if (!srh) + goto drop; + + err = seg6_do_srh_inline(skb, slwt->srh); + if (err) + goto drop; + + ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); + skb_set_transport_header(skb, sizeof(struct ipv6hdr)); + + lookup_nexthop(skb, NULL, 0); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return err; +} + +/* encapsulate within an outer IPv6 header and a specified SRH */ +static int input_action_end_b6_encap(struct sk_buff *skb, + struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + int err = -EINVAL; + + srh = get_and_validate_srh(skb); + if (!srh) + goto drop; + + advance_nextseg(srh, &ipv6_hdr(skb)->daddr); + + skb_reset_inner_headers(skb); + skb->encapsulation = 1; + + err = seg6_do_srh_encap(skb, slwt->srh, IPPROTO_IPV6); + if (err) + goto drop; + + ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); + skb_set_transport_header(skb, sizeof(struct ipv6hdr)); + + lookup_nexthop(skb, NULL, 0); + + return dst_input(skb); + +drop: + kfree_skb(skb); + return err; +} + +static struct seg6_action_desc seg6_action_table[] = { + { + .action = SEG6_LOCAL_ACTION_END, + .attrs = 0, + .input = input_action_end, + }, + { + .action = SEG6_LOCAL_ACTION_END_X, + .attrs = (1 << SEG6_LOCAL_NH6), + .input = input_action_end_x, + }, + { + .action = SEG6_LOCAL_ACTION_END_T, + .attrs = (1 << SEG6_LOCAL_TABLE), + .input = input_action_end_t, + }, + { + .action = SEG6_LOCAL_ACTION_END_DX2, + .attrs = (1 << SEG6_LOCAL_OIF), + .input = input_action_end_dx2, + }, + { + .action = SEG6_LOCAL_ACTION_END_DX6, + .attrs = (1 << SEG6_LOCAL_NH6), + .input = input_action_end_dx6, + }, + { + .action = SEG6_LOCAL_ACTION_END_DX4, + .attrs = (1 << SEG6_LOCAL_NH4), + .input = input_action_end_dx4, + }, + { + .action = SEG6_LOCAL_ACTION_END_DT6, + .attrs = (1 << SEG6_LOCAL_TABLE), + .input = input_action_end_dt6, + }, + { + .action = SEG6_LOCAL_ACTION_END_B6, + .attrs = (1 << SEG6_LOCAL_SRH), + .input = input_action_end_b6, + }, + { + .action = SEG6_LOCAL_ACTION_END_B6_ENCAP, + .attrs = (1 << SEG6_LOCAL_SRH), + .input = input_action_end_b6_encap, + .static_headroom = sizeof(struct ipv6hdr), + } +}; + +static struct seg6_action_desc *__get_action_desc(int action) +{ + struct seg6_action_desc *desc; + int i, count; + + count = sizeof(seg6_action_table) / sizeof(struct seg6_action_desc); + for (i = 0; i < count; i++) { + desc = &seg6_action_table[i]; + if (desc->action == action) + return desc; + } + + return NULL; +} + +static int seg6_local_input(struct sk_buff *skb) +{ + struct dst_entry *orig_dst = skb_dst(skb); + struct seg6_action_desc *desc; + struct seg6_local_lwt *slwt; + + if (skb->protocol != htons(ETH_P_IPV6)) { + kfree_skb(skb); + return -EINVAL; + } + + slwt = seg6_local_lwtunnel(orig_dst->lwtstate); + desc = slwt->desc; + + return desc->input(skb, slwt); +} + +static const struct nla_policy seg6_local_policy[SEG6_LOCAL_MAX + 1] = { + [SEG6_LOCAL_ACTION] = { .type = NLA_U32 }, + [SEG6_LOCAL_SRH] = { .type = NLA_BINARY }, + [SEG6_LOCAL_TABLE] = { .type = NLA_U32 }, + [SEG6_LOCAL_NH4] = { .type = NLA_BINARY, + .len = sizeof(struct in_addr) }, + [SEG6_LOCAL_NH6] = { .type = NLA_BINARY, + .len = sizeof(struct in6_addr) }, + [SEG6_LOCAL_IIF] = { .type = NLA_U32 }, + [SEG6_LOCAL_OIF] = { .type = NLA_U32 }, +}; + +static int parse_nla_srh(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + int len; + + srh = nla_data(attrs[SEG6_LOCAL_SRH]); + len = nla_len(attrs[SEG6_LOCAL_SRH]); + + /* SRH must contain at least one segment */ + if (len < sizeof(*srh) + sizeof(struct in6_addr)) + return -EINVAL; + + if (!seg6_validate_srh(srh, len)) + return -EINVAL; + + slwt->srh = kmalloc(len, GFP_KERNEL); + if (!slwt->srh) + return -ENOMEM; + + memcpy(slwt->srh, srh, len); + + slwt->headroom += len; + + return 0; +} + +static int put_nla_srh(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct ipv6_sr_hdr *srh; + struct nlattr *nla; + int len; + + srh = slwt->srh; + len = (srh->hdrlen + 1) << 3; + + nla = nla_reserve(skb, SEG6_LOCAL_SRH, len); + if (!nla) + return -EMSGSIZE; + + memcpy(nla_data(nla), srh, len); + + return 0; +} + +static int cmp_nla_srh(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + int len = (a->srh->hdrlen + 1) << 3; + + if (len != ((b->srh->hdrlen + 1) << 3)) + return 1; + + return memcmp(a->srh, b->srh, len); +} + +static int parse_nla_table(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + slwt->table = nla_get_u32(attrs[SEG6_LOCAL_TABLE]); + + return 0; +} + +static int put_nla_table(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + if (nla_put_u32(skb, SEG6_LOCAL_TABLE, slwt->table)) + return -EMSGSIZE; + + return 0; +} + +static int cmp_nla_table(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + if (a->table != b->table) + return 1; + + return 0; +} + +static int parse_nla_nh4(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + memcpy(&slwt->nh4, nla_data(attrs[SEG6_LOCAL_NH4]), + sizeof(struct in_addr)); + + return 0; +} + +static int put_nla_nh4(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct nlattr *nla; + + nla = nla_reserve(skb, SEG6_LOCAL_NH4, sizeof(struct in_addr)); + if (!nla) + return -EMSGSIZE; + + memcpy(nla_data(nla), &slwt->nh4, sizeof(struct in_addr)); + + return 0; +} + +static int cmp_nla_nh4(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + return memcmp(&a->nh4, &b->nh4, sizeof(struct in_addr)); +} + +static int parse_nla_nh6(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + memcpy(&slwt->nh6, nla_data(attrs[SEG6_LOCAL_NH6]), + sizeof(struct in6_addr)); + + return 0; +} + +static int put_nla_nh6(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + struct nlattr *nla; + + nla = nla_reserve(skb, SEG6_LOCAL_NH6, sizeof(struct in6_addr)); + if (!nla) + return -EMSGSIZE; + + memcpy(nla_data(nla), &slwt->nh6, sizeof(struct in6_addr)); + + return 0; +} + +static int cmp_nla_nh6(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + return memcmp(&a->nh6, &b->nh6, sizeof(struct in6_addr)); +} + +static int parse_nla_iif(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + slwt->iif = nla_get_u32(attrs[SEG6_LOCAL_IIF]); + + return 0; +} + +static int put_nla_iif(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + if (nla_put_u32(skb, SEG6_LOCAL_IIF, slwt->iif)) + return -EMSGSIZE; + + return 0; +} + +static int cmp_nla_iif(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + if (a->iif != b->iif) + return 1; + + return 0; +} + +static int parse_nla_oif(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + slwt->oif = nla_get_u32(attrs[SEG6_LOCAL_OIF]); + + return 0; +} + +static int put_nla_oif(struct sk_buff *skb, struct seg6_local_lwt *slwt) +{ + if (nla_put_u32(skb, SEG6_LOCAL_OIF, slwt->oif)) + return -EMSGSIZE; + + return 0; +} + +static int cmp_nla_oif(struct seg6_local_lwt *a, struct seg6_local_lwt *b) +{ + if (a->oif != b->oif) + return 1; + + return 0; +} + +struct seg6_action_param { + int (*parse)(struct nlattr **attrs, struct seg6_local_lwt *slwt); + int (*put)(struct sk_buff *skb, struct seg6_local_lwt *slwt); + int (*cmp)(struct seg6_local_lwt *a, struct seg6_local_lwt *b); +}; + +static struct seg6_action_param seg6_action_params[SEG6_LOCAL_MAX + 1] = { + [SEG6_LOCAL_SRH] = { .parse = parse_nla_srh, + .put = put_nla_srh, + .cmp = cmp_nla_srh }, + + [SEG6_LOCAL_TABLE] = { .parse = parse_nla_table, + .put = put_nla_table, + .cmp = cmp_nla_table }, + + [SEG6_LOCAL_NH4] = { .parse = parse_nla_nh4, + .put = put_nla_nh4, + .cmp = cmp_nla_nh4 }, + + [SEG6_LOCAL_NH6] = { .parse = parse_nla_nh6, + .put = put_nla_nh6, + .cmp = cmp_nla_nh6 }, + + [SEG6_LOCAL_IIF] = { .parse = parse_nla_iif, + .put = put_nla_iif, + .cmp = cmp_nla_iif }, + + [SEG6_LOCAL_OIF] = { .parse = parse_nla_oif, + .put = put_nla_oif, + .cmp = cmp_nla_oif }, +}; + +static int parse_nla_action(struct nlattr **attrs, struct seg6_local_lwt *slwt) +{ + struct seg6_action_param *param; + struct seg6_action_desc *desc; + int i, err; + + desc = __get_action_desc(slwt->action); + if (!desc) + return -EINVAL; + + if (!desc->input) + return -EOPNOTSUPP; + + slwt->desc = desc; + slwt->headroom += desc->static_headroom; + + for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { + if (desc->attrs & (1 << i)) { + if (!attrs[i]) + return -EINVAL; + + param = &seg6_action_params[i]; + + err = param->parse(attrs, slwt); + if (err < 0) + return err; + } + } + + return 0; +} + +static int seg6_local_build_state(struct nlattr *nla, unsigned int family, + const void *cfg, struct lwtunnel_state **ts, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[SEG6_LOCAL_MAX + 1]; + struct lwtunnel_state *newts; + struct seg6_local_lwt *slwt; + int err; + + if (family != AF_INET6) + return -EINVAL; + + err = nla_parse_nested(tb, SEG6_LOCAL_MAX, nla, seg6_local_policy, + extack); + + if (err < 0) + return err; + + if (!tb[SEG6_LOCAL_ACTION]) + return -EINVAL; + + newts = lwtunnel_state_alloc(sizeof(*slwt)); + if (!newts) + return -ENOMEM; + + slwt = seg6_local_lwtunnel(newts); + slwt->action = nla_get_u32(tb[SEG6_LOCAL_ACTION]); + + err = parse_nla_action(tb, slwt); + if (err < 0) + goto out_free; + + newts->type = LWTUNNEL_ENCAP_SEG6_LOCAL; + newts->flags = LWTUNNEL_STATE_INPUT_REDIRECT; + newts->headroom = slwt->headroom; + + *ts = newts; + + return 0; + +out_free: + kfree(slwt->srh); + kfree(newts); + return err; +} + +static void seg6_local_destroy_state(struct lwtunnel_state *lwt) +{ + struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); + + kfree(slwt->srh); +} + +static int seg6_local_fill_encap(struct sk_buff *skb, + struct lwtunnel_state *lwt) +{ + struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); + struct seg6_action_param *param; + int i, err; + + if (nla_put_u32(skb, SEG6_LOCAL_ACTION, slwt->action)) + return -EMSGSIZE; + + for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { + if (slwt->desc->attrs & (1 << i)) { + param = &seg6_action_params[i]; + err = param->put(skb, slwt); + if (err < 0) + return err; + } + } + + return 0; +} + +static int seg6_local_get_encap_size(struct lwtunnel_state *lwt) +{ + struct seg6_local_lwt *slwt = seg6_local_lwtunnel(lwt); + unsigned long attrs; + int nlsize; + + nlsize = nla_total_size(4); /* action */ + + attrs = slwt->desc->attrs; + + if (attrs & (1 << SEG6_LOCAL_SRH)) + nlsize += nla_total_size((slwt->srh->hdrlen + 1) << 3); + + if (attrs & (1 << SEG6_LOCAL_TABLE)) + nlsize += nla_total_size(4); + + if (attrs & (1 << SEG6_LOCAL_NH4)) + nlsize += nla_total_size(4); + + if (attrs & (1 << SEG6_LOCAL_NH6)) + nlsize += nla_total_size(16); + + if (attrs & (1 << SEG6_LOCAL_IIF)) + nlsize += nla_total_size(4); + + if (attrs & (1 << SEG6_LOCAL_OIF)) + nlsize += nla_total_size(4); + + return nlsize; +} + +static int seg6_local_cmp_encap(struct lwtunnel_state *a, + struct lwtunnel_state *b) +{ + struct seg6_local_lwt *slwt_a, *slwt_b; + struct seg6_action_param *param; + int i; + + slwt_a = seg6_local_lwtunnel(a); + slwt_b = seg6_local_lwtunnel(b); + + if (slwt_a->action != slwt_b->action) + return 1; + + if (slwt_a->desc->attrs != slwt_b->desc->attrs) + return 1; + + for (i = 0; i < SEG6_LOCAL_MAX + 1; i++) { + if (slwt_a->desc->attrs & (1 << i)) { + param = &seg6_action_params[i]; + if (param->cmp(slwt_a, slwt_b)) + return 1; + } + } + + return 0; +} + +static const struct lwtunnel_encap_ops seg6_local_ops = { + .build_state = seg6_local_build_state, + .destroy_state = seg6_local_destroy_state, + .input = seg6_local_input, + .fill_encap = seg6_local_fill_encap, + .get_encap_size = seg6_local_get_encap_size, + .cmp_encap = seg6_local_cmp_encap, + .owner = THIS_MODULE, +}; + +int __init seg6_local_init(void) +{ + return lwtunnel_encap_add_ops(&seg6_local_ops, + LWTUNNEL_ENCAP_SEG6_LOCAL); +} + +void seg6_local_exit(void) +{ + lwtunnel_encap_del_ops(&seg6_local_ops, LWTUNNEL_ENCAP_SEG6_LOCAL); +} diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 7b75b0620730..4e7817abc0b9 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -216,6 +216,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) treq->rcv_isn = ntohl(th->seq) - 1; treq->snt_isn = cookie; treq->ts_off = 0; + treq->txhash = net_tx_rndhash(); /* * We need to lookup the dst_entry to get the correct window size. diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 69c50e737c54..6fbf8ae5e52c 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -90,6 +90,13 @@ static struct ctl_table ipv6_table_template[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "flowlabel_reflect", + .data = &init_net.ipv6.sysctl.flowlabel_reflect, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; @@ -149,6 +156,7 @@ static int __net_init ipv6_sysctl_net_init(struct net *net) ipv6_table[6].data = &net->ipv6.sysctl.idgen_delay; ipv6_table[7].data = &net->ipv6.sysctl.flowlabel_state_ranges; ipv6_table[8].data = &net->ipv6.sysctl.ip_nonlocal_bind; + ipv6_table[9].data = &net->ipv6.sysctl.flowlabel_reflect; ipv6_route_table = ipv6_route_sysctl_init(net); if (!ipv6_route_table) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2521690d62d6..38f76d8b231e 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -350,7 +350,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, sk = __inet6_lookup_established(net, &tcp_hashinfo, &hdr->daddr, th->dest, &hdr->saddr, ntohs(th->source), - skb->dev->ifindex); + skb->dev->ifindex, inet6_sdif(skb)); if (!sk) { __ICMP6_INC_STATS(net, __in6_dev_get(skb->dev), @@ -918,7 +918,8 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb) &tcp_hashinfo, NULL, 0, &ipv6h->saddr, th->source, &ipv6h->daddr, - ntohs(th->source), tcp_v6_iif(skb)); + ntohs(th->source), tcp_v6_iif(skb), + tcp_v6_sdif(skb)); if (!sk1) goto out; @@ -1296,7 +1297,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) } } - tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len); + tcp_rcv_established(sk, skb, tcp_hdr(skb)); if (opt_skb) goto ipv6_pktoptions; return 0; @@ -1393,10 +1394,13 @@ static void tcp_v6_fill_cb(struct sk_buff *skb, const struct ipv6hdr *hdr, TCP_SKB_CB(skb)->tcp_tw_isn = 0; TCP_SKB_CB(skb)->ip_dsfield = ipv6_get_dsfield(hdr); TCP_SKB_CB(skb)->sacked = 0; + TCP_SKB_CB(skb)->has_rxtstamp = + skb->tstamp || skb_hwtstamps(skb)->hwtstamp; } static int tcp_v6_rcv(struct sk_buff *skb) { + int sdif = inet6_sdif(skb); const struct tcphdr *th; const struct ipv6hdr *hdr; bool refcounted; @@ -1430,7 +1434,7 @@ static int tcp_v6_rcv(struct sk_buff *skb) lookup: sk = __inet6_lookup_skb(&tcp_hashinfo, skb, __tcp_hdrlen(th), - th->source, th->dest, inet6_iif(skb), + th->source, th->dest, inet6_iif(skb), sdif, &refcounted); if (!sk) goto no_tcp_socket; @@ -1456,6 +1460,8 @@ process: } sock_hold(sk); refcounted = true; + if (tcp_filter(sk, skb)) + goto discard_and_relse; nsk = tcp_check_req(sk, skb, req, false); if (!nsk) { reqsk_put(req); @@ -1464,8 +1470,6 @@ process: if (nsk == sk) { reqsk_put(req); tcp_v6_restore_cb(skb); - } else if (tcp_filter(sk, skb)) { - goto discard_and_relse; } else if (tcp_child_process(sk, nsk, skb)) { tcp_v6_send_reset(nsk, skb); goto discard_and_relse; @@ -1505,8 +1509,7 @@ process: tcp_segs_in(tcp_sk(sk), skb); ret = 0; if (!sock_owned_by_user(sk)) { - if (!tcp_prequeue(sk, skb)) - ret = tcp_v6_do_rcv(sk, skb); + ret = tcp_v6_do_rcv(sk, skb); } else if (tcp_add_backlog(sk, skb)) { goto discard_and_relse; } @@ -1564,7 +1567,8 @@ do_time_wait: skb, __tcp_hdrlen(th), &ipv6_hdr(skb)->saddr, th->source, &ipv6_hdr(skb)->daddr, - ntohs(th->dest), tcp_v6_iif(skb)); + ntohs(th->dest), tcp_v6_iif(skb), + sdif); if (sk2) { struct inet_timewait_sock *tw = inet_twsk(sk); inet_twsk_deschedule_put(tw); @@ -1611,7 +1615,7 @@ static void tcp_v6_early_demux(struct sk_buff *skb) sk = __inet6_lookup_established(dev_net(skb->dev), &tcp_hashinfo, &hdr->saddr, th->source, &hdr->daddr, ntohs(th->dest), - inet6_iif(skb)); + inet6_iif(skb), inet6_sdif(skb)); if (sk) { skb->sk = sk; skb->destructor = sock_edemux; @@ -1945,6 +1949,9 @@ struct proto tcpv6_prot = { .diag_destroy = tcp_abort, }; +/* thinking of making this const? Don't. + * early_demux can change based on sysctl. + */ static struct inet6_protocol tcpv6_protocol = { .early_demux = tcp_v6_early_demux, .early_demux_handler = tcp_v6_early_demux, diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 4a3e65626e8b..e2ecfb137297 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -129,7 +129,7 @@ static void udp_v6_rehash(struct sock *sk) static int compute_score(struct sock *sk, struct net *net, const struct in6_addr *saddr, __be16 sport, const struct in6_addr *daddr, unsigned short hnum, - int dif, bool exact_dif) + int dif, int sdif, bool exact_dif) { int score; struct inet_sock *inet; @@ -161,9 +161,13 @@ static int compute_score(struct sock *sk, struct net *net, } if (sk->sk_bound_dev_if || exact_dif) { - if (sk->sk_bound_dev_if != dif) + bool dev_match = (sk->sk_bound_dev_if == dif || + sk->sk_bound_dev_if == sdif); + + if (exact_dif && !dev_match) return -1; - score++; + if (sk->sk_bound_dev_if && dev_match) + score++; } if (sk->sk_incoming_cpu == raw_smp_processor_id()) @@ -175,9 +179,9 @@ static int compute_score(struct sock *sk, struct net *net, /* called with rcu_read_lock() */ static struct sock *udp6_lib_lookup2(struct net *net, const struct in6_addr *saddr, __be16 sport, - const struct in6_addr *daddr, unsigned int hnum, int dif, - bool exact_dif, struct udp_hslot *hslot2, - struct sk_buff *skb) + const struct in6_addr *daddr, unsigned int hnum, + int dif, int sdif, bool exact_dif, + struct udp_hslot *hslot2, struct sk_buff *skb) { struct sock *sk, *result; int score, badness, matches = 0, reuseport = 0; @@ -187,7 +191,7 @@ static struct sock *udp6_lib_lookup2(struct net *net, badness = -1; udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { score = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, exact_dif); + daddr, hnum, dif, sdif, exact_dif); if (score > badness) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -214,10 +218,10 @@ static struct sock *udp6_lib_lookup2(struct net *net, /* rcu_read_lock() must be held */ struct sock *__udp6_lib_lookup(struct net *net, - const struct in6_addr *saddr, __be16 sport, - const struct in6_addr *daddr, __be16 dport, - int dif, struct udp_table *udptable, - struct sk_buff *skb) + const struct in6_addr *saddr, __be16 sport, + const struct in6_addr *daddr, __be16 dport, + int dif, int sdif, struct udp_table *udptable, + struct sk_buff *skb) { struct sock *sk, *result; unsigned short hnum = ntohs(dport); @@ -235,7 +239,7 @@ struct sock *__udp6_lib_lookup(struct net *net, goto begin; result = udp6_lib_lookup2(net, saddr, sport, - daddr, hnum, dif, exact_dif, + daddr, hnum, dif, sdif, exact_dif, hslot2, skb); if (!result) { unsigned int old_slot2 = slot2; @@ -250,7 +254,7 @@ struct sock *__udp6_lib_lookup(struct net *net, goto begin; result = udp6_lib_lookup2(net, saddr, sport, - daddr, hnum, dif, + daddr, hnum, dif, sdif, exact_dif, hslot2, skb); } @@ -261,7 +265,7 @@ begin: badness = -1; sk_for_each_rcu(sk, &hslot->head) { score = compute_score(sk, net, saddr, sport, daddr, hnum, dif, - exact_dif); + sdif, exact_dif); if (score > badness) { reuseport = sk->sk_reuseport; if (reuseport) { @@ -291,14 +295,10 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, struct udp_table *udptable) { const struct ipv6hdr *iph = ipv6_hdr(skb); - struct sock *sk; - sk = skb_steal_sock(skb); - if (unlikely(sk)) - return sk; return __udp6_lib_lookup(dev_net(skb->dev), &iph->saddr, sport, &iph->daddr, dport, inet6_iif(skb), - udptable, skb); + inet6_sdif(skb), udptable, skb); } struct sock *udp6_lib_lookup_skb(struct sk_buff *skb, @@ -308,7 +308,7 @@ struct sock *udp6_lib_lookup_skb(struct sk_buff *skb, return __udp6_lib_lookup(dev_net(skb->dev), &iph->saddr, sport, &iph->daddr, dport, inet6_iif(skb), - &udp_table, skb); + inet6_sdif(skb), &udp_table, skb); } EXPORT_SYMBOL_GPL(udp6_lib_lookup_skb); @@ -324,7 +324,7 @@ struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be struct sock *sk; sk = __udp6_lib_lookup(net, saddr, sport, daddr, dport, - dif, &udp_table, NULL); + dif, 0, &udp_table, NULL); if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; return sk; @@ -332,6 +332,15 @@ struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be EXPORT_SYMBOL_GPL(udp6_lib_lookup); #endif +/* do not use the scratch area len for jumbogram: their length execeeds the + * scratch area space; note that the IP6CB flags is still in the first + * cacheline, so checking for jumbograms is cheap + */ +static int udp6_skb_len(struct sk_buff *skb) +{ + return unlikely(inet6_is_jumbogram(skb)) ? skb->len : udp_skb_len(skb); +} + /* * This should be easy, if there is something there we * return it, otherwise we block. @@ -357,12 +366,13 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, return ipv6_recv_rxpmtu(sk, msg, len, addr_len); try_again: - peeking = off = sk_peek_offset(sk, flags); + peeking = flags & MSG_PEEK; + off = sk_peek_offset(sk, flags); skb = __skb_recv_udp(sk, flags, noblock, &peeked, &off, &err); if (!skb) return err; - ulen = udp_skb_len(skb); + ulen = udp6_skb_len(skb); copied = len; if (copied > ulen - off) copied = ulen - off; @@ -496,7 +506,7 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct net *net = dev_net(skb->dev); sk = __udp6_lib_lookup(net, daddr, uh->dest, saddr, uh->source, - inet6_iif(skb), udptable, skb); + inet6_iif(skb), 0, udptable, skb); if (!sk) { __ICMP6_INC_STATS(net, __in6_dev_get(skb->dev), ICMP6_MIB_INERRORS); @@ -569,8 +579,7 @@ static __inline__ void udpv6_err(struct sk_buff *skb, static struct static_key udpv6_encap_needed __read_mostly; void udpv6_encap_enable(void) { - if (!static_key_enabled(&udpv6_encap_needed)) - static_key_slow_inc(&udpv6_encap_needed); + static_key_enable(&udpv6_encap_needed); } EXPORT_SYMBOL(udpv6_encap_enable); @@ -762,6 +771,15 @@ start_lookup: return 0; } +static void udp6_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) +{ + if (udp_sk_rx_dst_set(sk, dst)) { + const struct rt6_info *rt = (const struct rt6_info *)dst; + + inet6_sk(sk)->rx_dst_cookie = rt6_get_cookie(rt); + } +} + int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, int proto) { @@ -804,6 +822,24 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (udp6_csum_init(skb, uh, proto)) goto csum_error; + /* Check if the socket is already available, e.g. due to early demux */ + sk = skb_steal_sock(skb); + if (sk) { + struct dst_entry *dst = skb_dst(skb); + int ret; + + if (unlikely(sk->sk_rx_dst != dst)) + udp6_sk_rx_dst_set(sk, dst); + + ret = udpv6_queue_rcv_skb(sk, skb); + sock_put(sk); + + /* a return value > 0 means to resubmit the input */ + if (ret > 0) + return ret; + return 0; + } + /* * Multicast receive code */ @@ -812,11 +848,6 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, saddr, daddr, udptable, proto); /* Unicast */ - - /* - * check socket cache ... must talk to Alan about his plans - * for sock caches... i'll skip this for now. - */ sk = __udp6_lib_lookup_skb(skb, uh->source, uh->dest, udptable); if (sk) { int ret; @@ -875,7 +906,7 @@ discard: static struct sock *__udp6_lib_demux_lookup(struct net *net, __be16 loc_port, const struct in6_addr *loc_addr, __be16 rmt_port, const struct in6_addr *rmt_addr, - int dif) + int dif, int sdif) { unsigned short hnum = ntohs(loc_port); unsigned int hash2 = udp6_portaddr_hash(net, loc_addr, hnum); @@ -886,7 +917,7 @@ static struct sock *__udp6_lib_demux_lookup(struct net *net, udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { if (sk->sk_state == TCP_ESTABLISHED && - INET6_MATCH(sk, net, rmt_addr, loc_addr, ports, dif)) + INET6_MATCH(sk, net, rmt_addr, loc_addr, ports, dif, sdif)) return sk; /* Only check first socket in chain */ break; @@ -901,6 +932,7 @@ static void udp_v6_early_demux(struct sk_buff *skb) struct sock *sk; struct dst_entry *dst; int dif = skb->dev->ifindex; + int sdif = inet6_sdif(skb); if (!pskb_may_pull(skb, skb_transport_offset(skb) + sizeof(struct udphdr))) @@ -912,7 +944,7 @@ static void udp_v6_early_demux(struct sk_buff *skb) sk = __udp6_lib_demux_lookup(net, uh->dest, &ipv6_hdr(skb)->daddr, uh->source, &ipv6_hdr(skb)->saddr, - dif); + dif, sdif); else return; @@ -1448,6 +1480,9 @@ int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, } #endif +/* thinking of making this const? Don't. + * early_demux can change based on sysctl. + */ static struct inet6_protocol udpv6_protocol = { .early_demux = udp_v6_early_demux, .early_demux_handler = udp_v6_early_demux, diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c index a2267f80febb..455fd4e39333 100644 --- a/net/ipv6/udp_offload.c +++ b/net/ipv6/udp_offload.c @@ -17,109 +17,15 @@ #include <net/ip6_checksum.h> #include "ip6_offload.h" -static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, - netdev_features_t features) +static struct sk_buff *udp6_tunnel_segment(struct sk_buff *skb, + netdev_features_t features) { struct sk_buff *segs = ERR_PTR(-EINVAL); - unsigned int mss; - unsigned int unfrag_ip6hlen, unfrag_len; - struct frag_hdr *fptr; - u8 *packet_start, *prevhdr; - u8 nexthdr; - u8 frag_hdr_sz = sizeof(struct frag_hdr); - __wsum csum; - int tnl_hlen; - int err; - - mss = skb_shinfo(skb)->gso_size; - if (unlikely(skb->len <= mss)) - goto out; - - if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { - /* Packet is from an untrusted source, reset gso_segs. */ - - skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss); - - /* Set the IPv6 fragment id if not set yet */ - if (!skb_shinfo(skb)->ip6_frag_id) - ipv6_proxy_select_ident(dev_net(skb->dev), skb); - - segs = NULL; - goto out; - } if (skb->encapsulation && skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL|SKB_GSO_UDP_TUNNEL_CSUM)) segs = skb_udp_tunnel_segment(skb, features, true); - else { - const struct ipv6hdr *ipv6h; - struct udphdr *uh; - - if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto out; - - /* Do software UFO. Complete and fill in the UDP checksum as HW cannot - * do checksum of UDP packets sent as multiple IP fragments. - */ - - uh = udp_hdr(skb); - ipv6h = ipv6_hdr(skb); - - uh->check = 0; - csum = skb_checksum(skb, 0, skb->len, 0); - uh->check = udp_v6_check(skb->len, &ipv6h->saddr, - &ipv6h->daddr, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - - skb->ip_summed = CHECKSUM_NONE; - - /* If there is no outer header we can fake a checksum offload - * due to the fact that we have already done the checksum in - * software prior to segmenting the frame. - */ - if (!skb->encap_hdr_csum) - features |= NETIF_F_HW_CSUM; - - /* Check if there is enough headroom to insert fragment header. */ - tnl_hlen = skb_tnl_header_len(skb); - if (skb->mac_header < (tnl_hlen + frag_hdr_sz)) { - if (gso_pskb_expand_head(skb, tnl_hlen + frag_hdr_sz)) - goto out; - } - - /* Find the unfragmentable header and shift it left by frag_hdr_sz - * bytes to insert fragment header. - */ - err = ip6_find_1stfragopt(skb, &prevhdr); - if (err < 0) - return ERR_PTR(err); - unfrag_ip6hlen = err; - nexthdr = *prevhdr; - *prevhdr = NEXTHDR_FRAGMENT; - unfrag_len = (skb_network_header(skb) - skb_mac_header(skb)) + - unfrag_ip6hlen + tnl_hlen; - packet_start = (u8 *) skb->head + SKB_GSO_CB(skb)->mac_offset; - memmove(packet_start-frag_hdr_sz, packet_start, unfrag_len); - - SKB_GSO_CB(skb)->mac_offset -= frag_hdr_sz; - skb->mac_header -= frag_hdr_sz; - skb->network_header -= frag_hdr_sz; - - fptr = (struct frag_hdr *)(skb_network_header(skb) + unfrag_ip6hlen); - fptr->nexthdr = nexthdr; - fptr->reserved = 0; - if (!skb_shinfo(skb)->ip6_frag_id) - ipv6_proxy_select_ident(dev_net(skb->dev), skb); - fptr->identification = skb_shinfo(skb)->ip6_frag_id; - - /* Fragment the skb. ipv6 header and the remaining fields of the - * fragment header are updated in ipv6_gso_segment() - */ - segs = skb_segment(skb, features); - } -out: return segs; } @@ -169,7 +75,7 @@ static int udp6_gro_complete(struct sk_buff *skb, int nhoff) static const struct net_offload udpv6_offload = { .callbacks = { - .gso_segment = udp6_ufo_fragment, + .gso_segment = udp6_tunnel_segment, .gro_receive = udp6_gro_receive, .gro_complete = udp6_gro_complete, }, diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index 3ef5d913e7a3..f95943a13abc 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -34,6 +34,7 @@ EXPORT_SYMBOL(xfrm6_rcv_spi); int xfrm6_transport_finish(struct sk_buff *skb, int async) { struct xfrm_offload *xo = xfrm_offload(skb); + int nhlen = skb->data - skb_network_header(skb); skb_network_header(skb)[IP6CB(skb)->nhoff] = XFRM_MODE_SKB_CB(skb)->protocol; @@ -43,8 +44,9 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async) return 1; #endif - __skb_push(skb, skb->data - skb_network_header(skb)); + __skb_push(skb, nhlen); ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); + skb_postpush_rcsum(skb, skb_network_header(skb), nhlen); if (xo && (xo->flags & XFRM_GRO)) { skb_mac_header_rebuild(skb); diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 79651bc71bf0..11d1314ab6c5 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -27,7 +27,8 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif, const xfrm_address_t *saddr, - const xfrm_address_t *daddr) + const xfrm_address_t *daddr, + u32 mark) { struct flowi6 fl6; struct dst_entry *dst; @@ -36,6 +37,7 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif, memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_oif = l3mdev_master_ifindex_by_index(net, oif); fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF; + fl6.flowi6_mark = mark; memcpy(&fl6.daddr, daddr, sizeof(fl6.daddr)); if (saddr) memcpy(&fl6.saddr, saddr, sizeof(fl6.saddr)); @@ -52,12 +54,13 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif, } static int xfrm6_get_saddr(struct net *net, int oif, - xfrm_address_t *saddr, xfrm_address_t *daddr) + xfrm_address_t *saddr, xfrm_address_t *daddr, + u32 mark) { struct dst_entry *dst; struct net_device *dev; - dst = xfrm6_dst_lookup(net, 0, oif, NULL, daddr); + dst = xfrm6_dst_lookup(net, 0, oif, NULL, daddr, mark); if (IS_ERR(dst)) return -EHOSTUNREACH; @@ -214,14 +217,6 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl, int reverse) } } -static inline int xfrm6_garbage_collect(struct dst_ops *ops) -{ - struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops); - - xfrm_garbage_collect_deferred(net); - return dst_entries_get_fast(ops) > ops->gc_thresh * 2; -} - static void xfrm6_update_pmtu(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb, u32 mtu) { @@ -279,14 +274,13 @@ static void xfrm6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, static struct dst_ops xfrm6_dst_ops_template = { .family = AF_INET6, - .gc = xfrm6_garbage_collect, .update_pmtu = xfrm6_update_pmtu, .redirect = xfrm6_redirect, .cow_metrics = dst_cow_metrics_generic, .destroy = xfrm6_dst_destroy, .ifdown = xfrm6_dst_ifdown, .local_out = __ip6_local_out, - .gc_thresh = INT_MAX, + .gc_thresh = 32768, }; static const struct xfrm_policy_afinfo xfrm6_policy_afinfo = { diff --git a/net/irda/Kconfig b/net/irda/Kconfig deleted file mode 100644 index c8671a7ffb3c..000000000000 --- a/net/irda/Kconfig +++ /dev/null @@ -1,96 +0,0 @@ -# -# IrDA protocol configuration -# - -menuconfig IRDA - depends on NET && !S390 - tristate "IrDA (infrared) subsystem support" - select CRC_CCITT - ---help--- - Say Y here if you want to build support for the IrDA (TM) protocols. - The Infrared Data Associations (tm) specifies standards for wireless - infrared communication and is supported by most laptops and PDA's. - - To use Linux support for the IrDA (tm) protocols, you will also need - some user-space utilities like irattach. For more information, see - the file <file:Documentation/networking/irda.txt>. You also want to - read the IR-HOWTO, available at - <http://www.tldp.org/docs.html#howto>. - - If you want to exchange bits of data (vCal, vCard) with a PDA, you - will need to install some OBEX application, such as OpenObex : - <http://sourceforge.net/projects/openobex/> - - To compile this support as a module, choose M here: the module will - be called irda. - -comment "IrDA protocols" - depends on IRDA - -source "net/irda/irlan/Kconfig" - -source "net/irda/irnet/Kconfig" - -source "net/irda/ircomm/Kconfig" - -config IRDA_ULTRA - bool "Ultra (connectionless) protocol" - depends on IRDA - help - Say Y here to support the connectionless Ultra IRDA protocol. - Ultra allows to exchange data over IrDA with really simple devices - (watch, beacon) without the overhead of the IrDA protocol (no handshaking, - no management frames, simple fixed header). - Ultra is available as a special socket : socket(AF_IRDA, SOCK_DGRAM, 1); - -comment "IrDA options" - depends on IRDA - -config IRDA_CACHE_LAST_LSAP - bool "Cache last LSAP" - depends on IRDA - help - Say Y here if you want IrLMP to cache the last LSAP used. This - makes sense since most frames will be sent/received on the same - connection. Enabling this option will save a hash-lookup per frame. - - If unsure, say Y. - -config IRDA_FAST_RR - bool "Fast RRs (low latency)" - depends on IRDA - ---help--- - Say Y here is you want IrLAP to send fast RR (Receive Ready) frames - when acting as a primary station. - Disabling this option will make latency over IrDA very bad. Enabling - this option will make the IrDA stack send more packet than strictly - necessary, thus reduce your battery life (but not that much). - - Fast RR will make IrLAP send out a RR frame immediately when - receiving a frame if its own transmit queue is currently empty. This - will give a lot of speed improvement when receiving much data since - the secondary station will not have to wait the max. turn around - time (usually 500ms) before it is allowed to transmit the next time. - If the transmit queue of the secondary is also empty, the primary will - start backing-off before sending another RR frame, waiting longer - each time until the back-off reaches the max. turn around time. - This back-off increase in controlled via - /proc/sys/net/irda/fast_poll_increase - - If unsure, say Y. - -config IRDA_DEBUG - bool "Debug information" - depends on IRDA - help - Say Y here if you want the IrDA subsystem to write debug information - to your syslog. You can change the debug level in - /proc/sys/net/irda/debug . - When this option is enabled, the IrDA also perform many extra internal - verifications which will usually prevent the kernel to crash in case of - bugs. - - If unsure, say Y (since it makes it easier to find the bugs). - -source "drivers/net/irda/Kconfig" - diff --git a/net/irda/Makefile b/net/irda/Makefile deleted file mode 100644 index 187f6c563a4b..000000000000 --- a/net/irda/Makefile +++ /dev/null @@ -1,15 +0,0 @@ -# -# Makefile for the Linux IrDA protocol layer. -# - -obj-$(CONFIG_IRDA) += irda.o -obj-$(CONFIG_IRLAN) += irlan/ -obj-$(CONFIG_IRNET) += irnet/ -obj-$(CONFIG_IRCOMM) += ircomm/ - -irda-y := iriap.o iriap_event.o irlmp.o irlmp_event.o irlmp_frame.o \ - irlap.o irlap_event.o irlap_frame.o timer.o qos.o irqueue.o \ - irttp.o irda_device.o irias_object.o wrapper.o af_irda.o \ - discovery.o parameters.o irnetlink.o irmod.o -irda-$(CONFIG_PROC_FS) += irproc.o -irda-$(CONFIG_SYSCTL) += irsysctl.o diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c deleted file mode 100644 index 2e6990f8b80b..000000000000 --- a/net/irda/af_irda.c +++ /dev/null @@ -1,2695 +0,0 @@ -/********************************************************************* - * - * Filename: af_irda.c - * Version: 0.9 - * Description: IrDA sockets implementation - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun May 31 10:12:43 1998 - * Modified at: Sat Dec 25 21:10:23 1999 - * Modified by: Dag Brattli <dag@brattli.net> - * Sources: af_netroom.c, af_ax25.c, af_rose.c, af_x25.c etc. - * - * Copyright (c) 1999 Dag Brattli <dagb@cs.uit.no> - * Copyright (c) 1999-2003 Jean Tourrilhes <jt@hpl.hp.com> - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - * Linux-IrDA now supports four different types of IrDA sockets: - * - * o SOCK_STREAM: TinyTP connections with SAR disabled. The - * max SDU size is 0 for conn. of this type - * o SOCK_SEQPACKET: TinyTP connections with SAR enabled. TTP may - * fragment the messages, but will preserve - * the message boundaries - * o SOCK_DGRAM: IRDAPROTO_UNITDATA: TinyTP connections with Unitdata - * (unreliable) transfers - * IRDAPROTO_ULTRA: Connectionless and unreliable data - * - ********************************************************************/ - -#include <linux/capability.h> -#include <linux/module.h> -#include <linux/types.h> -#include <linux/socket.h> -#include <linux/sockios.h> -#include <linux/slab.h> -#include <linux/sched/signal.h> -#include <linux/init.h> -#include <linux/net.h> -#include <linux/irda.h> -#include <linux/poll.h> - -#include <asm/ioctls.h> /* TIOCOUTQ, TIOCINQ */ -#include <linux/uaccess.h> - -#include <net/sock.h> -#include <net/tcp_states.h> - -#include <net/irda/af_irda.h> - -static int irda_create(struct net *net, struct socket *sock, int protocol, int kern); - -static const struct proto_ops irda_stream_ops; -static const struct proto_ops irda_seqpacket_ops; -static const struct proto_ops irda_dgram_ops; - -#ifdef CONFIG_IRDA_ULTRA -static const struct proto_ops irda_ultra_ops; -#define ULTRA_MAX_DATA 382 -#endif /* CONFIG_IRDA_ULTRA */ - -#define IRDA_MAX_HEADER (TTP_MAX_HEADER) - -/* - * Function irda_data_indication (instance, sap, skb) - * - * Received some data from TinyTP. Just queue it on the receive queue - * - */ -static int irda_data_indication(void *instance, void *sap, struct sk_buff *skb) -{ - struct irda_sock *self; - struct sock *sk; - int err; - - self = instance; - sk = instance; - - err = sock_queue_rcv_skb(sk, skb); - if (err) { - pr_debug("%s(), error: no more mem!\n", __func__); - self->rx_flow = FLOW_STOP; - - /* When we return error, TTP will need to requeue the skb */ - return err; - } - - return 0; -} - -/* - * Function irda_disconnect_indication (instance, sap, reason, skb) - * - * Connection has been closed. Check reason to find out why - * - */ -static void irda_disconnect_indication(void *instance, void *sap, - LM_REASON reason, struct sk_buff *skb) -{ - struct irda_sock *self; - struct sock *sk; - - self = instance; - - pr_debug("%s(%p)\n", __func__, self); - - /* Don't care about it, but let's not leak it */ - if(skb) - dev_kfree_skb(skb); - - sk = instance; - if (sk == NULL) { - pr_debug("%s(%p) : BUG : sk is NULL\n", - __func__, self); - return; - } - - /* Prevent race conditions with irda_release() and irda_shutdown() */ - bh_lock_sock(sk); - if (!sock_flag(sk, SOCK_DEAD) && sk->sk_state != TCP_CLOSE) { - sk->sk_state = TCP_CLOSE; - sk->sk_shutdown |= SEND_SHUTDOWN; - - sk->sk_state_change(sk); - - /* Close our TSAP. - * If we leave it open, IrLMP put it back into the list of - * unconnected LSAPs. The problem is that any incoming request - * can then be matched to this socket (and it will be, because - * it is at the head of the list). This would prevent any - * listening socket waiting on the same TSAP to get those - * requests. Some apps forget to close sockets, or hang to it - * a bit too long, so we may stay in this dead state long - * enough to be noticed... - * Note : all socket function do check sk->sk_state, so we are - * safe... - * Jean II - */ - if (self->tsap) { - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } - } - bh_unlock_sock(sk); - - /* Note : once we are there, there is not much you want to do - * with the socket anymore, apart from closing it. - * For example, bind() and connect() won't reset sk->sk_err, - * sk->sk_shutdown and sk->sk_flags to valid values... - * Jean II - */ -} - -/* - * Function irda_connect_confirm (instance, sap, qos, max_sdu_size, skb) - * - * Connections has been confirmed by the remote device - * - */ -static void irda_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, __u8 max_header_size, - struct sk_buff *skb) -{ - struct irda_sock *self; - struct sock *sk; - - self = instance; - - pr_debug("%s(%p)\n", __func__, self); - - sk = instance; - if (sk == NULL) { - dev_kfree_skb(skb); - return; - } - - dev_kfree_skb(skb); - // Should be ??? skb_queue_tail(&sk->sk_receive_queue, skb); - - /* How much header space do we need to reserve */ - self->max_header_size = max_header_size; - - /* IrTTP max SDU size in transmit direction */ - self->max_sdu_size_tx = max_sdu_size; - - /* Find out what the largest chunk of data that we can transmit is */ - switch (sk->sk_type) { - case SOCK_STREAM: - if (max_sdu_size != 0) { - net_err_ratelimited("%s: max_sdu_size must be 0\n", - __func__); - return; - } - self->max_data_size = irttp_get_max_seg_size(self->tsap); - break; - case SOCK_SEQPACKET: - if (max_sdu_size == 0) { - net_err_ratelimited("%s: max_sdu_size cannot be 0\n", - __func__); - return; - } - self->max_data_size = max_sdu_size; - break; - default: - self->max_data_size = irttp_get_max_seg_size(self->tsap); - } - - pr_debug("%s(), max_data_size=%d\n", __func__, - self->max_data_size); - - memcpy(&self->qos_tx, qos, sizeof(struct qos_info)); - - /* We are now connected! */ - sk->sk_state = TCP_ESTABLISHED; - sk->sk_state_change(sk); -} - -/* - * Function irda_connect_indication(instance, sap, qos, max_sdu_size, userdata) - * - * Incoming connection - * - */ -static void irda_connect_indication(void *instance, void *sap, - struct qos_info *qos, __u32 max_sdu_size, - __u8 max_header_size, struct sk_buff *skb) -{ - struct irda_sock *self; - struct sock *sk; - - self = instance; - - pr_debug("%s(%p)\n", __func__, self); - - sk = instance; - if (sk == NULL) { - dev_kfree_skb(skb); - return; - } - - /* How much header space do we need to reserve */ - self->max_header_size = max_header_size; - - /* IrTTP max SDU size in transmit direction */ - self->max_sdu_size_tx = max_sdu_size; - - /* Find out what the largest chunk of data that we can transmit is */ - switch (sk->sk_type) { - case SOCK_STREAM: - if (max_sdu_size != 0) { - net_err_ratelimited("%s: max_sdu_size must be 0\n", - __func__); - kfree_skb(skb); - return; - } - self->max_data_size = irttp_get_max_seg_size(self->tsap); - break; - case SOCK_SEQPACKET: - if (max_sdu_size == 0) { - net_err_ratelimited("%s: max_sdu_size cannot be 0\n", - __func__); - kfree_skb(skb); - return; - } - self->max_data_size = max_sdu_size; - break; - default: - self->max_data_size = irttp_get_max_seg_size(self->tsap); - } - - pr_debug("%s(), max_data_size=%d\n", __func__, - self->max_data_size); - - memcpy(&self->qos_tx, qos, sizeof(struct qos_info)); - - skb_queue_tail(&sk->sk_receive_queue, skb); - sk->sk_state_change(sk); -} - -/* - * Function irda_connect_response (handle) - * - * Accept incoming connection - * - */ -static void irda_connect_response(struct irda_sock *self) -{ - struct sk_buff *skb; - - skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER, GFP_KERNEL); - if (skb == NULL) { - pr_debug("%s() Unable to allocate sk_buff!\n", - __func__); - return; - } - - /* Reserve space for MUX_CONTROL and LAP header */ - skb_reserve(skb, IRDA_MAX_HEADER); - - irttp_connect_response(self->tsap, self->max_sdu_size_rx, skb); -} - -/* - * Function irda_flow_indication (instance, sap, flow) - * - * Used by TinyTP to tell us if it can accept more data or not - * - */ -static void irda_flow_indication(void *instance, void *sap, LOCAL_FLOW flow) -{ - struct irda_sock *self; - struct sock *sk; - - self = instance; - sk = instance; - BUG_ON(sk == NULL); - - switch (flow) { - case FLOW_STOP: - pr_debug("%s(), IrTTP wants us to slow down\n", - __func__); - self->tx_flow = flow; - break; - case FLOW_START: - self->tx_flow = flow; - pr_debug("%s(), IrTTP wants us to start again\n", - __func__); - wake_up_interruptible(sk_sleep(sk)); - break; - default: - pr_debug("%s(), Unknown flow command!\n", __func__); - /* Unknown flow command, better stop */ - self->tx_flow = flow; - break; - } -} - -/* - * Function irda_getvalue_confirm (obj_id, value, priv) - * - * Got answer from remote LM-IAS, just pass object to requester... - * - * Note : duplicate from above, but we need our own version that - * doesn't touch the dtsap_sel and save the full value structure... - */ -static void irda_getvalue_confirm(int result, __u16 obj_id, - struct ias_value *value, void *priv) -{ - struct irda_sock *self; - - self = priv; - if (!self) { - net_warn_ratelimited("%s: lost myself!\n", __func__); - return; - } - - pr_debug("%s(%p)\n", __func__, self); - - /* We probably don't need to make any more queries */ - iriap_close(self->iriap); - self->iriap = NULL; - - /* Check if request succeeded */ - if (result != IAS_SUCCESS) { - pr_debug("%s(), IAS query failed! (%d)\n", __func__, - result); - - self->errno = result; /* We really need it later */ - - /* Wake up any processes waiting for result */ - wake_up_interruptible(&self->query_wait); - - return; - } - - /* Pass the object to the caller (so the caller must delete it) */ - self->ias_result = value; - self->errno = 0; - - /* Wake up any processes waiting for result */ - wake_up_interruptible(&self->query_wait); -} - -/* - * Function irda_selective_discovery_indication (discovery) - * - * Got a selective discovery indication from IrLMP. - * - * IrLMP is telling us that this node is new and matching our hint bit - * filter. Wake up any process waiting for answer... - */ -static void irda_selective_discovery_indication(discinfo_t *discovery, - DISCOVERY_MODE mode, - void *priv) -{ - struct irda_sock *self; - - self = priv; - if (!self) { - net_warn_ratelimited("%s: lost myself!\n", __func__); - return; - } - - /* Pass parameter to the caller */ - self->cachedaddr = discovery->daddr; - - /* Wake up process if its waiting for device to be discovered */ - wake_up_interruptible(&self->query_wait); -} - -/* - * Function irda_discovery_timeout (priv) - * - * Timeout in the selective discovery process - * - * We were waiting for a node to be discovered, but nothing has come up - * so far. Wake up the user and tell him that we failed... - */ -static void irda_discovery_timeout(u_long priv) -{ - struct irda_sock *self; - - self = (struct irda_sock *) priv; - BUG_ON(self == NULL); - - /* Nothing for the caller */ - self->cachelog = NULL; - self->cachedaddr = 0; - self->errno = -ETIME; - - /* Wake up process if its still waiting... */ - wake_up_interruptible(&self->query_wait); -} - -/* - * Function irda_open_tsap (self) - * - * Open local Transport Service Access Point (TSAP) - * - */ -static int irda_open_tsap(struct irda_sock *self, __u8 tsap_sel, char *name) -{ - notify_t notify; - - if (self->tsap) { - pr_debug("%s: busy!\n", __func__); - return -EBUSY; - } - - /* Initialize callbacks to be used by the IrDA stack */ - irda_notify_init(¬ify); - notify.connect_confirm = irda_connect_confirm; - notify.connect_indication = irda_connect_indication; - notify.disconnect_indication = irda_disconnect_indication; - notify.data_indication = irda_data_indication; - notify.udata_indication = irda_data_indication; - notify.flow_indication = irda_flow_indication; - notify.instance = self; - strncpy(notify.name, name, NOTIFY_MAX_NAME); - - self->tsap = irttp_open_tsap(tsap_sel, DEFAULT_INITIAL_CREDIT, - ¬ify); - if (self->tsap == NULL) { - pr_debug("%s(), Unable to allocate TSAP!\n", - __func__); - return -ENOMEM; - } - /* Remember which TSAP selector we actually got */ - self->stsap_sel = self->tsap->stsap_sel; - - return 0; -} - -/* - * Function irda_open_lsap (self) - * - * Open local Link Service Access Point (LSAP). Used for opening Ultra - * sockets - */ -#ifdef CONFIG_IRDA_ULTRA -static int irda_open_lsap(struct irda_sock *self, int pid) -{ - notify_t notify; - - if (self->lsap) { - net_warn_ratelimited("%s(), busy!\n", __func__); - return -EBUSY; - } - - /* Initialize callbacks to be used by the IrDA stack */ - irda_notify_init(¬ify); - notify.udata_indication = irda_data_indication; - notify.instance = self; - strncpy(notify.name, "Ultra", NOTIFY_MAX_NAME); - - self->lsap = irlmp_open_lsap(LSAP_CONNLESS, ¬ify, pid); - if (self->lsap == NULL) { - pr_debug("%s(), Unable to allocate LSAP!\n", __func__); - return -ENOMEM; - } - - return 0; -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irda_find_lsap_sel (self, name) - * - * Try to lookup LSAP selector in remote LM-IAS - * - * Basically, we start a IAP query, and then go to sleep. When the query - * return, irda_getvalue_confirm will wake us up, and we can examine the - * result of the query... - * Note that in some case, the query fail even before we go to sleep, - * creating some races... - */ -static int irda_find_lsap_sel(struct irda_sock *self, char *name) -{ - pr_debug("%s(%p, %s)\n", __func__, self, name); - - if (self->iriap) { - net_warn_ratelimited("%s(): busy with a previous query\n", - __func__); - return -EBUSY; - } - - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - irda_getvalue_confirm); - if(self->iriap == NULL) - return -ENOMEM; - - /* Treat unexpected wakeup as disconnect */ - self->errno = -EHOSTUNREACH; - - /* Query remote LM-IAS */ - iriap_getvaluebyclass_request(self->iriap, self->saddr, self->daddr, - name, "IrDA:TinyTP:LsapSel"); - - /* Wait for answer, if not yet finished (or failed) */ - if (wait_event_interruptible(self->query_wait, (self->iriap==NULL))) - /* Treat signals as disconnect */ - return -EHOSTUNREACH; - - /* Check what happened */ - if (self->errno) - { - /* Requested object/attribute doesn't exist */ - if((self->errno == IAS_CLASS_UNKNOWN) || - (self->errno == IAS_ATTRIB_UNKNOWN)) - return -EADDRNOTAVAIL; - else - return -EHOSTUNREACH; - } - - /* Get the remote TSAP selector */ - switch (self->ias_result->type) { - case IAS_INTEGER: - pr_debug("%s() int=%d\n", - __func__, self->ias_result->t.integer); - - if (self->ias_result->t.integer != -1) - self->dtsap_sel = self->ias_result->t.integer; - else - self->dtsap_sel = 0; - break; - default: - self->dtsap_sel = 0; - pr_debug("%s(), bad type!\n", __func__); - break; - } - if (self->ias_result) - irias_delete_value(self->ias_result); - - if (self->dtsap_sel) - return 0; - - return -EADDRNOTAVAIL; -} - -/* - * Function irda_discover_daddr_and_lsap_sel (self, name) - * - * This try to find a device with the requested service. - * - * It basically look into the discovery log. For each address in the list, - * it queries the LM-IAS of the device to find if this device offer - * the requested service. - * If there is more than one node supporting the service, we complain - * to the user (it should move devices around). - * The, we set both the destination address and the lsap selector to point - * on the service on the unique device we have found. - * - * Note : this function fails if there is more than one device in range, - * because IrLMP doesn't disconnect the LAP when the last LSAP is closed. - * Moreover, we would need to wait the LAP disconnection... - */ -static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name) -{ - discinfo_t *discoveries; /* Copy of the discovery log */ - int number; /* Number of nodes in the log */ - int i; - int err = -ENETUNREACH; - __u32 daddr = DEV_ADDR_ANY; /* Address we found the service on */ - __u8 dtsap_sel = 0x0; /* TSAP associated with it */ - - pr_debug("%s(), name=%s\n", __func__, name); - - /* Ask lmp for the current discovery log - * Note : we have to use irlmp_get_discoveries(), as opposed - * to play with the cachelog directly, because while we are - * making our ias query, le log might change... */ - discoveries = irlmp_get_discoveries(&number, self->mask.word, - self->nslots); - /* Check if the we got some results */ - if (discoveries == NULL) - return -ENETUNREACH; /* No nodes discovered */ - - /* - * Now, check all discovered devices (if any), and connect - * client only about the services that the client is - * interested in... - */ - for(i = 0; i < number; i++) { - /* Try the address in the log */ - self->daddr = discoveries[i].daddr; - self->saddr = 0x0; - pr_debug("%s(), trying daddr = %08x\n", - __func__, self->daddr); - - /* Query remote LM-IAS for this service */ - err = irda_find_lsap_sel(self, name); - switch (err) { - case 0: - /* We found the requested service */ - if(daddr != DEV_ADDR_ANY) { - pr_debug("%s(), discovered service ''%s'' in two different devices !!!\n", - __func__, name); - self->daddr = DEV_ADDR_ANY; - kfree(discoveries); - return -ENOTUNIQ; - } - /* First time we found that one, save it ! */ - daddr = self->daddr; - dtsap_sel = self->dtsap_sel; - break; - case -EADDRNOTAVAIL: - /* Requested service simply doesn't exist on this node */ - break; - default: - /* Something bad did happen :-( */ - pr_debug("%s(), unexpected IAS query failure\n", - __func__); - self->daddr = DEV_ADDR_ANY; - kfree(discoveries); - return -EHOSTUNREACH; - } - } - /* Cleanup our copy of the discovery log */ - kfree(discoveries); - - /* Check out what we found */ - if(daddr == DEV_ADDR_ANY) { - pr_debug("%s(), cannot discover service ''%s'' in any device !!!\n", - __func__, name); - self->daddr = DEV_ADDR_ANY; - return -EADDRNOTAVAIL; - } - - /* Revert back to discovered device & service */ - self->daddr = daddr; - self->saddr = 0x0; - self->dtsap_sel = dtsap_sel; - - pr_debug("%s(), discovered requested service ''%s'' at address %08x\n", - __func__, name, self->daddr); - - return 0; -} - -/* - * Function irda_getname (sock, uaddr, uaddr_len, peer) - * - * Return the our own, or peers socket address (sockaddr_irda) - * - */ -static int irda_getname(struct socket *sock, struct sockaddr *uaddr, - int *uaddr_len, int peer) -{ - struct sockaddr_irda saddr; - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - - memset(&saddr, 0, sizeof(saddr)); - if (peer) { - if (sk->sk_state != TCP_ESTABLISHED) - return -ENOTCONN; - - saddr.sir_family = AF_IRDA; - saddr.sir_lsap_sel = self->dtsap_sel; - saddr.sir_addr = self->daddr; - } else { - saddr.sir_family = AF_IRDA; - saddr.sir_lsap_sel = self->stsap_sel; - saddr.sir_addr = self->saddr; - } - - pr_debug("%s(), tsap_sel = %#x\n", __func__, saddr.sir_lsap_sel); - pr_debug("%s(), addr = %08x\n", __func__, saddr.sir_addr); - - /* uaddr_len come to us uninitialised */ - *uaddr_len = sizeof (struct sockaddr_irda); - memcpy(uaddr, &saddr, *uaddr_len); - - return 0; -} - -/* - * Function irda_listen (sock, backlog) - * - * Just move to the listen state - * - */ -static int irda_listen(struct socket *sock, int backlog) -{ - struct sock *sk = sock->sk; - int err = -EOPNOTSUPP; - - lock_sock(sk); - - if ((sk->sk_type != SOCK_STREAM) && (sk->sk_type != SOCK_SEQPACKET) && - (sk->sk_type != SOCK_DGRAM)) - goto out; - - if (sk->sk_state != TCP_LISTEN) { - sk->sk_max_ack_backlog = backlog; - sk->sk_state = TCP_LISTEN; - - err = 0; - } -out: - release_sock(sk); - - return err; -} - -/* - * Function irda_bind (sock, uaddr, addr_len) - * - * Used by servers to register their well known TSAP - * - */ -static int irda_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) -{ - struct sock *sk = sock->sk; - struct sockaddr_irda *addr = (struct sockaddr_irda *) uaddr; - struct irda_sock *self = irda_sk(sk); - int err; - - pr_debug("%s(%p)\n", __func__, self); - - if (addr_len != sizeof(struct sockaddr_irda)) - return -EINVAL; - - lock_sock(sk); -#ifdef CONFIG_IRDA_ULTRA - /* Special care for Ultra sockets */ - if ((sk->sk_type == SOCK_DGRAM) && - (sk->sk_protocol == IRDAPROTO_ULTRA)) { - self->pid = addr->sir_lsap_sel; - err = -EOPNOTSUPP; - if (self->pid & 0x80) { - pr_debug("%s(), extension in PID not supp!\n", - __func__); - goto out; - } - err = irda_open_lsap(self, self->pid); - if (err < 0) - goto out; - - /* Pretend we are connected */ - sock->state = SS_CONNECTED; - sk->sk_state = TCP_ESTABLISHED; - err = 0; - - goto out; - } -#endif /* CONFIG_IRDA_ULTRA */ - - self->ias_obj = irias_new_object(addr->sir_name, jiffies); - err = -ENOMEM; - if (self->ias_obj == NULL) - goto out; - - err = irda_open_tsap(self, addr->sir_lsap_sel, addr->sir_name); - if (err < 0) { - irias_delete_object(self->ias_obj); - self->ias_obj = NULL; - goto out; - } - - /* Register with LM-IAS */ - irias_add_integer_attrib(self->ias_obj, "IrDA:TinyTP:LsapSel", - self->stsap_sel, IAS_KERNEL_ATTR); - irias_insert_object(self->ias_obj); - - err = 0; -out: - release_sock(sk); - return err; -} - -/* - * Function irda_accept (sock, newsock, flags) - * - * Wait for incoming connection - * - */ -static int irda_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) -{ - struct sock *sk = sock->sk; - struct irda_sock *new, *self = irda_sk(sk); - struct sock *newsk; - struct sk_buff *skb = NULL; - int err; - - err = irda_create(sock_net(sk), newsock, sk->sk_protocol, kern); - if (err) - return err; - - err = -EINVAL; - - lock_sock(sk); - if (sock->state != SS_UNCONNECTED) - goto out; - - err = -EOPNOTSUPP; - if ((sk->sk_type != SOCK_STREAM) && (sk->sk_type != SOCK_SEQPACKET) && - (sk->sk_type != SOCK_DGRAM)) - goto out; - - err = -EINVAL; - if (sk->sk_state != TCP_LISTEN) - goto out; - - /* - * The read queue this time is holding sockets ready to use - * hooked into the SABM we saved - */ - - /* - * We can perform the accept only if there is incoming data - * on the listening socket. - * So, we will block the caller until we receive any data. - * If the caller was waiting on select() or poll() before - * calling us, the data is waiting for us ;-) - * Jean II - */ - while (1) { - skb = skb_dequeue(&sk->sk_receive_queue); - if (skb) - break; - - /* Non blocking operation */ - err = -EWOULDBLOCK; - if (flags & O_NONBLOCK) - goto out; - - err = wait_event_interruptible(*(sk_sleep(sk)), - skb_peek(&sk->sk_receive_queue)); - if (err) - goto out; - } - - newsk = newsock->sk; - err = -EIO; - if (newsk == NULL) - goto out; - - newsk->sk_state = TCP_ESTABLISHED; - - new = irda_sk(newsk); - - /* Now attach up the new socket */ - new->tsap = irttp_dup(self->tsap, new); - err = -EPERM; /* value does not seem to make sense. -arnd */ - if (!new->tsap) { - pr_debug("%s(), dup failed!\n", __func__); - goto out; - } - - new->stsap_sel = new->tsap->stsap_sel; - new->dtsap_sel = new->tsap->dtsap_sel; - new->saddr = irttp_get_saddr(new->tsap); - new->daddr = irttp_get_daddr(new->tsap); - - new->max_sdu_size_tx = self->max_sdu_size_tx; - new->max_sdu_size_rx = self->max_sdu_size_rx; - new->max_data_size = self->max_data_size; - new->max_header_size = self->max_header_size; - - memcpy(&new->qos_tx, &self->qos_tx, sizeof(struct qos_info)); - - /* Clean up the original one to keep it in listen state */ - irttp_listen(self->tsap); - - sk->sk_ack_backlog--; - - newsock->state = SS_CONNECTED; - - irda_connect_response(new); - err = 0; -out: - kfree_skb(skb); - release_sock(sk); - return err; -} - -/* - * Function irda_connect (sock, uaddr, addr_len, flags) - * - * Connect to a IrDA device - * - * The main difference with a "standard" connect is that with IrDA we need - * to resolve the service name into a TSAP selector (in TCP, port number - * doesn't have to be resolved). - * Because of this service name resolution, we can offer "auto-connect", - * where we connect to a service without specifying a destination address. - * - * Note : by consulting "errno", the user space caller may learn the cause - * of the failure. Most of them are visible in the function, others may come - * from subroutines called and are listed here : - * o EBUSY : already processing a connect - * o EHOSTUNREACH : bad addr->sir_addr argument - * o EADDRNOTAVAIL : bad addr->sir_name argument - * o ENOTUNIQ : more than one node has addr->sir_name (auto-connect) - * o ENETUNREACH : no node found on the network (auto-connect) - */ -static int irda_connect(struct socket *sock, struct sockaddr *uaddr, - int addr_len, int flags) -{ - struct sock *sk = sock->sk; - struct sockaddr_irda *addr = (struct sockaddr_irda *) uaddr; - struct irda_sock *self = irda_sk(sk); - int err; - - pr_debug("%s(%p)\n", __func__, self); - - lock_sock(sk); - /* Don't allow connect for Ultra sockets */ - err = -ESOCKTNOSUPPORT; - if ((sk->sk_type == SOCK_DGRAM) && (sk->sk_protocol == IRDAPROTO_ULTRA)) - goto out; - - if (sk->sk_state == TCP_ESTABLISHED && sock->state == SS_CONNECTING) { - sock->state = SS_CONNECTED; - err = 0; - goto out; /* Connect completed during a ERESTARTSYS event */ - } - - if (sk->sk_state == TCP_CLOSE && sock->state == SS_CONNECTING) { - sock->state = SS_UNCONNECTED; - err = -ECONNREFUSED; - goto out; - } - - err = -EISCONN; /* No reconnect on a seqpacket socket */ - if (sk->sk_state == TCP_ESTABLISHED) - goto out; - - sk->sk_state = TCP_CLOSE; - sock->state = SS_UNCONNECTED; - - err = -EINVAL; - if (addr_len != sizeof(struct sockaddr_irda)) - goto out; - - /* Check if user supplied any destination device address */ - if ((!addr->sir_addr) || (addr->sir_addr == DEV_ADDR_ANY)) { - /* Try to find one suitable */ - err = irda_discover_daddr_and_lsap_sel(self, addr->sir_name); - if (err) { - pr_debug("%s(), auto-connect failed!\n", __func__); - goto out; - } - } else { - /* Use the one provided by the user */ - self->daddr = addr->sir_addr; - pr_debug("%s(), daddr = %08x\n", __func__, self->daddr); - - /* If we don't have a valid service name, we assume the - * user want to connect on a specific LSAP. Prevent - * the use of invalid LSAPs (IrLMP 1.1 p10). Jean II */ - if((addr->sir_name[0] != '\0') || - (addr->sir_lsap_sel >= 0x70)) { - /* Query remote LM-IAS using service name */ - err = irda_find_lsap_sel(self, addr->sir_name); - if (err) { - pr_debug("%s(), connect failed!\n", __func__); - goto out; - } - } else { - /* Directly connect to the remote LSAP - * specified by the sir_lsap field. - * Please use with caution, in IrDA LSAPs are - * dynamic and there is no "well-known" LSAP. */ - self->dtsap_sel = addr->sir_lsap_sel; - } - } - - /* Check if we have opened a local TSAP */ - if (!self->tsap) { - err = irda_open_tsap(self, LSAP_ANY, addr->sir_name); - if (err) - goto out; - } - - /* Move to connecting socket, start sending Connect Requests */ - sock->state = SS_CONNECTING; - sk->sk_state = TCP_SYN_SENT; - - /* Connect to remote device */ - err = irttp_connect_request(self->tsap, self->dtsap_sel, - self->saddr, self->daddr, NULL, - self->max_sdu_size_rx, NULL); - if (err) { - pr_debug("%s(), connect failed!\n", __func__); - goto out; - } - - /* Now the loop */ - err = -EINPROGRESS; - if (sk->sk_state != TCP_ESTABLISHED && (flags & O_NONBLOCK)) - goto out; - - err = -ERESTARTSYS; - if (wait_event_interruptible(*(sk_sleep(sk)), - (sk->sk_state != TCP_SYN_SENT))) - goto out; - - if (sk->sk_state != TCP_ESTABLISHED) { - sock->state = SS_UNCONNECTED; - err = sock_error(sk); - if (!err) - err = -ECONNRESET; - goto out; - } - - sock->state = SS_CONNECTED; - - /* At this point, IrLMP has assigned our source address */ - self->saddr = irttp_get_saddr(self->tsap); - err = 0; -out: - release_sock(sk); - return err; -} - -static struct proto irda_proto = { - .name = "IRDA", - .owner = THIS_MODULE, - .obj_size = sizeof(struct irda_sock), -}; - -/* - * Function irda_create (sock, protocol) - * - * Create IrDA socket - * - */ -static int irda_create(struct net *net, struct socket *sock, int protocol, - int kern) -{ - struct sock *sk; - struct irda_sock *self; - - if (protocol < 0 || protocol > SK_PROTOCOL_MAX) - return -EINVAL; - - if (net != &init_net) - return -EAFNOSUPPORT; - - /* Check for valid socket type */ - switch (sock->type) { - case SOCK_STREAM: /* For TTP connections with SAR disabled */ - case SOCK_SEQPACKET: /* For TTP connections with SAR enabled */ - case SOCK_DGRAM: /* For TTP Unitdata or LMP Ultra transfers */ - break; - default: - return -ESOCKTNOSUPPORT; - } - - /* Allocate networking socket */ - sk = sk_alloc(net, PF_IRDA, GFP_KERNEL, &irda_proto, kern); - if (sk == NULL) - return -ENOMEM; - - self = irda_sk(sk); - pr_debug("%s() : self is %p\n", __func__, self); - - init_waitqueue_head(&self->query_wait); - - switch (sock->type) { - case SOCK_STREAM: - sock->ops = &irda_stream_ops; - self->max_sdu_size_rx = TTP_SAR_DISABLE; - break; - case SOCK_SEQPACKET: - sock->ops = &irda_seqpacket_ops; - self->max_sdu_size_rx = TTP_SAR_UNBOUND; - break; - case SOCK_DGRAM: - switch (protocol) { -#ifdef CONFIG_IRDA_ULTRA - case IRDAPROTO_ULTRA: - sock->ops = &irda_ultra_ops; - /* Initialise now, because we may send on unbound - * sockets. Jean II */ - self->max_data_size = ULTRA_MAX_DATA - LMP_PID_HEADER; - self->max_header_size = IRDA_MAX_HEADER + LMP_PID_HEADER; - break; -#endif /* CONFIG_IRDA_ULTRA */ - case IRDAPROTO_UNITDATA: - sock->ops = &irda_dgram_ops; - /* We let Unitdata conn. be like seqpack conn. */ - self->max_sdu_size_rx = TTP_SAR_UNBOUND; - break; - default: - sk_free(sk); - return -ESOCKTNOSUPPORT; - } - break; - default: - sk_free(sk); - return -ESOCKTNOSUPPORT; - } - - /* Initialise networking socket struct */ - sock_init_data(sock, sk); /* Note : set sk->sk_refcnt to 1 */ - sk->sk_family = PF_IRDA; - sk->sk_protocol = protocol; - - /* Register as a client with IrLMP */ - self->ckey = irlmp_register_client(0, NULL, NULL, NULL); - self->mask.word = 0xffff; - self->rx_flow = self->tx_flow = FLOW_START; - self->nslots = DISCOVERY_DEFAULT_SLOTS; - self->daddr = DEV_ADDR_ANY; /* Until we get connected */ - self->saddr = 0x0; /* so IrLMP assign us any link */ - return 0; -} - -/* - * Function irda_destroy_socket (self) - * - * Destroy socket - * - */ -static void irda_destroy_socket(struct irda_sock *self) -{ - pr_debug("%s(%p)\n", __func__, self); - - /* Unregister with IrLMP */ - irlmp_unregister_client(self->ckey); - irlmp_unregister_service(self->skey); - - /* Unregister with LM-IAS */ - if (self->ias_obj) { - irias_delete_object(self->ias_obj); - self->ias_obj = NULL; - } - - if (self->iriap) { - iriap_close(self->iriap); - self->iriap = NULL; - } - - if (self->tsap) { - irttp_disconnect_request(self->tsap, NULL, P_NORMAL); - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } -#ifdef CONFIG_IRDA_ULTRA - if (self->lsap) { - irlmp_close_lsap(self->lsap); - self->lsap = NULL; - } -#endif /* CONFIG_IRDA_ULTRA */ -} - -/* - * Function irda_release (sock) - */ -static int irda_release(struct socket *sock) -{ - struct sock *sk = sock->sk; - - if (sk == NULL) - return 0; - - lock_sock(sk); - sk->sk_state = TCP_CLOSE; - sk->sk_shutdown |= SEND_SHUTDOWN; - sk->sk_state_change(sk); - - /* Destroy IrDA socket */ - irda_destroy_socket(irda_sk(sk)); - - sock_orphan(sk); - sock->sk = NULL; - release_sock(sk); - - /* Purge queues (see sock_init_data()) */ - skb_queue_purge(&sk->sk_receive_queue); - - /* Destroy networking socket if we are the last reference on it, - * i.e. if(sk->sk_refcnt == 0) -> sk_free(sk) */ - sock_put(sk); - - /* Notes on socket locking and deallocation... - Jean II - * In theory we should put pairs of sock_hold() / sock_put() to - * prevent the socket to be destroyed whenever there is an - * outstanding request or outstanding incoming packet or event. - * - * 1) This may include IAS request, both in connect and getsockopt. - * Unfortunately, the situation is a bit more messy than it looks, - * because we close iriap and kfree(self) above. - * - * 2) This may include selective discovery in getsockopt. - * Same stuff as above, irlmp registration and self are gone. - * - * Probably 1 and 2 may not matter, because it's all triggered - * by a process and the socket layer already prevent the - * socket to go away while a process is holding it, through - * sockfd_put() and fput()... - * - * 3) This may include deferred TSAP closure. In particular, - * we may receive a late irda_disconnect_indication() - * Fortunately, (tsap_cb *)->close_pend should protect us - * from that. - * - * I did some testing on SMP, and it looks solid. And the socket - * memory leak is now gone... - Jean II - */ - - return 0; -} - -/* - * Function irda_sendmsg (sock, msg, len) - * - * Send message down to TinyTP. This function is used for both STREAM and - * SEQPACK services. This is possible since it forces the client to - * fragment the message if necessary - */ -static int irda_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) -{ - struct sock *sk = sock->sk; - struct irda_sock *self; - struct sk_buff *skb; - int err = -EPIPE; - - pr_debug("%s(), len=%zd\n", __func__, len); - - /* Note : socket.c set MSG_EOR on SEQPACKET sockets */ - if (msg->msg_flags & ~(MSG_DONTWAIT | MSG_EOR | MSG_CMSG_COMPAT | - MSG_NOSIGNAL)) { - return -EINVAL; - } - - lock_sock(sk); - - if (sk->sk_shutdown & SEND_SHUTDOWN) - goto out_err; - - if (sk->sk_state != TCP_ESTABLISHED) { - err = -ENOTCONN; - goto out; - } - - self = irda_sk(sk); - - /* Check if IrTTP is wants us to slow down */ - - if (wait_event_interruptible(*(sk_sleep(sk)), - (self->tx_flow != FLOW_STOP || sk->sk_state != TCP_ESTABLISHED))) { - err = -ERESTARTSYS; - goto out; - } - - /* Check if we are still connected */ - if (sk->sk_state != TCP_ESTABLISHED) { - err = -ENOTCONN; - goto out; - } - - /* Check that we don't send out too big frames */ - if (len > self->max_data_size) { - pr_debug("%s(), Chopping frame from %zd to %d bytes!\n", - __func__, len, self->max_data_size); - len = self->max_data_size; - } - - skb = sock_alloc_send_skb(sk, len + self->max_header_size + 16, - msg->msg_flags & MSG_DONTWAIT, &err); - if (!skb) - goto out_err; - - skb_reserve(skb, self->max_header_size + 16); - skb_reset_transport_header(skb); - skb_put(skb, len); - err = memcpy_from_msg(skb_transport_header(skb), msg, len); - if (err) { - kfree_skb(skb); - goto out_err; - } - - /* - * Just send the message to TinyTP, and let it deal with possible - * errors. No need to duplicate all that here - */ - err = irttp_data_request(self->tsap, skb); - if (err) { - pr_debug("%s(), err=%d\n", __func__, err); - goto out_err; - } - - release_sock(sk); - /* Tell client how much data we actually sent */ - return len; - -out_err: - err = sk_stream_error(sk, msg->msg_flags, err); -out: - release_sock(sk); - return err; - -} - -/* - * Function irda_recvmsg_dgram (sock, msg, size, flags) - * - * Try to receive message and copy it to user. The frame is discarded - * after being read, regardless of how much the user actually read - */ -static int irda_recvmsg_dgram(struct socket *sock, struct msghdr *msg, - size_t size, int flags) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - struct sk_buff *skb; - size_t copied; - int err; - - skb = skb_recv_datagram(sk, flags & ~MSG_DONTWAIT, - flags & MSG_DONTWAIT, &err); - if (!skb) - return err; - - skb_reset_transport_header(skb); - copied = skb->len; - - if (copied > size) { - pr_debug("%s(), Received truncated frame (%zd < %zd)!\n", - __func__, copied, size); - copied = size; - msg->msg_flags |= MSG_TRUNC; - } - skb_copy_datagram_msg(skb, 0, msg, copied); - - skb_free_datagram(sk, skb); - - /* - * Check if we have previously stopped IrTTP and we know - * have more free space in our rx_queue. If so tell IrTTP - * to start delivering frames again before our rx_queue gets - * empty - */ - if (self->rx_flow == FLOW_STOP) { - if ((atomic_read(&sk->sk_rmem_alloc) << 2) <= sk->sk_rcvbuf) { - pr_debug("%s(), Starting IrTTP\n", __func__); - self->rx_flow = FLOW_START; - irttp_flow_request(self->tsap, FLOW_START); - } - } - - return copied; -} - -/* - * Function irda_recvmsg_stream (sock, msg, size, flags) - */ -static int irda_recvmsg_stream(struct socket *sock, struct msghdr *msg, - size_t size, int flags) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - int noblock = flags & MSG_DONTWAIT; - size_t copied = 0; - int target, err; - long timeo; - - if ((err = sock_error(sk)) < 0) - return err; - - if (sock->flags & __SO_ACCEPTCON) - return -EINVAL; - - err =-EOPNOTSUPP; - if (flags & MSG_OOB) - return -EOPNOTSUPP; - - err = 0; - target = sock_rcvlowat(sk, flags & MSG_WAITALL, size); - timeo = sock_rcvtimeo(sk, noblock); - - do { - int chunk; - struct sk_buff *skb = skb_dequeue(&sk->sk_receive_queue); - - if (skb == NULL) { - DEFINE_WAIT(wait); - err = 0; - - if (copied >= target) - break; - - prepare_to_wait_exclusive(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); - - /* - * POSIX 1003.1g mandates this order. - */ - err = sock_error(sk); - if (err) - ; - else if (sk->sk_shutdown & RCV_SHUTDOWN) - ; - else if (noblock) - err = -EAGAIN; - else if (signal_pending(current)) - err = sock_intr_errno(timeo); - else if (sk->sk_state != TCP_ESTABLISHED) - err = -ENOTCONN; - else if (skb_peek(&sk->sk_receive_queue) == NULL) - /* Wait process until data arrives */ - schedule(); - - finish_wait(sk_sleep(sk), &wait); - - if (err) - return err; - if (sk->sk_shutdown & RCV_SHUTDOWN) - break; - - continue; - } - - chunk = min_t(unsigned int, skb->len, size); - if (memcpy_to_msg(msg, skb->data, chunk)) { - skb_queue_head(&sk->sk_receive_queue, skb); - if (copied == 0) - copied = -EFAULT; - break; - } - copied += chunk; - size -= chunk; - - /* Mark read part of skb as used */ - if (!(flags & MSG_PEEK)) { - skb_pull(skb, chunk); - - /* put the skb back if we didn't use it up.. */ - if (skb->len) { - pr_debug("%s(), back on q!\n", - __func__); - skb_queue_head(&sk->sk_receive_queue, skb); - break; - } - - kfree_skb(skb); - } else { - pr_debug("%s() questionable!?\n", __func__); - - /* put message back and return */ - skb_queue_head(&sk->sk_receive_queue, skb); - break; - } - } while (size); - - /* - * Check if we have previously stopped IrTTP and we know - * have more free space in our rx_queue. If so tell IrTTP - * to start delivering frames again before our rx_queue gets - * empty - */ - if (self->rx_flow == FLOW_STOP) { - if ((atomic_read(&sk->sk_rmem_alloc) << 2) <= sk->sk_rcvbuf) { - pr_debug("%s(), Starting IrTTP\n", __func__); - self->rx_flow = FLOW_START; - irttp_flow_request(self->tsap, FLOW_START); - } - } - - return copied; -} - -/* - * Function irda_sendmsg_dgram (sock, msg, len) - * - * Send message down to TinyTP for the unreliable sequenced - * packet service... - * - */ -static int irda_sendmsg_dgram(struct socket *sock, struct msghdr *msg, - size_t len) -{ - struct sock *sk = sock->sk; - struct irda_sock *self; - struct sk_buff *skb; - int err; - - pr_debug("%s(), len=%zd\n", __func__, len); - - if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_CMSG_COMPAT)) - return -EINVAL; - - lock_sock(sk); - - if (sk->sk_shutdown & SEND_SHUTDOWN) { - send_sig(SIGPIPE, current, 0); - err = -EPIPE; - goto out; - } - - err = -ENOTCONN; - if (sk->sk_state != TCP_ESTABLISHED) - goto out; - - self = irda_sk(sk); - - /* - * Check that we don't send out too big frames. This is an unreliable - * service, so we have no fragmentation and no coalescence - */ - if (len > self->max_data_size) { - pr_debug("%s(), Warning too much data! Chopping frame from %zd to %d bytes!\n", - __func__, len, self->max_data_size); - len = self->max_data_size; - } - - skb = sock_alloc_send_skb(sk, len + self->max_header_size, - msg->msg_flags & MSG_DONTWAIT, &err); - err = -ENOBUFS; - if (!skb) - goto out; - - skb_reserve(skb, self->max_header_size); - skb_reset_transport_header(skb); - - pr_debug("%s(), appending user data\n", __func__); - skb_put(skb, len); - err = memcpy_from_msg(skb_transport_header(skb), msg, len); - if (err) { - kfree_skb(skb); - goto out; - } - - /* - * Just send the message to TinyTP, and let it deal with possible - * errors. No need to duplicate all that here - */ - err = irttp_udata_request(self->tsap, skb); - if (err) { - pr_debug("%s(), err=%d\n", __func__, err); - goto out; - } - - release_sock(sk); - return len; - -out: - release_sock(sk); - return err; -} - -/* - * Function irda_sendmsg_ultra (sock, msg, len) - * - * Send message down to IrLMP for the unreliable Ultra - * packet service... - */ -#ifdef CONFIG_IRDA_ULTRA -static int irda_sendmsg_ultra(struct socket *sock, struct msghdr *msg, - size_t len) -{ - struct sock *sk = sock->sk; - struct irda_sock *self; - __u8 pid = 0; - int bound = 0; - struct sk_buff *skb; - int err; - - pr_debug("%s(), len=%zd\n", __func__, len); - - err = -EINVAL; - if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_CMSG_COMPAT)) - return -EINVAL; - - lock_sock(sk); - - err = -EPIPE; - if (sk->sk_shutdown & SEND_SHUTDOWN) { - send_sig(SIGPIPE, current, 0); - goto out; - } - - self = irda_sk(sk); - - /* Check if an address was specified with sendto. Jean II */ - if (msg->msg_name) { - DECLARE_SOCKADDR(struct sockaddr_irda *, addr, msg->msg_name); - err = -EINVAL; - /* Check address, extract pid. Jean II */ - if (msg->msg_namelen < sizeof(*addr)) - goto out; - if (addr->sir_family != AF_IRDA) - goto out; - - pid = addr->sir_lsap_sel; - if (pid & 0x80) { - pr_debug("%s(), extension in PID not supp!\n", - __func__); - err = -EOPNOTSUPP; - goto out; - } - } else { - /* Check that the socket is properly bound to an Ultra - * port. Jean II */ - if ((self->lsap == NULL) || - (sk->sk_state != TCP_ESTABLISHED)) { - pr_debug("%s(), socket not bound to Ultra PID.\n", - __func__); - err = -ENOTCONN; - goto out; - } - /* Use PID from socket */ - bound = 1; - } - - /* - * Check that we don't send out too big frames. This is an unreliable - * service, so we have no fragmentation and no coalescence - */ - if (len > self->max_data_size) { - pr_debug("%s(), Warning too much data! Chopping frame from %zd to %d bytes!\n", - __func__, len, self->max_data_size); - len = self->max_data_size; - } - - skb = sock_alloc_send_skb(sk, len + self->max_header_size, - msg->msg_flags & MSG_DONTWAIT, &err); - err = -ENOBUFS; - if (!skb) - goto out; - - skb_reserve(skb, self->max_header_size); - skb_reset_transport_header(skb); - - pr_debug("%s(), appending user data\n", __func__); - skb_put(skb, len); - err = memcpy_from_msg(skb_transport_header(skb), msg, len); - if (err) { - kfree_skb(skb); - goto out; - } - - err = irlmp_connless_data_request((bound ? self->lsap : NULL), - skb, pid); - if (err) - pr_debug("%s(), err=%d\n", __func__, err); -out: - release_sock(sk); - return err ? : len; -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irda_shutdown (sk, how) - */ -static int irda_shutdown(struct socket *sock, int how) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - - pr_debug("%s(%p)\n", __func__, self); - - lock_sock(sk); - - sk->sk_state = TCP_CLOSE; - sk->sk_shutdown |= SEND_SHUTDOWN; - sk->sk_state_change(sk); - - if (self->iriap) { - iriap_close(self->iriap); - self->iriap = NULL; - } - - if (self->tsap) { - irttp_disconnect_request(self->tsap, NULL, P_NORMAL); - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } - - /* A few cleanup so the socket look as good as new... */ - self->rx_flow = self->tx_flow = FLOW_START; /* needed ??? */ - self->daddr = DEV_ADDR_ANY; /* Until we get re-connected */ - self->saddr = 0x0; /* so IrLMP assign us any link */ - - release_sock(sk); - - return 0; -} - -/* - * Function irda_poll (file, sock, wait) - */ -static unsigned int irda_poll(struct file * file, struct socket *sock, - poll_table *wait) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - unsigned int mask; - - poll_wait(file, sk_sleep(sk), wait); - mask = 0; - - /* Exceptional events? */ - if (sk->sk_err) - mask |= POLLERR; - if (sk->sk_shutdown & RCV_SHUTDOWN) { - pr_debug("%s(), POLLHUP\n", __func__); - mask |= POLLHUP; - } - - /* Readable? */ - if (!skb_queue_empty(&sk->sk_receive_queue)) { - pr_debug("Socket is readable\n"); - mask |= POLLIN | POLLRDNORM; - } - - /* Connection-based need to check for termination and startup */ - switch (sk->sk_type) { - case SOCK_STREAM: - if (sk->sk_state == TCP_CLOSE) { - pr_debug("%s(), POLLHUP\n", __func__); - mask |= POLLHUP; - } - - if (sk->sk_state == TCP_ESTABLISHED) { - if ((self->tx_flow == FLOW_START) && - sock_writeable(sk)) - { - mask |= POLLOUT | POLLWRNORM | POLLWRBAND; - } - } - break; - case SOCK_SEQPACKET: - if ((self->tx_flow == FLOW_START) && - sock_writeable(sk)) - { - mask |= POLLOUT | POLLWRNORM | POLLWRBAND; - } - break; - case SOCK_DGRAM: - if (sock_writeable(sk)) - mask |= POLLOUT | POLLWRNORM | POLLWRBAND; - break; - default: - break; - } - - return mask; -} - -/* - * Function irda_ioctl (sock, cmd, arg) - */ -static int irda_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) -{ - struct sock *sk = sock->sk; - int err; - - pr_debug("%s(), cmd=%#x\n", __func__, cmd); - - err = -EINVAL; - switch (cmd) { - case TIOCOUTQ: { - long amount; - - amount = sk->sk_sndbuf - sk_wmem_alloc_get(sk); - if (amount < 0) - amount = 0; - err = put_user(amount, (unsigned int __user *)arg); - break; - } - - case TIOCINQ: { - struct sk_buff *skb; - long amount = 0L; - /* These two are safe on a single CPU system as only user tasks fiddle here */ - if ((skb = skb_peek(&sk->sk_receive_queue)) != NULL) - amount = skb->len; - err = put_user(amount, (unsigned int __user *)arg); - break; - } - - case SIOCGSTAMP: - if (sk != NULL) - err = sock_get_timestamp(sk, (struct timeval __user *)arg); - break; - - case SIOCGIFADDR: - case SIOCSIFADDR: - case SIOCGIFDSTADDR: - case SIOCSIFDSTADDR: - case SIOCGIFBRDADDR: - case SIOCSIFBRDADDR: - case SIOCGIFNETMASK: - case SIOCSIFNETMASK: - case SIOCGIFMETRIC: - case SIOCSIFMETRIC: - break; - default: - pr_debug("%s(), doing device ioctl!\n", __func__); - err = -ENOIOCTLCMD; - } - - return err; -} - -#ifdef CONFIG_COMPAT -/* - * Function irda_ioctl (sock, cmd, arg) - */ -static int irda_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) -{ - /* - * All IRDA's ioctl are standard ones. - */ - return -ENOIOCTLCMD; -} -#endif - -/* - * Function irda_setsockopt (sock, level, optname, optval, optlen) - * - * Set some options for the socket - * - */ -static int irda_setsockopt(struct socket *sock, int level, int optname, - char __user *optval, unsigned int optlen) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - struct irda_ias_set *ias_opt; - struct ias_object *ias_obj; - struct ias_attrib * ias_attr; /* Attribute in IAS object */ - int opt, free_ias = 0, err = 0; - - pr_debug("%s(%p)\n", __func__, self); - - if (level != SOL_IRLMP) - return -ENOPROTOOPT; - - lock_sock(sk); - - switch (optname) { - case IRLMP_IAS_SET: - /* The user want to add an attribute to an existing IAS object - * (in the IAS database) or to create a new object with this - * attribute. - * We first query IAS to know if the object exist, and then - * create the right attribute... - */ - - if (optlen != sizeof(struct irda_ias_set)) { - err = -EINVAL; - goto out; - } - - /* Copy query to the driver. */ - ias_opt = memdup_user(optval, optlen); - if (IS_ERR(ias_opt)) { - err = PTR_ERR(ias_opt); - goto out; - } - - /* Find the object we target. - * If the user gives us an empty string, we use the object - * associated with this socket. This will workaround - * duplicated class name - Jean II */ - if(ias_opt->irda_class_name[0] == '\0') { - if(self->ias_obj == NULL) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - ias_obj = self->ias_obj; - } else - ias_obj = irias_find_object(ias_opt->irda_class_name); - - /* Only ROOT can mess with the global IAS database. - * Users can only add attributes to the object associated - * with the socket they own - Jean II */ - if((!capable(CAP_NET_ADMIN)) && - ((ias_obj == NULL) || (ias_obj != self->ias_obj))) { - kfree(ias_opt); - err = -EPERM; - goto out; - } - - /* If the object doesn't exist, create it */ - if(ias_obj == (struct ias_object *) NULL) { - /* Create a new object */ - ias_obj = irias_new_object(ias_opt->irda_class_name, - jiffies); - if (ias_obj == NULL) { - kfree(ias_opt); - err = -ENOMEM; - goto out; - } - free_ias = 1; - } - - /* Do we have the attribute already ? */ - if(irias_find_attrib(ias_obj, ias_opt->irda_attrib_name)) { - kfree(ias_opt); - if (free_ias) { - kfree(ias_obj->name); - kfree(ias_obj); - } - err = -EINVAL; - goto out; - } - - /* Look at the type */ - switch(ias_opt->irda_attrib_type) { - case IAS_INTEGER: - /* Add an integer attribute */ - irias_add_integer_attrib( - ias_obj, - ias_opt->irda_attrib_name, - ias_opt->attribute.irda_attrib_int, - IAS_USER_ATTR); - break; - case IAS_OCT_SEQ: - /* Check length */ - if(ias_opt->attribute.irda_attrib_octet_seq.len > - IAS_MAX_OCTET_STRING) { - kfree(ias_opt); - if (free_ias) { - kfree(ias_obj->name); - kfree(ias_obj); - } - - err = -EINVAL; - goto out; - } - /* Add an octet sequence attribute */ - irias_add_octseq_attrib( - ias_obj, - ias_opt->irda_attrib_name, - ias_opt->attribute.irda_attrib_octet_seq.octet_seq, - ias_opt->attribute.irda_attrib_octet_seq.len, - IAS_USER_ATTR); - break; - case IAS_STRING: - /* Should check charset & co */ - /* Check length */ - /* The length is encoded in a __u8, and - * IAS_MAX_STRING == 256, so there is no way - * userspace can pass us a string too large. - * Jean II */ - /* NULL terminate the string (avoid troubles) */ - ias_opt->attribute.irda_attrib_string.string[ias_opt->attribute.irda_attrib_string.len] = '\0'; - /* Add a string attribute */ - irias_add_string_attrib( - ias_obj, - ias_opt->irda_attrib_name, - ias_opt->attribute.irda_attrib_string.string, - IAS_USER_ATTR); - break; - default : - kfree(ias_opt); - if (free_ias) { - kfree(ias_obj->name); - kfree(ias_obj); - } - err = -EINVAL; - goto out; - } - irias_insert_object(ias_obj); - kfree(ias_opt); - break; - case IRLMP_IAS_DEL: - /* The user want to delete an object from our local IAS - * database. We just need to query the IAS, check is the - * object is not owned by the kernel and delete it. - */ - - if (optlen != sizeof(struct irda_ias_set)) { - err = -EINVAL; - goto out; - } - - /* Copy query to the driver. */ - ias_opt = memdup_user(optval, optlen); - if (IS_ERR(ias_opt)) { - err = PTR_ERR(ias_opt); - goto out; - } - - /* Find the object we target. - * If the user gives us an empty string, we use the object - * associated with this socket. This will workaround - * duplicated class name - Jean II */ - if(ias_opt->irda_class_name[0] == '\0') - ias_obj = self->ias_obj; - else - ias_obj = irias_find_object(ias_opt->irda_class_name); - if(ias_obj == (struct ias_object *) NULL) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - - /* Only ROOT can mess with the global IAS database. - * Users can only del attributes from the object associated - * with the socket they own - Jean II */ - if((!capable(CAP_NET_ADMIN)) && - ((ias_obj == NULL) || (ias_obj != self->ias_obj))) { - kfree(ias_opt); - err = -EPERM; - goto out; - } - - /* Find the attribute (in the object) we target */ - ias_attr = irias_find_attrib(ias_obj, - ias_opt->irda_attrib_name); - if(ias_attr == (struct ias_attrib *) NULL) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - - /* Check is the user space own the object */ - if(ias_attr->value->owner != IAS_USER_ATTR) { - pr_debug("%s(), attempting to delete a kernel attribute\n", - __func__); - kfree(ias_opt); - err = -EPERM; - goto out; - } - - /* Remove the attribute (and maybe the object) */ - irias_delete_attrib(ias_obj, ias_attr, 1); - kfree(ias_opt); - break; - case IRLMP_MAX_SDU_SIZE: - if (optlen < sizeof(int)) { - err = -EINVAL; - goto out; - } - - if (get_user(opt, (int __user *)optval)) { - err = -EFAULT; - goto out; - } - - /* Only possible for a seqpacket service (TTP with SAR) */ - if (sk->sk_type != SOCK_SEQPACKET) { - pr_debug("%s(), setting max_sdu_size = %d\n", - __func__, opt); - self->max_sdu_size_rx = opt; - } else { - net_warn_ratelimited("%s: not allowed to set MAXSDUSIZE for this socket type!\n", - __func__); - err = -ENOPROTOOPT; - goto out; - } - break; - case IRLMP_HINTS_SET: - if (optlen < sizeof(int)) { - err = -EINVAL; - goto out; - } - - /* The input is really a (__u8 hints[2]), easier as an int */ - if (get_user(opt, (int __user *)optval)) { - err = -EFAULT; - goto out; - } - - /* Unregister any old registration */ - irlmp_unregister_service(self->skey); - - self->skey = irlmp_register_service((__u16) opt); - break; - case IRLMP_HINT_MASK_SET: - /* As opposed to the previous case which set the hint bits - * that we advertise, this one set the filter we use when - * making a discovery (nodes which don't match any hint - * bit in the mask are not reported). - */ - if (optlen < sizeof(int)) { - err = -EINVAL; - goto out; - } - - /* The input is really a (__u8 hints[2]), easier as an int */ - if (get_user(opt, (int __user *)optval)) { - err = -EFAULT; - goto out; - } - - /* Set the new hint mask */ - self->mask.word = (__u16) opt; - /* Mask out extension bits */ - self->mask.word &= 0x7f7f; - /* Check if no bits */ - if(!self->mask.word) - self->mask.word = 0xFFFF; - - break; - default: - err = -ENOPROTOOPT; - break; - } - -out: - release_sock(sk); - - return err; -} - -/* - * Function irda_extract_ias_value(ias_opt, ias_value) - * - * Translate internal IAS value structure to the user space representation - * - * The external representation of IAS values, as we exchange them with - * user space program is quite different from the internal representation, - * as stored in the IAS database (because we need a flat structure for - * crossing kernel boundary). - * This function transform the former in the latter. We also check - * that the value type is valid. - */ -static int irda_extract_ias_value(struct irda_ias_set *ias_opt, - struct ias_value *ias_value) -{ - /* Look at the type */ - switch (ias_value->type) { - case IAS_INTEGER: - /* Copy the integer */ - ias_opt->attribute.irda_attrib_int = ias_value->t.integer; - break; - case IAS_OCT_SEQ: - /* Set length */ - ias_opt->attribute.irda_attrib_octet_seq.len = ias_value->len; - /* Copy over */ - memcpy(ias_opt->attribute.irda_attrib_octet_seq.octet_seq, - ias_value->t.oct_seq, ias_value->len); - break; - case IAS_STRING: - /* Set length */ - ias_opt->attribute.irda_attrib_string.len = ias_value->len; - ias_opt->attribute.irda_attrib_string.charset = ias_value->charset; - /* Copy over */ - memcpy(ias_opt->attribute.irda_attrib_string.string, - ias_value->t.string, ias_value->len); - /* NULL terminate the string (avoid troubles) */ - ias_opt->attribute.irda_attrib_string.string[ias_value->len] = '\0'; - break; - case IAS_MISSING: - default : - return -EINVAL; - } - - /* Copy type over */ - ias_opt->irda_attrib_type = ias_value->type; - - return 0; -} - -/* - * Function irda_getsockopt (sock, level, optname, optval, optlen) - */ -static int irda_getsockopt(struct socket *sock, int level, int optname, - char __user *optval, int __user *optlen) -{ - struct sock *sk = sock->sk; - struct irda_sock *self = irda_sk(sk); - struct irda_device_list list; - struct irda_device_info *discoveries; - struct irda_ias_set * ias_opt; /* IAS get/query params */ - struct ias_object * ias_obj; /* Object in IAS */ - struct ias_attrib * ias_attr; /* Attribute in IAS object */ - int daddr = DEV_ADDR_ANY; /* Dest address for IAS queries */ - int val = 0; - int len = 0; - int err = 0; - int offset, total; - - pr_debug("%s(%p)\n", __func__, self); - - if (level != SOL_IRLMP) - return -ENOPROTOOPT; - - if (get_user(len, optlen)) - return -EFAULT; - - if(len < 0) - return -EINVAL; - - lock_sock(sk); - - switch (optname) { - case IRLMP_ENUMDEVICES: - - /* Offset to first device entry */ - offset = sizeof(struct irda_device_list) - - sizeof(struct irda_device_info); - - if (len < offset) { - err = -EINVAL; - goto out; - } - - /* Ask lmp for the current discovery log */ - discoveries = irlmp_get_discoveries(&list.len, self->mask.word, - self->nslots); - /* Check if the we got some results */ - if (discoveries == NULL) { - err = -EAGAIN; - goto out; /* Didn't find any devices */ - } - - /* Write total list length back to client */ - if (copy_to_user(optval, &list, offset)) - err = -EFAULT; - - /* Copy the list itself - watch for overflow */ - if (list.len > 2048) { - err = -EINVAL; - goto bed; - } - total = offset + (list.len * sizeof(struct irda_device_info)); - if (total > len) - total = len; - if (copy_to_user(optval+offset, discoveries, total - offset)) - err = -EFAULT; - - /* Write total number of bytes used back to client */ - if (put_user(total, optlen)) - err = -EFAULT; -bed: - /* Free up our buffer */ - kfree(discoveries); - break; - case IRLMP_MAX_SDU_SIZE: - val = self->max_data_size; - len = sizeof(int); - if (put_user(len, optlen)) { - err = -EFAULT; - goto out; - } - - if (copy_to_user(optval, &val, len)) { - err = -EFAULT; - goto out; - } - - break; - case IRLMP_IAS_GET: - /* The user want an object from our local IAS database. - * We just need to query the IAS and return the value - * that we found */ - - /* Check that the user has allocated the right space for us */ - if (len != sizeof(struct irda_ias_set)) { - err = -EINVAL; - goto out; - } - - /* Copy query to the driver. */ - ias_opt = memdup_user(optval, len); - if (IS_ERR(ias_opt)) { - err = PTR_ERR(ias_opt); - goto out; - } - - /* Find the object we target. - * If the user gives us an empty string, we use the object - * associated with this socket. This will workaround - * duplicated class name - Jean II */ - if(ias_opt->irda_class_name[0] == '\0') - ias_obj = self->ias_obj; - else - ias_obj = irias_find_object(ias_opt->irda_class_name); - if(ias_obj == (struct ias_object *) NULL) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - - /* Find the attribute (in the object) we target */ - ias_attr = irias_find_attrib(ias_obj, - ias_opt->irda_attrib_name); - if(ias_attr == (struct ias_attrib *) NULL) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - - /* Translate from internal to user structure */ - err = irda_extract_ias_value(ias_opt, ias_attr->value); - if(err) { - kfree(ias_opt); - goto out; - } - - /* Copy reply to the user */ - if (copy_to_user(optval, ias_opt, - sizeof(struct irda_ias_set))) { - kfree(ias_opt); - err = -EFAULT; - goto out; - } - /* Note : don't need to put optlen, we checked it */ - kfree(ias_opt); - break; - case IRLMP_IAS_QUERY: - /* The user want an object from a remote IAS database. - * We need to use IAP to query the remote database and - * then wait for the answer to come back. */ - - /* Check that the user has allocated the right space for us */ - if (len != sizeof(struct irda_ias_set)) { - err = -EINVAL; - goto out; - } - - /* Copy query to the driver. */ - ias_opt = memdup_user(optval, len); - if (IS_ERR(ias_opt)) { - err = PTR_ERR(ias_opt); - goto out; - } - - /* At this point, there are two cases... - * 1) the socket is connected - that's the easy case, we - * just query the device we are connected to... - * 2) the socket is not connected - the user doesn't want - * to connect and/or may not have a valid service name - * (so can't create a fake connection). In this case, - * we assume that the user pass us a valid destination - * address in the requesting structure... - */ - if(self->daddr != DEV_ADDR_ANY) { - /* We are connected - reuse known daddr */ - daddr = self->daddr; - } else { - /* We are not connected, we must specify a valid - * destination address */ - daddr = ias_opt->daddr; - if((!daddr) || (daddr == DEV_ADDR_ANY)) { - kfree(ias_opt); - err = -EINVAL; - goto out; - } - } - - /* Check that we can proceed with IAP */ - if (self->iriap) { - net_warn_ratelimited("%s: busy with a previous query\n", - __func__); - kfree(ias_opt); - err = -EBUSY; - goto out; - } - - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - irda_getvalue_confirm); - - if (self->iriap == NULL) { - kfree(ias_opt); - err = -ENOMEM; - goto out; - } - - /* Treat unexpected wakeup as disconnect */ - self->errno = -EHOSTUNREACH; - - /* Query remote LM-IAS */ - iriap_getvaluebyclass_request(self->iriap, - self->saddr, daddr, - ias_opt->irda_class_name, - ias_opt->irda_attrib_name); - - /* Wait for answer, if not yet finished (or failed) */ - if (wait_event_interruptible(self->query_wait, - (self->iriap == NULL))) { - /* pending request uses copy of ias_opt-content - * we can free it regardless! */ - kfree(ias_opt); - /* Treat signals as disconnect */ - err = -EHOSTUNREACH; - goto out; - } - - /* Check what happened */ - if (self->errno) - { - kfree(ias_opt); - /* Requested object/attribute doesn't exist */ - if((self->errno == IAS_CLASS_UNKNOWN) || - (self->errno == IAS_ATTRIB_UNKNOWN)) - err = -EADDRNOTAVAIL; - else - err = -EHOSTUNREACH; - - goto out; - } - - /* Translate from internal to user structure */ - err = irda_extract_ias_value(ias_opt, self->ias_result); - if (self->ias_result) - irias_delete_value(self->ias_result); - if (err) { - kfree(ias_opt); - goto out; - } - - /* Copy reply to the user */ - if (copy_to_user(optval, ias_opt, - sizeof(struct irda_ias_set))) { - kfree(ias_opt); - err = -EFAULT; - goto out; - } - /* Note : don't need to put optlen, we checked it */ - kfree(ias_opt); - break; - case IRLMP_WAITDEVICE: - /* This function is just another way of seeing life ;-) - * IRLMP_ENUMDEVICES assumes that you have a static network, - * and that you just want to pick one of the devices present. - * On the other hand, in here we assume that no device is - * present and that at some point in the future a device will - * come into range. When this device arrive, we just wake - * up the caller, so that he has time to connect to it before - * the device goes away... - * Note : once the node has been discovered for more than a - * few second, it won't trigger this function, unless it - * goes away and come back changes its hint bits (so we - * might call it IRLMP_WAITNEWDEVICE). - */ - - /* Check that the user is passing us an int */ - if (len != sizeof(int)) { - err = -EINVAL; - goto out; - } - /* Get timeout in ms (max time we block the caller) */ - if (get_user(val, (int __user *)optval)) { - err = -EFAULT; - goto out; - } - - /* Tell IrLMP we want to be notified */ - irlmp_update_client(self->ckey, self->mask.word, - irda_selective_discovery_indication, - NULL, (void *) self); - - /* Do some discovery (and also return cached results) */ - irlmp_discovery_request(self->nslots); - - /* Wait until a node is discovered */ - if (!self->cachedaddr) { - pr_debug("%s(), nothing discovered yet, going to sleep...\n", - __func__); - - /* Set watchdog timer to expire in <val> ms. */ - self->errno = 0; - setup_timer(&self->watchdog, irda_discovery_timeout, - (unsigned long)self); - mod_timer(&self->watchdog, - jiffies + msecs_to_jiffies(val)); - - /* Wait for IR-LMP to call us back */ - err = __wait_event_interruptible(self->query_wait, - (self->cachedaddr != 0 || self->errno == -ETIME)); - - /* If watchdog is still activated, kill it! */ - del_timer(&(self->watchdog)); - - pr_debug("%s(), ...waking up !\n", __func__); - - if (err != 0) - goto out; - } - else - pr_debug("%s(), found immediately !\n", - __func__); - - /* Tell IrLMP that we have been notified */ - irlmp_update_client(self->ckey, self->mask.word, - NULL, NULL, NULL); - - /* Check if the we got some results */ - if (!self->cachedaddr) { - err = -EAGAIN; /* Didn't find any devices */ - goto out; - } - daddr = self->cachedaddr; - /* Cleanup */ - self->cachedaddr = 0; - - /* We return the daddr of the device that trigger the - * wakeup. As irlmp pass us only the new devices, we - * are sure that it's not an old device. - * If the user want more details, he should query - * the whole discovery log and pick one device... - */ - if (put_user(daddr, (int __user *)optval)) { - err = -EFAULT; - goto out; - } - - break; - default: - err = -ENOPROTOOPT; - } - -out: - - release_sock(sk); - - return err; -} - -static const struct net_proto_family irda_family_ops = { - .family = PF_IRDA, - .create = irda_create, - .owner = THIS_MODULE, -}; - -static const struct proto_ops irda_stream_ops = { - .family = PF_IRDA, - .owner = THIS_MODULE, - .release = irda_release, - .bind = irda_bind, - .connect = irda_connect, - .socketpair = sock_no_socketpair, - .accept = irda_accept, - .getname = irda_getname, - .poll = irda_poll, - .ioctl = irda_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = irda_compat_ioctl, -#endif - .listen = irda_listen, - .shutdown = irda_shutdown, - .setsockopt = irda_setsockopt, - .getsockopt = irda_getsockopt, - .sendmsg = irda_sendmsg, - .recvmsg = irda_recvmsg_stream, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage, -}; - -static const struct proto_ops irda_seqpacket_ops = { - .family = PF_IRDA, - .owner = THIS_MODULE, - .release = irda_release, - .bind = irda_bind, - .connect = irda_connect, - .socketpair = sock_no_socketpair, - .accept = irda_accept, - .getname = irda_getname, - .poll = datagram_poll, - .ioctl = irda_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = irda_compat_ioctl, -#endif - .listen = irda_listen, - .shutdown = irda_shutdown, - .setsockopt = irda_setsockopt, - .getsockopt = irda_getsockopt, - .sendmsg = irda_sendmsg, - .recvmsg = irda_recvmsg_dgram, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage, -}; - -static const struct proto_ops irda_dgram_ops = { - .family = PF_IRDA, - .owner = THIS_MODULE, - .release = irda_release, - .bind = irda_bind, - .connect = irda_connect, - .socketpair = sock_no_socketpair, - .accept = irda_accept, - .getname = irda_getname, - .poll = datagram_poll, - .ioctl = irda_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = irda_compat_ioctl, -#endif - .listen = irda_listen, - .shutdown = irda_shutdown, - .setsockopt = irda_setsockopt, - .getsockopt = irda_getsockopt, - .sendmsg = irda_sendmsg_dgram, - .recvmsg = irda_recvmsg_dgram, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage, -}; - -#ifdef CONFIG_IRDA_ULTRA -static const struct proto_ops irda_ultra_ops = { - .family = PF_IRDA, - .owner = THIS_MODULE, - .release = irda_release, - .bind = irda_bind, - .connect = sock_no_connect, - .socketpair = sock_no_socketpair, - .accept = sock_no_accept, - .getname = irda_getname, - .poll = datagram_poll, - .ioctl = irda_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = irda_compat_ioctl, -#endif - .listen = sock_no_listen, - .shutdown = irda_shutdown, - .setsockopt = irda_setsockopt, - .getsockopt = irda_getsockopt, - .sendmsg = irda_sendmsg_ultra, - .recvmsg = irda_recvmsg_dgram, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage, -}; -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irsock_init (pro) - * - * Initialize IrDA protocol - * - */ -int __init irsock_init(void) -{ - int rc = proto_register(&irda_proto, 0); - - if (rc == 0) - rc = sock_register(&irda_family_ops); - - return rc; -} - -/* - * Function irsock_cleanup (void) - * - * Remove IrDA protocol - * - */ -void irsock_cleanup(void) -{ - sock_unregister(PF_IRDA); - proto_unregister(&irda_proto); -} diff --git a/net/irda/discovery.c b/net/irda/discovery.c deleted file mode 100644 index 364d70aed068..000000000000 --- a/net/irda/discovery.c +++ /dev/null @@ -1,417 +0,0 @@ -/********************************************************************* - * - * Filename: discovery.c - * Version: 0.1 - * Description: Routines for handling discoveries at the IrLMP layer - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Apr 6 15:33:50 1999 - * Modified at: Sat Oct 9 17:11:31 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Modified at: Fri May 28 3:11 CST 1999 - * Modified by: Horst von Brand <vonbrand@sleipnir.valparaiso.cl> - * - * Copyright (c) 1999 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/string.h> -#include <linux/socket.h> -#include <linux/fs.h> -#include <linux/seq_file.h> -#include <linux/slab.h> -#include <linux/export.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> - -#include <net/irda/discovery.h> - -#include <asm/unaligned.h> - -/* - * Function irlmp_add_discovery (cachelog, discovery) - * - * Add a new discovery to the cachelog, and remove any old discoveries - * from the same device - * - * Note : we try to preserve the time this device was *first* discovered - * (as opposed to the time of last discovery used for cleanup). This is - * used by clients waiting for discovery events to tell if the device - * discovered is "new" or just the same old one. They can't rely there - * on a binary flag (new/old), because not all discovery events are - * propagated to them, and they might not always listen, so they would - * miss some new devices popping up... - * Jean II - */ -void irlmp_add_discovery(hashbin_t *cachelog, discovery_t *new) -{ - discovery_t *discovery, *node; - unsigned long flags; - - /* Set time of first discovery if node is new (see below) */ - new->firststamp = new->timestamp; - - spin_lock_irqsave(&cachelog->hb_spinlock, flags); - - /* - * Remove all discoveries of devices that has previously been - * discovered on the same link with the same name (info), or the - * same daddr. We do this since some devices (mostly PDAs) change - * their device address between every discovery. - */ - discovery = (discovery_t *) hashbin_get_first(cachelog); - while (discovery != NULL ) { - node = discovery; - - /* Be sure to stay one item ahead */ - discovery = (discovery_t *) hashbin_get_next(cachelog); - - if ((node->data.saddr == new->data.saddr) && - ((node->data.daddr == new->data.daddr) || - (strcmp(node->data.info, new->data.info) == 0))) - { - /* This discovery is a previous discovery - * from the same device, so just remove it - */ - hashbin_remove_this(cachelog, (irda_queue_t *) node); - /* Check if hints bits are unchanged */ - if (get_unaligned((__u16 *)node->data.hints) == get_unaligned((__u16 *)new->data.hints)) - /* Set time of first discovery for this node */ - new->firststamp = node->firststamp; - kfree(node); - } - } - - /* Insert the new and updated version */ - hashbin_insert(cachelog, (irda_queue_t *) new, new->data.daddr, NULL); - - spin_unlock_irqrestore(&cachelog->hb_spinlock, flags); -} - -/* - * Function irlmp_add_discovery_log (cachelog, log) - * - * Merge a disovery log into the cachelog. - * - */ -void irlmp_add_discovery_log(hashbin_t *cachelog, hashbin_t *log) -{ - discovery_t *discovery; - - /* - * If log is missing this means that IrLAP was unable to perform the - * discovery, so restart discovery again with just the half timeout - * of the normal one. - */ - /* Well... It means that there was nobody out there - Jean II */ - if (log == NULL) { - /* irlmp_start_discovery_timer(irlmp, 150); */ - return; - } - - /* - * Locking : we are the only owner of this discovery log, so - * no need to lock it. - * We just need to lock the global log in irlmp_add_discovery(). - */ - discovery = (discovery_t *) hashbin_remove_first(log); - while (discovery != NULL) { - irlmp_add_discovery(cachelog, discovery); - - discovery = (discovery_t *) hashbin_remove_first(log); - } - - /* Delete the now empty log */ - hashbin_delete(log, (FREE_FUNC) kfree); -} - -/* - * Function irlmp_expire_discoveries (log, saddr, force) - * - * Go through all discoveries and expire all that has stayed too long - * - * Note : this assume that IrLAP won't change its saddr, which - * currently is a valid assumption... - */ -void irlmp_expire_discoveries(hashbin_t *log, __u32 saddr, int force) -{ - discovery_t * discovery; - discovery_t * curr; - unsigned long flags; - discinfo_t * buffer = NULL; - int n; /* Size of the full log */ - int i = 0; /* How many we expired */ - - IRDA_ASSERT(log != NULL, return;); - spin_lock_irqsave(&log->hb_spinlock, flags); - - discovery = (discovery_t *) hashbin_get_first(log); - while (discovery != NULL) { - /* Be sure to be one item ahead */ - curr = discovery; - discovery = (discovery_t *) hashbin_get_next(log); - - /* Test if it's time to expire this discovery */ - if ((curr->data.saddr == saddr) && - (force || - ((jiffies - curr->timestamp) > DISCOVERY_EXPIRE_TIMEOUT))) - { - /* Create buffer as needed. - * As this function get called a lot and most time - * we don't have anything to put in the log (we are - * quite picky), we can save a lot of overhead - * by not calling kmalloc. Jean II */ - if(buffer == NULL) { - /* Create the client specific buffer */ - n = HASHBIN_GET_SIZE(log); - buffer = kmalloc(n * sizeof(struct irda_device_info), GFP_ATOMIC); - if (buffer == NULL) { - spin_unlock_irqrestore(&log->hb_spinlock, flags); - return; - } - - } - - /* Copy discovery information */ - memcpy(&(buffer[i]), &(curr->data), - sizeof(discinfo_t)); - i++; - - /* Remove it from the log */ - curr = hashbin_remove_this(log, (irda_queue_t *) curr); - kfree(curr); - } - } - - /* Drop the spinlock before calling the higher layers, as - * we can't guarantee they won't call us back and create a - * deadlock. We will work on our own private data, so we - * don't care to be interrupted. - Jean II */ - spin_unlock_irqrestore(&log->hb_spinlock, flags); - - if(buffer == NULL) - return; - - /* Tell IrLMP and registered clients about it */ - irlmp_discovery_expiry(buffer, i); - - /* Free up our buffer */ - kfree(buffer); -} - -#if 0 -/* - * Function irlmp_dump_discoveries (log) - * - * Print out all discoveries in log - * - */ -void irlmp_dump_discoveries(hashbin_t *log) -{ - discovery_t *discovery; - - IRDA_ASSERT(log != NULL, return;); - - discovery = (discovery_t *) hashbin_get_first(log); - while (discovery != NULL) { - pr_debug("Discovery:\n"); - pr_debug(" daddr=%08x\n", discovery->data.daddr); - pr_debug(" saddr=%08x\n", discovery->data.saddr); - pr_debug(" nickname=%s\n", discovery->data.info); - - discovery = (discovery_t *) hashbin_get_next(log); - } -} -#endif - -/* - * Function irlmp_copy_discoveries (log, pn, mask) - * - * Copy all discoveries in a buffer - * - * This function implement a safe way for lmp clients to access the - * discovery log. The basic problem is that we don't want the log - * to change (add/remove) while the client is reading it. If the - * lmp client manipulate directly the hashbin, he is sure to get - * into troubles... - * The idea is that we copy all the current discovery log in a buffer - * which is specific to the client and pass this copy to him. As we - * do this operation with the spinlock grabbed, we are safe... - * Note : we don't want those clients to grab the spinlock, because - * we have no control on how long they will hold it... - * Note : we choose to copy the log in "struct irda_device_info" to - * save space... - * Note : the client must kfree himself() the log... - * Jean II - */ -struct irda_device_info *irlmp_copy_discoveries(hashbin_t *log, int *pn, - __u16 mask, int old_entries) -{ - discovery_t * discovery; - unsigned long flags; - discinfo_t * buffer = NULL; - int j_timeout = (sysctl_discovery_timeout * HZ); - int n; /* Size of the full log */ - int i = 0; /* How many we picked */ - - IRDA_ASSERT(pn != NULL, return NULL;); - IRDA_ASSERT(log != NULL, return NULL;); - - /* Save spin lock */ - spin_lock_irqsave(&log->hb_spinlock, flags); - - discovery = (discovery_t *) hashbin_get_first(log); - while (discovery != NULL) { - /* Mask out the ones we don't want : - * We want to match the discovery mask, and to get only - * the most recent one (unless we want old ones) */ - if ((get_unaligned((__u16 *)discovery->data.hints) & mask) && - ((old_entries) || - ((jiffies - discovery->firststamp) < j_timeout))) { - /* Create buffer as needed. - * As this function get called a lot and most time - * we don't have anything to put in the log (we are - * quite picky), we can save a lot of overhead - * by not calling kmalloc. Jean II */ - if(buffer == NULL) { - /* Create the client specific buffer */ - n = HASHBIN_GET_SIZE(log); - buffer = kmalloc(n * sizeof(struct irda_device_info), GFP_ATOMIC); - if (buffer == NULL) { - spin_unlock_irqrestore(&log->hb_spinlock, flags); - return NULL; - } - - } - - /* Copy discovery information */ - memcpy(&(buffer[i]), &(discovery->data), - sizeof(discinfo_t)); - i++; - } - discovery = (discovery_t *) hashbin_get_next(log); - } - - spin_unlock_irqrestore(&log->hb_spinlock, flags); - - /* Get the actual number of device in the buffer and return */ - *pn = i; - return buffer; -} - -#ifdef CONFIG_PROC_FS -static inline discovery_t *discovery_seq_idx(loff_t pos) - -{ - discovery_t *discovery; - - for (discovery = (discovery_t *) hashbin_get_first(irlmp->cachelog); - discovery != NULL; - discovery = (discovery_t *) hashbin_get_next(irlmp->cachelog)) { - if (pos-- == 0) - break; - } - - return discovery; -} - -static void *discovery_seq_start(struct seq_file *seq, loff_t *pos) -{ - spin_lock_irq(&irlmp->cachelog->hb_spinlock); - return *pos ? discovery_seq_idx(*pos - 1) : SEQ_START_TOKEN; -} - -static void *discovery_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - ++*pos; - return (v == SEQ_START_TOKEN) - ? (void *) hashbin_get_first(irlmp->cachelog) - : (void *) hashbin_get_next(irlmp->cachelog); -} - -static void discovery_seq_stop(struct seq_file *seq, void *v) -{ - spin_unlock_irq(&irlmp->cachelog->hb_spinlock); -} - -static int discovery_seq_show(struct seq_file *seq, void *v) -{ - if (v == SEQ_START_TOKEN) - seq_puts(seq, "IrLMP: Discovery log:\n\n"); - else { - const discovery_t *discovery = v; - - seq_printf(seq, "nickname: %s, hint: 0x%02x%02x", - discovery->data.info, - discovery->data.hints[0], - discovery->data.hints[1]); -#if 0 - if ( discovery->data.hints[0] & HINT_PNP) - seq_puts(seq, "PnP Compatible "); - if ( discovery->data.hints[0] & HINT_PDA) - seq_puts(seq, "PDA/Palmtop "); - if ( discovery->data.hints[0] & HINT_COMPUTER) - seq_puts(seq, "Computer "); - if ( discovery->data.hints[0] & HINT_PRINTER) - seq_puts(seq, "Printer "); - if ( discovery->data.hints[0] & HINT_MODEM) - seq_puts(seq, "Modem "); - if ( discovery->data.hints[0] & HINT_FAX) - seq_puts(seq, "Fax "); - if ( discovery->data.hints[0] & HINT_LAN) - seq_puts(seq, "LAN Access "); - - if ( discovery->data.hints[1] & HINT_TELEPHONY) - seq_puts(seq, "Telephony "); - if ( discovery->data.hints[1] & HINT_FILE_SERVER) - seq_puts(seq, "File Server "); - if ( discovery->data.hints[1] & HINT_COMM) - seq_puts(seq, "IrCOMM "); - if ( discovery->data.hints[1] & HINT_OBEX) - seq_puts(seq, "IrOBEX "); -#endif - seq_printf(seq,", saddr: 0x%08x, daddr: 0x%08x\n\n", - discovery->data.saddr, - discovery->data.daddr); - - seq_putc(seq, '\n'); - } - return 0; -} - -static const struct seq_operations discovery_seq_ops = { - .start = discovery_seq_start, - .next = discovery_seq_next, - .stop = discovery_seq_stop, - .show = discovery_seq_show, -}; - -static int discovery_seq_open(struct inode *inode, struct file *file) -{ - IRDA_ASSERT(irlmp != NULL, return -EINVAL;); - - return seq_open(file, &discovery_seq_ops); -} - -const struct file_operations discovery_seq_fops = { - .owner = THIS_MODULE, - .open = discovery_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; -#endif diff --git a/net/irda/ircomm/Kconfig b/net/irda/ircomm/Kconfig deleted file mode 100644 index 19492c1707b7..000000000000 --- a/net/irda/ircomm/Kconfig +++ /dev/null @@ -1,12 +0,0 @@ -config IRCOMM - tristate "IrCOMM protocol" - depends on IRDA && TTY - help - Say Y here if you want to build support for the IrCOMM protocol. - To compile it as modules, choose M here: the modules will be - called ircomm and ircomm_tty. - IrCOMM implements serial port emulation, and makes it possible to - use all existing applications that understands TTY's with an - infrared link. Thus you should be able to use application like PPP, - minicom and others. - diff --git a/net/irda/ircomm/Makefile b/net/irda/ircomm/Makefile deleted file mode 100644 index ab23b5ba7e33..000000000000 --- a/net/irda/ircomm/Makefile +++ /dev/null @@ -1,8 +0,0 @@ -# -# Makefile for the Linux IrDA IrCOMM protocol layer. -# - -obj-$(CONFIG_IRCOMM) += ircomm.o ircomm-tty.o - -ircomm-y := ircomm_core.o ircomm_event.o ircomm_lmp.o ircomm_ttp.o -ircomm-tty-y := ircomm_tty.o ircomm_tty_attach.o ircomm_tty_ioctl.o ircomm_param.o diff --git a/net/irda/ircomm/ircomm_core.c b/net/irda/ircomm/ircomm_core.c deleted file mode 100644 index 3af219545f6d..000000000000 --- a/net/irda/ircomm/ircomm_core.c +++ /dev/null @@ -1,563 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_core.c - * Version: 1.0 - * Description: IrCOMM service interface - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Jun 6 20:37:34 1999 - * Modified at: Tue Dec 21 13:26:41 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/module.h> -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/init.h> -#include <linux/slab.h> - -#include <net/irda/irda.h> -#include <net/irda/irmod.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/irttp.h> -#include <net/irda/irias_object.h> - -#include <net/irda/ircomm_event.h> -#include <net/irda/ircomm_lmp.h> -#include <net/irda/ircomm_ttp.h> -#include <net/irda/ircomm_param.h> -#include <net/irda/ircomm_core.h> - -static int __ircomm_close(struct ircomm_cb *self); -static void ircomm_control_indication(struct ircomm_cb *self, - struct sk_buff *skb, int clen); - -#ifdef CONFIG_PROC_FS -extern struct proc_dir_entry *proc_irda; -static int ircomm_seq_open(struct inode *, struct file *); - -static const struct file_operations ircomm_proc_fops = { - .owner = THIS_MODULE, - .open = ircomm_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; -#endif /* CONFIG_PROC_FS */ - -hashbin_t *ircomm = NULL; - -static int __init ircomm_init(void) -{ - ircomm = hashbin_new(HB_LOCK); - if (ircomm == NULL) { - net_err_ratelimited("%s(), can't allocate hashbin!\n", - __func__); - return -ENOMEM; - } - -#ifdef CONFIG_PROC_FS - { struct proc_dir_entry *ent; - ent = proc_create("ircomm", 0, proc_irda, &ircomm_proc_fops); - if (!ent) { - printk(KERN_ERR "ircomm_init: can't create /proc entry!\n"); - return -ENODEV; - } - } -#endif /* CONFIG_PROC_FS */ - - net_info_ratelimited("IrCOMM protocol (Dag Brattli)\n"); - - return 0; -} - -static void __exit ircomm_cleanup(void) -{ - hashbin_delete(ircomm, (FREE_FUNC) __ircomm_close); - -#ifdef CONFIG_PROC_FS - remove_proc_entry("ircomm", proc_irda); -#endif /* CONFIG_PROC_FS */ -} - -/* - * Function ircomm_open (client_notify) - * - * Start a new IrCOMM instance - * - */ -struct ircomm_cb *ircomm_open(notify_t *notify, __u8 service_type, int line) -{ - struct ircomm_cb *self = NULL; - int ret; - - pr_debug("%s(), service_type=0x%02x\n", __func__ , - service_type); - - IRDA_ASSERT(ircomm != NULL, return NULL;); - - self = kzalloc(sizeof(struct ircomm_cb), GFP_KERNEL); - if (self == NULL) - return NULL; - - self->notify = *notify; - self->magic = IRCOMM_MAGIC; - - /* Check if we should use IrLMP or IrTTP */ - if (service_type & IRCOMM_3_WIRE_RAW) { - self->flow_status = FLOW_START; - ret = ircomm_open_lsap(self); - } else - ret = ircomm_open_tsap(self); - - if (ret < 0) { - kfree(self); - return NULL; - } - - self->service_type = service_type; - self->line = line; - - hashbin_insert(ircomm, (irda_queue_t *) self, line, NULL); - - ircomm_next_state(self, IRCOMM_IDLE); - - return self; -} - -EXPORT_SYMBOL(ircomm_open); - -/* - * Function ircomm_close_instance (self) - * - * Remove IrCOMM instance - * - */ -static int __ircomm_close(struct ircomm_cb *self) -{ - /* Disconnect link if any */ - ircomm_do_event(self, IRCOMM_DISCONNECT_REQUEST, NULL, NULL); - - /* Remove TSAP */ - if (self->tsap) { - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } - - /* Remove LSAP */ - if (self->lsap) { - irlmp_close_lsap(self->lsap); - self->lsap = NULL; - } - self->magic = 0; - - kfree(self); - - return 0; -} - -/* - * Function ircomm_close (self) - * - * Closes and removes the specified IrCOMM instance - * - */ -int ircomm_close(struct ircomm_cb *self) -{ - struct ircomm_cb *entry; - - IRDA_ASSERT(self != NULL, return -EIO;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -EIO;); - - entry = hashbin_remove(ircomm, self->line, NULL); - - IRDA_ASSERT(entry == self, return -1;); - - return __ircomm_close(self); -} - -EXPORT_SYMBOL(ircomm_close); - -/* - * Function ircomm_connect_request (self, service_type) - * - * Impl. of this function is differ from one of the reference. This - * function does discovery as well as sending connect request - * - */ -int ircomm_connect_request(struct ircomm_cb *self, __u8 dlsap_sel, - __u32 saddr, __u32 daddr, struct sk_buff *skb, - __u8 service_type) -{ - struct ircomm_info info; - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -1;); - - self->service_type= service_type; - - info.dlsap_sel = dlsap_sel; - info.saddr = saddr; - info.daddr = daddr; - - ret = ircomm_do_event(self, IRCOMM_CONNECT_REQUEST, skb, &info); - - return ret; -} - -EXPORT_SYMBOL(ircomm_connect_request); - -/* - * Function ircomm_connect_indication (self, qos, skb) - * - * Notify user layer about the incoming connection - * - */ -void ircomm_connect_indication(struct ircomm_cb *self, struct sk_buff *skb, - struct ircomm_info *info) -{ - /* - * If there are any data hiding in the control channel, we must - * deliver it first. The side effect is that the control channel - * will be removed from the skb - */ - if (self->notify.connect_indication) - self->notify.connect_indication(self->notify.instance, self, - info->qos, info->max_data_size, - info->max_header_size, skb); - else { - pr_debug("%s(), missing handler\n", __func__); - } -} - -/* - * Function ircomm_connect_response (self, userdata, max_sdu_size) - * - * User accepts connection - * - */ -int ircomm_connect_response(struct ircomm_cb *self, struct sk_buff *userdata) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -1;); - - ret = ircomm_do_event(self, IRCOMM_CONNECT_RESPONSE, userdata, NULL); - - return ret; -} - -EXPORT_SYMBOL(ircomm_connect_response); - -/* - * Function connect_confirm (self, skb) - * - * Notify user layer that the link is now connected - * - */ -void ircomm_connect_confirm(struct ircomm_cb *self, struct sk_buff *skb, - struct ircomm_info *info) -{ - if (self->notify.connect_confirm ) - self->notify.connect_confirm(self->notify.instance, - self, info->qos, - info->max_data_size, - info->max_header_size, skb); - else { - pr_debug("%s(), missing handler\n", __func__); - } -} - -/* - * Function ircomm_data_request (self, userdata) - * - * Send IrCOMM data to peer device - * - */ -int ircomm_data_request(struct ircomm_cb *self, struct sk_buff *skb) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -EFAULT;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -EFAULT;); - IRDA_ASSERT(skb != NULL, return -EFAULT;); - - ret = ircomm_do_event(self, IRCOMM_DATA_REQUEST, skb, NULL); - - return ret; -} - -EXPORT_SYMBOL(ircomm_data_request); - -/* - * Function ircomm_data_indication (self, skb) - * - * Data arrived, so deliver it to user - * - */ -void ircomm_data_indication(struct ircomm_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(skb->len > 0, return;); - - if (self->notify.data_indication) - self->notify.data_indication(self->notify.instance, self, skb); - else { - pr_debug("%s(), missing handler\n", __func__); - } -} - -/* - * Function ircomm_process_data (self, skb) - * - * Data arrived which may contain control channel data - * - */ -void ircomm_process_data(struct ircomm_cb *self, struct sk_buff *skb) -{ - int clen; - - IRDA_ASSERT(skb->len > 0, return;); - - clen = skb->data[0]; - - /* - * Input validation check: a stir4200/mcp2150 combinations sometimes - * results in frames with clen > remaining packet size. These are - * illegal; if we throw away just this frame then it seems to carry on - * fine - */ - if (unlikely(skb->len < (clen + 1))) { - pr_debug("%s() throwing away illegal frame\n", - __func__); - return; - } - - /* - * If there are any data hiding in the control channel, we must - * deliver it first. The side effect is that the control channel - * will be removed from the skb - */ - if (clen > 0) - ircomm_control_indication(self, skb, clen); - - /* Remove control channel from data channel */ - skb_pull(skb, clen+1); - - if (skb->len) - ircomm_data_indication(self, skb); - else { - pr_debug("%s(), data was control info only!\n", - __func__); - } -} - -/* - * Function ircomm_control_request (self, params) - * - * Send control data to peer device - * - */ -int ircomm_control_request(struct ircomm_cb *self, struct sk_buff *skb) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -EFAULT;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -EFAULT;); - IRDA_ASSERT(skb != NULL, return -EFAULT;); - - ret = ircomm_do_event(self, IRCOMM_CONTROL_REQUEST, skb, NULL); - - return ret; -} - -EXPORT_SYMBOL(ircomm_control_request); - -/* - * Function ircomm_control_indication (self, skb) - * - * Data has arrived on the control channel - * - */ -static void ircomm_control_indication(struct ircomm_cb *self, - struct sk_buff *skb, int clen) -{ - /* Use udata for delivering data on the control channel */ - if (self->notify.udata_indication) { - struct sk_buff *ctrl_skb; - - /* We don't own the skb, so clone it */ - ctrl_skb = skb_clone(skb, GFP_ATOMIC); - if (!ctrl_skb) - return; - - /* Remove data channel from control channel */ - skb_trim(ctrl_skb, clen+1); - - self->notify.udata_indication(self->notify.instance, self, - ctrl_skb); - - /* Drop reference count - - * see ircomm_tty_control_indication(). */ - dev_kfree_skb(ctrl_skb); - } else { - pr_debug("%s(), missing handler\n", __func__); - } -} - -/* - * Function ircomm_disconnect_request (self, userdata, priority) - * - * User layer wants to disconnect the IrCOMM connection - * - */ -int ircomm_disconnect_request(struct ircomm_cb *self, struct sk_buff *userdata) -{ - struct ircomm_info info; - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -1;); - - ret = ircomm_do_event(self, IRCOMM_DISCONNECT_REQUEST, userdata, - &info); - return ret; -} - -EXPORT_SYMBOL(ircomm_disconnect_request); - -/* - * Function disconnect_indication (self, skb) - * - * Tell user that the link has been disconnected - * - */ -void ircomm_disconnect_indication(struct ircomm_cb *self, struct sk_buff *skb, - struct ircomm_info *info) -{ - IRDA_ASSERT(info != NULL, return;); - - if (self->notify.disconnect_indication) { - self->notify.disconnect_indication(self->notify.instance, self, - info->reason, skb); - } else { - pr_debug("%s(), missing handler\n", __func__); - } -} - -/* - * Function ircomm_flow_request (self, flow) - * - * - * - */ -void ircomm_flow_request(struct ircomm_cb *self, LOCAL_FLOW flow) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - - if (self->service_type == IRCOMM_3_WIRE_RAW) - return; - - irttp_flow_request(self->tsap, flow); -} - -EXPORT_SYMBOL(ircomm_flow_request); - -#ifdef CONFIG_PROC_FS -static void *ircomm_seq_start(struct seq_file *seq, loff_t *pos) -{ - struct ircomm_cb *self; - loff_t off = 0; - - spin_lock_irq(&ircomm->hb_spinlock); - - for (self = (struct ircomm_cb *) hashbin_get_first(ircomm); - self != NULL; - self = (struct ircomm_cb *) hashbin_get_next(ircomm)) { - if (off++ == *pos) - break; - - } - return self; -} - -static void *ircomm_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - ++*pos; - - return (void *) hashbin_get_next(ircomm); -} - -static void ircomm_seq_stop(struct seq_file *seq, void *v) -{ - spin_unlock_irq(&ircomm->hb_spinlock); -} - -static int ircomm_seq_show(struct seq_file *seq, void *v) -{ - const struct ircomm_cb *self = v; - - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -EINVAL; ); - - if(self->line < 0x10) - seq_printf(seq, "ircomm%d", self->line); - else - seq_printf(seq, "irlpt%d", self->line - 0x10); - - seq_printf(seq, - " state: %s, slsap_sel: %#02x, dlsap_sel: %#02x, mode:", - ircomm_state[ self->state], - self->slsap_sel, self->dlsap_sel); - - if(self->service_type & IRCOMM_3_WIRE_RAW) - seq_printf(seq, " 3-wire-raw"); - if(self->service_type & IRCOMM_3_WIRE) - seq_printf(seq, " 3-wire"); - if(self->service_type & IRCOMM_9_WIRE) - seq_printf(seq, " 9-wire"); - if(self->service_type & IRCOMM_CENTRONICS) - seq_printf(seq, " Centronics"); - seq_putc(seq, '\n'); - - return 0; -} - -static const struct seq_operations ircomm_seq_ops = { - .start = ircomm_seq_start, - .next = ircomm_seq_next, - .stop = ircomm_seq_stop, - .show = ircomm_seq_show, -}; - -static int ircomm_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &ircomm_seq_ops); -} -#endif /* CONFIG_PROC_FS */ - -MODULE_AUTHOR("Dag Brattli <dag@brattli.net>"); -MODULE_DESCRIPTION("IrCOMM protocol"); -MODULE_LICENSE("GPL"); - -module_init(ircomm_init); -module_exit(ircomm_cleanup); diff --git a/net/irda/ircomm/ircomm_event.c b/net/irda/ircomm/ircomm_event.c deleted file mode 100644 index b0730ac9f388..000000000000 --- a/net/irda/ircomm/ircomm_event.c +++ /dev/null @@ -1,246 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_event.c - * Version: 1.0 - * Description: IrCOMM layer state machine - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Jun 6 20:33:11 1999 - * Modified at: Sun Dec 12 13:44:32 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/proc_fs.h> -#include <linux/init.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/irttp.h> -#include <net/irda/irias_object.h> - -#include <net/irda/ircomm_core.h> -#include <net/irda/ircomm_event.h> - -static int ircomm_state_idle(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info); -static int ircomm_state_waiti(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info); -static int ircomm_state_waitr(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info); -static int ircomm_state_conn(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info); - -const char *const ircomm_state[] = { - "IRCOMM_IDLE", - "IRCOMM_WAITI", - "IRCOMM_WAITR", - "IRCOMM_CONN", -}; - -static const char *const ircomm_event[] __maybe_unused = { - "IRCOMM_CONNECT_REQUEST", - "IRCOMM_CONNECT_RESPONSE", - "IRCOMM_TTP_CONNECT_INDICATION", - "IRCOMM_LMP_CONNECT_INDICATION", - "IRCOMM_TTP_CONNECT_CONFIRM", - "IRCOMM_LMP_CONNECT_CONFIRM", - - "IRCOMM_LMP_DISCONNECT_INDICATION", - "IRCOMM_TTP_DISCONNECT_INDICATION", - "IRCOMM_DISCONNECT_REQUEST", - - "IRCOMM_TTP_DATA_INDICATION", - "IRCOMM_LMP_DATA_INDICATION", - "IRCOMM_DATA_REQUEST", - "IRCOMM_CONTROL_REQUEST", - "IRCOMM_CONTROL_INDICATION", -}; - -static int (*state[])(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) = -{ - ircomm_state_idle, - ircomm_state_waiti, - ircomm_state_waitr, - ircomm_state_conn, -}; - -/* - * Function ircomm_state_idle (self, event, skb) - * - * IrCOMM is currently idle - * - */ -static int ircomm_state_idle(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) -{ - int ret = 0; - - switch (event) { - case IRCOMM_CONNECT_REQUEST: - ircomm_next_state(self, IRCOMM_WAITI); - ret = self->issue.connect_request(self, skb, info); - break; - case IRCOMM_TTP_CONNECT_INDICATION: - case IRCOMM_LMP_CONNECT_INDICATION: - ircomm_next_state(self, IRCOMM_WAITR); - ircomm_connect_indication(self, skb, info); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_state_waiti (self, event, skb) - * - * The IrCOMM user has requested an IrCOMM connection to the remote - * device and is awaiting confirmation - */ -static int ircomm_state_waiti(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) -{ - int ret = 0; - - switch (event) { - case IRCOMM_TTP_CONNECT_CONFIRM: - case IRCOMM_LMP_CONNECT_CONFIRM: - ircomm_next_state(self, IRCOMM_CONN); - ircomm_connect_confirm(self, skb, info); - break; - case IRCOMM_TTP_DISCONNECT_INDICATION: - case IRCOMM_LMP_DISCONNECT_INDICATION: - ircomm_next_state(self, IRCOMM_IDLE); - ircomm_disconnect_indication(self, skb, info); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_state_waitr (self, event, skb) - * - * IrCOMM has received an incoming connection request and is awaiting - * response from the user - */ -static int ircomm_state_waitr(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) -{ - int ret = 0; - - switch (event) { - case IRCOMM_CONNECT_RESPONSE: - ircomm_next_state(self, IRCOMM_CONN); - ret = self->issue.connect_response(self, skb); - break; - case IRCOMM_DISCONNECT_REQUEST: - ircomm_next_state(self, IRCOMM_IDLE); - ret = self->issue.disconnect_request(self, skb, info); - break; - case IRCOMM_TTP_DISCONNECT_INDICATION: - case IRCOMM_LMP_DISCONNECT_INDICATION: - ircomm_next_state(self, IRCOMM_IDLE); - ircomm_disconnect_indication(self, skb, info); - break; - default: - pr_debug("%s(), unknown event = %s\n", __func__ , - ircomm_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_state_conn (self, event, skb) - * - * IrCOMM is connected to the peer IrCOMM device - * - */ -static int ircomm_state_conn(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) -{ - int ret = 0; - - switch (event) { - case IRCOMM_DATA_REQUEST: - ret = self->issue.data_request(self, skb, 0); - break; - case IRCOMM_TTP_DATA_INDICATION: - ircomm_process_data(self, skb); - break; - case IRCOMM_LMP_DATA_INDICATION: - ircomm_data_indication(self, skb); - break; - case IRCOMM_CONTROL_REQUEST: - /* Just send a separate frame for now */ - ret = self->issue.data_request(self, skb, skb->len); - break; - case IRCOMM_TTP_DISCONNECT_INDICATION: - case IRCOMM_LMP_DISCONNECT_INDICATION: - ircomm_next_state(self, IRCOMM_IDLE); - ircomm_disconnect_indication(self, skb, info); - break; - case IRCOMM_DISCONNECT_REQUEST: - ircomm_next_state(self, IRCOMM_IDLE); - ret = self->issue.disconnect_request(self, skb, info); - break; - default: - pr_debug("%s(), unknown event = %s\n", __func__ , - ircomm_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_do_event (self, event, skb) - * - * Process event - * - */ -int ircomm_do_event(struct ircomm_cb *self, IRCOMM_EVENT event, - struct sk_buff *skb, struct ircomm_info *info) -{ - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_state[self->state], ircomm_event[event]); - - return (*state[self->state])(self, event, skb, info); -} - -/* - * Function ircomm_next_state (self, state) - * - * Switch state - * - */ -void ircomm_next_state(struct ircomm_cb *self, IRCOMM_STATE state) -{ - self->state = state; - - pr_debug("%s: next state=%s, service type=%d\n", __func__ , - ircomm_state[self->state], self->service_type); -} diff --git a/net/irda/ircomm/ircomm_lmp.c b/net/irda/ircomm/ircomm_lmp.c deleted file mode 100644 index e4cc847bb933..000000000000 --- a/net/irda/ircomm/ircomm_lmp.c +++ /dev/null @@ -1,350 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_lmp.c - * Version: 1.0 - * Description: Interface between IrCOMM and IrLMP - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Jun 6 20:48:27 1999 - * Modified at: Sun Dec 12 13:44:17 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Sources: Previous IrLPT work by Thomas Davis - * - * Copyright (c) 1999 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/init.h> -#include <linux/gfp.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/irda_device.h> /* struct irda_skb_cb */ - -#include <net/irda/ircomm_event.h> -#include <net/irda/ircomm_lmp.h> - - -/* - * Function ircomm_lmp_connect_request (self, userdata) - * - * - * - */ -static int ircomm_lmp_connect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info) -{ - int ret = 0; - - /* Don't forget to refcount it - should be NULL anyway */ - if(userdata) - skb_get(userdata); - - ret = irlmp_connect_request(self->lsap, info->dlsap_sel, - info->saddr, info->daddr, NULL, userdata); - return ret; -} - -/* - * Function ircomm_lmp_connect_response (self, skb) - * - * - * - */ -static int ircomm_lmp_connect_response(struct ircomm_cb *self, - struct sk_buff *userdata) -{ - struct sk_buff *tx_skb; - - /* Any userdata supplied? */ - if (userdata == NULL) { - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* Reserve space for MUX and LAP header */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - } else { - /* - * Check that the client has reserved enough space for - * headers - */ - IRDA_ASSERT(skb_headroom(userdata) >= LMP_MAX_HEADER, - return -1;); - - /* Don't forget to refcount it - should be NULL anyway */ - skb_get(userdata); - tx_skb = userdata; - } - - return irlmp_connect_response(self->lsap, tx_skb); -} - -static int ircomm_lmp_disconnect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info) -{ - struct sk_buff *tx_skb; - int ret; - - if (!userdata) { - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* Reserve space for MUX and LAP header */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - userdata = tx_skb; - } else { - /* Don't forget to refcount it - should be NULL anyway */ - skb_get(userdata); - } - - ret = irlmp_disconnect_request(self->lsap, userdata); - - return ret; -} - -/* - * Function ircomm_lmp_flow_control (skb) - * - * This function is called when a data frame we have sent to IrLAP has - * been deallocated. We do this to make sure we don't flood IrLAP with - * frames, since we are not using the IrTTP flow control mechanism - */ -static void ircomm_lmp_flow_control(struct sk_buff *skb) -{ - struct irda_skb_cb *cb; - struct ircomm_cb *self; - int line; - - IRDA_ASSERT(skb != NULL, return;); - - cb = (struct irda_skb_cb *) skb->cb; - - line = cb->line; - - self = (struct ircomm_cb *) hashbin_lock_find(ircomm, line, NULL); - if (!self) { - pr_debug("%s(), didn't find myself\n", __func__); - return; - } - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - - self->pkt_count--; - - if ((self->pkt_count < 2) && (self->flow_status == FLOW_STOP)) { - pr_debug("%s(), asking TTY to start again!\n", __func__); - self->flow_status = FLOW_START; - if (self->notify.flow_indication) - self->notify.flow_indication(self->notify.instance, - self, FLOW_START); - } -} - -/* - * Function ircomm_lmp_data_request (self, userdata) - * - * Send data frame to peer device - * - */ -static int ircomm_lmp_data_request(struct ircomm_cb *self, - struct sk_buff *skb, - int not_used) -{ - struct irda_skb_cb *cb; - int ret; - - IRDA_ASSERT(skb != NULL, return -1;); - - cb = (struct irda_skb_cb *) skb->cb; - - cb->line = self->line; - - pr_debug("%s(), sending frame\n", __func__); - - /* Don't forget to refcount it - see ircomm_tty_do_softint() */ - skb_get(skb); - - skb_orphan(skb); - skb->destructor = ircomm_lmp_flow_control; - - if ((self->pkt_count++ > 7) && (self->flow_status == FLOW_START)) { - pr_debug("%s(), asking TTY to slow down!\n", __func__); - self->flow_status = FLOW_STOP; - if (self->notify.flow_indication) - self->notify.flow_indication(self->notify.instance, - self, FLOW_STOP); - } - ret = irlmp_data_request(self->lsap, skb); - if (ret) { - net_err_ratelimited("%s(), failed\n", __func__); - /* irlmp_data_request already free the packet */ - } - - return ret; -} - -/* - * Function ircomm_lmp_data_indication (instance, sap, skb) - * - * Incoming data which we must deliver to the state machine, to check - * we are still connected. - */ -static int ircomm_lmp_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - ircomm_do_event(self, IRCOMM_LMP_DATA_INDICATION, skb, NULL); - - /* Drop reference count - see ircomm_tty_data_indication(). */ - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function ircomm_lmp_connect_confirm (instance, sap, qos, max_sdu_size, - * max_header_size, skb) - * - * Connection has been confirmed by peer device - * - */ -static void ircomm_lmp_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_seg_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(qos != NULL, return;); - - info.max_data_size = max_seg_size; - info.max_header_size = max_header_size; - info.qos = qos; - - ircomm_do_event(self, IRCOMM_LMP_CONNECT_CONFIRM, skb, &info); - - /* Drop reference count - see ircomm_tty_connect_confirm(). */ - dev_kfree_skb(skb); -} - -/* - * Function ircomm_lmp_connect_indication (instance, sap, qos, max_sdu_size, - * max_header_size, skb) - * - * Peer device wants to make a connection with us - * - */ -static void ircomm_lmp_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_seg_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *)instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(qos != NULL, return;); - - info.max_data_size = max_seg_size; - info.max_header_size = max_header_size; - info.qos = qos; - - ircomm_do_event(self, IRCOMM_LMP_CONNECT_INDICATION, skb, &info); - - /* Drop reference count - see ircomm_tty_connect_indication(). */ - dev_kfree_skb(skb); -} - -/* - * Function ircomm_lmp_disconnect_indication (instance, sap, reason, skb) - * - * Peer device has closed the connection, or the link went down for some - * other reason - */ -static void ircomm_lmp_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - - info.reason = reason; - - ircomm_do_event(self, IRCOMM_LMP_DISCONNECT_INDICATION, skb, &info); - - /* Drop reference count - see ircomm_tty_disconnect_indication(). */ - if(skb) - dev_kfree_skb(skb); -} -/* - * Function ircomm_open_lsap (self) - * - * Open LSAP. This function will only be used when using "raw" services - * - */ -int ircomm_open_lsap(struct ircomm_cb *self) -{ - notify_t notify; - - /* Register callbacks */ - irda_notify_init(¬ify); - notify.data_indication = ircomm_lmp_data_indication; - notify.connect_confirm = ircomm_lmp_connect_confirm; - notify.connect_indication = ircomm_lmp_connect_indication; - notify.disconnect_indication = ircomm_lmp_disconnect_indication; - notify.instance = self; - strlcpy(notify.name, "IrCOMM", sizeof(notify.name)); - - self->lsap = irlmp_open_lsap(LSAP_ANY, ¬ify, 0); - if (!self->lsap) { - pr_debug("%sfailed to allocate tsap\n", __func__); - return -1; - } - self->slsap_sel = self->lsap->slsap_sel; - - /* - * Initialize the call-table for issuing commands - */ - self->issue.data_request = ircomm_lmp_data_request; - self->issue.connect_request = ircomm_lmp_connect_request; - self->issue.connect_response = ircomm_lmp_connect_response; - self->issue.disconnect_request = ircomm_lmp_disconnect_request; - - return 0; -} diff --git a/net/irda/ircomm/ircomm_param.c b/net/irda/ircomm/ircomm_param.c deleted file mode 100644 index 5728e76ca6d5..000000000000 --- a/net/irda/ircomm/ircomm_param.c +++ /dev/null @@ -1,501 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_param.c - * Version: 1.0 - * Description: Parameter handling for the IrCOMM protocol - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Jun 7 10:25:11 1999 - * Modified at: Sun Jan 30 14:32:03 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/gfp.h> -#include <linux/workqueue.h> -#include <linux/interrupt.h> - -#include <net/irda/irda.h> -#include <net/irda/parameters.h> - -#include <net/irda/ircomm_core.h> -#include <net/irda/ircomm_tty_attach.h> -#include <net/irda/ircomm_tty.h> - -#include <net/irda/ircomm_param.h> - -static int ircomm_param_service_type(void *instance, irda_param_t *param, - int get); -static int ircomm_param_port_type(void *instance, irda_param_t *param, - int get); -static int ircomm_param_port_name(void *instance, irda_param_t *param, - int get); -static int ircomm_param_service_type(void *instance, irda_param_t *param, - int get); -static int ircomm_param_data_rate(void *instance, irda_param_t *param, - int get); -static int ircomm_param_data_format(void *instance, irda_param_t *param, - int get); -static int ircomm_param_flow_control(void *instance, irda_param_t *param, - int get); -static int ircomm_param_xon_xoff(void *instance, irda_param_t *param, int get); -static int ircomm_param_enq_ack(void *instance, irda_param_t *param, int get); -static int ircomm_param_line_status(void *instance, irda_param_t *param, - int get); -static int ircomm_param_dte(void *instance, irda_param_t *param, int get); -static int ircomm_param_dce(void *instance, irda_param_t *param, int get); -static int ircomm_param_poll(void *instance, irda_param_t *param, int get); - -static const pi_minor_info_t pi_minor_call_table_common[] = { - { ircomm_param_service_type, PV_INT_8_BITS }, - { ircomm_param_port_type, PV_INT_8_BITS }, - { ircomm_param_port_name, PV_STRING } -}; -static const pi_minor_info_t pi_minor_call_table_non_raw[] = { - { ircomm_param_data_rate, PV_INT_32_BITS | PV_BIG_ENDIAN }, - { ircomm_param_data_format, PV_INT_8_BITS }, - { ircomm_param_flow_control, PV_INT_8_BITS }, - { ircomm_param_xon_xoff, PV_INT_16_BITS }, - { ircomm_param_enq_ack, PV_INT_16_BITS }, - { ircomm_param_line_status, PV_INT_8_BITS } -}; -static const pi_minor_info_t pi_minor_call_table_9_wire[] = { - { ircomm_param_dte, PV_INT_8_BITS }, - { ircomm_param_dce, PV_INT_8_BITS }, - { ircomm_param_poll, PV_NO_VALUE }, -}; - -static const pi_major_info_t pi_major_call_table[] = { - { pi_minor_call_table_common, 3 }, - { pi_minor_call_table_non_raw, 6 }, - { pi_minor_call_table_9_wire, 3 } -/* { pi_minor_call_table_centronics } */ -}; - -pi_param_info_t ircomm_param_info = { pi_major_call_table, 3, 0x0f, 4 }; - -/* - * Function ircomm_param_request (self, pi, flush) - * - * Queue a parameter for the control channel - * - */ -int ircomm_param_request(struct ircomm_tty_cb *self, __u8 pi, int flush) -{ - unsigned long flags; - struct sk_buff *skb; - int count; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - /* Make sure we don't send parameters for raw mode */ - if (self->service_type == IRCOMM_3_WIRE_RAW) - return 0; - - spin_lock_irqsave(&self->spinlock, flags); - - skb = self->ctrl_skb; - if (!skb) { - skb = alloc_skb(256, GFP_ATOMIC); - if (!skb) { - spin_unlock_irqrestore(&self->spinlock, flags); - return -ENOMEM; - } - - skb_reserve(skb, self->max_header_size); - self->ctrl_skb = skb; - } - /* - * Inserting is a little bit tricky since we don't know how much - * room we will need. But this should hopefully work OK - */ - count = irda_param_insert(self, pi, skb_tail_pointer(skb), - skb_tailroom(skb), &ircomm_param_info); - if (count < 0) { - net_warn_ratelimited("%s(), no room for parameter!\n", - __func__); - spin_unlock_irqrestore(&self->spinlock, flags); - return -1; - } - skb_put(skb, count); - pr_debug("%s(), skb->len=%d\n", __func__, skb->len); - - spin_unlock_irqrestore(&self->spinlock, flags); - - if (flush) { - /* ircomm_tty_do_softint will take care of the rest */ - schedule_work(&self->tqueue); - } - - return count; -} - -/* - * Function ircomm_param_service_type (self, buf, len) - * - * Handle service type, this function will both be called after the LM-IAS - * query and then the remote device sends its initial parameters - * - */ -static int ircomm_param_service_type(void *instance, irda_param_t *param, - int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - __u8 service_type = (__u8) param->pv.i; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) { - param->pv.i = self->settings.service_type; - return 0; - } - - /* Find all common service types */ - service_type &= self->service_type; - if (!service_type) { - pr_debug("%s(), No common service type to use!\n", __func__); - return -1; - } - pr_debug("%s(), services in common=%02x\n", __func__ , - service_type); - - /* - * Now choose a preferred service type of those available - */ - if (service_type & IRCOMM_CENTRONICS) - self->settings.service_type = IRCOMM_CENTRONICS; - else if (service_type & IRCOMM_9_WIRE) - self->settings.service_type = IRCOMM_9_WIRE; - else if (service_type & IRCOMM_3_WIRE) - self->settings.service_type = IRCOMM_3_WIRE; - else if (service_type & IRCOMM_3_WIRE_RAW) - self->settings.service_type = IRCOMM_3_WIRE_RAW; - - pr_debug("%s(), resulting service type=0x%02x\n", __func__ , - self->settings.service_type); - - /* - * Now the line is ready for some communication. Check if we are a - * server, and send over some initial parameters. - * Client do it in ircomm_tty_state_setup(). - * Note : we may get called from ircomm_tty_getvalue_confirm(), - * therefore before we even have open any socket. And self->client - * is initialised to TRUE only later. So, we check if the link is - * really initialised. - Jean II - */ - if ((self->max_header_size != IRCOMM_TTY_HDR_UNINITIALISED) && - (!self->client) && - (self->settings.service_type != IRCOMM_3_WIRE_RAW)) - { - /* Init connection */ - ircomm_tty_send_initial_parameters(self); - ircomm_tty_link_established(self); - } - - return 0; -} - -/* - * Function ircomm_param_port_type (self, param) - * - * The port type parameter tells if the devices are serial or parallel. - * Since we only advertise serial service, this parameter should only - * be equal to IRCOMM_SERIAL. - */ -static int ircomm_param_port_type(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) - param->pv.i = IRCOMM_SERIAL; - else { - self->settings.port_type = (__u8) param->pv.i; - - pr_debug("%s(), port type=%d\n", __func__ , - self->settings.port_type); - } - return 0; -} - -/* - * Function ircomm_param_port_name (self, param) - * - * Exchange port name - * - */ -static int ircomm_param_port_name(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) { - pr_debug("%s(), not imp!\n", __func__); - } else { - pr_debug("%s(), port-name=%s\n", __func__ , param->pv.c); - strncpy(self->settings.port_name, param->pv.c, 32); - } - - return 0; -} - -/* - * Function ircomm_param_data_rate (self, param) - * - * Exchange data rate to be used in this settings - * - */ -static int ircomm_param_data_rate(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) - param->pv.i = self->settings.data_rate; - else - self->settings.data_rate = param->pv.i; - - pr_debug("%s(), data rate = %d\n", __func__ , param->pv.i); - - return 0; -} - -/* - * Function ircomm_param_data_format (self, param) - * - * Exchange data format to be used in this settings - * - */ -static int ircomm_param_data_format(void *instance, irda_param_t *param, - int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) - param->pv.i = self->settings.data_format; - else - self->settings.data_format = (__u8) param->pv.i; - - return 0; -} - -/* - * Function ircomm_param_flow_control (self, param) - * - * Exchange flow control settings to be used in this settings - * - */ -static int ircomm_param_flow_control(void *instance, irda_param_t *param, - int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) - param->pv.i = self->settings.flow_control; - else - self->settings.flow_control = (__u8) param->pv.i; - - pr_debug("%s(), flow control = 0x%02x\n", __func__ , (__u8)param->pv.i); - - return 0; -} - -/* - * Function ircomm_param_xon_xoff (self, param) - * - * Exchange XON/XOFF characters - * - */ -static int ircomm_param_xon_xoff(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) { - param->pv.i = self->settings.xonxoff[0]; - param->pv.i |= self->settings.xonxoff[1] << 8; - } else { - self->settings.xonxoff[0] = (__u16) param->pv.i & 0xff; - self->settings.xonxoff[1] = (__u16) param->pv.i >> 8; - } - - pr_debug("%s(), XON/XOFF = 0x%02x,0x%02x\n", __func__ , - param->pv.i & 0xff, param->pv.i >> 8); - - return 0; -} - -/* - * Function ircomm_param_enq_ack (self, param) - * - * Exchange ENQ/ACK characters - * - */ -static int ircomm_param_enq_ack(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) { - param->pv.i = self->settings.enqack[0]; - param->pv.i |= self->settings.enqack[1] << 8; - } else { - self->settings.enqack[0] = (__u16) param->pv.i & 0xff; - self->settings.enqack[1] = (__u16) param->pv.i >> 8; - } - - pr_debug("%s(), ENQ/ACK = 0x%02x,0x%02x\n", __func__ , - param->pv.i & 0xff, param->pv.i >> 8); - - return 0; -} - -/* - * Function ircomm_param_line_status (self, param) - * - * - * - */ -static int ircomm_param_line_status(void *instance, irda_param_t *param, - int get) -{ - pr_debug("%s(), not impl.\n", __func__); - - return 0; -} - -/* - * Function ircomm_param_dte (instance, param) - * - * If we get here, there must be some sort of null-modem connection, and - * we are probably working in server mode as well. - */ -static int ircomm_param_dte(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - __u8 dte; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (get) - param->pv.i = self->settings.dte; - else { - dte = (__u8) param->pv.i; - - self->settings.dce = 0; - - if (dte & IRCOMM_DELTA_DTR) - self->settings.dce |= (IRCOMM_DELTA_DSR| - IRCOMM_DELTA_RI | - IRCOMM_DELTA_CD); - if (dte & IRCOMM_DTR) - self->settings.dce |= (IRCOMM_DSR| - IRCOMM_RI | - IRCOMM_CD); - - if (dte & IRCOMM_DELTA_RTS) - self->settings.dce |= IRCOMM_DELTA_CTS; - if (dte & IRCOMM_RTS) - self->settings.dce |= IRCOMM_CTS; - - /* Take appropriate actions */ - ircomm_tty_check_modem_status(self); - - /* Null modem cable emulator */ - self->settings.null_modem = TRUE; - } - - return 0; -} - -/* - * Function ircomm_param_dce (instance, param) - * - * - * - */ -static int ircomm_param_dce(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - __u8 dce; - - pr_debug("%s(), dce = 0x%02x\n", __func__ , (__u8)param->pv.i); - - dce = (__u8) param->pv.i; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - self->settings.dce = dce; - - /* Check if any of the settings have changed */ - if (dce & 0x0f) { - if (dce & IRCOMM_DELTA_CTS) { - pr_debug("%s(), CTS\n", __func__); - } - } - - ircomm_tty_check_modem_status(self); - - return 0; -} - -/* - * Function ircomm_param_poll (instance, param) - * - * Called when the peer device is polling for the line settings - * - */ -static int ircomm_param_poll(void *instance, irda_param_t *param, int get) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - /* Poll parameters are always of length 0 (just a signal) */ - if (!get) { - /* Respond with DTE line settings */ - ircomm_param_request(self, IRCOMM_DTE, TRUE); - } - return 0; -} - - - - - diff --git a/net/irda/ircomm/ircomm_ttp.c b/net/irda/ircomm/ircomm_ttp.c deleted file mode 100644 index 4b81e0934770..000000000000 --- a/net/irda/ircomm/ircomm_ttp.c +++ /dev/null @@ -1,350 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_ttp.c - * Version: 1.0 - * Description: Interface between IrCOMM and IrTTP - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Jun 6 20:48:27 1999 - * Modified at: Mon Dec 13 11:35:13 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/init.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/irttp.h> - -#include <net/irda/ircomm_event.h> -#include <net/irda/ircomm_ttp.h> - -static int ircomm_ttp_data_indication(void *instance, void *sap, - struct sk_buff *skb); -static void ircomm_ttp_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb); -static void ircomm_ttp_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb); -static void ircomm_ttp_flow_indication(void *instance, void *sap, - LOCAL_FLOW cmd); -static void ircomm_ttp_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *skb); -static int ircomm_ttp_data_request(struct ircomm_cb *self, - struct sk_buff *skb, - int clen); -static int ircomm_ttp_connect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info); -static int ircomm_ttp_connect_response(struct ircomm_cb *self, - struct sk_buff *userdata); -static int ircomm_ttp_disconnect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info); - -/* - * Function ircomm_open_tsap (self) - * - * - * - */ -int ircomm_open_tsap(struct ircomm_cb *self) -{ - notify_t notify; - - /* Register callbacks */ - irda_notify_init(¬ify); - notify.data_indication = ircomm_ttp_data_indication; - notify.connect_confirm = ircomm_ttp_connect_confirm; - notify.connect_indication = ircomm_ttp_connect_indication; - notify.flow_indication = ircomm_ttp_flow_indication; - notify.disconnect_indication = ircomm_ttp_disconnect_indication; - notify.instance = self; - strlcpy(notify.name, "IrCOMM", sizeof(notify.name)); - - self->tsap = irttp_open_tsap(LSAP_ANY, DEFAULT_INITIAL_CREDIT, - ¬ify); - if (!self->tsap) { - pr_debug("%sfailed to allocate tsap\n", __func__); - return -1; - } - self->slsap_sel = self->tsap->stsap_sel; - - /* - * Initialize the call-table for issuing commands - */ - self->issue.data_request = ircomm_ttp_data_request; - self->issue.connect_request = ircomm_ttp_connect_request; - self->issue.connect_response = ircomm_ttp_connect_response; - self->issue.disconnect_request = ircomm_ttp_disconnect_request; - - return 0; -} - -/* - * Function ircomm_ttp_connect_request (self, userdata) - * - * - * - */ -static int ircomm_ttp_connect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info) -{ - int ret = 0; - - /* Don't forget to refcount it - should be NULL anyway */ - if(userdata) - skb_get(userdata); - - ret = irttp_connect_request(self->tsap, info->dlsap_sel, - info->saddr, info->daddr, NULL, - TTP_SAR_DISABLE, userdata); - - return ret; -} - -/* - * Function ircomm_ttp_connect_response (self, skb) - * - * - * - */ -static int ircomm_ttp_connect_response(struct ircomm_cb *self, - struct sk_buff *userdata) -{ - int ret; - - /* Don't forget to refcount it - should be NULL anyway */ - if(userdata) - skb_get(userdata); - - ret = irttp_connect_response(self->tsap, TTP_SAR_DISABLE, userdata); - - return ret; -} - -/* - * Function ircomm_ttp_data_request (self, userdata) - * - * Send IrCOMM data to IrTTP layer. Currently we do not try to combine - * control data with pure data, so they will be sent as separate frames. - * Should not be a big problem though, since control frames are rare. But - * some of them are sent after connection establishment, so this can - * increase the latency a bit. - */ -static int ircomm_ttp_data_request(struct ircomm_cb *self, - struct sk_buff *skb, - int clen) -{ - int ret; - - IRDA_ASSERT(skb != NULL, return -1;); - - pr_debug("%s(), clen=%d\n", __func__ , clen); - - /* - * Insert clen field, currently we either send data only, or control - * only frames, to make things easier and avoid queueing - */ - IRDA_ASSERT(skb_headroom(skb) >= IRCOMM_HEADER_SIZE, return -1;); - - /* Don't forget to refcount it - see ircomm_tty_do_softint() */ - skb_get(skb); - - skb_push(skb, IRCOMM_HEADER_SIZE); - - skb->data[0] = clen; - - ret = irttp_data_request(self->tsap, skb); - if (ret) { - net_err_ratelimited("%s(), failed\n", __func__); - /* irttp_data_request already free the packet */ - } - - return ret; -} - -/* - * Function ircomm_ttp_data_indication (instance, sap, skb) - * - * Incoming data - * - */ -static int ircomm_ttp_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - ircomm_do_event(self, IRCOMM_TTP_DATA_INDICATION, skb, NULL); - - /* Drop reference count - see ircomm_tty_data_indication(). */ - dev_kfree_skb(skb); - - return 0; -} - -static void ircomm_ttp_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(qos != NULL, goto out;); - - if (max_sdu_size != TTP_SAR_DISABLE) { - net_err_ratelimited("%s(), SAR not allowed for IrCOMM!\n", - __func__); - goto out; - } - - info.max_data_size = irttp_get_max_seg_size(self->tsap) - - IRCOMM_HEADER_SIZE; - info.max_header_size = max_header_size + IRCOMM_HEADER_SIZE; - info.qos = qos; - - ircomm_do_event(self, IRCOMM_TTP_CONNECT_CONFIRM, skb, &info); - -out: - /* Drop reference count - see ircomm_tty_connect_confirm(). */ - dev_kfree_skb(skb); -} - -/* - * Function ircomm_ttp_connect_indication (instance, sap, qos, max_sdu_size, - * max_header_size, skb) - * - * - * - */ -static void ircomm_ttp_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *)instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(qos != NULL, goto out;); - - if (max_sdu_size != TTP_SAR_DISABLE) { - net_err_ratelimited("%s(), SAR not allowed for IrCOMM!\n", - __func__); - goto out; - } - - info.max_data_size = irttp_get_max_seg_size(self->tsap) - - IRCOMM_HEADER_SIZE; - info.max_header_size = max_header_size + IRCOMM_HEADER_SIZE; - info.qos = qos; - - ircomm_do_event(self, IRCOMM_TTP_CONNECT_INDICATION, skb, &info); - -out: - /* Drop reference count - see ircomm_tty_connect_indication(). */ - dev_kfree_skb(skb); -} - -/* - * Function ircomm_ttp_disconnect_request (self, userdata, info) - * - * - * - */ -static int ircomm_ttp_disconnect_request(struct ircomm_cb *self, - struct sk_buff *userdata, - struct ircomm_info *info) -{ - int ret; - - /* Don't forget to refcount it - should be NULL anyway */ - if(userdata) - skb_get(userdata); - - ret = irttp_disconnect_request(self->tsap, userdata, P_NORMAL); - - return ret; -} - -/* - * Function ircomm_ttp_disconnect_indication (instance, sap, reason, skb) - * - * - * - */ -static void ircomm_ttp_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *skb) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - struct ircomm_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - - info.reason = reason; - - ircomm_do_event(self, IRCOMM_TTP_DISCONNECT_INDICATION, skb, &info); - - /* Drop reference count - see ircomm_tty_disconnect_indication(). */ - if(skb) - dev_kfree_skb(skb); -} - -/* - * Function ircomm_ttp_flow_indication (instance, sap, cmd) - * - * Layer below is telling us to start or stop the flow of data - * - */ -static void ircomm_ttp_flow_indication(void *instance, void *sap, - LOCAL_FLOW cmd) -{ - struct ircomm_cb *self = (struct ircomm_cb *) instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_MAGIC, return;); - - if (self->notify.flow_indication) - self->notify.flow_indication(self->notify.instance, self, cmd); -} - - diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c deleted file mode 100644 index ec157c3419b5..000000000000 --- a/net/irda/ircomm/ircomm_tty.c +++ /dev/null @@ -1,1329 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_tty.c - * Version: 1.0 - * Description: IrCOMM serial TTY driver - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Jun 6 21:00:56 1999 - * Modified at: Wed Feb 23 00:09:02 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Sources: serial.c and previous IrCOMM work by Takahide Higuchi - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/init.h> -#include <linux/module.h> -#include <linux/fs.h> -#include <linux/slab.h> -#include <linux/sched/signal.h> -#include <linux/seq_file.h> -#include <linux/termios.h> -#include <linux/tty.h> -#include <linux/tty_flip.h> -#include <linux/interrupt.h> -#include <linux/device.h> /* for MODULE_ALIAS_CHARDEV_MAJOR */ - -#include <linux/uaccess.h> - -#include <net/irda/irda.h> -#include <net/irda/irmod.h> - -#include <net/irda/ircomm_core.h> -#include <net/irda/ircomm_param.h> -#include <net/irda/ircomm_tty_attach.h> -#include <net/irda/ircomm_tty.h> - -static int ircomm_tty_install(struct tty_driver *driver, - struct tty_struct *tty); -static int ircomm_tty_open(struct tty_struct *tty, struct file *filp); -static void ircomm_tty_close(struct tty_struct * tty, struct file *filp); -static int ircomm_tty_write(struct tty_struct * tty, - const unsigned char *buf, int count); -static int ircomm_tty_write_room(struct tty_struct *tty); -static void ircomm_tty_throttle(struct tty_struct *tty); -static void ircomm_tty_unthrottle(struct tty_struct *tty); -static int ircomm_tty_chars_in_buffer(struct tty_struct *tty); -static void ircomm_tty_flush_buffer(struct tty_struct *tty); -static void ircomm_tty_send_xchar(struct tty_struct *tty, char ch); -static void ircomm_tty_wait_until_sent(struct tty_struct *tty, int timeout); -static void ircomm_tty_hangup(struct tty_struct *tty); -static void ircomm_tty_do_softint(struct work_struct *work); -static void ircomm_tty_shutdown(struct ircomm_tty_cb *self); -static void ircomm_tty_stop(struct tty_struct *tty); - -static int ircomm_tty_data_indication(void *instance, void *sap, - struct sk_buff *skb); -static int ircomm_tty_control_indication(void *instance, void *sap, - struct sk_buff *skb); -static void ircomm_tty_flow_indication(void *instance, void *sap, - LOCAL_FLOW cmd); -#ifdef CONFIG_PROC_FS -static const struct file_operations ircomm_tty_proc_fops; -#endif /* CONFIG_PROC_FS */ -static struct tty_driver *driver; - -static hashbin_t *ircomm_tty = NULL; - -static const struct tty_operations ops = { - .install = ircomm_tty_install, - .open = ircomm_tty_open, - .close = ircomm_tty_close, - .write = ircomm_tty_write, - .write_room = ircomm_tty_write_room, - .chars_in_buffer = ircomm_tty_chars_in_buffer, - .flush_buffer = ircomm_tty_flush_buffer, - .ioctl = ircomm_tty_ioctl, /* ircomm_tty_ioctl.c */ - .tiocmget = ircomm_tty_tiocmget, /* ircomm_tty_ioctl.c */ - .tiocmset = ircomm_tty_tiocmset, /* ircomm_tty_ioctl.c */ - .throttle = ircomm_tty_throttle, - .unthrottle = ircomm_tty_unthrottle, - .send_xchar = ircomm_tty_send_xchar, - .set_termios = ircomm_tty_set_termios, - .stop = ircomm_tty_stop, - .start = ircomm_tty_start, - .hangup = ircomm_tty_hangup, - .wait_until_sent = ircomm_tty_wait_until_sent, -#ifdef CONFIG_PROC_FS - .proc_fops = &ircomm_tty_proc_fops, -#endif /* CONFIG_PROC_FS */ -}; - -static void ircomm_port_raise_dtr_rts(struct tty_port *port, int raise) -{ - struct ircomm_tty_cb *self = container_of(port, struct ircomm_tty_cb, - port); - /* - * Here, we use to lock those two guys, but as ircomm_param_request() - * does it itself, I don't see the point (and I see the deadlock). - * Jean II - */ - if (raise) - self->settings.dte |= IRCOMM_RTS | IRCOMM_DTR; - else - self->settings.dte &= ~(IRCOMM_RTS | IRCOMM_DTR); - - ircomm_param_request(self, IRCOMM_DTE, TRUE); -} - -static int ircomm_port_carrier_raised(struct tty_port *port) -{ - struct ircomm_tty_cb *self = container_of(port, struct ircomm_tty_cb, - port); - return self->settings.dce & IRCOMM_CD; -} - -static const struct tty_port_operations ircomm_port_ops = { - .dtr_rts = ircomm_port_raise_dtr_rts, - .carrier_raised = ircomm_port_carrier_raised, -}; - -/* - * Function ircomm_tty_init() - * - * Init IrCOMM TTY layer/driver - * - */ -static int __init ircomm_tty_init(void) -{ - driver = alloc_tty_driver(IRCOMM_TTY_PORTS); - if (!driver) - return -ENOMEM; - ircomm_tty = hashbin_new(HB_LOCK); - if (ircomm_tty == NULL) { - net_err_ratelimited("%s(), can't allocate hashbin!\n", - __func__); - put_tty_driver(driver); - return -ENOMEM; - } - - driver->driver_name = "ircomm"; - driver->name = "ircomm"; - driver->major = IRCOMM_TTY_MAJOR; - driver->minor_start = IRCOMM_TTY_MINOR; - driver->type = TTY_DRIVER_TYPE_SERIAL; - driver->subtype = SERIAL_TYPE_NORMAL; - driver->init_termios = tty_std_termios; - driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; - driver->flags = TTY_DRIVER_REAL_RAW; - tty_set_operations(driver, &ops); - if (tty_register_driver(driver)) { - net_err_ratelimited("%s(): Couldn't register serial driver\n", - __func__); - put_tty_driver(driver); - return -1; - } - return 0; -} - -static void __exit __ircomm_tty_cleanup(struct ircomm_tty_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - ircomm_tty_shutdown(self); - - self->magic = 0; - tty_port_destroy(&self->port); - kfree(self); -} - -/* - * Function ircomm_tty_cleanup () - * - * Remove IrCOMM TTY layer/driver - * - */ -static void __exit ircomm_tty_cleanup(void) -{ - int ret; - - ret = tty_unregister_driver(driver); - if (ret) { - net_err_ratelimited("%s(), failed to unregister driver\n", - __func__); - return; - } - - hashbin_delete(ircomm_tty, (FREE_FUNC) __ircomm_tty_cleanup); - put_tty_driver(driver); -} - -/* - * Function ircomm_startup (self) - * - * - * - */ -static int ircomm_tty_startup(struct ircomm_tty_cb *self) -{ - notify_t notify; - int ret = -ENODEV; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - /* Check if already open */ - if (tty_port_initialized(&self->port)) { - pr_debug("%s(), already open so break out!\n", __func__); - return 0; - } - tty_port_set_initialized(&self->port, 1); - - /* Register with IrCOMM */ - irda_notify_init(¬ify); - /* These callbacks we must handle ourselves */ - notify.data_indication = ircomm_tty_data_indication; - notify.udata_indication = ircomm_tty_control_indication; - notify.flow_indication = ircomm_tty_flow_indication; - - /* Use the ircomm_tty interface for these ones */ - notify.disconnect_indication = ircomm_tty_disconnect_indication; - notify.connect_confirm = ircomm_tty_connect_confirm; - notify.connect_indication = ircomm_tty_connect_indication; - strlcpy(notify.name, "ircomm_tty", sizeof(notify.name)); - notify.instance = self; - - if (!self->ircomm) { - self->ircomm = ircomm_open(¬ify, self->service_type, - self->line); - } - if (!self->ircomm) - goto err; - - self->slsap_sel = self->ircomm->slsap_sel; - - /* Connect IrCOMM link with remote device */ - ret = ircomm_tty_attach_cable(self); - if (ret < 0) { - net_err_ratelimited("%s(), error attaching cable!\n", __func__); - goto err; - } - - return 0; -err: - tty_port_set_initialized(&self->port, 0); - return ret; -} - -/* - * Function ircomm_block_til_ready (self, filp) - * - * - * - */ -static int ircomm_tty_block_til_ready(struct ircomm_tty_cb *self, - struct tty_struct *tty, struct file *filp) -{ - struct tty_port *port = &self->port; - DECLARE_WAITQUEUE(wait, current); - int retval; - int do_clocal = 0; - unsigned long flags; - - /* - * If non-blocking mode is set, or the port is not enabled, - * then make the check up front and then exit. - */ - if (tty_io_error(tty)) { - tty_port_set_active(port, 1); - return 0; - } - - if (filp->f_flags & O_NONBLOCK) { - /* nonblock mode is set */ - if (C_BAUD(tty)) - tty_port_raise_dtr_rts(port); - tty_port_set_active(port, 1); - pr_debug("%s(), O_NONBLOCK requested!\n", __func__); - return 0; - } - - if (C_CLOCAL(tty)) { - pr_debug("%s(), doing CLOCAL!\n", __func__); - do_clocal = 1; - } - - /* Wait for carrier detect and the line to become - * free (i.e., not in use by the callout). While we are in - * this loop, port->count is dropped by one, so that - * mgsl_close() knows when to free things. We restore it upon - * exit, either normal or abnormal. - */ - - retval = 0; - add_wait_queue(&port->open_wait, &wait); - - pr_debug("%s(%d):block_til_ready before block on %s open_count=%d\n", - __FILE__, __LINE__, tty->driver->name, port->count); - - spin_lock_irqsave(&port->lock, flags); - port->count--; - port->blocked_open++; - spin_unlock_irqrestore(&port->lock, flags); - - while (1) { - if (C_BAUD(tty) && tty_port_initialized(port)) - tty_port_raise_dtr_rts(port); - - set_current_state(TASK_INTERRUPTIBLE); - - if (tty_hung_up_p(filp) || !tty_port_initialized(port)) { - retval = (port->flags & ASYNC_HUP_NOTIFY) ? - -EAGAIN : -ERESTARTSYS; - break; - } - - /* - * Check if link is ready now. Even if CLOCAL is - * specified, we cannot return before the IrCOMM link is - * ready - */ - if ((do_clocal || tty_port_carrier_raised(port)) && - self->state == IRCOMM_TTY_READY) - { - break; - } - - if (signal_pending(current)) { - retval = -ERESTARTSYS; - break; - } - - pr_debug("%s(%d):block_til_ready blocking on %s open_count=%d\n", - __FILE__, __LINE__, tty->driver->name, port->count); - - schedule(); - } - - __set_current_state(TASK_RUNNING); - remove_wait_queue(&port->open_wait, &wait); - - spin_lock_irqsave(&port->lock, flags); - if (!tty_hung_up_p(filp)) - port->count++; - port->blocked_open--; - spin_unlock_irqrestore(&port->lock, flags); - - pr_debug("%s(%d):block_til_ready after blocking on %s open_count=%d\n", - __FILE__, __LINE__, tty->driver->name, port->count); - - if (!retval) - tty_port_set_active(port, 1); - - return retval; -} - - -static int ircomm_tty_install(struct tty_driver *driver, struct tty_struct *tty) -{ - struct ircomm_tty_cb *self; - unsigned int line = tty->index; - - /* Check if instance already exists */ - self = hashbin_lock_find(ircomm_tty, line, NULL); - if (!self) { - /* No, so make new instance */ - self = kzalloc(sizeof(struct ircomm_tty_cb), GFP_KERNEL); - if (self == NULL) - return -ENOMEM; - - tty_port_init(&self->port); - self->port.ops = &ircomm_port_ops; - self->magic = IRCOMM_TTY_MAGIC; - self->flow = FLOW_STOP; - - self->line = line; - INIT_WORK(&self->tqueue, ircomm_tty_do_softint); - self->max_header_size = IRCOMM_TTY_HDR_UNINITIALISED; - self->max_data_size = IRCOMM_TTY_DATA_UNINITIALISED; - - /* Init some important stuff */ - init_timer(&self->watchdog_timer); - spin_lock_init(&self->spinlock); - - /* - * Force TTY into raw mode by default which is usually what - * we want for IrCOMM and IrLPT. This way applications will - * not have to twiddle with printcap etc. - * - * Note this is completely usafe and doesn't work properly - */ - tty->termios.c_iflag = 0; - tty->termios.c_oflag = 0; - - /* Insert into hash */ - hashbin_insert(ircomm_tty, (irda_queue_t *) self, line, NULL); - } - - tty->driver_data = self; - - return tty_port_install(&self->port, driver, tty); -} - -/* - * Function ircomm_tty_open (tty, filp) - * - * This routine is called when a particular tty device is opened. This - * routine is mandatory; if this routine is not filled in, the attempted - * open will fail with ENODEV. - */ -static int ircomm_tty_open(struct tty_struct *tty, struct file *filp) -{ - struct ircomm_tty_cb *self = tty->driver_data; - unsigned long flags; - int ret; - - /* ++ is not atomic, so this should be protected - Jean II */ - spin_lock_irqsave(&self->port.lock, flags); - self->port.count++; - spin_unlock_irqrestore(&self->port.lock, flags); - tty_port_tty_set(&self->port, tty); - - pr_debug("%s(), %s%d, count = %d\n", __func__ , tty->driver->name, - self->line, self->port.count); - - /* Not really used by us, but lets do it anyway */ - self->port.low_latency = (self->port.flags & ASYNC_LOW_LATENCY) ? 1 : 0; - - /* Check if this is a "normal" ircomm device, or an irlpt device */ - if (self->line < 0x10) { - self->service_type = IRCOMM_3_WIRE | IRCOMM_9_WIRE; - self->settings.service_type = IRCOMM_9_WIRE; /* 9 wire as default */ - /* Jan Kiszka -> add DSR/RI -> Conform to IrCOMM spec */ - self->settings.dce = IRCOMM_CTS | IRCOMM_CD | IRCOMM_DSR | IRCOMM_RI; /* Default line settings */ - pr_debug("%s(), IrCOMM device\n", __func__); - } else { - pr_debug("%s(), IrLPT device\n", __func__); - self->service_type = IRCOMM_3_WIRE_RAW; - self->settings.service_type = IRCOMM_3_WIRE_RAW; /* Default */ - } - - ret = ircomm_tty_startup(self); - if (ret) - return ret; - - ret = ircomm_tty_block_til_ready(self, tty, filp); - if (ret) { - pr_debug("%s(), returning after block_til_ready with %d\n", - __func__, ret); - - return ret; - } - return 0; -} - -/* - * Function ircomm_tty_close (tty, filp) - * - * This routine is called when a particular tty device is closed. - * - */ -static void ircomm_tty_close(struct tty_struct *tty, struct file *filp) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - struct tty_port *port = &self->port; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - if (tty_port_close_start(port, tty, filp) == 0) - return; - - ircomm_tty_shutdown(self); - - tty_driver_flush_buffer(tty); - - tty_port_close_end(port, tty); - tty_port_tty_set(port, NULL); -} - -/* - * Function ircomm_tty_flush_buffer (tty) - * - * - * - */ -static void ircomm_tty_flush_buffer(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* - * Let do_softint() do this to avoid race condition with - * do_softint() ;-) - */ - schedule_work(&self->tqueue); -} - -/* - * Function ircomm_tty_do_softint (work) - * - * We use this routine to give the write wakeup to the user at at a - * safe time (as fast as possible after write have completed). This - * can be compared to the Tx interrupt. - */ -static void ircomm_tty_do_softint(struct work_struct *work) -{ - struct ircomm_tty_cb *self = - container_of(work, struct ircomm_tty_cb, tqueue); - struct tty_struct *tty; - unsigned long flags; - struct sk_buff *skb, *ctrl_skb; - - if (!self || self->magic != IRCOMM_TTY_MAGIC) - return; - - tty = tty_port_tty_get(&self->port); - if (!tty) - return; - - /* Unlink control buffer */ - spin_lock_irqsave(&self->spinlock, flags); - - ctrl_skb = self->ctrl_skb; - self->ctrl_skb = NULL; - - spin_unlock_irqrestore(&self->spinlock, flags); - - /* Flush control buffer if any */ - if(ctrl_skb) { - if(self->flow == FLOW_START) - ircomm_control_request(self->ircomm, ctrl_skb); - /* Drop reference count - see ircomm_ttp_data_request(). */ - dev_kfree_skb(ctrl_skb); - } - - if (tty->hw_stopped) - goto put; - - /* Unlink transmit buffer */ - spin_lock_irqsave(&self->spinlock, flags); - - skb = self->tx_skb; - self->tx_skb = NULL; - - spin_unlock_irqrestore(&self->spinlock, flags); - - /* Flush transmit buffer if any */ - if (skb) { - ircomm_tty_do_event(self, IRCOMM_TTY_DATA_REQUEST, skb, NULL); - /* Drop reference count - see ircomm_ttp_data_request(). */ - dev_kfree_skb(skb); - } - - /* Check if user (still) wants to be waken up */ - tty_wakeup(tty); -put: - tty_kref_put(tty); -} - -/* - * Function ircomm_tty_write (tty, buf, count) - * - * This routine is called by the kernel to write a series of characters - * to the tty device. The characters may come from user space or kernel - * space. This routine will return the number of characters actually - * accepted for writing. This routine is mandatory. - */ -static int ircomm_tty_write(struct tty_struct *tty, - const unsigned char *buf, int count) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned long flags; - struct sk_buff *skb; - int tailroom = 0; - int len = 0; - int size; - - pr_debug("%s(), count=%d, hw_stopped=%d\n", __func__ , count, - tty->hw_stopped); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - /* We may receive packets from the TTY even before we have finished - * our setup. Not cool. - * The problem is that we don't know the final header and data size - * to create the proper skb, so any skb we would create would have - * bogus header and data size, so need care. - * We use a bogus header size to safely detect this condition. - * Another problem is that hw_stopped was set to 0 way before it - * should be, so we would drop this skb. It should now be fixed. - * One option is to not accept data until we are properly setup. - * But, I suspect that when it happens, the ppp line discipline - * just "drops" the data, which might screw up connect scripts. - * The second option is to create a "safe skb", with large header - * and small size (see ircomm_tty_open() for values). - * We just need to make sure that when the real values get filled, - * we don't mess up the original "safe skb" (see tx_data_size). - * Jean II */ - if (self->max_header_size == IRCOMM_TTY_HDR_UNINITIALISED) { - pr_debug("%s() : not initialised\n", __func__); -#ifdef IRCOMM_NO_TX_BEFORE_INIT - /* We didn't consume anything, TTY will retry */ - return 0; -#endif - } - - if (count < 1) - return 0; - - /* Protect our manipulation of self->tx_skb and related */ - spin_lock_irqsave(&self->spinlock, flags); - - /* Fetch current transmit buffer */ - skb = self->tx_skb; - - /* - * Send out all the data we get, possibly as multiple fragmented - * frames, but this will only happen if the data is larger than the - * max data size. The normal case however is just the opposite, and - * this function may be called multiple times, and will then actually - * defragment the data and send it out as one packet as soon as - * possible, but at a safer point in time - */ - while (count) { - size = count; - - /* Adjust data size to the max data size */ - if (size > self->max_data_size) - size = self->max_data_size; - - /* - * Do we already have a buffer ready for transmit, or do - * we need to allocate a new frame - */ - if (skb) { - /* - * Any room for more data at the end of the current - * transmit buffer? Cannot use skb_tailroom, since - * dev_alloc_skb gives us a larger skb than we - * requested - * Note : use tx_data_size, because max_data_size - * may have changed and we don't want to overwrite - * the skb. - Jean II - */ - if ((tailroom = (self->tx_data_size - skb->len)) > 0) { - /* Adjust data to tailroom */ - if (size > tailroom) - size = tailroom; - } else { - /* - * Current transmit frame is full, so break - * out, so we can send it as soon as possible - */ - break; - } - } else { - /* Prepare a full sized frame */ - skb = alloc_skb(self->max_data_size+ - self->max_header_size, - GFP_ATOMIC); - if (!skb) { - spin_unlock_irqrestore(&self->spinlock, flags); - return -ENOBUFS; - } - skb_reserve(skb, self->max_header_size); - self->tx_skb = skb; - /* Remember skb size because max_data_size may - * change later on - Jean II */ - self->tx_data_size = self->max_data_size; - } - - /* Copy data */ - skb_put_data(skb, buf + len, size); - - count -= size; - len += size; - } - - spin_unlock_irqrestore(&self->spinlock, flags); - - /* - * Schedule a new thread which will transmit the frame as soon - * as possible, but at a safe point in time. We do this so the - * "user" can give us data multiple times, as PPP does (because of - * its 256 byte tx buffer). We will then defragment and send out - * all this data as one single packet. - */ - schedule_work(&self->tqueue); - - return len; -} - -/* - * Function ircomm_tty_write_room (tty) - * - * This routine returns the numbers of characters the tty driver will - * accept for queuing to be written. This number is subject to change as - * output buffers get emptied, or if the output flow control is acted. - */ -static int ircomm_tty_write_room(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned long flags; - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - -#ifdef IRCOMM_NO_TX_BEFORE_INIT - /* max_header_size tells us if the channel is initialised or not. */ - if (self->max_header_size == IRCOMM_TTY_HDR_UNINITIALISED) - /* Don't bother us yet */ - return 0; -#endif - - /* Check if we are allowed to transmit any data. - * hw_stopped is the regular flow control. - * Jean II */ - if (tty->hw_stopped) - ret = 0; - else { - spin_lock_irqsave(&self->spinlock, flags); - if (self->tx_skb) - ret = self->tx_data_size - self->tx_skb->len; - else - ret = self->max_data_size; - spin_unlock_irqrestore(&self->spinlock, flags); - } - pr_debug("%s(), ret=%d\n", __func__ , ret); - - return ret; -} - -/* - * Function ircomm_tty_wait_until_sent (tty, timeout) - * - * This routine waits until the device has written out all of the - * characters in its transmitter FIFO. - */ -static void ircomm_tty_wait_until_sent(struct tty_struct *tty, int timeout) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned long orig_jiffies, poll_time; - unsigned long flags; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - orig_jiffies = jiffies; - - /* Set poll time to 200 ms */ - poll_time = msecs_to_jiffies(200); - if (timeout) - poll_time = min_t(unsigned long, timeout, poll_time); - - spin_lock_irqsave(&self->spinlock, flags); - while (self->tx_skb && self->tx_skb->len) { - spin_unlock_irqrestore(&self->spinlock, flags); - schedule_timeout_interruptible(poll_time); - spin_lock_irqsave(&self->spinlock, flags); - if (signal_pending(current)) - break; - if (timeout && time_after(jiffies, orig_jiffies + timeout)) - break; - } - spin_unlock_irqrestore(&self->spinlock, flags); - __set_current_state(TASK_RUNNING); -} - -/* - * Function ircomm_tty_throttle (tty) - * - * This routine notifies the tty driver that input buffers for the line - * discipline are close to full, and it should somehow signal that no - * more characters should be sent to the tty. - */ -static void ircomm_tty_throttle(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* Software flow control? */ - if (I_IXOFF(tty)) - ircomm_tty_send_xchar(tty, STOP_CHAR(tty)); - - /* Hardware flow control? */ - if (C_CRTSCTS(tty)) { - self->settings.dte &= ~IRCOMM_RTS; - self->settings.dte |= IRCOMM_DELTA_RTS; - - ircomm_param_request(self, IRCOMM_DTE, TRUE); - } - - ircomm_flow_request(self->ircomm, FLOW_STOP); -} - -/* - * Function ircomm_tty_unthrottle (tty) - * - * This routine notifies the tty drivers that it should signals that - * characters can now be sent to the tty without fear of overrunning the - * input buffers of the line disciplines. - */ -static void ircomm_tty_unthrottle(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* Using software flow control? */ - if (I_IXOFF(tty)) - ircomm_tty_send_xchar(tty, START_CHAR(tty)); - - /* Using hardware flow control? */ - if (C_CRTSCTS(tty)) { - self->settings.dte |= (IRCOMM_RTS|IRCOMM_DELTA_RTS); - - ircomm_param_request(self, IRCOMM_DTE, TRUE); - pr_debug("%s(), FLOW_START\n", __func__); - } - ircomm_flow_request(self->ircomm, FLOW_START); -} - -/* - * Function ircomm_tty_chars_in_buffer (tty) - * - * Indicates if there are any data in the buffer - * - */ -static int ircomm_tty_chars_in_buffer(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned long flags; - int len = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - spin_lock_irqsave(&self->spinlock, flags); - - if (self->tx_skb) - len = self->tx_skb->len; - - spin_unlock_irqrestore(&self->spinlock, flags); - - return len; -} - -static void ircomm_tty_shutdown(struct ircomm_tty_cb *self) -{ - unsigned long flags; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - if (!tty_port_initialized(&self->port)) - return; - tty_port_set_initialized(&self->port, 0); - - ircomm_tty_detach_cable(self); - - spin_lock_irqsave(&self->spinlock, flags); - - del_timer(&self->watchdog_timer); - - /* Free parameter buffer */ - if (self->ctrl_skb) { - dev_kfree_skb(self->ctrl_skb); - self->ctrl_skb = NULL; - } - - /* Free transmit buffer */ - if (self->tx_skb) { - dev_kfree_skb(self->tx_skb); - self->tx_skb = NULL; - } - - if (self->ircomm) { - ircomm_close(self->ircomm); - self->ircomm = NULL; - } - - spin_unlock_irqrestore(&self->spinlock, flags); -} - -/* - * Function ircomm_tty_hangup (tty) - * - * This routine notifies the tty driver that it should hangup the tty - * device. - * - */ -static void ircomm_tty_hangup(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - struct tty_port *port = &self->port; - unsigned long flags; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* ircomm_tty_flush_buffer(tty); */ - ircomm_tty_shutdown(self); - - spin_lock_irqsave(&port->lock, flags); - if (port->tty) { - set_bit(TTY_IO_ERROR, &port->tty->flags); - tty_kref_put(port->tty); - } - port->tty = NULL; - port->count = 0; - spin_unlock_irqrestore(&port->lock, flags); - tty_port_set_active(port, 0); - - wake_up_interruptible(&port->open_wait); -} - -/* - * Function ircomm_tty_send_xchar (tty, ch) - * - * This routine is used to send a high-priority XON/XOFF character to - * the device. - */ -static void ircomm_tty_send_xchar(struct tty_struct *tty, char ch) -{ - pr_debug("%s(), not impl\n", __func__); -} - -/* - * Function ircomm_tty_start (tty) - * - * This routine notifies the tty driver that it resume sending - * characters to the tty device. - */ -void ircomm_tty_start(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - ircomm_flow_request(self->ircomm, FLOW_START); -} - -/* - * Function ircomm_tty_stop (tty) - * - * This routine notifies the tty driver that it should stop outputting - * characters to the tty device. - */ -static void ircomm_tty_stop(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - ircomm_flow_request(self->ircomm, FLOW_STOP); -} - -/* - * Function ircomm_check_modem_status (self) - * - * Check for any changes in the DCE's line settings. This function should - * be called whenever the dce parameter settings changes, to update the - * flow control settings and other things - */ -void ircomm_tty_check_modem_status(struct ircomm_tty_cb *self) -{ - struct tty_struct *tty; - int status; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - tty = tty_port_tty_get(&self->port); - - status = self->settings.dce; - - if (status & IRCOMM_DCE_DELTA_ANY) { - /*wake_up_interruptible(&self->delta_msr_wait);*/ - } - if (tty_port_check_carrier(&self->port) && (status & IRCOMM_DELTA_CD)) { - pr_debug("%s(), ircomm%d CD now %s...\n", __func__ , self->line, - (status & IRCOMM_CD) ? "on" : "off"); - - if (status & IRCOMM_CD) { - wake_up_interruptible(&self->port.open_wait); - } else { - pr_debug("%s(), Doing serial hangup..\n", __func__); - if (tty) - tty_hangup(tty); - - /* Hangup will remote the tty, so better break out */ - goto put; - } - } - if (tty && tty_port_cts_enabled(&self->port)) { - if (tty->hw_stopped) { - if (status & IRCOMM_CTS) { - pr_debug("%s(), CTS tx start...\n", __func__); - tty->hw_stopped = 0; - - /* Wake up processes blocked on open */ - wake_up_interruptible(&self->port.open_wait); - - schedule_work(&self->tqueue); - goto put; - } - } else { - if (!(status & IRCOMM_CTS)) { - pr_debug("%s(), CTS tx stop...\n", __func__); - tty->hw_stopped = 1; - } - } - } -put: - tty_kref_put(tty); -} - -/* - * Function ircomm_tty_data_indication (instance, sap, skb) - * - * Handle incoming data, and deliver it to the line discipline - * - */ -static int ircomm_tty_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - struct tty_struct *tty; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - tty = tty_port_tty_get(&self->port); - if (!tty) { - pr_debug("%s(), no tty!\n", __func__); - return 0; - } - - /* - * If we receive data when hardware is stopped then something is wrong. - * We try to poll the peers line settings to check if we are up todate. - * Devices like WinCE can do this, and since they don't send any - * params, we can just as well declare the hardware for running. - */ - if (tty->hw_stopped && (self->flow == FLOW_START)) { - pr_debug("%s(), polling for line settings!\n", __func__); - ircomm_param_request(self, IRCOMM_POLL, TRUE); - - /* We can just as well declare the hardware for running */ - ircomm_tty_send_initial_parameters(self); - ircomm_tty_link_established(self); - } - tty_kref_put(tty); - - /* - * Use flip buffer functions since the code may be called from interrupt - * context - */ - tty_insert_flip_string(&self->port, skb->data, skb->len); - tty_flip_buffer_push(&self->port); - - /* No need to kfree_skb - see ircomm_ttp_data_indication() */ - - return 0; -} - -/* - * Function ircomm_tty_control_indication (instance, sap, skb) - * - * Parse all incoming parameters (easy!) - * - */ -static int ircomm_tty_control_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - int clen; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - clen = skb->data[0]; - - irda_param_extract_all(self, skb->data+1, IRDA_MIN(skb->len-1, clen), - &ircomm_param_info); - - /* No need to kfree_skb - see ircomm_control_indication() */ - - return 0; -} - -/* - * Function ircomm_tty_flow_indication (instance, sap, cmd) - * - * This function is called by IrTTP when it wants us to slow down the - * transmission of data. We just mark the hardware as stopped, and wait - * for IrTTP to notify us that things are OK again. - */ -static void ircomm_tty_flow_indication(void *instance, void *sap, - LOCAL_FLOW cmd) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - struct tty_struct *tty; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - tty = tty_port_tty_get(&self->port); - - switch (cmd) { - case FLOW_START: - pr_debug("%s(), hw start!\n", __func__); - if (tty) - tty->hw_stopped = 0; - - /* ircomm_tty_do_softint will take care of the rest */ - schedule_work(&self->tqueue); - break; - default: /* If we get here, something is very wrong, better stop */ - case FLOW_STOP: - pr_debug("%s(), hw stopped!\n", __func__); - if (tty) - tty->hw_stopped = 1; - break; - } - - tty_kref_put(tty); - self->flow = cmd; -} - -#ifdef CONFIG_PROC_FS -static void ircomm_tty_line_info(struct ircomm_tty_cb *self, struct seq_file *m) -{ - struct tty_struct *tty; - char sep; - - seq_printf(m, "State: %s\n", ircomm_tty_state[self->state]); - - seq_puts(m, "Service type: "); - if (self->service_type & IRCOMM_9_WIRE) - seq_puts(m, "9_WIRE"); - else if (self->service_type & IRCOMM_3_WIRE) - seq_puts(m, "3_WIRE"); - else if (self->service_type & IRCOMM_3_WIRE_RAW) - seq_puts(m, "3_WIRE_RAW"); - else - seq_puts(m, "No common service type!\n"); - seq_putc(m, '\n'); - - seq_printf(m, "Port name: %s\n", self->settings.port_name); - - seq_printf(m, "DTE status:"); - sep = ' '; - if (self->settings.dte & IRCOMM_RTS) { - seq_printf(m, "%cRTS", sep); - sep = '|'; - } - if (self->settings.dte & IRCOMM_DTR) { - seq_printf(m, "%cDTR", sep); - sep = '|'; - } - seq_putc(m, '\n'); - - seq_puts(m, "DCE status:"); - sep = ' '; - if (self->settings.dce & IRCOMM_CTS) { - seq_printf(m, "%cCTS", sep); - sep = '|'; - } - if (self->settings.dce & IRCOMM_DSR) { - seq_printf(m, "%cDSR", sep); - sep = '|'; - } - if (self->settings.dce & IRCOMM_CD) { - seq_printf(m, "%cCD", sep); - sep = '|'; - } - if (self->settings.dce & IRCOMM_RI) { - seq_printf(m, "%cRI", sep); - sep = '|'; - } - seq_putc(m, '\n'); - - seq_puts(m, "Configuration: "); - if (!self->settings.null_modem) - seq_puts(m, "DTE <-> DCE\n"); - else - seq_puts(m, "DTE <-> DTE (null modem emulation)\n"); - - seq_printf(m, "Data rate: %d\n", self->settings.data_rate); - - seq_puts(m, "Flow control:"); - sep = ' '; - if (self->settings.flow_control & IRCOMM_XON_XOFF_IN) { - seq_printf(m, "%cXON_XOFF_IN", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_XON_XOFF_OUT) { - seq_printf(m, "%cXON_XOFF_OUT", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_RTS_CTS_IN) { - seq_printf(m, "%cRTS_CTS_IN", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_RTS_CTS_OUT) { - seq_printf(m, "%cRTS_CTS_OUT", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_DSR_DTR_IN) { - seq_printf(m, "%cDSR_DTR_IN", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_DSR_DTR_OUT) { - seq_printf(m, "%cDSR_DTR_OUT", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_ENQ_ACK_IN) { - seq_printf(m, "%cENQ_ACK_IN", sep); - sep = '|'; - } - if (self->settings.flow_control & IRCOMM_ENQ_ACK_OUT) { - seq_printf(m, "%cENQ_ACK_OUT", sep); - sep = '|'; - } - seq_putc(m, '\n'); - - seq_puts(m, "Flags:"); - sep = ' '; - if (tty_port_cts_enabled(&self->port)) { - seq_printf(m, "%cASYNC_CTS_FLOW", sep); - sep = '|'; - } - if (tty_port_check_carrier(&self->port)) { - seq_printf(m, "%cASYNC_CHECK_CD", sep); - sep = '|'; - } - if (tty_port_initialized(&self->port)) { - seq_printf(m, "%cASYNC_INITIALIZED", sep); - sep = '|'; - } - if (self->port.flags & ASYNC_LOW_LATENCY) { - seq_printf(m, "%cASYNC_LOW_LATENCY", sep); - sep = '|'; - } - if (tty_port_active(&self->port)) { - seq_printf(m, "%cASYNC_NORMAL_ACTIVE", sep); - sep = '|'; - } - seq_putc(m, '\n'); - - seq_printf(m, "Role: %s\n", self->client ? "client" : "server"); - seq_printf(m, "Open count: %d\n", self->port.count); - seq_printf(m, "Max data size: %d\n", self->max_data_size); - seq_printf(m, "Max header size: %d\n", self->max_header_size); - - tty = tty_port_tty_get(&self->port); - if (tty) { - seq_printf(m, "Hardware: %s\n", - tty->hw_stopped ? "Stopped" : "Running"); - tty_kref_put(tty); - } -} - -static int ircomm_tty_proc_show(struct seq_file *m, void *v) -{ - struct ircomm_tty_cb *self; - unsigned long flags; - - spin_lock_irqsave(&ircomm_tty->hb_spinlock, flags); - - self = (struct ircomm_tty_cb *) hashbin_get_first(ircomm_tty); - while (self != NULL) { - if (self->magic != IRCOMM_TTY_MAGIC) - break; - - ircomm_tty_line_info(self, m); - self = (struct ircomm_tty_cb *) hashbin_get_next(ircomm_tty); - } - spin_unlock_irqrestore(&ircomm_tty->hb_spinlock, flags); - return 0; -} - -static int ircomm_tty_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, ircomm_tty_proc_show, NULL); -} - -static const struct file_operations ircomm_tty_proc_fops = { - .owner = THIS_MODULE, - .open = ircomm_tty_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; -#endif /* CONFIG_PROC_FS */ - -MODULE_AUTHOR("Dag Brattli <dagb@cs.uit.no>"); -MODULE_DESCRIPTION("IrCOMM serial TTY driver"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_CHARDEV_MAJOR(IRCOMM_TTY_MAJOR); - -module_init(ircomm_tty_init); -module_exit(ircomm_tty_cleanup); diff --git a/net/irda/ircomm/ircomm_tty_attach.c b/net/irda/ircomm/ircomm_tty_attach.c deleted file mode 100644 index 0a411019c098..000000000000 --- a/net/irda/ircomm/ircomm_tty_attach.c +++ /dev/null @@ -1,987 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_tty_attach.c - * Version: - * Description: Code for attaching the serial driver to IrCOMM - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sat Jun 5 17:42:00 1999 - * Modified at: Tue Jan 4 14:20:49 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/init.h> -#include <linux/sched.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/irttp.h> -#include <net/irda/irias_object.h> -#include <net/irda/parameters.h> - -#include <net/irda/ircomm_core.h> -#include <net/irda/ircomm_param.h> -#include <net/irda/ircomm_event.h> - -#include <net/irda/ircomm_tty.h> -#include <net/irda/ircomm_tty_attach.h> - -static void ircomm_tty_ias_register(struct ircomm_tty_cb *self); -static void ircomm_tty_discovery_indication(discinfo_t *discovery, - DISCOVERY_MODE mode, - void *priv); -static void ircomm_tty_getvalue_confirm(int result, __u16 obj_id, - struct ias_value *value, void *priv); -static void ircomm_tty_start_watchdog_timer(struct ircomm_tty_cb *self, - int timeout); -static void ircomm_tty_watchdog_timer_expired(void *data); - -static int ircomm_tty_state_idle(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); -static int ircomm_tty_state_search(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); -static int ircomm_tty_state_query_parameters(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); -static int ircomm_tty_state_query_lsap_sel(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); -static int ircomm_tty_state_setup(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); -static int ircomm_tty_state_ready(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info); - -const char *const ircomm_tty_state[] = { - "IRCOMM_TTY_IDLE", - "IRCOMM_TTY_SEARCH", - "IRCOMM_TTY_QUERY_PARAMETERS", - "IRCOMM_TTY_QUERY_LSAP_SEL", - "IRCOMM_TTY_SETUP", - "IRCOMM_TTY_READY", - "*** ERROR *** ", -}; - -static const char *const ircomm_tty_event[] __maybe_unused = { - "IRCOMM_TTY_ATTACH_CABLE", - "IRCOMM_TTY_DETACH_CABLE", - "IRCOMM_TTY_DATA_REQUEST", - "IRCOMM_TTY_DATA_INDICATION", - "IRCOMM_TTY_DISCOVERY_REQUEST", - "IRCOMM_TTY_DISCOVERY_INDICATION", - "IRCOMM_TTY_CONNECT_CONFIRM", - "IRCOMM_TTY_CONNECT_INDICATION", - "IRCOMM_TTY_DISCONNECT_REQUEST", - "IRCOMM_TTY_DISCONNECT_INDICATION", - "IRCOMM_TTY_WD_TIMER_EXPIRED", - "IRCOMM_TTY_GOT_PARAMETERS", - "IRCOMM_TTY_GOT_LSAPSEL", - "*** ERROR ****", -}; - -static int (*state[])(struct ircomm_tty_cb *self, IRCOMM_TTY_EVENT event, - struct sk_buff *skb, struct ircomm_tty_info *info) = -{ - ircomm_tty_state_idle, - ircomm_tty_state_search, - ircomm_tty_state_query_parameters, - ircomm_tty_state_query_lsap_sel, - ircomm_tty_state_setup, - ircomm_tty_state_ready, -}; - -/* - * Function ircomm_tty_attach_cable (driver) - * - * Try to attach cable (IrCOMM link). This function will only return - * when the link has been connected, or if an error condition occurs. - * If success, the return value is the resulting service type. - */ -int ircomm_tty_attach_cable(struct ircomm_tty_cb *self) -{ - struct tty_struct *tty; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - /* Check if somebody has already connected to us */ - if (ircomm_is_connected(self->ircomm)) { - pr_debug("%s(), already connected!\n", __func__); - return 0; - } - - /* Make sure nobody tries to write before the link is up */ - tty = tty_port_tty_get(&self->port); - if (tty) { - tty->hw_stopped = 1; - tty_kref_put(tty); - } - - ircomm_tty_ias_register(self); - - ircomm_tty_do_event(self, IRCOMM_TTY_ATTACH_CABLE, NULL, NULL); - - return 0; -} - -/* - * Function ircomm_detach_cable (driver) - * - * Detach cable, or cable has been detached by peer - * - */ -void ircomm_tty_detach_cable(struct ircomm_tty_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - del_timer(&self->watchdog_timer); - - /* Remove discovery handler */ - if (self->ckey) { - irlmp_unregister_client(self->ckey); - self->ckey = NULL; - } - /* Remove IrCOMM hint bits */ - if (self->skey) { - irlmp_unregister_service(self->skey); - self->skey = NULL; - } - - if (self->iriap) { - iriap_close(self->iriap); - self->iriap = NULL; - } - - /* Remove LM-IAS object */ - if (self->obj) { - irias_delete_object(self->obj); - self->obj = NULL; - } - - ircomm_tty_do_event(self, IRCOMM_TTY_DETACH_CABLE, NULL, NULL); - - /* Reset some values */ - self->daddr = self->saddr = 0; - self->dlsap_sel = self->slsap_sel = 0; - - memset(&self->settings, 0, sizeof(struct ircomm_params)); -} - -/* - * Function ircomm_tty_ias_register (self) - * - * Register with LM-IAS depending on which service type we are - * - */ -static void ircomm_tty_ias_register(struct ircomm_tty_cb *self) -{ - __u8 oct_seq[6]; - __u16 hints; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* Compute hint bits based on service */ - hints = irlmp_service_to_hint(S_COMM); - if (self->service_type & IRCOMM_3_WIRE_RAW) - hints |= irlmp_service_to_hint(S_PRINTER); - - /* Advertise IrCOMM hint bit in discovery */ - if (!self->skey) - self->skey = irlmp_register_service(hints); - /* Set up a discovery handler */ - if (!self->ckey) - self->ckey = irlmp_register_client(hints, - ircomm_tty_discovery_indication, - NULL, (void *) self); - - /* If already done, no need to do it again */ - if (self->obj) - return; - - if (self->service_type & IRCOMM_3_WIRE_RAW) { - /* Register IrLPT with LM-IAS */ - self->obj = irias_new_object("IrLPT", IAS_IRLPT_ID); - irias_add_integer_attrib(self->obj, "IrDA:IrLMP:LsapSel", - self->slsap_sel, IAS_KERNEL_ATTR); - } else { - /* Register IrCOMM with LM-IAS */ - self->obj = irias_new_object("IrDA:IrCOMM", IAS_IRCOMM_ID); - irias_add_integer_attrib(self->obj, "IrDA:TinyTP:LsapSel", - self->slsap_sel, IAS_KERNEL_ATTR); - - /* Code the parameters into the buffer */ - irda_param_pack(oct_seq, "bbbbbb", - IRCOMM_SERVICE_TYPE, 1, self->service_type, - IRCOMM_PORT_TYPE, 1, IRCOMM_SERIAL); - - /* Register parameters with LM-IAS */ - irias_add_octseq_attrib(self->obj, "Parameters", oct_seq, 6, - IAS_KERNEL_ATTR); - } - irias_insert_object(self->obj); -} - -/* - * Function ircomm_tty_ias_unregister (self) - * - * Remove our IAS object and client hook while connected. - * - */ -static void ircomm_tty_ias_unregister(struct ircomm_tty_cb *self) -{ - /* Remove LM-IAS object now so it is not reused. - * IrCOMM deals very poorly with multiple incoming connections. - * It should looks a lot more like IrNET, and "dup" a server TSAP - * to the application TSAP (based on various rules). - * This is a cheap workaround allowing multiple clients to - * connect to us. It will not always work. - * Each IrCOMM socket has an IAS entry. Incoming connection will - * pick the first one found. So, when we are fully connected, - * we remove our IAS entries so that the next IAS entry is used. - * We do that for *both* client and server, because a server - * can also create client instances. - * Jean II */ - if (self->obj) { - irias_delete_object(self->obj); - self->obj = NULL; - } - -#if 0 - /* Remove discovery handler. - * While we are connected, we no longer need to receive - * discovery events. This would be the case if there is - * multiple IrLAP interfaces. Jean II */ - if (self->ckey) { - irlmp_unregister_client(self->ckey); - self->ckey = NULL; - } -#endif -} - -/* - * Function ircomm_send_initial_parameters (self) - * - * Send initial parameters to the remote IrCOMM device. These parameters - * must be sent before any data. - */ -int ircomm_tty_send_initial_parameters(struct ircomm_tty_cb *self) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (self->service_type & IRCOMM_3_WIRE_RAW) - return 0; - - /* - * Set default values, but only if the application for some reason - * haven't set them already - */ - pr_debug("%s(), data-rate = %d\n", __func__ , - self->settings.data_rate); - if (!self->settings.data_rate) - self->settings.data_rate = 9600; - pr_debug("%s(), data-format = %d\n", __func__ , - self->settings.data_format); - if (!self->settings.data_format) - self->settings.data_format = IRCOMM_WSIZE_8; /* 8N1 */ - - pr_debug("%s(), flow-control = %d\n", __func__ , - self->settings.flow_control); - /*self->settings.flow_control = IRCOMM_RTS_CTS_IN|IRCOMM_RTS_CTS_OUT;*/ - - /* Do not set delta values for the initial parameters */ - self->settings.dte = IRCOMM_DTR | IRCOMM_RTS; - - /* Only send service type parameter when we are the client */ - if (self->client) - ircomm_param_request(self, IRCOMM_SERVICE_TYPE, FALSE); - ircomm_param_request(self, IRCOMM_DATA_RATE, FALSE); - ircomm_param_request(self, IRCOMM_DATA_FORMAT, FALSE); - - /* For a 3 wire service, we just flush the last parameter and return */ - if (self->settings.service_type == IRCOMM_3_WIRE) { - ircomm_param_request(self, IRCOMM_FLOW_CONTROL, TRUE); - return 0; - } - - /* Only 9-wire service types continue here */ - ircomm_param_request(self, IRCOMM_FLOW_CONTROL, FALSE); -#if 0 - ircomm_param_request(self, IRCOMM_XON_XOFF, FALSE); - ircomm_param_request(self, IRCOMM_ENQ_ACK, FALSE); -#endif - /* Notify peer that we are ready to receive data */ - ircomm_param_request(self, IRCOMM_DTE, TRUE); - - return 0; -} - -/* - * Function ircomm_tty_discovery_indication (discovery) - * - * Remote device is discovered, try query the remote IAS to see which - * device it is, and which services it has. - * - */ -static void ircomm_tty_discovery_indication(discinfo_t *discovery, - DISCOVERY_MODE mode, - void *priv) -{ - struct ircomm_tty_cb *self; - struct ircomm_tty_info info; - - /* Important note : - * We need to drop all passive discoveries. - * The LSAP management of IrComm is deficient and doesn't deal - * with the case of two instance connecting to each other - * simultaneously (it will deadlock in LMP). - * The proper fix would be to use the same technique as in IrNET, - * to have one server socket and separate instances for the - * connecting/connected socket. - * The workaround is to drop passive discovery, which drastically - * reduce the probability of this happening. - * Jean II */ - if(mode == DISCOVERY_PASSIVE) - return; - - info.daddr = discovery->daddr; - info.saddr = discovery->saddr; - - self = priv; - ircomm_tty_do_event(self, IRCOMM_TTY_DISCOVERY_INDICATION, - NULL, &info); -} - -/* - * Function ircomm_tty_disconnect_indication (instance, sap, reason, skb) - * - * Link disconnected - * - */ -void ircomm_tty_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *skb) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - struct tty_struct *tty; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - tty = tty_port_tty_get(&self->port); - if (!tty) - return; - - /* This will stop control data transfers */ - self->flow = FLOW_STOP; - - /* Stop data transfers */ - tty->hw_stopped = 1; - - ircomm_tty_do_event(self, IRCOMM_TTY_DISCONNECT_INDICATION, NULL, - NULL); - tty_kref_put(tty); -} - -/* - * Function ircomm_tty_getvalue_confirm (result, obj_id, value, priv) - * - * Got result from the IAS query we make - * - */ -static void ircomm_tty_getvalue_confirm(int result, __u16 obj_id, - struct ias_value *value, - void *priv) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) priv; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - /* We probably don't need to make any more queries */ - iriap_close(self->iriap); - self->iriap = NULL; - - /* Check if request succeeded */ - if (result != IAS_SUCCESS) { - pr_debug("%s(), got NULL value!\n", __func__); - return; - } - - switch (value->type) { - case IAS_OCT_SEQ: - pr_debug("%s(), got octet sequence\n", __func__); - - irda_param_extract_all(self, value->t.oct_seq, value->len, - &ircomm_param_info); - - ircomm_tty_do_event(self, IRCOMM_TTY_GOT_PARAMETERS, NULL, - NULL); - break; - case IAS_INTEGER: - /* Got LSAP selector */ - pr_debug("%s(), got lsapsel = %d\n", __func__ , - value->t.integer); - - if (value->t.integer == -1) { - pr_debug("%s(), invalid value!\n", __func__); - } else - self->dlsap_sel = value->t.integer; - - ircomm_tty_do_event(self, IRCOMM_TTY_GOT_LSAPSEL, NULL, NULL); - break; - case IAS_MISSING: - pr_debug("%s(), got IAS_MISSING\n", __func__); - break; - default: - pr_debug("%s(), got unknown type!\n", __func__); - break; - } - irias_delete_value(value); -} - -/* - * Function ircomm_tty_connect_confirm (instance, sap, qos, max_sdu_size, skb) - * - * Connection confirmed - * - */ -void ircomm_tty_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_data_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - self->client = TRUE; - self->max_data_size = max_data_size; - self->max_header_size = max_header_size; - self->flow = FLOW_START; - - ircomm_tty_do_event(self, IRCOMM_TTY_CONNECT_CONFIRM, NULL, NULL); - - /* No need to kfree_skb - see ircomm_ttp_connect_confirm() */ -} - -/* - * Function ircomm_tty_connect_indication (instance, sap, qos, max_sdu_size, - * skb) - * - * we are discovered and being requested to connect by remote device ! - * - */ -void ircomm_tty_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_data_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance; - int clen; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - self->client = FALSE; - self->max_data_size = max_data_size; - self->max_header_size = max_header_size; - self->flow = FLOW_START; - - clen = skb->data[0]; - if (clen) - irda_param_extract_all(self, skb->data+1, - IRDA_MIN(skb->len, clen), - &ircomm_param_info); - - ircomm_tty_do_event(self, IRCOMM_TTY_CONNECT_INDICATION, NULL, NULL); - - /* No need to kfree_skb - see ircomm_ttp_connect_indication() */ -} - -/* - * Function ircomm_tty_link_established (self) - * - * Called when the IrCOMM link is established - * - */ -void ircomm_tty_link_established(struct ircomm_tty_cb *self) -{ - struct tty_struct *tty; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - tty = tty_port_tty_get(&self->port); - if (!tty) - return; - - del_timer(&self->watchdog_timer); - - /* - * IrCOMM link is now up, and if we are not using hardware - * flow-control, then declare the hardware as running. Otherwise we - * will have to wait for the peer device (DCE) to raise the CTS - * line. - */ - if (tty_port_cts_enabled(&self->port) && - ((self->settings.dce & IRCOMM_CTS) == 0)) { - pr_debug("%s(), waiting for CTS ...\n", __func__); - goto put; - } else { - pr_debug("%s(), starting hardware!\n", __func__); - - tty->hw_stopped = 0; - - /* Wake up processes blocked on open */ - wake_up_interruptible(&self->port.open_wait); - } - - schedule_work(&self->tqueue); -put: - tty_kref_put(tty); -} - -/* - * Function ircomm_tty_start_watchdog_timer (self, timeout) - * - * Start the watchdog timer. This timer is used to make sure that any - * connection attempt is successful, and if not, we will retry after - * the timeout - */ -static void ircomm_tty_start_watchdog_timer(struct ircomm_tty_cb *self, - int timeout) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - irda_start_timer(&self->watchdog_timer, timeout, (void *) self, - ircomm_tty_watchdog_timer_expired); -} - -/* - * Function ircomm_tty_watchdog_timer_expired (data) - * - * Called when the connect procedure have taken to much time. - * - */ -static void ircomm_tty_watchdog_timer_expired(void *data) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - ircomm_tty_do_event(self, IRCOMM_TTY_WD_TIMER_EXPIRED, NULL, NULL); -} - - -/* - * Function ircomm_tty_do_event (self, event, skb) - * - * Process event - * - */ -int ircomm_tty_do_event(struct ircomm_tty_cb *self, IRCOMM_TTY_EVENT event, - struct sk_buff *skb, struct ircomm_tty_info *info) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - - return (*state[self->state])(self, event, skb, info); -} - -/* - * Function ircomm_tty_next_state (self, state) - * - * Switch state - * - */ -static inline void ircomm_tty_next_state(struct ircomm_tty_cb *self, IRCOMM_TTY_STATE state) -{ - /* - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return;); - - pr_debug("%s: next state=%s, service type=%d\n", __func__ , - ircomm_tty_state[self->state], self->service_type); - */ - self->state = state; -} - -/* - * Function ircomm_tty_state_idle (self, event, skb, info) - * - * Just hanging around - * - */ -static int ircomm_tty_state_idle(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - switch (event) { - case IRCOMM_TTY_ATTACH_CABLE: - /* Try to discover any remote devices */ - ircomm_tty_start_watchdog_timer(self, 3*HZ); - ircomm_tty_next_state(self, IRCOMM_TTY_SEARCH); - - irlmp_discovery_request(DISCOVERY_DEFAULT_SLOTS); - break; - case IRCOMM_TTY_DISCOVERY_INDICATION: - self->daddr = info->daddr; - self->saddr = info->saddr; - - if (self->iriap) { - net_warn_ratelimited("%s(), busy with a previous query\n", - __func__); - return -EBUSY; - } - - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - ircomm_tty_getvalue_confirm); - - iriap_getvaluebyclass_request(self->iriap, - self->saddr, self->daddr, - "IrDA:IrCOMM", "Parameters"); - - ircomm_tty_start_watchdog_timer(self, 3*HZ); - ircomm_tty_next_state(self, IRCOMM_TTY_QUERY_PARAMETERS); - break; - case IRCOMM_TTY_CONNECT_INDICATION: - del_timer(&self->watchdog_timer); - - /* Accept connection */ - ircomm_connect_response(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_WD_TIMER_EXPIRED: - /* Just stay idle */ - break; - case IRCOMM_TTY_DETACH_CABLE: - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_tty_state_search (self, event, skb, info) - * - * Trying to discover an IrCOMM device - * - */ -static int ircomm_tty_state_search(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - - switch (event) { - case IRCOMM_TTY_DISCOVERY_INDICATION: - self->daddr = info->daddr; - self->saddr = info->saddr; - - if (self->iriap) { - net_warn_ratelimited("%s(), busy with a previous query\n", - __func__); - return -EBUSY; - } - - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - ircomm_tty_getvalue_confirm); - - if (self->service_type == IRCOMM_3_WIRE_RAW) { - iriap_getvaluebyclass_request(self->iriap, self->saddr, - self->daddr, "IrLPT", - "IrDA:IrLMP:LsapSel"); - ircomm_tty_next_state(self, IRCOMM_TTY_QUERY_LSAP_SEL); - } else { - iriap_getvaluebyclass_request(self->iriap, self->saddr, - self->daddr, - "IrDA:IrCOMM", - "Parameters"); - - ircomm_tty_next_state(self, IRCOMM_TTY_QUERY_PARAMETERS); - } - ircomm_tty_start_watchdog_timer(self, 3*HZ); - break; - case IRCOMM_TTY_CONNECT_INDICATION: - del_timer(&self->watchdog_timer); - ircomm_tty_ias_unregister(self); - - /* Accept connection */ - ircomm_connect_response(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_WD_TIMER_EXPIRED: -#if 1 - /* Give up */ -#else - /* Try to discover any remote devices */ - ircomm_tty_start_watchdog_timer(self, 3*HZ); - irlmp_discovery_request(DISCOVERY_DEFAULT_SLOTS); -#endif - break; - case IRCOMM_TTY_DETACH_CABLE: - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_tty_state_query (self, event, skb, info) - * - * Querying the remote LM-IAS for IrCOMM parameters - * - */ -static int ircomm_tty_state_query_parameters(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - - switch (event) { - case IRCOMM_TTY_GOT_PARAMETERS: - if (self->iriap) { - net_warn_ratelimited("%s(), busy with a previous query\n", - __func__); - return -EBUSY; - } - - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - ircomm_tty_getvalue_confirm); - - iriap_getvaluebyclass_request(self->iriap, self->saddr, - self->daddr, "IrDA:IrCOMM", - "IrDA:TinyTP:LsapSel"); - - ircomm_tty_start_watchdog_timer(self, 3*HZ); - ircomm_tty_next_state(self, IRCOMM_TTY_QUERY_LSAP_SEL); - break; - case IRCOMM_TTY_WD_TIMER_EXPIRED: - /* Go back to search mode */ - ircomm_tty_next_state(self, IRCOMM_TTY_SEARCH); - ircomm_tty_start_watchdog_timer(self, 3*HZ); - break; - case IRCOMM_TTY_CONNECT_INDICATION: - del_timer(&self->watchdog_timer); - ircomm_tty_ias_unregister(self); - - /* Accept connection */ - ircomm_connect_response(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_DETACH_CABLE: - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_tty_state_query_lsap_sel (self, event, skb, info) - * - * Query remote LM-IAS for the LSAP selector which we can connect to - * - */ -static int ircomm_tty_state_query_lsap_sel(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - - switch (event) { - case IRCOMM_TTY_GOT_LSAPSEL: - /* Connect to remote device */ - ret = ircomm_connect_request(self->ircomm, self->dlsap_sel, - self->saddr, self->daddr, - NULL, self->service_type); - ircomm_tty_start_watchdog_timer(self, 3*HZ); - ircomm_tty_next_state(self, IRCOMM_TTY_SETUP); - break; - case IRCOMM_TTY_WD_TIMER_EXPIRED: - /* Go back to search mode */ - ircomm_tty_next_state(self, IRCOMM_TTY_SEARCH); - ircomm_tty_start_watchdog_timer(self, 3*HZ); - break; - case IRCOMM_TTY_CONNECT_INDICATION: - del_timer(&self->watchdog_timer); - ircomm_tty_ias_unregister(self); - - /* Accept connection */ - ircomm_connect_response(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_DETACH_CABLE: - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_tty_state_setup (self, event, skb, info) - * - * Trying to connect - * - */ -static int ircomm_tty_state_setup(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - pr_debug("%s: state=%s, event=%s\n", __func__ , - ircomm_tty_state[self->state], ircomm_tty_event[event]); - - switch (event) { - case IRCOMM_TTY_CONNECT_CONFIRM: - del_timer(&self->watchdog_timer); - ircomm_tty_ias_unregister(self); - - /* - * Send initial parameters. This will also send out queued - * parameters waiting for the connection to come up - */ - ircomm_tty_send_initial_parameters(self); - ircomm_tty_link_established(self); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_CONNECT_INDICATION: - del_timer(&self->watchdog_timer); - ircomm_tty_ias_unregister(self); - - /* Accept connection */ - ircomm_connect_response(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_READY); - break; - case IRCOMM_TTY_WD_TIMER_EXPIRED: - /* Go back to search mode */ - ircomm_tty_next_state(self, IRCOMM_TTY_SEARCH); - ircomm_tty_start_watchdog_timer(self, 3*HZ); - break; - case IRCOMM_TTY_DETACH_CABLE: - /* ircomm_disconnect_request(self->ircomm, NULL); */ - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - -/* - * Function ircomm_tty_state_ready (self, event, skb, info) - * - * IrCOMM is now connected - * - */ -static int ircomm_tty_state_ready(struct ircomm_tty_cb *self, - IRCOMM_TTY_EVENT event, - struct sk_buff *skb, - struct ircomm_tty_info *info) -{ - int ret = 0; - - switch (event) { - case IRCOMM_TTY_DATA_REQUEST: - ret = ircomm_data_request(self->ircomm, skb); - break; - case IRCOMM_TTY_DETACH_CABLE: - ircomm_disconnect_request(self->ircomm, NULL); - ircomm_tty_next_state(self, IRCOMM_TTY_IDLE); - break; - case IRCOMM_TTY_DISCONNECT_INDICATION: - ircomm_tty_ias_register(self); - ircomm_tty_next_state(self, IRCOMM_TTY_SEARCH); - ircomm_tty_start_watchdog_timer(self, 3*HZ); - - if (tty_port_check_carrier(&self->port)) { - /* Drop carrier */ - self->settings.dce = IRCOMM_DELTA_CD; - ircomm_tty_check_modem_status(self); - } else { - pr_debug("%s(), hanging up!\n", __func__); - tty_port_tty_hangup(&self->port, false); - } - break; - default: - pr_debug("%s(), unknown event: %s\n", __func__ , - ircomm_tty_event[event]); - ret = -EINVAL; - } - return ret; -} - diff --git a/net/irda/ircomm/ircomm_tty_ioctl.c b/net/irda/ircomm/ircomm_tty_ioctl.c deleted file mode 100644 index 171c3dee760e..000000000000 --- a/net/irda/ircomm/ircomm_tty_ioctl.c +++ /dev/null @@ -1,291 +0,0 @@ -/********************************************************************* - * - * Filename: ircomm_tty_ioctl.c - * Version: - * Description: - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Thu Jun 10 14:39:09 1999 - * Modified at: Wed Jan 5 14:45:43 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/init.h> -#include <linux/fs.h> -#include <linux/termios.h> -#include <linux/tty.h> -#include <linux/serial.h> - -#include <linux/uaccess.h> - -#include <net/irda/irda.h> -#include <net/irda/irmod.h> - -#include <net/irda/ircomm_core.h> -#include <net/irda/ircomm_param.h> -#include <net/irda/ircomm_tty_attach.h> -#include <net/irda/ircomm_tty.h> - -#define RELEVANT_IFLAG(iflag) (iflag & (IGNBRK|BRKINT|IGNPAR|PARMRK|INPCK)) - -/* - * Function ircomm_tty_change_speed (driver) - * - * Change speed of the driver. If the remote device is a DCE, then this - * should make it change the speed of its serial port - */ -static void ircomm_tty_change_speed(struct ircomm_tty_cb *self, - struct tty_struct *tty) -{ - unsigned int cflag, cval; - int baud; - - if (!self->ircomm) - return; - - cflag = tty->termios.c_cflag; - - /* byte size and parity */ - switch (cflag & CSIZE) { - case CS5: cval = IRCOMM_WSIZE_5; break; - case CS6: cval = IRCOMM_WSIZE_6; break; - case CS7: cval = IRCOMM_WSIZE_7; break; - case CS8: cval = IRCOMM_WSIZE_8; break; - default: cval = IRCOMM_WSIZE_5; break; - } - if (cflag & CSTOPB) - cval |= IRCOMM_2_STOP_BIT; - - if (cflag & PARENB) - cval |= IRCOMM_PARITY_ENABLE; - if (!(cflag & PARODD)) - cval |= IRCOMM_PARITY_EVEN; - - /* Determine divisor based on baud rate */ - baud = tty_get_baud_rate(tty); - if (!baud) - baud = 9600; /* B0 transition handled in rs_set_termios */ - - self->settings.data_rate = baud; - ircomm_param_request(self, IRCOMM_DATA_RATE, FALSE); - - /* CTS flow control flag and modem status interrupts */ - tty_port_set_cts_flow(&self->port, cflag & CRTSCTS); - if (cflag & CRTSCTS) { - self->settings.flow_control |= IRCOMM_RTS_CTS_IN; - /* This got me. Bummer. Jean II */ - if (self->service_type == IRCOMM_3_WIRE_RAW) - net_warn_ratelimited("%s(), enabling RTS/CTS on link that doesn't support it (3-wire-raw)\n", - __func__); - } else { - self->settings.flow_control &= ~IRCOMM_RTS_CTS_IN; - } - tty_port_set_check_carrier(&self->port, ~cflag & CLOCAL); - - self->settings.data_format = cval; - - ircomm_param_request(self, IRCOMM_DATA_FORMAT, FALSE); - ircomm_param_request(self, IRCOMM_FLOW_CONTROL, TRUE); -} - -/* - * Function ircomm_tty_set_termios (tty, old_termios) - * - * This routine allows the tty driver to be notified when device's - * termios settings have changed. Note that a well-designed tty driver - * should be prepared to accept the case where old == NULL, and try to - * do something rational. - */ -void ircomm_tty_set_termios(struct tty_struct *tty, - struct ktermios *old_termios) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned int cflag = tty->termios.c_cflag; - - if ((cflag == old_termios->c_cflag) && - (RELEVANT_IFLAG(tty->termios.c_iflag) == - RELEVANT_IFLAG(old_termios->c_iflag))) - { - return; - } - - ircomm_tty_change_speed(self, tty); - - /* Handle transition to B0 status */ - if ((old_termios->c_cflag & CBAUD) && !(cflag & CBAUD)) { - self->settings.dte &= ~(IRCOMM_DTR|IRCOMM_RTS); - ircomm_param_request(self, IRCOMM_DTE, TRUE); - } - - /* Handle transition away from B0 status */ - if (!(old_termios->c_cflag & CBAUD) && (cflag & CBAUD)) { - self->settings.dte |= IRCOMM_DTR; - if (!C_CRTSCTS(tty) || !tty_throttled(tty)) - self->settings.dte |= IRCOMM_RTS; - ircomm_param_request(self, IRCOMM_DTE, TRUE); - } - - /* Handle turning off CRTSCTS */ - if ((old_termios->c_cflag & CRTSCTS) && !C_CRTSCTS(tty)) - { - tty->hw_stopped = 0; - ircomm_tty_start(tty); - } -} - -/* - * Function ircomm_tty_tiocmget (tty) - * - * - * - */ -int ircomm_tty_tiocmget(struct tty_struct *tty) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - unsigned int result; - - if (tty_io_error(tty)) - return -EIO; - - result = ((self->settings.dte & IRCOMM_RTS) ? TIOCM_RTS : 0) - | ((self->settings.dte & IRCOMM_DTR) ? TIOCM_DTR : 0) - | ((self->settings.dce & IRCOMM_CD) ? TIOCM_CAR : 0) - | ((self->settings.dce & IRCOMM_RI) ? TIOCM_RNG : 0) - | ((self->settings.dce & IRCOMM_DSR) ? TIOCM_DSR : 0) - | ((self->settings.dce & IRCOMM_CTS) ? TIOCM_CTS : 0); - return result; -} - -/* - * Function ircomm_tty_tiocmset (tty, set, clear) - * - * - * - */ -int ircomm_tty_tiocmset(struct tty_struct *tty, - unsigned int set, unsigned int clear) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - - if (tty_io_error(tty)) - return -EIO; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRCOMM_TTY_MAGIC, return -1;); - - if (set & TIOCM_RTS) - self->settings.dte |= IRCOMM_RTS; - if (set & TIOCM_DTR) - self->settings.dte |= IRCOMM_DTR; - - if (clear & TIOCM_RTS) - self->settings.dte &= ~IRCOMM_RTS; - if (clear & TIOCM_DTR) - self->settings.dte &= ~IRCOMM_DTR; - - if ((set|clear) & TIOCM_RTS) - self->settings.dte |= IRCOMM_DELTA_RTS; - if ((set|clear) & TIOCM_DTR) - self->settings.dte |= IRCOMM_DELTA_DTR; - - ircomm_param_request(self, IRCOMM_DTE, TRUE); - - return 0; -} - -/* - * Function get_serial_info (driver, retinfo) - * - * - * - */ -static int ircomm_tty_get_serial_info(struct ircomm_tty_cb *self, - struct serial_struct __user *retinfo) -{ - struct serial_struct info; - - memset(&info, 0, sizeof(info)); - info.line = self->line; - info.flags = self->port.flags; - info.baud_base = self->settings.data_rate; - info.close_delay = self->port.close_delay; - info.closing_wait = self->port.closing_wait; - - /* For compatibility */ - info.type = PORT_16550A; - - if (copy_to_user(retinfo, &info, sizeof(*retinfo))) - return -EFAULT; - - return 0; -} - -/* - * Function set_serial_info (driver, new_info) - * - * - * - */ -static int ircomm_tty_set_serial_info(struct ircomm_tty_cb *self, - struct serial_struct __user *new_info) -{ - return 0; -} - -/* - * Function ircomm_tty_ioctl (tty, cmd, arg) - * - * - * - */ -int ircomm_tty_ioctl(struct tty_struct *tty, - unsigned int cmd, unsigned long arg) -{ - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; - int ret = 0; - - if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) && - (cmd != TIOCSERCONFIG) && (cmd != TIOCSERGSTRUCT) && - (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) { - if (tty_io_error(tty)) - return -EIO; - } - - switch (cmd) { - case TIOCGSERIAL: - ret = ircomm_tty_get_serial_info(self, (struct serial_struct __user *) arg); - break; - case TIOCSSERIAL: - ret = ircomm_tty_set_serial_info(self, (struct serial_struct __user *) arg); - break; - case TIOCMIWAIT: - pr_debug("(), TIOCMIWAIT, not impl!\n"); - break; - - case TIOCGICOUNT: - pr_debug("%s(), TIOCGICOUNT not impl!\n", __func__); - return 0; - default: - ret = -ENOIOCTLCMD; /* ioctls which we must ignore */ - } - return ret; -} - - - diff --git a/net/irda/irda_device.c b/net/irda/irda_device.c deleted file mode 100644 index 890b90d055d5..000000000000 --- a/net/irda/irda_device.c +++ /dev/null @@ -1,316 +0,0 @@ -/********************************************************************* - * - * Filename: irda_device.c - * Version: 0.9 - * Description: Utility functions used by the device drivers - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sat Oct 9 09:22:27 1999 - * Modified at: Sun Jan 23 17:41:24 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2001 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/string.h> -#include <linux/proc_fs.h> -#include <linux/skbuff.h> -#include <linux/capability.h> -#include <linux/if.h> -#include <linux/if_ether.h> -#include <linux/if_arp.h> -#include <linux/netdevice.h> -#include <linux/init.h> -#include <linux/tty.h> -#include <linux/kmod.h> -#include <linux/spinlock.h> -#include <linux/slab.h> -#include <linux/export.h> - -#include <asm/ioctls.h> -#include <linux/uaccess.h> -#include <asm/dma.h> -#include <asm/io.h> - -#include <net/irda/irda_device.h> -#include <net/irda/irlap.h> -#include <net/irda/timer.h> -#include <net/irda/wrapper.h> - -static void __irda_task_delete(struct irda_task *task); - -static hashbin_t *dongles = NULL; -static hashbin_t *tasks = NULL; - -static void irda_task_timer_expired(void *data); - -int __init irda_device_init( void) -{ - dongles = hashbin_new(HB_NOLOCK); - if (dongles == NULL) { - net_warn_ratelimited("IrDA: Can't allocate dongles hashbin!\n"); - return -ENOMEM; - } - spin_lock_init(&dongles->hb_spinlock); - - tasks = hashbin_new(HB_LOCK); - if (tasks == NULL) { - net_warn_ratelimited("IrDA: Can't allocate tasks hashbin!\n"); - hashbin_delete(dongles, NULL); - return -ENOMEM; - } - - /* We no longer initialise the driver ourselves here, we let - * the system do it for us... - Jean II */ - - return 0; -} - -static void leftover_dongle(void *arg) -{ - struct dongle_reg *reg = arg; - net_warn_ratelimited("IrDA: Dongle type %x not unregistered\n", - reg->type); -} - -void irda_device_cleanup(void) -{ - hashbin_delete(tasks, (FREE_FUNC) __irda_task_delete); - - hashbin_delete(dongles, leftover_dongle); -} - -/* - * Function irda_device_set_media_busy (self, status) - * - * Called when we have detected that another station is transmitting - * in contention mode. - */ -void irda_device_set_media_busy(struct net_device *dev, int status) -{ - struct irlap_cb *self; - - pr_debug("%s(%s)\n", __func__, status ? "TRUE" : "FALSE"); - - self = (struct irlap_cb *) dev->atalk_ptr; - - /* Some drivers may enable the receive interrupt before calling - * irlap_open(), or they may disable the receive interrupt - * after calling irlap_close(). - * The IrDA stack is protected from this in irlap_driver_rcv(). - * However, the driver calls directly the wrapper, that calls - * us directly. Make sure we protect ourselves. - * Jean II */ - if (!self || self->magic != LAP_MAGIC) - return; - - if (status) { - self->media_busy = TRUE; - if (status == SMALL) - irlap_start_mbusy_timer(self, SMALLBUSY_TIMEOUT); - else - irlap_start_mbusy_timer(self, MEDIABUSY_TIMEOUT); - pr_debug("Media busy!\n"); - } else { - self->media_busy = FALSE; - irlap_stop_mbusy_timer(self); - } -} -EXPORT_SYMBOL(irda_device_set_media_busy); - - -/* - * Function irda_device_is_receiving (dev) - * - * Check if the device driver is currently receiving data - * - */ -int irda_device_is_receiving(struct net_device *dev) -{ - struct if_irda_req req; - int ret; - - if (!dev->netdev_ops->ndo_do_ioctl) { - net_err_ratelimited("%s: do_ioctl not impl. by device driver\n", - __func__); - return -1; - } - - ret = (dev->netdev_ops->ndo_do_ioctl)(dev, (struct ifreq *) &req, - SIOCGRECEIVING); - if (ret < 0) - return ret; - - return req.ifr_receiving; -} - -static void __irda_task_delete(struct irda_task *task) -{ - del_timer(&task->timer); - - kfree(task); -} - -static void irda_task_delete(struct irda_task *task) -{ - /* Unregister task */ - hashbin_remove(tasks, (long) task, NULL); - - __irda_task_delete(task); -} - -/* - * Function irda_task_kick (task) - * - * Tries to execute a task possible multiple times until the task is either - * finished, or askes for a timeout. When a task is finished, we do post - * processing, and notify the parent task, that is waiting for this task - * to complete. - */ -static int irda_task_kick(struct irda_task *task) -{ - int finished = TRUE; - int count = 0; - int timeout; - - IRDA_ASSERT(task != NULL, return -1;); - IRDA_ASSERT(task->magic == IRDA_TASK_MAGIC, return -1;); - - /* Execute task until it's finished, or askes for a timeout */ - do { - timeout = task->function(task); - if (count++ > 100) { - net_err_ratelimited("%s: error in task handler!\n", - __func__); - irda_task_delete(task); - return TRUE; - } - } while ((timeout == 0) && (task->state != IRDA_TASK_DONE)); - - if (timeout < 0) { - net_err_ratelimited("%s: Error executing task!\n", __func__); - irda_task_delete(task); - return TRUE; - } - - /* Check if we are finished */ - if (task->state == IRDA_TASK_DONE) { - del_timer(&task->timer); - - /* Do post processing */ - if (task->finished) - task->finished(task); - - /* Notify parent */ - if (task->parent) { - /* Check if parent is waiting for us to complete */ - if (task->parent->state == IRDA_TASK_CHILD_WAIT) { - task->parent->state = IRDA_TASK_CHILD_DONE; - - /* Stop timer now that we are here */ - del_timer(&task->parent->timer); - - /* Kick parent task */ - irda_task_kick(task->parent); - } - } - irda_task_delete(task); - } else if (timeout > 0) { - irda_start_timer(&task->timer, timeout, (void *) task, - irda_task_timer_expired); - finished = FALSE; - } else { - pr_debug("%s(), not finished, and no timeout!\n", - __func__); - finished = FALSE; - } - - return finished; -} - -/* - * Function irda_task_timer_expired (data) - * - * Task time has expired. We now try to execute task (again), and restart - * the timer if the task has not finished yet - */ -static void irda_task_timer_expired(void *data) -{ - struct irda_task *task; - - task = data; - - irda_task_kick(task); -} - -/* - * Function irda_device_setup (dev) - * - * This function should be used by low level device drivers in a similar way - * as ether_setup() is used by normal network device drivers - */ -static void irda_device_setup(struct net_device *dev) -{ - dev->hard_header_len = 0; - dev->addr_len = LAP_ALEN; - - dev->type = ARPHRD_IRDA; - dev->tx_queue_len = 8; /* Window size + 1 s-frame */ - - memset(dev->broadcast, 0xff, LAP_ALEN); - - dev->mtu = 2048; - dev->flags = IFF_NOARP; -} - -/* - * Funciton alloc_irdadev - * Allocates and sets up an IRDA device in a manner similar to - * alloc_etherdev. - */ -struct net_device *alloc_irdadev(int sizeof_priv) -{ - return alloc_netdev(sizeof_priv, "irda%d", NET_NAME_UNKNOWN, - irda_device_setup); -} -EXPORT_SYMBOL(alloc_irdadev); - -#ifdef CONFIG_ISA_DMA_API -/* - * Function setup_dma (idev, buffer, count, mode) - * - * Setup the DMA channel. Commonly used by LPC FIR drivers - * - */ -void irda_setup_dma(int channel, dma_addr_t buffer, int count, int mode) -{ - unsigned long flags; - - flags = claim_dma_lock(); - - disable_dma(channel); - clear_dma_ff(channel); - set_dma_mode(channel, mode); - set_dma_addr(channel, buffer); - set_dma_count(channel, count); - enable_dma(channel); - - release_dma_lock(flags); -} -EXPORT_SYMBOL(irda_setup_dma); -#endif diff --git a/net/irda/iriap.c b/net/irda/iriap.c deleted file mode 100644 index 1138eaf5c682..000000000000 --- a/net/irda/iriap.c +++ /dev/null @@ -1,1085 +0,0 @@ -/********************************************************************* - * - * Filename: iriap.c - * Version: 0.8 - * Description: Information Access Protocol (IAP) - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Thu Aug 21 00:02:07 1997 - * Modified at: Sat Dec 25 16:42:42 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/skbuff.h> -#include <linux/fs.h> -#include <linux/string.h> -#include <linux/init.h> -#include <linux/seq_file.h> -#include <linux/slab.h> - -#include <asm/byteorder.h> -#include <asm/unaligned.h> - -#include <net/irda/irda.h> -#include <net/irda/irttp.h> -#include <net/irda/irlmp.h> -#include <net/irda/irias_object.h> -#include <net/irda/iriap_event.h> -#include <net/irda/iriap.h> - -/* FIXME: This one should go in irlmp.c */ -static const char *const ias_charset_types[] __maybe_unused = { - "CS_ASCII", - "CS_ISO_8859_1", - "CS_ISO_8859_2", - "CS_ISO_8859_3", - "CS_ISO_8859_4", - "CS_ISO_8859_5", - "CS_ISO_8859_6", - "CS_ISO_8859_7", - "CS_ISO_8859_8", - "CS_ISO_8859_9", - "CS_UNICODE" -}; - -static hashbin_t *iriap = NULL; -static void *service_handle; - -static void __iriap_close(struct iriap_cb *self); -static int iriap_register_lsap(struct iriap_cb *self, __u8 slsap_sel, int mode); -static void iriap_disconnect_indication(void *instance, void *sap, - LM_REASON reason, struct sk_buff *skb); -static void iriap_connect_indication(void *instance, void *sap, - struct qos_info *qos, __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb); -static void iriap_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, __u8 max_header_size, - struct sk_buff *skb); -static int iriap_data_indication(void *instance, void *sap, - struct sk_buff *skb); - -static void iriap_watchdog_timer_expired(void *data); - -static inline void iriap_start_watchdog_timer(struct iriap_cb *self, - int timeout) -{ - irda_start_timer(&self->watchdog_timer, timeout, self, - iriap_watchdog_timer_expired); -} - -static struct lock_class_key irias_objects_key; - -/* - * Function iriap_init (void) - * - * Initializes the IrIAP layer, called by the module initialization code - * in irmod.c - */ -int __init iriap_init(void) -{ - struct ias_object *obj; - struct iriap_cb *server; - __u8 oct_seq[6]; - __u16 hints; - - /* Allocate master array */ - iriap = hashbin_new(HB_LOCK); - if (!iriap) - return -ENOMEM; - - /* Object repository - defined in irias_object.c */ - irias_objects = hashbin_new(HB_LOCK); - if (!irias_objects) { - net_warn_ratelimited("%s: Can't allocate irias_objects hashbin!\n", - __func__); - hashbin_delete(iriap, NULL); - return -ENOMEM; - } - - lockdep_set_class_and_name(&irias_objects->hb_spinlock, &irias_objects_key, - "irias_objects"); - - /* - * Register some default services for IrLMP - */ - hints = irlmp_service_to_hint(S_COMPUTER); - service_handle = irlmp_register_service(hints); - - /* Register the Device object with LM-IAS */ - obj = irias_new_object("Device", IAS_DEVICE_ID); - irias_add_string_attrib(obj, "DeviceName", "Linux", IAS_KERNEL_ATTR); - - oct_seq[0] = 0x01; /* Version 1 */ - oct_seq[1] = 0x00; /* IAS support bits */ - oct_seq[2] = 0x00; /* LM-MUX support bits */ -#ifdef CONFIG_IRDA_ULTRA - oct_seq[2] |= 0x04; /* Connectionless Data support */ -#endif - irias_add_octseq_attrib(obj, "IrLMPSupport", oct_seq, 3, - IAS_KERNEL_ATTR); - irias_insert_object(obj); - - /* - * Register server support with IrLMP so we can accept incoming - * connections - */ - server = iriap_open(LSAP_IAS, IAS_SERVER, NULL, NULL); - if (!server) { - pr_debug("%s(), unable to open server\n", __func__); - return -1; - } - iriap_register_lsap(server, LSAP_IAS, IAS_SERVER); - - return 0; -} - -/* - * Function iriap_cleanup (void) - * - * Initializes the IrIAP layer, called by the module cleanup code in - * irmod.c - */ -void iriap_cleanup(void) -{ - irlmp_unregister_service(service_handle); - - hashbin_delete(iriap, (FREE_FUNC) __iriap_close); - hashbin_delete(irias_objects, (FREE_FUNC) __irias_delete_object); -} - -/* - * Function iriap_open (void) - * - * Opens an instance of the IrIAP layer, and registers with IrLMP - */ -struct iriap_cb *iriap_open(__u8 slsap_sel, int mode, void *priv, - CONFIRM_CALLBACK callback) -{ - struct iriap_cb *self; - - self = kzalloc(sizeof(*self), GFP_ATOMIC); - if (!self) - return NULL; - - /* - * Initialize instance - */ - - self->magic = IAS_MAGIC; - self->mode = mode; - if (mode == IAS_CLIENT) { - if (iriap_register_lsap(self, slsap_sel, mode)) { - kfree(self); - return NULL; - } - } - - self->confirm = callback; - self->priv = priv; - - /* iriap_getvaluebyclass_request() will construct packets before - * we connect, so this must have a sane value... Jean II */ - self->max_header_size = LMP_MAX_HEADER; - - init_timer(&self->watchdog_timer); - - hashbin_insert(iriap, (irda_queue_t *) self, (long) self, NULL); - - /* Initialize state machines */ - iriap_next_client_state(self, S_DISCONNECT); - iriap_next_call_state(self, S_MAKE_CALL); - iriap_next_server_state(self, R_DISCONNECT); - iriap_next_r_connect_state(self, R_WAITING); - - return self; -} -EXPORT_SYMBOL(iriap_open); - -/* - * Function __iriap_close (self) - * - * Removes (deallocates) the IrIAP instance - * - */ -static void __iriap_close(struct iriap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - del_timer(&self->watchdog_timer); - - if (self->request_skb) - dev_kfree_skb(self->request_skb); - - self->magic = 0; - - kfree(self); -} - -/* - * Function iriap_close (void) - * - * Closes IrIAP and deregisters with IrLMP - */ -void iriap_close(struct iriap_cb *self) -{ - struct iriap_cb *entry; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - if (self->lsap) { - irlmp_close_lsap(self->lsap); - self->lsap = NULL; - } - - entry = (struct iriap_cb *) hashbin_remove(iriap, (long) self, NULL); - IRDA_ASSERT(entry == self, return;); - - __iriap_close(self); -} -EXPORT_SYMBOL(iriap_close); - -static int iriap_register_lsap(struct iriap_cb *self, __u8 slsap_sel, int mode) -{ - notify_t notify; - - irda_notify_init(¬ify); - notify.connect_confirm = iriap_connect_confirm; - notify.connect_indication = iriap_connect_indication; - notify.disconnect_indication = iriap_disconnect_indication; - notify.data_indication = iriap_data_indication; - notify.instance = self; - if (mode == IAS_CLIENT) - strcpy(notify.name, "IrIAS cli"); - else - strcpy(notify.name, "IrIAS srv"); - - self->lsap = irlmp_open_lsap(slsap_sel, ¬ify, 0); - if (self->lsap == NULL) { - net_err_ratelimited("%s: Unable to allocated LSAP!\n", - __func__); - return -1; - } - self->slsap_sel = self->lsap->slsap_sel; - - return 0; -} - -/* - * Function iriap_disconnect_indication (handle, reason) - * - * Got disconnect, so clean up everything associated with this connection - * - */ -static void iriap_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *skb) -{ - struct iriap_cb *self; - - pr_debug("%s(), reason=%s [%d]\n", __func__, - irlmp_reason_str(reason), reason); - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - IRDA_ASSERT(iriap != NULL, return;); - - del_timer(&self->watchdog_timer); - - /* Not needed */ - if (skb) - dev_kfree_skb(skb); - - if (self->mode == IAS_CLIENT) { - pr_debug("%s(), disconnect as client\n", __func__); - - - iriap_do_client_event(self, IAP_LM_DISCONNECT_INDICATION, - NULL); - /* - * Inform service user that the request failed by sending - * it a NULL value. Warning, the client might close us, so - * remember no to use self anymore after calling confirm - */ - if (self->confirm) - self->confirm(IAS_DISCONNECT, 0, NULL, self->priv); - } else { - pr_debug("%s(), disconnect as server\n", __func__); - iriap_do_server_event(self, IAP_LM_DISCONNECT_INDICATION, - NULL); - iriap_close(self); - } -} - -/* - * Function iriap_disconnect_request (handle) - */ -static void iriap_disconnect_request(struct iriap_cb *self) -{ - struct sk_buff *tx_skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (tx_skb == NULL) { - pr_debug("%s(), Could not allocate an sk_buff of length %d\n", - __func__, LMP_MAX_HEADER); - return; - } - - /* - * Reserve space for MUX control and LAP header - */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - - irlmp_disconnect_request(self->lsap, tx_skb); -} - -/* - * Function iriap_getvaluebyclass (addr, name, attr) - * - * Retrieve all values from attribute in all objects with given class - * name - */ -int iriap_getvaluebyclass_request(struct iriap_cb *self, - __u32 saddr, __u32 daddr, - char *name, char *attr) -{ - struct sk_buff *tx_skb; - int name_len, attr_len, skb_len; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return -1;); - - /* Client must supply the destination device address */ - if (!daddr) - return -1; - - self->daddr = daddr; - self->saddr = saddr; - - /* - * Save operation, so we know what the later indication is about - */ - self->operation = GET_VALUE_BY_CLASS; - - /* Give ourselves 10 secs to finish this operation */ - iriap_start_watchdog_timer(self, 10*HZ); - - name_len = strlen(name); /* Up to IAS_MAX_CLASSNAME = 60 */ - attr_len = strlen(attr); /* Up to IAS_MAX_ATTRIBNAME = 60 */ - - skb_len = self->max_header_size+2+name_len+1+attr_len+4; - tx_skb = alloc_skb(skb_len, GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* Reserve space for MUX and LAP header */ - skb_reserve(tx_skb, self->max_header_size); - skb_put(tx_skb, 3+name_len+attr_len); - frame = tx_skb->data; - - /* Build frame */ - frame[0] = IAP_LST | GET_VALUE_BY_CLASS; - frame[1] = name_len; /* Insert length of name */ - memcpy(frame+2, name, name_len); /* Insert name */ - frame[2+name_len] = attr_len; /* Insert length of attr */ - memcpy(frame+3+name_len, attr, attr_len); /* Insert attr */ - - iriap_do_client_event(self, IAP_CALL_REQUEST_GVBC, tx_skb); - - /* Drop reference count - see state_s_disconnect(). */ - dev_kfree_skb(tx_skb); - - return 0; -} -EXPORT_SYMBOL(iriap_getvaluebyclass_request); - -/* - * Function iriap_getvaluebyclass_confirm (self, skb) - * - * Got result from GetValueByClass command. Parse it and return result - * to service user. - * - */ -static void iriap_getvaluebyclass_confirm(struct iriap_cb *self, - struct sk_buff *skb) -{ - struct ias_value *value; - int charset; - __u32 value_len; - __u32 tmp_cpu32; - __u16 obj_id; - __u16 len; - __u8 type; - __u8 *fp; - int n; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Initialize variables */ - fp = skb->data; - n = 2; - - /* Get length, MSB first */ - len = get_unaligned_be16(fp + n); - n += 2; - - pr_debug("%s(), len=%d\n", __func__, len); - - /* Get object ID, MSB first */ - obj_id = get_unaligned_be16(fp + n); - n += 2; - - type = fp[n++]; - pr_debug("%s(), Value type = %d\n", __func__, type); - - switch (type) { - case IAS_INTEGER: - memcpy(&tmp_cpu32, fp+n, 4); n += 4; - be32_to_cpus(&tmp_cpu32); - value = irias_new_integer_value(tmp_cpu32); - - /* Legal values restricted to 0x01-0x6f, page 15 irttp */ - pr_debug("%s(), lsap=%d\n", __func__, value->t.integer); - break; - case IAS_STRING: - charset = fp[n++]; - - switch (charset) { - case CS_ASCII: - break; -/* case CS_ISO_8859_1: */ -/* case CS_ISO_8859_2: */ -/* case CS_ISO_8859_3: */ -/* case CS_ISO_8859_4: */ -/* case CS_ISO_8859_5: */ -/* case CS_ISO_8859_6: */ -/* case CS_ISO_8859_7: */ -/* case CS_ISO_8859_8: */ -/* case CS_ISO_8859_9: */ -/* case CS_UNICODE: */ - default: - pr_debug("%s(), charset [%d] %s, not supported\n", - __func__, charset, - charset < ARRAY_SIZE(ias_charset_types) ? - ias_charset_types[charset] : - "(unknown)"); - - /* Aborting, close connection! */ - iriap_disconnect_request(self); - return; - /* break; */ - } - value_len = fp[n++]; - pr_debug("%s(), strlen=%d\n", __func__, value_len); - - /* Make sure the string is null-terminated */ - if (n + value_len < skb->len) - fp[n + value_len] = 0x00; - pr_debug("Got string %s\n", fp+n); - - /* Will truncate to IAS_MAX_STRING bytes */ - value = irias_new_string_value(fp+n); - break; - case IAS_OCT_SEQ: - value_len = get_unaligned_be16(fp + n); - n += 2; - - /* Will truncate to IAS_MAX_OCTET_STRING bytes */ - value = irias_new_octseq_value(fp+n, value_len); - break; - default: - value = irias_new_missing_value(); - break; - } - - /* Finished, close connection! */ - iriap_disconnect_request(self); - - /* Warning, the client might close us, so remember no to use self - * anymore after calling confirm - */ - if (self->confirm) - self->confirm(IAS_SUCCESS, obj_id, value, self->priv); - else { - pr_debug("%s(), missing handler!\n", __func__); - irias_delete_value(value); - } -} - -/* - * Function iriap_getvaluebyclass_response () - * - * Send answer back to remote LM-IAS - * - */ -static void iriap_getvaluebyclass_response(struct iriap_cb *self, - __u16 obj_id, - __u8 ret_code, - struct ias_value *value) -{ - struct sk_buff *tx_skb; - int n; - __be32 tmp_be32; - __be16 tmp_be16; - __u8 *fp; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - IRDA_ASSERT(value != NULL, return;); - IRDA_ASSERT(value->len <= 1024, return;); - - /* Initialize variables */ - n = 0; - - /* - * We must adjust the size of the response after the length of the - * value. We add 32 bytes because of the 6 bytes for the frame and - * max 5 bytes for the value coding. - */ - tx_skb = alloc_skb(value->len + self->max_header_size + 32, - GFP_ATOMIC); - if (!tx_skb) - return; - - /* Reserve space for MUX and LAP header */ - skb_reserve(tx_skb, self->max_header_size); - skb_put(tx_skb, 6); - - fp = tx_skb->data; - - /* Build frame */ - fp[n++] = GET_VALUE_BY_CLASS | IAP_LST; - fp[n++] = ret_code; - - /* Insert list length (MSB first) */ - tmp_be16 = htons(0x0001); - memcpy(fp+n, &tmp_be16, 2); n += 2; - - /* Insert object identifier ( MSB first) */ - tmp_be16 = cpu_to_be16(obj_id); - memcpy(fp+n, &tmp_be16, 2); n += 2; - - switch (value->type) { - case IAS_STRING: - skb_put(tx_skb, 3 + value->len); - fp[n++] = value->type; - fp[n++] = 0; /* ASCII */ - fp[n++] = (__u8) value->len; - memcpy(fp+n, value->t.string, value->len); n+=value->len; - break; - case IAS_INTEGER: - skb_put(tx_skb, 5); - fp[n++] = value->type; - - tmp_be32 = cpu_to_be32(value->t.integer); - memcpy(fp+n, &tmp_be32, 4); n += 4; - break; - case IAS_OCT_SEQ: - skb_put(tx_skb, 3 + value->len); - fp[n++] = value->type; - - tmp_be16 = cpu_to_be16(value->len); - memcpy(fp+n, &tmp_be16, 2); n += 2; - memcpy(fp+n, value->t.oct_seq, value->len); n+=value->len; - break; - case IAS_MISSING: - pr_debug("%s: sending IAS_MISSING\n", __func__); - skb_put(tx_skb, 1); - fp[n++] = value->type; - break; - default: - pr_debug("%s(), type not implemented!\n", __func__); - break; - } - iriap_do_r_connect_event(self, IAP_CALL_RESPONSE, tx_skb); - - /* Drop reference count - see state_r_execute(). */ - dev_kfree_skb(tx_skb); -} - -/* - * Function iriap_getvaluebyclass_indication (self, skb) - * - * getvaluebyclass is requested from peer LM-IAS - * - */ -static void iriap_getvaluebyclass_indication(struct iriap_cb *self, - struct sk_buff *skb) -{ - struct ias_object *obj; - struct ias_attrib *attrib; - int name_len; - int attr_len; - char name[IAS_MAX_CLASSNAME + 1]; /* 60 bytes */ - char attr[IAS_MAX_ATTRIBNAME + 1]; /* 60 bytes */ - __u8 *fp; - int n; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - fp = skb->data; - n = 1; - - name_len = fp[n++]; - - IRDA_ASSERT(name_len < IAS_MAX_CLASSNAME + 1, return;); - - memcpy(name, fp+n, name_len); n+=name_len; - name[name_len] = '\0'; - - attr_len = fp[n++]; - - IRDA_ASSERT(attr_len < IAS_MAX_ATTRIBNAME + 1, return;); - - memcpy(attr, fp+n, attr_len); n+=attr_len; - attr[attr_len] = '\0'; - - pr_debug("LM-IAS: Looking up %s: %s\n", name, attr); - obj = irias_find_object(name); - - if (obj == NULL) { - pr_debug("LM-IAS: Object %s not found\n", name); - iriap_getvaluebyclass_response(self, 0x1235, IAS_CLASS_UNKNOWN, - &irias_missing); - return; - } - pr_debug("LM-IAS: found %s, id=%d\n", obj->name, obj->id); - - attrib = irias_find_attrib(obj, attr); - if (attrib == NULL) { - pr_debug("LM-IAS: Attribute %s not found\n", attr); - iriap_getvaluebyclass_response(self, obj->id, - IAS_ATTRIB_UNKNOWN, - &irias_missing); - return; - } - - /* We have a match; send the value. */ - iriap_getvaluebyclass_response(self, obj->id, IAS_SUCCESS, - attrib->value); -} - -/* - * Function iriap_send_ack (void) - * - * Currently not used - * - */ -void iriap_send_ack(struct iriap_cb *self) -{ - struct sk_buff *tx_skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - tx_skb = alloc_skb(LMP_MAX_HEADER + 1, GFP_ATOMIC); - if (!tx_skb) - return; - - /* Reserve space for MUX and LAP header */ - skb_reserve(tx_skb, self->max_header_size); - skb_put(tx_skb, 1); - frame = tx_skb->data; - - /* Build frame */ - frame[0] = IAP_LST | IAP_ACK | self->operation; - - irlmp_data_request(self->lsap, tx_skb); -} - -void iriap_connect_request(struct iriap_cb *self) -{ - int ret; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - ret = irlmp_connect_request(self->lsap, LSAP_IAS, - self->saddr, self->daddr, - NULL, NULL); - if (ret < 0) { - pr_debug("%s(), connect failed!\n", __func__); - self->confirm(IAS_DISCONNECT, 0, NULL, self->priv); - } -} - -/* - * Function iriap_connect_confirm (handle, skb) - * - * LSAP connection confirmed! - * - */ -static void iriap_connect_confirm(void *instance, void *sap, - struct qos_info *qos, __u32 max_seg_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct iriap_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - self->max_data_size = max_seg_size; - self->max_header_size = max_header_size; - - del_timer(&self->watchdog_timer); - - iriap_do_client_event(self, IAP_LM_CONNECT_CONFIRM, skb); - - /* Drop reference count - see state_s_make_call(). */ - dev_kfree_skb(skb); -} - -/* - * Function iriap_connect_indication ( handle, skb) - * - * Remote LM-IAS is requesting connection - * - */ -static void iriap_connect_indication(void *instance, void *sap, - struct qos_info *qos, __u32 max_seg_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct iriap_cb *self, *new; - - self = instance; - - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(self != NULL, goto out;); - IRDA_ASSERT(self->magic == IAS_MAGIC, goto out;); - - /* Start new server */ - new = iriap_open(LSAP_IAS, IAS_SERVER, NULL, NULL); - if (!new) { - pr_debug("%s(), open failed\n", __func__); - goto out; - } - - /* Now attach up the new "socket" */ - new->lsap = irlmp_dup(self->lsap, new); - if (!new->lsap) { - pr_debug("%s(), dup failed!\n", __func__); - goto out; - } - - new->max_data_size = max_seg_size; - new->max_header_size = max_header_size; - - /* Clean up the original one to keep it in listen state */ - irlmp_listen(self->lsap); - - iriap_do_server_event(new, IAP_LM_CONNECT_INDICATION, skb); - -out: - /* Drop reference count - see state_r_disconnect(). */ - dev_kfree_skb(skb); -} - -/* - * Function iriap_data_indication (handle, skb) - * - * Receives data from connection identified by handle from IrLMP - * - */ -static int iriap_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct iriap_cb *self; - __u8 *frame; - __u8 opcode; - - self = instance; - - IRDA_ASSERT(skb != NULL, return 0;); - IRDA_ASSERT(self != NULL, goto out;); - IRDA_ASSERT(self->magic == IAS_MAGIC, goto out;); - - frame = skb->data; - - if (self->mode == IAS_SERVER) { - /* Call server */ - pr_debug("%s(), Calling server!\n", __func__); - iriap_do_r_connect_event(self, IAP_RECV_F_LST, skb); - goto out; - } - opcode = frame[0]; - if (~opcode & IAP_LST) { - net_warn_ratelimited("%s:, IrIAS multiframe commands or results is not implemented yet!\n", - __func__); - goto out; - } - - /* Check for ack frames since they don't contain any data */ - if (opcode & IAP_ACK) { - pr_debug("%s() Got ack frame!\n", __func__); - goto out; - } - - opcode &= ~IAP_LST; /* Mask away LST bit */ - - switch (opcode) { - case GET_INFO_BASE: - pr_debug("IrLMP GetInfoBaseDetails not implemented!\n"); - break; - case GET_VALUE_BY_CLASS: - iriap_do_call_event(self, IAP_RECV_F_LST, NULL); - - switch (frame[1]) { - case IAS_SUCCESS: - iriap_getvaluebyclass_confirm(self, skb); - break; - case IAS_CLASS_UNKNOWN: - pr_debug("%s(), No such class!\n", __func__); - /* Finished, close connection! */ - iriap_disconnect_request(self); - - /* - * Warning, the client might close us, so remember - * no to use self anymore after calling confirm - */ - if (self->confirm) - self->confirm(IAS_CLASS_UNKNOWN, 0, NULL, - self->priv); - break; - case IAS_ATTRIB_UNKNOWN: - pr_debug("%s(), No such attribute!\n", __func__); - /* Finished, close connection! */ - iriap_disconnect_request(self); - - /* - * Warning, the client might close us, so remember - * no to use self anymore after calling confirm - */ - if (self->confirm) - self->confirm(IAS_ATTRIB_UNKNOWN, 0, NULL, - self->priv); - break; - } - break; - default: - pr_debug("%s(), Unknown op-code: %02x\n", __func__, - opcode); - break; - } - -out: - /* Cleanup - sub-calls will have done skb_get() as needed. */ - dev_kfree_skb(skb); - return 0; -} - -/* - * Function iriap_call_indication (self, skb) - * - * Received call to server from peer LM-IAS - * - */ -void iriap_call_indication(struct iriap_cb *self, struct sk_buff *skb) -{ - __u8 *fp; - __u8 opcode; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - fp = skb->data; - - opcode = fp[0]; - if (~opcode & 0x80) { - net_warn_ratelimited("%s: IrIAS multiframe commands or results is not implemented yet!\n", - __func__); - return; - } - opcode &= 0x7f; /* Mask away LST bit */ - - switch (opcode) { - case GET_INFO_BASE: - net_warn_ratelimited("%s: GetInfoBaseDetails not implemented yet!\n", - __func__); - break; - case GET_VALUE_BY_CLASS: - iriap_getvaluebyclass_indication(self, skb); - break; - } - /* skb will be cleaned up in iriap_data_indication */ -} - -/* - * Function iriap_watchdog_timer_expired (data) - * - * Query has taken too long time, so abort - * - */ -static void iriap_watchdog_timer_expired(void *data) -{ - struct iriap_cb *self = (struct iriap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - /* iriap_close(self); */ -} - -#ifdef CONFIG_PROC_FS - -static const char *const ias_value_types[] = { - "IAS_MISSING", - "IAS_INTEGER", - "IAS_OCT_SEQ", - "IAS_STRING" -}; - -static inline struct ias_object *irias_seq_idx(loff_t pos) -{ - struct ias_object *obj; - - for (obj = (struct ias_object *) hashbin_get_first(irias_objects); - obj; obj = (struct ias_object *) hashbin_get_next(irias_objects)) { - if (pos-- == 0) - break; - } - - return obj; -} - -static void *irias_seq_start(struct seq_file *seq, loff_t *pos) -{ - spin_lock_irq(&irias_objects->hb_spinlock); - - return *pos ? irias_seq_idx(*pos - 1) : SEQ_START_TOKEN; -} - -static void *irias_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - ++*pos; - - return (v == SEQ_START_TOKEN) - ? (void *) hashbin_get_first(irias_objects) - : (void *) hashbin_get_next(irias_objects); -} - -static void irias_seq_stop(struct seq_file *seq, void *v) -{ - spin_unlock_irq(&irias_objects->hb_spinlock); -} - -static int irias_seq_show(struct seq_file *seq, void *v) -{ - if (v == SEQ_START_TOKEN) - seq_puts(seq, "LM-IAS Objects:\n"); - else { - struct ias_object *obj = v; - struct ias_attrib *attrib; - - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return -EINVAL;); - - seq_printf(seq, "name: %s, id=%d\n", - obj->name, obj->id); - - /* Careful for priority inversions here ! - * All other uses of attrib spinlock are independent of - * the object spinlock, so we are safe. Jean II */ - spin_lock(&obj->attribs->hb_spinlock); - - /* List all attributes for this object */ - for (attrib = (struct ias_attrib *) hashbin_get_first(obj->attribs); - attrib != NULL; - attrib = (struct ias_attrib *) hashbin_get_next(obj->attribs)) { - - IRDA_ASSERT(attrib->magic == IAS_ATTRIB_MAGIC, - goto outloop; ); - - seq_printf(seq, " - Attribute name: \"%s\", ", - attrib->name); - seq_printf(seq, "value[%s]: ", - ias_value_types[attrib->value->type]); - - switch (attrib->value->type) { - case IAS_INTEGER: - seq_printf(seq, "%d\n", - attrib->value->t.integer); - break; - case IAS_STRING: - seq_printf(seq, "\"%s\"\n", - attrib->value->t.string); - break; - case IAS_OCT_SEQ: - seq_printf(seq, "octet sequence (%d bytes)\n", - attrib->value->len); - break; - case IAS_MISSING: - seq_puts(seq, "missing\n"); - break; - default: - seq_printf(seq, "type %d?\n", - attrib->value->type); - } - seq_putc(seq, '\n'); - - } - IRDA_ASSERT_LABEL(outloop:) - spin_unlock(&obj->attribs->hb_spinlock); - } - - return 0; -} - -static const struct seq_operations irias_seq_ops = { - .start = irias_seq_start, - .next = irias_seq_next, - .stop = irias_seq_stop, - .show = irias_seq_show, -}; - -static int irias_seq_open(struct inode *inode, struct file *file) -{ - IRDA_ASSERT( irias_objects != NULL, return -EINVAL;); - - return seq_open(file, &irias_seq_ops); -} - -const struct file_operations irias_seq_fops = { - .owner = THIS_MODULE, - .open = irias_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -#endif /* PROC_FS */ diff --git a/net/irda/iriap_event.c b/net/irda/iriap_event.c deleted file mode 100644 index e6098b2e048a..000000000000 --- a/net/irda/iriap_event.c +++ /dev/null @@ -1,496 +0,0 @@ -/********************************************************************* - * - * Filename: iriap_event.c - * Version: 0.1 - * Description: IAP Finite State Machine - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Thu Aug 21 00:02:07 1997 - * Modified at: Wed Mar 1 11:28:34 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1997, 1999-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/slab.h> - -#include <net/irda/irda.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/iriap_event.h> - -static void state_s_disconnect (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_connecting (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_call (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); - -static void state_s_make_call (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_calling (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_outstanding (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_replying (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_wait_for_call(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_s_wait_active (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); - -static void state_r_disconnect (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_call (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_waiting (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_wait_active (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_receiving (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_execute (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); -static void state_r_returning (struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb); - -static void (*iriap_state[])(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) = { - /* Client FSM */ - state_s_disconnect, - state_s_connecting, - state_s_call, - - /* S-Call FSM */ - state_s_make_call, - state_s_calling, - state_s_outstanding, - state_s_replying, - state_s_wait_for_call, - state_s_wait_active, - - /* Server FSM */ - state_r_disconnect, - state_r_call, - - /* R-Connect FSM */ - state_r_waiting, - state_r_wait_active, - state_r_receiving, - state_r_execute, - state_r_returning, -}; - -void iriap_next_client_state(struct iriap_cb *self, IRIAP_STATE state) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - self->client_state = state; -} - -void iriap_next_call_state(struct iriap_cb *self, IRIAP_STATE state) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - self->call_state = state; -} - -void iriap_next_server_state(struct iriap_cb *self, IRIAP_STATE state) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - self->server_state = state; -} - -void iriap_next_r_connect_state(struct iriap_cb *self, IRIAP_STATE state) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - self->r_connect_state = state; -} - -void iriap_do_client_event(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - (*iriap_state[ self->client_state]) (self, event, skb); -} - -void iriap_do_call_event(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - (*iriap_state[ self->call_state]) (self, event, skb); -} - -void iriap_do_server_event(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - (*iriap_state[ self->server_state]) (self, event, skb); -} - -void iriap_do_r_connect_event(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - (*iriap_state[ self->r_connect_state]) (self, event, skb); -} - - -/* - * Function state_s_disconnect (event, skb) - * - * S-Disconnect, The device has no LSAP connection to a particular - * remote device. - */ -static void state_s_disconnect(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - switch (event) { - case IAP_CALL_REQUEST_GVBC: - iriap_next_client_state(self, S_CONNECTING); - IRDA_ASSERT(self->request_skb == NULL, return;); - /* Don't forget to refcount it - - * see iriap_getvaluebyclass_request(). */ - skb_get(skb); - self->request_skb = skb; - iriap_connect_request(self); - break; - case IAP_LM_DISCONNECT_INDICATION: - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__, event); - break; - } -} - -/* - * Function state_s_connecting (self, event, skb) - * - * S-Connecting - * - */ -static void state_s_connecting(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - switch (event) { - case IAP_LM_CONNECT_CONFIRM: - /* - * Jump to S-Call FSM - */ - iriap_do_call_event(self, IAP_CALL_REQUEST, skb); - /* iriap_call_request(self, 0,0,0); */ - iriap_next_client_state(self, S_CALL); - break; - case IAP_LM_DISCONNECT_INDICATION: - /* Abort calls */ - iriap_next_call_state(self, S_MAKE_CALL); - iriap_next_client_state(self, S_DISCONNECT); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__, event); - break; - } -} - -/* - * Function state_s_call (self, event, skb) - * - * S-Call, The device can process calls to a specific remote - * device. Whenever the LSAP connection is disconnected, this state - * catches that event and clears up - */ -static void state_s_call(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - - switch (event) { - case IAP_LM_DISCONNECT_INDICATION: - /* Abort calls */ - iriap_next_call_state(self, S_MAKE_CALL); - iriap_next_client_state(self, S_DISCONNECT); - break; - default: - pr_debug("state_s_call: Unknown event %d\n", event); - break; - } -} - -/* - * Function state_s_make_call (event, skb) - * - * S-Make-Call - * - */ -static void state_s_make_call(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - - IRDA_ASSERT(self != NULL, return;); - - switch (event) { - case IAP_CALL_REQUEST: - /* Already refcounted - see state_s_disconnect() */ - tx_skb = self->request_skb; - self->request_skb = NULL; - - irlmp_data_request(self->lsap, tx_skb); - iriap_next_call_state(self, S_OUTSTANDING); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__, event); - break; - } -} - -/* - * Function state_s_calling (event, skb) - * - * S-Calling - * - */ -static void state_s_calling(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - -/* - * Function state_s_outstanding (event, skb) - * - * S-Outstanding, The device is waiting for a response to a command - * - */ -static void state_s_outstanding(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - - switch (event) { - case IAP_RECV_F_LST: - /*iriap_send_ack(self);*/ - /*LM_Idle_request(idle); */ - - iriap_next_call_state(self, S_WAIT_FOR_CALL); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__, event); - break; - } -} - -/* - * Function state_s_replying (event, skb) - * - * S-Replying, The device is collecting a multiple part response - */ -static void state_s_replying(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - -/* - * Function state_s_wait_for_call (event, skb) - * - * S-Wait-for-Call - * - */ -static void state_s_wait_for_call(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - - -/* - * Function state_s_wait_active (event, skb) - * - * S-Wait-Active - * - */ -static void state_s_wait_active(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - -/************************************************************************** - * - * Server FSM - * - **************************************************************************/ - -/* - * Function state_r_disconnect (self, event, skb) - * - * LM-IAS server is disconnected (not processing any requests!) - * - */ -static void state_r_disconnect(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - - switch (event) { - case IAP_LM_CONNECT_INDICATION: - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (tx_skb == NULL) - return; - - /* Reserve space for MUX_CONTROL and LAP header */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - - irlmp_connect_response(self->lsap, tx_skb); - /*LM_Idle_request(idle); */ - - iriap_next_server_state(self, R_CALL); - - /* - * Jump to R-Connect FSM, we skip R-Waiting since we do not - * care about LM_Idle_request()! - */ - iriap_next_r_connect_state(self, R_RECEIVING); - break; - default: - pr_debug("%s(), unknown event %d\n", __func__, event); - break; - } -} - -/* - * Function state_r_call (self, event, skb) - */ -static void state_r_call(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - switch (event) { - case IAP_LM_DISCONNECT_INDICATION: - /* Abort call */ - iriap_next_server_state(self, R_DISCONNECT); - iriap_next_r_connect_state(self, R_WAITING); - break; - default: - pr_debug("%s(), unknown event!\n", __func__); - break; - } -} - -/* - * R-Connect FSM - */ - -/* - * Function state_r_waiting (self, event, skb) - */ -static void state_r_waiting(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - -static void state_r_wait_active(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), Not implemented\n", __func__); -} - -/* - * Function state_r_receiving (self, event, skb) - * - * We are receiving a command - * - */ -static void state_r_receiving(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - switch (event) { - case IAP_RECV_F_LST: - iriap_next_r_connect_state(self, R_EXECUTE); - - iriap_call_indication(self, skb); - break; - default: - pr_debug("%s(), unknown event!\n", __func__); - break; - } -} - -/* - * Function state_r_execute (self, event, skb) - * - * The server is processing the request - * - */ -static void state_r_execute(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IAS_MAGIC, return;); - - switch (event) { - case IAP_CALL_RESPONSE: - /* - * Since we don't implement the Waiting state, we return - * to state Receiving instead, DB. - */ - iriap_next_r_connect_state(self, R_RECEIVING); - - /* Don't forget to refcount it - see - * iriap_getvaluebyclass_response(). */ - skb_get(skb); - - irlmp_data_request(self->lsap, skb); - break; - default: - pr_debug("%s(), unknown event!\n", __func__); - break; - } -} - -static void state_r_returning(struct iriap_cb *self, IRIAP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), event=%d\n", __func__, event); - - switch (event) { - case IAP_RECV_F_LST: - break; - default: - break; - } -} diff --git a/net/irda/irias_object.c b/net/irda/irias_object.c deleted file mode 100644 index 53b86d0e1630..000000000000 --- a/net/irda/irias_object.c +++ /dev/null @@ -1,555 +0,0 @@ -/********************************************************************* - * - * Filename: irias_object.c - * Version: 0.3 - * Description: IAS object database and functions - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Thu Oct 1 22:50:04 1998 - * Modified at: Wed Dec 15 11:23:16 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/socket.h> -#include <linux/module.h> - -#include <net/irda/irda.h> -#include <net/irda/irias_object.h> - -hashbin_t *irias_objects; - -/* - * Used when a missing value needs to be returned - */ -struct ias_value irias_missing = { IAS_MISSING, 0, 0, 0, {0}}; - - -/* - * Function ias_new_object (name, id) - * - * Create a new IAS object - * - */ -struct ias_object *irias_new_object( char *name, int id) -{ - struct ias_object *obj; - - obj = kzalloc(sizeof(struct ias_object), GFP_ATOMIC); - if (obj == NULL) { - net_warn_ratelimited("%s(), Unable to allocate object!\n", - __func__); - return NULL; - } - - obj->magic = IAS_OBJECT_MAGIC; - obj->name = kstrndup(name, IAS_MAX_CLASSNAME, GFP_ATOMIC); - if (!obj->name) { - net_warn_ratelimited("%s(), Unable to allocate name!\n", - __func__); - kfree(obj); - return NULL; - } - obj->id = id; - - /* Locking notes : the attrib spinlock has lower precendence - * than the objects spinlock. Never grap the objects spinlock - * while holding any attrib spinlock (risk of deadlock). Jean II */ - obj->attribs = hashbin_new(HB_LOCK); - - if (obj->attribs == NULL) { - net_warn_ratelimited("%s(), Unable to allocate attribs!\n", - __func__); - kfree(obj->name); - kfree(obj); - return NULL; - } - - return obj; -} -EXPORT_SYMBOL(irias_new_object); - -/* - * Function irias_delete_attrib (attrib) - * - * Delete given attribute and deallocate all its memory - * - */ -static void __irias_delete_attrib(struct ias_attrib *attrib) -{ - IRDA_ASSERT(attrib != NULL, return;); - IRDA_ASSERT(attrib->magic == IAS_ATTRIB_MAGIC, return;); - - kfree(attrib->name); - - irias_delete_value(attrib->value); - attrib->magic = ~IAS_ATTRIB_MAGIC; - - kfree(attrib); -} - -void __irias_delete_object(struct ias_object *obj) -{ - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - - kfree(obj->name); - - hashbin_delete(obj->attribs, (FREE_FUNC) __irias_delete_attrib); - - obj->magic = ~IAS_OBJECT_MAGIC; - - kfree(obj); -} - -/* - * Function irias_delete_object (obj) - * - * Remove object from hashbin and deallocate all attributes associated with - * with this object and the object itself - * - */ -int irias_delete_object(struct ias_object *obj) -{ - struct ias_object *node; - - IRDA_ASSERT(obj != NULL, return -1;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return -1;); - - /* Remove from list */ - node = hashbin_remove_this(irias_objects, (irda_queue_t *) obj); - if (!node) - pr_debug("%s(), object already removed!\n", - __func__); - - /* Destroy */ - __irias_delete_object(obj); - - return 0; -} -EXPORT_SYMBOL(irias_delete_object); - -/* - * Function irias_delete_attrib (obj) - * - * Remove attribute from hashbin and, if it was the last attribute of - * the object, remove the object as well. - * - */ -int irias_delete_attrib(struct ias_object *obj, struct ias_attrib *attrib, - int cleanobject) -{ - struct ias_attrib *node; - - IRDA_ASSERT(obj != NULL, return -1;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return -1;); - IRDA_ASSERT(attrib != NULL, return -1;); - - /* Remove attribute from object */ - node = hashbin_remove_this(obj->attribs, (irda_queue_t *) attrib); - if (!node) - return 0; /* Already removed or non-existent */ - - /* Deallocate attribute */ - __irias_delete_attrib(node); - - /* Check if object has still some attributes, destroy it if none. - * At first glance, this look dangerous, as the kernel reference - * various IAS objects. However, we only use this function on - * user attributes, not kernel attributes, so there is no risk - * of deleting a kernel object this way. Jean II */ - node = (struct ias_attrib *) hashbin_get_first(obj->attribs); - if (cleanobject && !node) - irias_delete_object(obj); - - return 0; -} - -/* - * Function irias_insert_object (obj) - * - * Insert an object into the LM-IAS database - * - */ -void irias_insert_object(struct ias_object *obj) -{ - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - - hashbin_insert(irias_objects, (irda_queue_t *) obj, 0, obj->name); -} -EXPORT_SYMBOL(irias_insert_object); - -/* - * Function irias_find_object (name) - * - * Find object with given name - * - */ -struct ias_object *irias_find_object(char *name) -{ - IRDA_ASSERT(name != NULL, return NULL;); - - /* Unsafe (locking), object might change */ - return hashbin_lock_find(irias_objects, 0, name); -} -EXPORT_SYMBOL(irias_find_object); - -/* - * Function irias_find_attrib (obj, name) - * - * Find named attribute in object - * - */ -struct ias_attrib *irias_find_attrib(struct ias_object *obj, char *name) -{ - struct ias_attrib *attrib; - - IRDA_ASSERT(obj != NULL, return NULL;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return NULL;); - IRDA_ASSERT(name != NULL, return NULL;); - - attrib = hashbin_lock_find(obj->attribs, 0, name); - if (attrib == NULL) - return NULL; - - /* Unsafe (locking), attrib might change */ - return attrib; -} - -/* - * Function irias_add_attribute (obj, attrib) - * - * Add attribute to object - * - */ -static void irias_add_attrib(struct ias_object *obj, struct ias_attrib *attrib, - int owner) -{ - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - - IRDA_ASSERT(attrib != NULL, return;); - IRDA_ASSERT(attrib->magic == IAS_ATTRIB_MAGIC, return;); - - /* Set if attrib is owned by kernel or user space */ - attrib->value->owner = owner; - - hashbin_insert(obj->attribs, (irda_queue_t *) attrib, 0, attrib->name); -} - -/* - * Function irias_object_change_attribute (obj_name, attrib_name, new_value) - * - * Change the value of an objects attribute. - * - */ -int irias_object_change_attribute(char *obj_name, char *attrib_name, - struct ias_value *new_value) -{ - struct ias_object *obj; - struct ias_attrib *attrib; - unsigned long flags; - - /* Find object */ - obj = hashbin_lock_find(irias_objects, 0, obj_name); - if (obj == NULL) { - net_warn_ratelimited("%s: Unable to find object: %s\n", - __func__, obj_name); - return -1; - } - - /* Slightly unsafe (obj might get removed under us) */ - spin_lock_irqsave(&obj->attribs->hb_spinlock, flags); - - /* Find attribute */ - attrib = hashbin_find(obj->attribs, 0, attrib_name); - if (attrib == NULL) { - net_warn_ratelimited("%s: Unable to find attribute: %s\n", - __func__, attrib_name); - spin_unlock_irqrestore(&obj->attribs->hb_spinlock, flags); - return -1; - } - - if ( attrib->value->type != new_value->type) { - pr_debug("%s(), changing value type not allowed!\n", - __func__); - spin_unlock_irqrestore(&obj->attribs->hb_spinlock, flags); - return -1; - } - - /* Delete old value */ - irias_delete_value(attrib->value); - - /* Insert new value */ - attrib->value = new_value; - - /* Success */ - spin_unlock_irqrestore(&obj->attribs->hb_spinlock, flags); - return 0; -} -EXPORT_SYMBOL(irias_object_change_attribute); - -/* - * Function irias_object_add_integer_attrib (obj, name, value) - * - * Add an integer attribute to an LM-IAS object - * - */ -void irias_add_integer_attrib(struct ias_object *obj, char *name, int value, - int owner) -{ - struct ias_attrib *attrib; - - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - IRDA_ASSERT(name != NULL, return;); - - attrib = kzalloc(sizeof(struct ias_attrib), GFP_ATOMIC); - if (attrib == NULL) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - return; - } - - attrib->magic = IAS_ATTRIB_MAGIC; - attrib->name = kstrndup(name, IAS_MAX_ATTRIBNAME, GFP_ATOMIC); - - /* Insert value */ - attrib->value = irias_new_integer_value(value); - if (!attrib->name || !attrib->value) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - if (attrib->value) - irias_delete_value(attrib->value); - kfree(attrib->name); - kfree(attrib); - return; - } - - irias_add_attrib(obj, attrib, owner); -} -EXPORT_SYMBOL(irias_add_integer_attrib); - - /* - * Function irias_add_octseq_attrib (obj, name, octet_seq, len) - * - * Add a octet sequence attribute to an LM-IAS object - * - */ - -void irias_add_octseq_attrib(struct ias_object *obj, char *name, __u8 *octets, - int len, int owner) -{ - struct ias_attrib *attrib; - - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - - IRDA_ASSERT(name != NULL, return;); - IRDA_ASSERT(octets != NULL, return;); - - attrib = kzalloc(sizeof(struct ias_attrib), GFP_ATOMIC); - if (attrib == NULL) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - return; - } - - attrib->magic = IAS_ATTRIB_MAGIC; - attrib->name = kstrndup(name, IAS_MAX_ATTRIBNAME, GFP_ATOMIC); - - attrib->value = irias_new_octseq_value( octets, len); - if (!attrib->name || !attrib->value) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - if (attrib->value) - irias_delete_value(attrib->value); - kfree(attrib->name); - kfree(attrib); - return; - } - - irias_add_attrib(obj, attrib, owner); -} -EXPORT_SYMBOL(irias_add_octseq_attrib); - -/* - * Function irias_object_add_string_attrib (obj, string) - * - * Add a string attribute to an LM-IAS object - * - */ -void irias_add_string_attrib(struct ias_object *obj, char *name, char *value, - int owner) -{ - struct ias_attrib *attrib; - - IRDA_ASSERT(obj != NULL, return;); - IRDA_ASSERT(obj->magic == IAS_OBJECT_MAGIC, return;); - - IRDA_ASSERT(name != NULL, return;); - IRDA_ASSERT(value != NULL, return;); - - attrib = kzalloc(sizeof( struct ias_attrib), GFP_ATOMIC); - if (attrib == NULL) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - return; - } - - attrib->magic = IAS_ATTRIB_MAGIC; - attrib->name = kstrndup(name, IAS_MAX_ATTRIBNAME, GFP_ATOMIC); - - attrib->value = irias_new_string_value(value); - if (!attrib->name || !attrib->value) { - net_warn_ratelimited("%s: Unable to allocate attribute!\n", - __func__); - if (attrib->value) - irias_delete_value(attrib->value); - kfree(attrib->name); - kfree(attrib); - return; - } - - irias_add_attrib(obj, attrib, owner); -} -EXPORT_SYMBOL(irias_add_string_attrib); - -/* - * Function irias_new_integer_value (integer) - * - * Create new IAS integer value - * - */ -struct ias_value *irias_new_integer_value(int integer) -{ - struct ias_value *value; - - value = kzalloc(sizeof(struct ias_value), GFP_ATOMIC); - if (value == NULL) - return NULL; - - value->type = IAS_INTEGER; - value->len = 4; - value->t.integer = integer; - - return value; -} -EXPORT_SYMBOL(irias_new_integer_value); - -/* - * Function irias_new_string_value (string) - * - * Create new IAS string value - * - * Per IrLMP 1.1, 4.3.3.2, strings are up to 256 chars - Jean II - */ -struct ias_value *irias_new_string_value(char *string) -{ - struct ias_value *value; - - value = kzalloc(sizeof(struct ias_value), GFP_ATOMIC); - if (value == NULL) - return NULL; - - value->type = IAS_STRING; - value->charset = CS_ASCII; - value->t.string = kstrndup(string, IAS_MAX_STRING, GFP_ATOMIC); - if (!value->t.string) { - net_warn_ratelimited("%s: Unable to kmalloc!\n", __func__); - kfree(value); - return NULL; - } - - value->len = strlen(value->t.string); - - return value; -} - -/* - * Function irias_new_octseq_value (octets, len) - * - * Create new IAS octet-sequence value - * - * Per IrLMP 1.1, 4.3.3.2, octet-sequence are up to 1024 bytes - Jean II - */ -struct ias_value *irias_new_octseq_value(__u8 *octseq , int len) -{ - struct ias_value *value; - - value = kzalloc(sizeof(struct ias_value), GFP_ATOMIC); - if (value == NULL) - return NULL; - - value->type = IAS_OCT_SEQ; - /* Check length */ - if(len > IAS_MAX_OCTET_STRING) - len = IAS_MAX_OCTET_STRING; - value->len = len; - - value->t.oct_seq = kmemdup(octseq, len, GFP_ATOMIC); - if (value->t.oct_seq == NULL){ - net_warn_ratelimited("%s: Unable to kmalloc!\n", __func__); - kfree(value); - return NULL; - } - return value; -} - -struct ias_value *irias_new_missing_value(void) -{ - struct ias_value *value; - - value = kzalloc(sizeof(struct ias_value), GFP_ATOMIC); - if (value == NULL) - return NULL; - - value->type = IAS_MISSING; - - return value; -} - -/* - * Function irias_delete_value (value) - * - * Delete IAS value - * - */ -void irias_delete_value(struct ias_value *value) -{ - IRDA_ASSERT(value != NULL, return;); - - switch (value->type) { - case IAS_INTEGER: /* Fallthrough */ - case IAS_MISSING: - /* No need to deallocate */ - break; - case IAS_STRING: - /* Deallocate string */ - kfree(value->t.string); - break; - case IAS_OCT_SEQ: - /* Deallocate byte stream */ - kfree(value->t.oct_seq); - break; - default: - pr_debug("%s(), Unknown value type!\n", __func__); - break; - } - kfree(value); -} -EXPORT_SYMBOL(irias_delete_value); diff --git a/net/irda/irlan/Kconfig b/net/irda/irlan/Kconfig deleted file mode 100644 index 951abc2e3a7f..000000000000 --- a/net/irda/irlan/Kconfig +++ /dev/null @@ -1,14 +0,0 @@ -config IRLAN - tristate "IrLAN protocol" - depends on IRDA - help - Say Y here if you want to build support for the IrLAN protocol. - To compile it as a module, choose M here: the module will be called - irlan. IrLAN emulates an Ethernet and makes it possible to put up - a wireless LAN using infrared beams. - - The IrLAN protocol can be used to talk with infrared access points - like the HP NetbeamIR, or the ESI JetEye NET. You can also connect - to another Linux machine running the IrLAN protocol for ad-hoc - networking! - diff --git a/net/irda/irlan/Makefile b/net/irda/irlan/Makefile deleted file mode 100644 index 94eefbc8e6b9..000000000000 --- a/net/irda/irlan/Makefile +++ /dev/null @@ -1,7 +0,0 @@ -# -# Makefile for the Linux IrDA IrLAN protocol layer. -# - -obj-$(CONFIG_IRLAN) += irlan.o - -irlan-y := irlan_common.o irlan_eth.o irlan_event.o irlan_client.o irlan_provider.o irlan_filter.o irlan_provider_event.o irlan_client_event.o diff --git a/net/irda/irlan/irlan_client.c b/net/irda/irlan/irlan_client.c deleted file mode 100644 index c5837a40c78e..000000000000 --- a/net/irda/irlan/irlan_client.c +++ /dev/null @@ -1,559 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_client.c - * Version: 0.9 - * Description: IrDA LAN Access Protocol (IrLAN) Client - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:37 1997 - * Modified at: Tue Dec 14 15:47:02 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Sources: skeleton.c by Donald Becker <becker@CESDIS.gsfc.nasa.gov> - * slip.c by Laurence Culhane, <loz@holmes.demon.co.uk> - * Fred N. van Kempen, <waltje@uwalt.nl.mugnet.org> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/slab.h> -#include <linux/errno.h> -#include <linux/init.h> -#include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/if_arp.h> -#include <linux/bitops.h> -#include <net/arp.h> - -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/irttp.h> -#include <net/irda/irlmp.h> -#include <net/irda/irias_object.h> -#include <net/irda/iriap.h> -#include <net/irda/timer.h> - -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_event.h> -#include <net/irda/irlan_eth.h> -#include <net/irda/irlan_provider.h> -#include <net/irda/irlan_client.h> - -#undef CONFIG_IRLAN_GRATUITOUS_ARP - -static void irlan_client_ctrl_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *); -static int irlan_client_ctrl_data_indication(void *instance, void *sap, - struct sk_buff *skb); -static void irlan_client_ctrl_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *); -static void irlan_check_response_param(struct irlan_cb *self, char *param, - char *value, int val_len); -static void irlan_client_open_ctrl_tsap(struct irlan_cb *self); - -static void irlan_client_kick_timer_expired(void *data) -{ - struct irlan_cb *self = (struct irlan_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* - * If we are in peer mode, the client may not have got the discovery - * indication it needs to make progress. If the client is still in - * IDLE state, we must kick it to, but only if the provider is not IDLE - */ - if ((self->provider.access_type == ACCESS_PEER) && - (self->client.state == IRLAN_IDLE) && - (self->provider.state != IRLAN_IDLE)) { - irlan_client_wakeup(self, self->saddr, self->daddr); - } -} - -static void irlan_client_start_kick_timer(struct irlan_cb *self, int timeout) -{ - irda_start_timer(&self->client.kick_timer, timeout, (void *) self, - irlan_client_kick_timer_expired); -} - -/* - * Function irlan_client_wakeup (self, saddr, daddr) - * - * Wake up client - * - */ -void irlan_client_wakeup(struct irlan_cb *self, __u32 saddr, __u32 daddr) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* - * Check if we are already awake, or if we are a provider in direct - * mode (in that case we must leave the client idle - */ - if ((self->client.state != IRLAN_IDLE) || - (self->provider.access_type == ACCESS_DIRECT)) - { - pr_debug("%s(), already awake!\n", __func__); - return; - } - - /* Addresses may have changed! */ - self->saddr = saddr; - self->daddr = daddr; - - if (self->disconnect_reason == LM_USER_REQUEST) { - pr_debug("%s(), still stopped by user\n", __func__); - return; - } - - /* Open TSAPs */ - irlan_client_open_ctrl_tsap(self); - irlan_open_data_tsap(self); - - irlan_do_client_event(self, IRLAN_DISCOVERY_INDICATION, NULL); - - /* Start kick timer */ - irlan_client_start_kick_timer(self, 2*HZ); -} - -/* - * Function irlan_discovery_indication (daddr) - * - * Remote device with IrLAN server support discovered - * - */ -void irlan_client_discovery_indication(discinfo_t *discovery, - DISCOVERY_MODE mode, - void *priv) -{ - struct irlan_cb *self; - __u32 saddr, daddr; - - IRDA_ASSERT(discovery != NULL, return;); - - /* - * I didn't check it, but I bet that IrLAN suffer from the same - * deficiency as IrComm and doesn't handle two instances - * simultaneously connecting to each other. - * Same workaround, drop passive discoveries. - * Jean II */ - if(mode == DISCOVERY_PASSIVE) - return; - - saddr = discovery->saddr; - daddr = discovery->daddr; - - /* Find instance */ - rcu_read_lock(); - self = irlan_get_any(); - if (self) { - IRDA_ASSERT(self->magic == IRLAN_MAGIC, goto out;); - - pr_debug("%s(), Found instance (%08x)!\n", __func__ , - daddr); - - irlan_client_wakeup(self, saddr, daddr); - } -IRDA_ASSERT_LABEL(out:) - rcu_read_unlock(); -} - -/* - * Function irlan_client_data_indication (handle, skb) - * - * This function gets the data that is received on the control channel - * - */ -static int irlan_client_ctrl_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct irlan_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - irlan_do_client_event(self, IRLAN_DATA_INDICATION, skb); - - /* Ready for a new command */ - pr_debug("%s(), clearing tx_busy\n", __func__); - self->client.tx_busy = FALSE; - - /* Check if we have some queued commands waiting to be sent */ - irlan_run_ctrl_tx_queue(self); - - return 0; -} - -static void irlan_client_ctrl_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *userdata) -{ - struct irlan_cb *self; - struct tsap_cb *tsap; - struct sk_buff *skb; - - pr_debug("%s(), reason=%d\n", __func__ , reason); - - self = instance; - tsap = sap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - IRDA_ASSERT(tsap != NULL, return;); - IRDA_ASSERT(tsap->magic == TTP_TSAP_MAGIC, return;); - - IRDA_ASSERT(tsap == self->client.tsap_ctrl, return;); - - /* Remove frames queued on the control channel */ - while ((skb = skb_dequeue(&self->client.txq)) != NULL) { - dev_kfree_skb(skb); - } - self->client.tx_busy = FALSE; - - irlan_do_client_event(self, IRLAN_LMP_DISCONNECT, NULL); -} - -/* - * Function irlan_client_open_tsaps (self) - * - * Initialize callbacks and open IrTTP TSAPs - * - */ -static void irlan_client_open_ctrl_tsap(struct irlan_cb *self) -{ - struct tsap_cb *tsap; - notify_t notify; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Check if already open */ - if (self->client.tsap_ctrl) - return; - - irda_notify_init(¬ify); - - /* Set up callbacks */ - notify.data_indication = irlan_client_ctrl_data_indication; - notify.connect_confirm = irlan_client_ctrl_connect_confirm; - notify.disconnect_indication = irlan_client_ctrl_disconnect_indication; - notify.instance = self; - strlcpy(notify.name, "IrLAN ctrl (c)", sizeof(notify.name)); - - tsap = irttp_open_tsap(LSAP_ANY, DEFAULT_INITIAL_CREDIT, ¬ify); - if (!tsap) { - pr_debug("%s(), Got no tsap!\n", __func__); - return; - } - self->client.tsap_ctrl = tsap; -} - -/* - * Function irlan_client_connect_confirm (handle, skb) - * - * Connection to peer IrLAN laye confirmed - * - */ -static void irlan_client_ctrl_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct irlan_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - self->client.max_sdu_size = max_sdu_size; - self->client.max_header_size = max_header_size; - - /* TODO: we could set the MTU depending on the max_sdu_size */ - - irlan_do_client_event(self, IRLAN_CONNECT_COMPLETE, NULL); -} - -/* - * Function print_ret_code (code) - * - * Print return code of request to peer IrLAN layer. - * - */ -static void print_ret_code(__u8 code) -{ - switch(code) { - case 0: - printk(KERN_INFO "Success\n"); - break; - case 1: - net_warn_ratelimited("IrLAN: Insufficient resources\n"); - break; - case 2: - net_warn_ratelimited("IrLAN: Invalid command format\n"); - break; - case 3: - net_warn_ratelimited("IrLAN: Command not supported\n"); - break; - case 4: - net_warn_ratelimited("IrLAN: Parameter not supported\n"); - break; - case 5: - net_warn_ratelimited("IrLAN: Value not supported\n"); - break; - case 6: - net_warn_ratelimited("IrLAN: Not open\n"); - break; - case 7: - net_warn_ratelimited("IrLAN: Authentication required\n"); - break; - case 8: - net_warn_ratelimited("IrLAN: Invalid password\n"); - break; - case 9: - net_warn_ratelimited("IrLAN: Protocol error\n"); - break; - case 255: - net_warn_ratelimited("IrLAN: Asynchronous status\n"); - break; - } -} - -/* - * Function irlan_client_parse_response (self, skb) - * - * Extract all parameters from received buffer, then feed them to - * check_params for parsing - */ -void irlan_client_parse_response(struct irlan_cb *self, struct sk_buff *skb) -{ - __u8 *frame; - __u8 *ptr; - int count; - int ret; - __u16 val_len; - int i; - char *name; - char *value; - - IRDA_ASSERT(skb != NULL, return;); - - pr_debug("%s() skb->len=%d\n", __func__ , (int)skb->len); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - if (!skb) { - net_err_ratelimited("%s(), Got NULL skb!\n", __func__); - return; - } - frame = skb->data; - - /* - * Check return code and print it if not success - */ - if (frame[0]) { - print_ret_code(frame[0]); - return; - } - - name = kmalloc(255, GFP_ATOMIC); - if (!name) - return; - value = kmalloc(1016, GFP_ATOMIC); - if (!value) { - kfree(name); - return; - } - - /* How many parameters? */ - count = frame[1]; - - pr_debug("%s(), got %d parameters\n", __func__ , count); - - ptr = frame+2; - - /* For all parameters */ - for (i=0; i<count;i++) { - ret = irlan_extract_param(ptr, name, value, &val_len); - if (ret < 0) { - pr_debug("%s(), IrLAN, Error!\n", __func__); - break; - } - ptr += ret; - irlan_check_response_param(self, name, value, val_len); - } - /* Cleanup */ - kfree(name); - kfree(value); -} - -/* - * Function irlan_check_response_param (self, param, value, val_len) - * - * Check which parameter is received and update local variables - * - */ -static void irlan_check_response_param(struct irlan_cb *self, char *param, - char *value, int val_len) -{ - __u16 tmp_cpu; /* Temporary value in host order */ - __u8 *bytes; - int i; - - pr_debug("%s(), parm=%s\n", __func__ , param); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Media type */ - if (strcmp(param, "MEDIA") == 0) { - if (strcmp(value, "802.3") == 0) - self->media = MEDIA_802_3; - else - self->media = MEDIA_802_5; - return; - } - if (strcmp(param, "FILTER_TYPE") == 0) { - if (strcmp(value, "DIRECTED") == 0) - self->client.filter_type |= IRLAN_DIRECTED; - else if (strcmp(value, "FUNCTIONAL") == 0) - self->client.filter_type |= IRLAN_FUNCTIONAL; - else if (strcmp(value, "GROUP") == 0) - self->client.filter_type |= IRLAN_GROUP; - else if (strcmp(value, "MAC_FRAME") == 0) - self->client.filter_type |= IRLAN_MAC_FRAME; - else if (strcmp(value, "MULTICAST") == 0) - self->client.filter_type |= IRLAN_MULTICAST; - else if (strcmp(value, "BROADCAST") == 0) - self->client.filter_type |= IRLAN_BROADCAST; - else if (strcmp(value, "IPX_SOCKET") == 0) - self->client.filter_type |= IRLAN_IPX_SOCKET; - - } - if (strcmp(param, "ACCESS_TYPE") == 0) { - if (strcmp(value, "DIRECT") == 0) - self->client.access_type = ACCESS_DIRECT; - else if (strcmp(value, "PEER") == 0) - self->client.access_type = ACCESS_PEER; - else if (strcmp(value, "HOSTED") == 0) - self->client.access_type = ACCESS_HOSTED; - else { - pr_debug("%s(), unknown access type!\n", __func__); - } - } - /* IRLAN version */ - if (strcmp(param, "IRLAN_VER") == 0) { - pr_debug("IrLAN version %d.%d\n", (__u8)value[0], - (__u8)value[1]); - - self->version[0] = value[0]; - self->version[1] = value[1]; - return; - } - /* Which remote TSAP to use for data channel */ - if (strcmp(param, "DATA_CHAN") == 0) { - self->dtsap_sel_data = value[0]; - pr_debug("Data TSAP = %02x\n", self->dtsap_sel_data); - return; - } - if (strcmp(param, "CON_ARB") == 0) { - memcpy(&tmp_cpu, value, 2); /* Align value */ - le16_to_cpus(&tmp_cpu); /* Convert to host order */ - self->client.recv_arb_val = tmp_cpu; - pr_debug("%s(), receive arb val=%d\n", __func__ , - self->client.recv_arb_val); - } - if (strcmp(param, "MAX_FRAME") == 0) { - memcpy(&tmp_cpu, value, 2); /* Align value */ - le16_to_cpus(&tmp_cpu); /* Convert to host order */ - self->client.max_frame = tmp_cpu; - pr_debug("%s(), max frame=%d\n", __func__ , - self->client.max_frame); - } - - /* RECONNECT_KEY, in case the link goes down! */ - if (strcmp(param, "RECONNECT_KEY") == 0) { - pr_debug("Got reconnect key: "); - /* for (i = 0; i < val_len; i++) */ -/* printk("%02x", value[i]); */ - memcpy(self->client.reconnect_key, value, val_len); - self->client.key_len = val_len; - pr_debug("\n"); - } - /* FILTER_ENTRY, have we got an ethernet address? */ - if (strcmp(param, "FILTER_ENTRY") == 0) { - bytes = value; - pr_debug("Ethernet address = %pM\n", bytes); - for (i = 0; i < 6; i++) - self->dev->dev_addr[i] = bytes[i]; - } -} - -/* - * Function irlan_client_get_value_confirm (obj_id, value) - * - * Got results from remote LM-IAS - * - */ -void irlan_client_get_value_confirm(int result, __u16 obj_id, - struct ias_value *value, void *priv) -{ - struct irlan_cb *self; - - IRDA_ASSERT(priv != NULL, return;); - - self = priv; - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* We probably don't need to make any more queries */ - iriap_close(self->client.iriap); - self->client.iriap = NULL; - - /* Check if request succeeded */ - if (result != IAS_SUCCESS) { - pr_debug("%s(), got NULL value!\n", __func__); - irlan_do_client_event(self, IRLAN_IAS_PROVIDER_NOT_AVAIL, - NULL); - return; - } - - switch (value->type) { - case IAS_INTEGER: - self->dtsap_sel_ctrl = value->t.integer; - - if (value->t.integer != -1) { - irlan_do_client_event(self, IRLAN_IAS_PROVIDER_AVAIL, - NULL); - return; - } - irias_delete_value(value); - break; - default: - pr_debug("%s(), unknown type!\n", __func__); - break; - } - irlan_do_client_event(self, IRLAN_IAS_PROVIDER_NOT_AVAIL, NULL); -} diff --git a/net/irda/irlan/irlan_client_event.c b/net/irda/irlan/irlan_client_event.c deleted file mode 100644 index cc93fabbbb19..000000000000 --- a/net/irda/irlan/irlan_client_event.c +++ /dev/null @@ -1,511 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_client_event.c - * Version: 0.9 - * Description: IrLAN client state machine - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:37 1997 - * Modified at: Sun Dec 26 21:52:24 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> - -#include <net/irda/irda.h> -#include <net/irda/timer.h> -#include <net/irda/irmod.h> -#include <net/irda/iriap.h> -#include <net/irda/irlmp.h> -#include <net/irda/irttp.h> - -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_client.h> -#include <net/irda/irlan_event.h> - -static int irlan_client_state_idle (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_query(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_conn (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_info (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_media(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_open (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_wait (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_arb (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_data (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_close(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_client_state_sync (struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); - -static int (*state[])(struct irlan_cb *, IRLAN_EVENT event, struct sk_buff *) = -{ - irlan_client_state_idle, - irlan_client_state_query, - irlan_client_state_conn, - irlan_client_state_info, - irlan_client_state_media, - irlan_client_state_open, - irlan_client_state_wait, - irlan_client_state_arb, - irlan_client_state_data, - irlan_client_state_close, - irlan_client_state_sync -}; - -void irlan_do_client_event(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - (*state[ self->client.state]) (self, event, skb); -} - -/* - * Function irlan_client_state_idle (event, skb, info) - * - * IDLE, We are waiting for an indication that there is a provider - * available. - */ -static int irlan_client_state_idle(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - switch (event) { - case IRLAN_DISCOVERY_INDICATION: - if (self->client.iriap) { - net_warn_ratelimited("%s(), busy with a previous query\n", - __func__); - return -EBUSY; - } - - self->client.iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - irlan_client_get_value_confirm); - /* Get some values from peer IAS */ - irlan_next_client_state(self, IRLAN_QUERY); - iriap_getvaluebyclass_request(self->client.iriap, - self->saddr, self->daddr, - "IrLAN", "IrDA:TinyTP:LsapSel"); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_query (event, skb, info) - * - * QUERY, We have queryed the remote IAS and is ready to connect - * to provider, just waiting for the confirm. - * - */ -static int irlan_client_state_query(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - switch(event) { - case IRLAN_IAS_PROVIDER_AVAIL: - IRDA_ASSERT(self->dtsap_sel_ctrl != 0, return -1;); - - self->client.open_retries = 0; - - irttp_connect_request(self->client.tsap_ctrl, - self->dtsap_sel_ctrl, - self->saddr, self->daddr, NULL, - IRLAN_MTU, NULL); - irlan_next_client_state(self, IRLAN_CONN); - break; - case IRLAN_IAS_PROVIDER_NOT_AVAIL: - pr_debug("%s(), IAS_PROVIDER_NOT_AVAIL\n", __func__); - irlan_next_client_state(self, IRLAN_IDLE); - - /* Give the client a kick! */ - if ((self->provider.access_type == ACCESS_PEER) && - (self->provider.state != IRLAN_IDLE)) - irlan_client_wakeup(self, self->saddr, self->daddr); - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_conn (event, skb, info) - * - * CONN, We have connected to a provider but has not issued any - * commands yet. - * - */ -static int irlan_client_state_conn(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch (event) { - case IRLAN_CONNECT_COMPLETE: - /* Send getinfo cmd */ - irlan_get_provider_info(self); - irlan_next_client_state(self, IRLAN_INFO); - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_info (self, event, skb, info) - * - * INFO, We have issued a GetInfo command and is awaiting a reply. - */ -static int irlan_client_state_info(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch (event) { - case IRLAN_DATA_INDICATION: - IRDA_ASSERT(skb != NULL, return -1;); - - irlan_client_parse_response(self, skb); - - irlan_next_client_state(self, IRLAN_MEDIA); - - irlan_get_media_char(self); - break; - - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_media (self, event, skb, info) - * - * MEDIA, The irlan_client has issued a GetMedia command and is awaiting a - * reply. - * - */ -static int irlan_client_state_media(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_DATA_INDICATION: - irlan_client_parse_response(self, skb); - irlan_open_data_channel(self); - irlan_next_client_state(self, IRLAN_OPEN); - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_open (self, event, skb, info) - * - * OPEN, The irlan_client has issued a OpenData command and is awaiting a - * reply - * - */ -static int irlan_client_state_open(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - struct qos_info qos; - - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_DATA_INDICATION: - irlan_client_parse_response(self, skb); - - /* - * Check if we have got the remote TSAP for data - * communications - */ - IRDA_ASSERT(self->dtsap_sel_data != 0, return -1;); - - /* Check which access type we are dealing with */ - switch (self->client.access_type) { - case ACCESS_PEER: - if (self->provider.state == IRLAN_OPEN) { - - irlan_next_client_state(self, IRLAN_ARB); - irlan_do_client_event(self, IRLAN_CHECK_CON_ARB, - NULL); - } else { - - irlan_next_client_state(self, IRLAN_WAIT); - } - break; - case ACCESS_DIRECT: - case ACCESS_HOSTED: - qos.link_disc_time.bits = 0x01; /* 3 secs */ - - irttp_connect_request(self->tsap_data, - self->dtsap_sel_data, - self->saddr, self->daddr, &qos, - IRLAN_MTU, NULL); - - irlan_next_client_state(self, IRLAN_DATA); - break; - default: - pr_debug("%s(), unknown access type!\n", __func__); - break; - } - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_wait (self, event, skb, info) - * - * WAIT, The irlan_client is waiting for the local provider to enter the - * provider OPEN state. - * - */ -static int irlan_client_state_wait(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_PROVIDER_SIGNAL: - irlan_next_client_state(self, IRLAN_ARB); - irlan_do_client_event(self, IRLAN_CHECK_CON_ARB, NULL); - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -static int irlan_client_state_arb(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - struct qos_info qos; - - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_CHECK_CON_ARB: - if (self->client.recv_arb_val == self->provider.send_arb_val) { - irlan_next_client_state(self, IRLAN_CLOSE); - irlan_close_data_channel(self); - } else if (self->client.recv_arb_val < - self->provider.send_arb_val) - { - qos.link_disc_time.bits = 0x01; /* 3 secs */ - - irlan_next_client_state(self, IRLAN_DATA); - irttp_connect_request(self->tsap_data, - self->dtsap_sel_data, - self->saddr, self->daddr, &qos, - IRLAN_MTU, NULL); - } else if (self->client.recv_arb_val > - self->provider.send_arb_val) - { - pr_debug("%s(), lost the battle :-(\n", __func__); - } - break; - case IRLAN_DATA_CONNECT_INDICATION: - irlan_next_client_state(self, IRLAN_DATA); - break; - case IRLAN_LMP_DISCONNECT: - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - case IRLAN_WATCHDOG_TIMEOUT: - pr_debug("%s(), IRLAN_WATCHDOG_TIMEOUT\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_data (self, event, skb, info) - * - * DATA, The data channel is connected, allowing data transfers between - * the local and remote machines. - * - */ -static int irlan_client_state_data(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - switch(event) { - case IRLAN_DATA_INDICATION: - irlan_client_parse_response(self, skb); - break; - case IRLAN_LMP_DISCONNECT: /* FALLTHROUGH */ - case IRLAN_LAP_DISCONNECT: - irlan_next_client_state(self, IRLAN_IDLE); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_close (self, event, skb, info) - * - * - * - */ -static int irlan_client_state_close(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_client_state_sync (self, event, skb, info) - * - * - * - */ -static int irlan_client_state_sync(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - if (skb) - dev_kfree_skb(skb); - - return 0; -} - - - - - - - - - - - - - diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c deleted file mode 100644 index 481bbc2a4349..000000000000 --- a/net/irda/irlan/irlan_common.c +++ /dev/null @@ -1,1176 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_common.c - * Version: 0.9 - * Description: IrDA LAN Access Protocol Implementation - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:37 1997 - * Modified at: Sun Dec 26 21:53:10 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1997, 1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/module.h> - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/gfp.h> -#include <linux/init.h> -#include <linux/errno.h> -#include <linux/proc_fs.h> -#include <linux/sched.h> -#include <linux/seq_file.h> -#include <linux/random.h> -#include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/rtnetlink.h> -#include <linux/moduleparam.h> -#include <linux/bitops.h> - -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/irttp.h> -#include <net/irda/irlmp.h> -#include <net/irda/iriap.h> -#include <net/irda/timer.h> - -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_client.h> -#include <net/irda/irlan_provider.h> -#include <net/irda/irlan_eth.h> -#include <net/irda/irlan_filter.h> - - -/* extern char sysctl_devname[]; */ - -/* - * Master structure - */ -static LIST_HEAD(irlans); - -static void *ckey; -static void *skey; - -/* Module parameters */ -static bool eth; /* Use "eth" or "irlan" name for devices */ -static int access = ACCESS_PEER; /* PEER, DIRECT or HOSTED */ - -#ifdef CONFIG_PROC_FS -static const char *const irlan_access[] = { - "UNKNOWN", - "DIRECT", - "PEER", - "HOSTED" -}; - -static const char *const irlan_media[] = { - "UNKNOWN", - "802.3", - "802.5" -}; - -extern struct proc_dir_entry *proc_irda; - -static int irlan_seq_open(struct inode *inode, struct file *file); - -static const struct file_operations irlan_fops = { - .owner = THIS_MODULE, - .open = irlan_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -extern struct proc_dir_entry *proc_irda; -#endif /* CONFIG_PROC_FS */ - -static struct irlan_cb __init *irlan_open(__u32 saddr, __u32 daddr); -static void __irlan_close(struct irlan_cb *self); -static int __irlan_insert_param(struct sk_buff *skb, char *param, int type, - __u8 value_byte, __u16 value_short, - __u8 *value_array, __u16 value_len); -static void irlan_open_unicast_addr(struct irlan_cb *self); -static void irlan_get_unicast_addr(struct irlan_cb *self); -void irlan_close_tsaps(struct irlan_cb *self); - -/* - * Function irlan_init (void) - * - * Initialize IrLAN layer - * - */ -static int __init irlan_init(void) -{ - struct irlan_cb *new; - __u16 hints; - -#ifdef CONFIG_PROC_FS - { struct proc_dir_entry *proc; - proc = proc_create("irlan", 0, proc_irda, &irlan_fops); - if (!proc) { - printk(KERN_ERR "irlan_init: can't create /proc entry!\n"); - return -ENODEV; - } - } -#endif /* CONFIG_PROC_FS */ - - hints = irlmp_service_to_hint(S_LAN); - - /* Register with IrLMP as a client */ - ckey = irlmp_register_client(hints, &irlan_client_discovery_indication, - NULL, NULL); - if (!ckey) - goto err_ckey; - - /* Register with IrLMP as a service */ - skey = irlmp_register_service(hints); - if (!skey) - goto err_skey; - - /* Start the master IrLAN instance (the only one for now) */ - new = irlan_open(DEV_ADDR_ANY, DEV_ADDR_ANY); - if (!new) - goto err_open; - - /* The master will only open its (listen) control TSAP */ - irlan_provider_open_ctrl_tsap(new); - - /* Do some fast discovery! */ - irlmp_discovery_request(DISCOVERY_DEFAULT_SLOTS); - - return 0; - -err_open: - irlmp_unregister_service(skey); -err_skey: - irlmp_unregister_client(ckey); -err_ckey: -#ifdef CONFIG_PROC_FS - remove_proc_entry("irlan", proc_irda); -#endif /* CONFIG_PROC_FS */ - - return -ENOMEM; -} - -static void __exit irlan_cleanup(void) -{ - struct irlan_cb *self, *next; - - irlmp_unregister_client(ckey); - irlmp_unregister_service(skey); - -#ifdef CONFIG_PROC_FS - remove_proc_entry("irlan", proc_irda); -#endif /* CONFIG_PROC_FS */ - - /* Cleanup any leftover network devices */ - rtnl_lock(); - list_for_each_entry_safe(self, next, &irlans, dev_list) { - __irlan_close(self); - } - rtnl_unlock(); -} - -/* - * Function irlan_open (void) - * - * Open new instance of a client/provider, we should only register the - * network device if this instance is ment for a particular client/provider - */ -static struct irlan_cb __init *irlan_open(__u32 saddr, __u32 daddr) -{ - struct net_device *dev; - struct irlan_cb *self; - - /* Create network device with irlan */ - dev = alloc_irlandev(eth ? "eth%d" : "irlan%d"); - if (!dev) - return NULL; - - self = netdev_priv(dev); - self->dev = dev; - - /* - * Initialize local device structure - */ - self->magic = IRLAN_MAGIC; - self->saddr = saddr; - self->daddr = daddr; - - /* Provider access can only be PEER, DIRECT, or HOSTED */ - self->provider.access_type = access; - if (access == ACCESS_DIRECT) { - /* - * Since we are emulating an IrLAN sever we will have to - * give ourself an ethernet address! - */ - dev->dev_addr[0] = 0x40; - dev->dev_addr[1] = 0x00; - dev->dev_addr[2] = 0x00; - dev->dev_addr[3] = 0x00; - get_random_bytes(dev->dev_addr+4, 1); - get_random_bytes(dev->dev_addr+5, 1); - } - - self->media = MEDIA_802_3; - self->disconnect_reason = LM_USER_REQUEST; - init_timer(&self->watchdog_timer); - init_timer(&self->client.kick_timer); - init_waitqueue_head(&self->open_wait); - - skb_queue_head_init(&self->client.txq); - - irlan_next_client_state(self, IRLAN_IDLE); - irlan_next_provider_state(self, IRLAN_IDLE); - - if (register_netdev(dev)) { - pr_debug("%s(), register_netdev() failed!\n", - __func__); - self = NULL; - free_netdev(dev); - } else { - rtnl_lock(); - list_add_rcu(&self->dev_list, &irlans); - rtnl_unlock(); - } - - return self; -} -/* - * Function __irlan_close (self) - * - * This function closes and deallocates the IrLAN client instances. Be - * aware that other functions which calls client_close() must - * remove self from irlans list first. - */ -static void __irlan_close(struct irlan_cb *self) -{ - ASSERT_RTNL(); - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - del_timer_sync(&self->watchdog_timer); - del_timer_sync(&self->client.kick_timer); - - /* Close all open connections and remove TSAPs */ - irlan_close_tsaps(self); - - if (self->client.iriap) - iriap_close(self->client.iriap); - - /* Remove frames queued on the control channel */ - skb_queue_purge(&self->client.txq); - - /* Unregister and free self via destructor */ - unregister_netdevice(self->dev); -} - -/* Find any instance of irlan, used for client discovery wakeup */ -struct irlan_cb *irlan_get_any(void) -{ - struct irlan_cb *self; - - list_for_each_entry_rcu(self, &irlans, dev_list) { - return self; - } - return NULL; -} - -/* - * Function irlan_connect_indication (instance, sap, qos, max_sdu_size, skb) - * - * Here we receive the connect indication for the data channel - * - */ -static void irlan_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct irlan_cb *self; - struct tsap_cb *tsap; - - self = instance; - tsap = sap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - IRDA_ASSERT(tsap == self->tsap_data,return;); - - self->max_sdu_size = max_sdu_size; - self->max_header_size = max_header_size; - - pr_debug("%s: We are now connected!\n", __func__); - - del_timer(&self->watchdog_timer); - - /* If you want to pass the skb to *both* state machines, you will - * need to skb_clone() it, so that you don't free it twice. - * As the state machines don't need it, git rid of it here... - * Jean II */ - if (skb) - dev_kfree_skb(skb); - - irlan_do_provider_event(self, IRLAN_DATA_CONNECT_INDICATION, NULL); - irlan_do_client_event(self, IRLAN_DATA_CONNECT_INDICATION, NULL); - - if (self->provider.access_type == ACCESS_PEER) { - /* - * Data channel is open, so we are now allowed to - * configure the remote filter - */ - irlan_get_unicast_addr(self); - irlan_open_unicast_addr(self); - } - /* Ready to transfer Ethernet frames (at last) */ - netif_start_queue(self->dev); /* Clear reason */ -} - -static void irlan_connect_confirm(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct irlan_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - self->max_sdu_size = max_sdu_size; - self->max_header_size = max_header_size; - - /* TODO: we could set the MTU depending on the max_sdu_size */ - - pr_debug("%s: We are now connected!\n", __func__); - del_timer(&self->watchdog_timer); - - /* - * Data channel is open, so we are now allowed to configure the remote - * filter - */ - irlan_get_unicast_addr(self); - irlan_open_unicast_addr(self); - - /* Open broadcast and multicast filter by default */ - irlan_set_broadcast_filter(self, TRUE); - irlan_set_multicast_filter(self, TRUE); - - /* Ready to transfer Ethernet frames */ - netif_start_queue(self->dev); - self->disconnect_reason = 0; /* Clear reason */ - wake_up_interruptible(&self->open_wait); -} - -/* - * Function irlan_client_disconnect_indication (handle) - * - * Callback function for the IrTTP layer. Indicates a disconnection of - * the specified connection (handle) - */ -static void irlan_disconnect_indication(void *instance, - void *sap, LM_REASON reason, - struct sk_buff *userdata) -{ - struct irlan_cb *self; - struct tsap_cb *tsap; - - pr_debug("%s(), reason=%d\n", __func__ , reason); - - self = instance; - tsap = sap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - IRDA_ASSERT(tsap != NULL, return;); - IRDA_ASSERT(tsap->magic == TTP_TSAP_MAGIC, return;); - - IRDA_ASSERT(tsap == self->tsap_data, return;); - - pr_debug("IrLAN, data channel disconnected by peer!\n"); - - /* Save reason so we know if we should try to reconnect or not */ - self->disconnect_reason = reason; - - switch (reason) { - case LM_USER_REQUEST: /* User request */ - pr_debug("%s(), User requested\n", __func__); - break; - case LM_LAP_DISCONNECT: /* Unexpected IrLAP disconnect */ - pr_debug("%s(), Unexpected IrLAP disconnect\n", __func__); - break; - case LM_CONNECT_FAILURE: /* Failed to establish IrLAP connection */ - pr_debug("%s(), IrLAP connect failed\n", __func__); - break; - case LM_LAP_RESET: /* IrLAP reset */ - pr_debug("%s(), IrLAP reset\n", __func__); - break; - case LM_INIT_DISCONNECT: - pr_debug("%s(), IrLMP connect failed\n", __func__); - break; - default: - net_err_ratelimited("%s(), Unknown disconnect reason\n", - __func__); - break; - } - - /* If you want to pass the skb to *both* state machines, you will - * need to skb_clone() it, so that you don't free it twice. - * As the state machines don't need it, git rid of it here... - * Jean II */ - if (userdata) - dev_kfree_skb(userdata); - - irlan_do_client_event(self, IRLAN_LMP_DISCONNECT, NULL); - irlan_do_provider_event(self, IRLAN_LMP_DISCONNECT, NULL); - - wake_up_interruptible(&self->open_wait); -} - -void irlan_open_data_tsap(struct irlan_cb *self) -{ - struct tsap_cb *tsap; - notify_t notify; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Check if already open */ - if (self->tsap_data) - return; - - irda_notify_init(¬ify); - - notify.data_indication = irlan_eth_receive; - notify.udata_indication = irlan_eth_receive; - notify.connect_indication = irlan_connect_indication; - notify.connect_confirm = irlan_connect_confirm; - notify.flow_indication = irlan_eth_flow_indication; - notify.disconnect_indication = irlan_disconnect_indication; - notify.instance = self; - strlcpy(notify.name, "IrLAN data", sizeof(notify.name)); - - tsap = irttp_open_tsap(LSAP_ANY, DEFAULT_INITIAL_CREDIT, ¬ify); - if (!tsap) { - pr_debug("%s(), Got no tsap!\n", __func__); - return; - } - self->tsap_data = tsap; - - /* - * This is the data TSAP selector which we will pass to the client - * when the client ask for it. - */ - self->stsap_sel_data = self->tsap_data->stsap_sel; -} - -void irlan_close_tsaps(struct irlan_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Disconnect and close all open TSAP connections */ - if (self->tsap_data) { - irttp_disconnect_request(self->tsap_data, NULL, P_NORMAL); - irttp_close_tsap(self->tsap_data); - self->tsap_data = NULL; - } - if (self->client.tsap_ctrl) { - irttp_disconnect_request(self->client.tsap_ctrl, NULL, - P_NORMAL); - irttp_close_tsap(self->client.tsap_ctrl); - self->client.tsap_ctrl = NULL; - } - if (self->provider.tsap_ctrl) { - irttp_disconnect_request(self->provider.tsap_ctrl, NULL, - P_NORMAL); - irttp_close_tsap(self->provider.tsap_ctrl); - self->provider.tsap_ctrl = NULL; - } - self->disconnect_reason = LM_USER_REQUEST; -} - -/* - * Function irlan_ias_register (self, tsap_sel) - * - * Register with LM-IAS - * - */ -void irlan_ias_register(struct irlan_cb *self, __u8 tsap_sel) -{ - struct ias_object *obj; - struct ias_value *new_value; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* - * Check if object has already been registered by a previous provider. - * If that is the case, we just change the value of the attribute - */ - if (!irias_find_object("IrLAN")) { - obj = irias_new_object("IrLAN", IAS_IRLAN_ID); - irias_add_integer_attrib(obj, "IrDA:TinyTP:LsapSel", tsap_sel, - IAS_KERNEL_ATTR); - irias_insert_object(obj); - } else { - new_value = irias_new_integer_value(tsap_sel); - irias_object_change_attribute("IrLAN", "IrDA:TinyTP:LsapSel", - new_value); - } - - /* Register PnP object only if not registered before */ - if (!irias_find_object("PnP")) { - obj = irias_new_object("PnP", IAS_PNP_ID); -#if 0 - irias_add_string_attrib(obj, "Name", sysctl_devname, - IAS_KERNEL_ATTR); -#else - irias_add_string_attrib(obj, "Name", "Linux", IAS_KERNEL_ATTR); -#endif - irias_add_string_attrib(obj, "DeviceID", "HWP19F0", - IAS_KERNEL_ATTR); - irias_add_integer_attrib(obj, "CompCnt", 1, IAS_KERNEL_ATTR); - if (self->provider.access_type == ACCESS_PEER) - irias_add_string_attrib(obj, "Comp#01", "PNP8389", - IAS_KERNEL_ATTR); - else - irias_add_string_attrib(obj, "Comp#01", "PNP8294", - IAS_KERNEL_ATTR); - - irias_add_string_attrib(obj, "Manufacturer", - "Linux-IrDA Project", IAS_KERNEL_ATTR); - irias_insert_object(obj); - } -} - -/* - * Function irlan_run_ctrl_tx_queue (self) - * - * Try to send the next command in the control transmit queue - * - */ -int irlan_run_ctrl_tx_queue(struct irlan_cb *self) -{ - struct sk_buff *skb; - - if (irda_lock(&self->client.tx_busy) == FALSE) - return -EBUSY; - - skb = skb_dequeue(&self->client.txq); - if (!skb) { - self->client.tx_busy = FALSE; - return 0; - } - - /* Check that it's really possible to send commands */ - if ((self->client.tsap_ctrl == NULL) || - (self->client.state == IRLAN_IDLE)) - { - self->client.tx_busy = FALSE; - dev_kfree_skb(skb); - return -1; - } - pr_debug("%s(), sending ...\n", __func__); - - return irttp_data_request(self->client.tsap_ctrl, skb); -} - -/* - * Function irlan_ctrl_data_request (self, skb) - * - * This function makes sure that commands on the control channel is being - * sent in a command/response fashion - */ -static void irlan_ctrl_data_request(struct irlan_cb *self, struct sk_buff *skb) -{ - /* Queue command */ - skb_queue_tail(&self->client.txq, skb); - - /* Try to send command */ - irlan_run_ctrl_tx_queue(self); -} - -/* - * Function irlan_get_provider_info (self) - * - * Send Get Provider Information command to peer IrLAN layer - * - */ -void irlan_get_provider_info(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER, - GFP_ATOMIC); - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - frame[0] = CMD_GET_PROVIDER_INFO; - frame[1] = 0x00; /* Zero parameters */ - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_open_data_channel (self) - * - * Send an Open Data Command to provider - * - */ -void irlan_open_data_channel(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_STRING_PARAMETER_LEN("MEDIA", "802.3") + - IRLAN_STRING_PARAMETER_LEN("ACCESS_TYPE", "DIRECT"), - GFP_ATOMIC); - if (!skb) - return; - - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - /* Build frame */ - frame[0] = CMD_OPEN_DATA_CHANNEL; - frame[1] = 0x02; /* Two parameters */ - - irlan_insert_string_param(skb, "MEDIA", "802.3"); - irlan_insert_string_param(skb, "ACCESS_TYPE", "DIRECT"); - /* irlan_insert_string_param(skb, "MODE", "UNRELIABLE"); */ - -/* self->use_udata = TRUE; */ - - irlan_ctrl_data_request(self, skb); -} - -void irlan_close_data_channel(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Check if the TSAP is still there */ - if (self->client.tsap_ctrl == NULL) - return; - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN"), - GFP_ATOMIC); - if (!skb) - return; - - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - /* Build frame */ - frame[0] = CMD_CLOSE_DATA_CHAN; - frame[1] = 0x01; /* One parameter */ - - irlan_insert_byte_param(skb, "DATA_CHAN", self->dtsap_sel_data); - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_open_unicast_addr (self) - * - * Make IrLAN provider accept ethernet frames addressed to the unicast - * address. - * - */ -static void irlan_open_unicast_addr(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") + - IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "FILTER"), - GFP_ATOMIC); - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - frame[0] = CMD_FILTER_OPERATION; - frame[1] = 0x03; /* Three parameters */ - irlan_insert_byte_param(skb, "DATA_CHAN" , self->dtsap_sel_data); - irlan_insert_string_param(skb, "FILTER_TYPE", "DIRECTED"); - irlan_insert_string_param(skb, "FILTER_MODE", "FILTER"); - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_set_broadcast_filter (self, status) - * - * Make IrLAN provider accept ethernet frames addressed to the broadcast - * address. Be careful with the use of this one, since there may be a lot - * of broadcast traffic out there. We can still function without this - * one but then _we_ have to initiate all communication with other - * hosts, since ARP request for this host will not be answered. - */ -void irlan_set_broadcast_filter(struct irlan_cb *self, int status) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "BROADCAST") + - /* We may waste one byte here...*/ - IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "FILTER"), - GFP_ATOMIC); - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - frame[0] = CMD_FILTER_OPERATION; - frame[1] = 0x03; /* Three parameters */ - irlan_insert_byte_param(skb, "DATA_CHAN", self->dtsap_sel_data); - irlan_insert_string_param(skb, "FILTER_TYPE", "BROADCAST"); - if (status) - irlan_insert_string_param(skb, "FILTER_MODE", "FILTER"); - else - irlan_insert_string_param(skb, "FILTER_MODE", "NONE"); - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_set_multicast_filter (self, status) - * - * Make IrLAN provider accept ethernet frames addressed to the multicast - * address. - * - */ -void irlan_set_multicast_filter(struct irlan_cb *self, int status) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "MULTICAST") + - /* We may waste one byte here...*/ - IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "NONE"), - GFP_ATOMIC); - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - frame[0] = CMD_FILTER_OPERATION; - frame[1] = 0x03; /* Three parameters */ - irlan_insert_byte_param(skb, "DATA_CHAN", self->dtsap_sel_data); - irlan_insert_string_param(skb, "FILTER_TYPE", "MULTICAST"); - if (status) - irlan_insert_string_param(skb, "FILTER_MODE", "ALL"); - else - irlan_insert_string_param(skb, "FILTER_MODE", "NONE"); - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_get_unicast_addr (self) - * - * Retrieves the unicast address from the IrLAN provider. This address - * will be inserted into the devices structure, so the ethernet layer - * can construct its packets. - * - */ -static void irlan_get_unicast_addr(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") + - IRLAN_STRING_PARAMETER_LEN("FILTER_OPERATION", - "DYNAMIC"), - GFP_ATOMIC); - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - frame[0] = CMD_FILTER_OPERATION; - frame[1] = 0x03; /* Three parameters */ - irlan_insert_byte_param(skb, "DATA_CHAN", self->dtsap_sel_data); - irlan_insert_string_param(skb, "FILTER_TYPE", "DIRECTED"); - irlan_insert_string_param(skb, "FILTER_OPERATION", "DYNAMIC"); - - irlan_ctrl_data_request(self, skb); -} - -/* - * Function irlan_get_media_char (self) - * - * - * - */ -void irlan_get_media_char(struct irlan_cb *self) -{ - struct sk_buff *skb; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - IRLAN_STRING_PARAMETER_LEN("MEDIA", "802.3"), - GFP_ATOMIC); - - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->client.max_header_size); - skb_put(skb, 2); - - frame = skb->data; - - /* Build frame */ - frame[0] = CMD_GET_MEDIA_CHAR; - frame[1] = 0x01; /* One parameter */ - - irlan_insert_string_param(skb, "MEDIA", "802.3"); - irlan_ctrl_data_request(self, skb); -} - -/* - * Function insert_byte_param (skb, param, value) - * - * Insert byte parameter into frame - * - */ -int irlan_insert_byte_param(struct sk_buff *skb, char *param, __u8 value) -{ - return __irlan_insert_param(skb, param, IRLAN_BYTE, value, 0, NULL, 0); -} - -int irlan_insert_short_param(struct sk_buff *skb, char *param, __u16 value) -{ - return __irlan_insert_param(skb, param, IRLAN_SHORT, 0, value, NULL, 0); -} - -/* - * Function insert_string (skb, param, value) - * - * Insert string parameter into frame - * - */ -int irlan_insert_string_param(struct sk_buff *skb, char *param, char *string) -{ - int string_len = strlen(string); - - return __irlan_insert_param(skb, param, IRLAN_ARRAY, 0, 0, string, - string_len); -} - -/* - * Function insert_array_param(skb, param, value, len_value) - * - * Insert array parameter into frame - * - */ -int irlan_insert_array_param(struct sk_buff *skb, char *name, __u8 *array, - __u16 array_len) -{ - return __irlan_insert_param(skb, name, IRLAN_ARRAY, 0, 0, array, - array_len); -} - -/* - * Function insert_param (skb, param, value, byte) - * - * Insert parameter at end of buffer, structure of a parameter is: - * - * ----------------------------------------------------------------------- - * | Name Length[1] | Param Name[1..255] | Val Length[2] | Value[0..1016]| - * ----------------------------------------------------------------------- - */ -static int __irlan_insert_param(struct sk_buff *skb, char *param, int type, - __u8 value_byte, __u16 value_short, - __u8 *value_array, __u16 value_len) -{ - __u8 *frame; - __u8 param_len; - __le16 tmp_le; /* Temporary value in little endian format */ - int n=0; - - if (skb == NULL) { - pr_debug("%s(), Got NULL skb\n", __func__); - return 0; - } - - param_len = strlen(param); - switch (type) { - case IRLAN_BYTE: - value_len = 1; - break; - case IRLAN_SHORT: - value_len = 2; - break; - case IRLAN_ARRAY: - IRDA_ASSERT(value_array != NULL, return 0;); - IRDA_ASSERT(value_len > 0, return 0;); - break; - default: - pr_debug("%s(), Unknown parameter type!\n", __func__); - return 0; - } - - /* Insert at end of sk-buffer */ - frame = skb_tail_pointer(skb); - - /* Make space for data */ - if (skb_tailroom(skb) < (param_len+value_len+3)) { - pr_debug("%s(), No more space at end of skb\n", __func__); - return 0; - } - skb_put(skb, param_len+value_len+3); - - /* Insert parameter length */ - frame[n++] = param_len; - - /* Insert parameter */ - memcpy(frame+n, param, param_len); n += param_len; - - /* Insert value length (2 byte little endian format, LSB first) */ - tmp_le = cpu_to_le16(value_len); - memcpy(frame+n, &tmp_le, 2); n += 2; /* To avoid alignment problems */ - - /* Insert value */ - switch (type) { - case IRLAN_BYTE: - frame[n++] = value_byte; - break; - case IRLAN_SHORT: - tmp_le = cpu_to_le16(value_short); - memcpy(frame+n, &tmp_le, 2); n += 2; - break; - case IRLAN_ARRAY: - memcpy(frame+n, value_array, value_len); n+=value_len; - break; - default: - break; - } - IRDA_ASSERT(n == (param_len+value_len+3), return 0;); - - return param_len+value_len+3; -} - -/* - * Function irlan_extract_param (buf, name, value, len) - * - * Extracts a single parameter name/value pair from buffer and updates - * the buffer pointer to point to the next name/value pair. - */ -int irlan_extract_param(__u8 *buf, char *name, char *value, __u16 *len) -{ - __u8 name_len; - __u16 val_len; - int n=0; - - /* get length of parameter name (1 byte) */ - name_len = buf[n++]; - - if (name_len > 254) { - pr_debug("%s(), name_len > 254\n", __func__); - return -RSP_INVALID_COMMAND_FORMAT; - } - - /* get parameter name */ - memcpy(name, buf+n, name_len); - name[name_len] = '\0'; - n+=name_len; - - /* - * Get length of parameter value (2 bytes in little endian - * format) - */ - memcpy(&val_len, buf+n, 2); /* To avoid alignment problems */ - le16_to_cpus(&val_len); n+=2; - - if (val_len >= 1016) { - pr_debug("%s(), parameter length to long\n", __func__); - return -RSP_INVALID_COMMAND_FORMAT; - } - *len = val_len; - - /* get parameter value */ - memcpy(value, buf+n, val_len); - value[val_len] = '\0'; - n+=val_len; - - pr_debug("Parameter: %s ", name); - pr_debug("Value: %s\n", value); - - return n; -} - -#ifdef CONFIG_PROC_FS - -/* - * Start of reading /proc entries. - * Return entry at pos, - * or start_token to indicate print header line - * or NULL if end of file - */ -static void *irlan_seq_start(struct seq_file *seq, loff_t *pos) -{ - rcu_read_lock(); - return seq_list_start_head(&irlans, *pos); -} - -/* Return entry after v, and increment pos */ -static void *irlan_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - return seq_list_next(v, &irlans, pos); -} - -/* End of reading /proc file */ -static void irlan_seq_stop(struct seq_file *seq, void *v) -{ - rcu_read_unlock(); -} - - -/* - * Show one entry in /proc file. - */ -static int irlan_seq_show(struct seq_file *seq, void *v) -{ - if (v == &irlans) - seq_puts(seq, "IrLAN instances:\n"); - else { - struct irlan_cb *self = list_entry(v, struct irlan_cb, dev_list); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - seq_printf(seq,"ifname: %s,\n", - self->dev->name); - seq_printf(seq,"client state: %s, ", - irlan_state[ self->client.state]); - seq_printf(seq,"provider state: %s,\n", - irlan_state[ self->provider.state]); - seq_printf(seq,"saddr: %#08x, ", - self->saddr); - seq_printf(seq,"daddr: %#08x\n", - self->daddr); - seq_printf(seq,"version: %d.%d,\n", - self->version[1], self->version[0]); - seq_printf(seq,"access type: %s\n", - irlan_access[self->client.access_type]); - seq_printf(seq,"media: %s\n", - irlan_media[self->media]); - - seq_printf(seq,"local filter:\n"); - seq_printf(seq,"remote filter: "); - irlan_print_filter(seq, self->client.filter_type); - seq_printf(seq,"tx busy: %s\n", - netif_queue_stopped(self->dev) ? "TRUE" : "FALSE"); - - seq_putc(seq,'\n'); - } - return 0; -} - -static const struct seq_operations irlan_seq_ops = { - .start = irlan_seq_start, - .next = irlan_seq_next, - .stop = irlan_seq_stop, - .show = irlan_seq_show, -}; - -static int irlan_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &irlan_seq_ops); -} -#endif - -MODULE_AUTHOR("Dag Brattli <dagb@cs.uit.no>"); -MODULE_DESCRIPTION("The Linux IrDA LAN protocol"); -MODULE_LICENSE("GPL"); - -module_param(eth, bool, 0); -MODULE_PARM_DESC(eth, "Name devices ethX (0) or irlanX (1)"); -module_param(access, int, 0); -MODULE_PARM_DESC(access, "Access type DIRECT=1, PEER=2, HOSTED=3"); - -module_init(irlan_init); -module_exit(irlan_cleanup); - diff --git a/net/irda/irlan/irlan_eth.c b/net/irda/irlan/irlan_eth.c deleted file mode 100644 index 3be852808a9d..000000000000 --- a/net/irda/irlan/irlan_eth.c +++ /dev/null @@ -1,340 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_eth.c - * Version: - * Description: - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Thu Oct 15 08:37:58 1998 - * Modified at: Tue Mar 21 09:06:41 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Sources: skeleton.c by Donald Becker <becker@CESDIS.gsfc.nasa.gov> - * slip.c by Laurence Culhane, <loz@holmes.demon.co.uk> - * Fred N. van Kempen, <waltje@uwalt.nl.mugnet.org> - * - * Copyright (c) 1998-2000 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/inetdevice.h> -#include <linux/if_arp.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <net/arp.h> - -#include <net/irda/irda.h> -#include <net/irda/irmod.h> -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_client.h> -#include <net/irda/irlan_event.h> -#include <net/irda/irlan_eth.h> - -static int irlan_eth_open(struct net_device *dev); -static int irlan_eth_close(struct net_device *dev); -static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, - struct net_device *dev); -static void irlan_eth_set_multicast_list(struct net_device *dev); - -static const struct net_device_ops irlan_eth_netdev_ops = { - .ndo_open = irlan_eth_open, - .ndo_stop = irlan_eth_close, - .ndo_start_xmit = irlan_eth_xmit, - .ndo_set_rx_mode = irlan_eth_set_multicast_list, - .ndo_validate_addr = eth_validate_addr, -}; - -/* - * Function irlan_eth_setup (dev) - * - * The network device initialization function. - * - */ -static void irlan_eth_setup(struct net_device *dev) -{ - ether_setup(dev); - - dev->netdev_ops = &irlan_eth_netdev_ops; - dev->needs_free_netdev = true; - dev->min_mtu = 0; - dev->max_mtu = ETH_MAX_MTU; - - /* - * Lets do all queueing in IrTTP instead of this device driver. - * Queueing here as well can introduce some strange latency - * problems, which we will avoid by setting the queue size to 0. - */ - /* - * The bugs in IrTTP and IrLAN that created this latency issue - * have now been fixed, and we can propagate flow control properly - * to the network layer. However, this requires a minimal queue of - * packets for the device. - * Without flow control, the Tx Queue is 14 (ttp) + 0 (dev) = 14 - * With flow control, the Tx Queue is 7 (ttp) + 4 (dev) = 11 - * See irlan_eth_flow_indication()... - * Note : this number was randomly selected and would need to - * be adjusted. - * Jean II */ - dev->tx_queue_len = 4; -} - -/* - * Function alloc_irlandev - * - * Allocate network device and control block - * - */ -struct net_device *alloc_irlandev(const char *name) -{ - return alloc_netdev(sizeof(struct irlan_cb), name, NET_NAME_UNKNOWN, - irlan_eth_setup); -} - -/* - * Function irlan_eth_open (dev) - * - * Network device has been opened by user - * - */ -static int irlan_eth_open(struct net_device *dev) -{ - struct irlan_cb *self = netdev_priv(dev); - - /* Ready to play! */ - netif_stop_queue(dev); /* Wait until data link is ready */ - - /* We are now open, so time to do some work */ - self->disconnect_reason = 0; - irlan_client_wakeup(self, self->saddr, self->daddr); - - /* Make sure we have a hardware address before we return, - so DHCP clients gets happy */ - return wait_event_interruptible(self->open_wait, - !self->tsap_data->connected); -} - -/* - * Function irlan_eth_close (dev) - * - * Stop the ether network device, his function will usually be called by - * ifconfig down. We should now disconnect the link, We start the - * close timer, so that the instance will be removed if we are unable - * to discover the remote device after the disconnect. - */ -static int irlan_eth_close(struct net_device *dev) -{ - struct irlan_cb *self = netdev_priv(dev); - - /* Stop device */ - netif_stop_queue(dev); - - irlan_close_data_channel(self); - irlan_close_tsaps(self); - - irlan_do_client_event(self, IRLAN_LMP_DISCONNECT, NULL); - irlan_do_provider_event(self, IRLAN_LMP_DISCONNECT, NULL); - - /* Remove frames queued on the control channel */ - skb_queue_purge(&self->client.txq); - - self->client.tx_busy = 0; - - return 0; -} - -/* - * Function irlan_eth_tx (skb) - * - * Transmits ethernet frames over IrDA link. - * - */ -static netdev_tx_t irlan_eth_xmit(struct sk_buff *skb, - struct net_device *dev) -{ - struct irlan_cb *self = netdev_priv(dev); - int ret; - unsigned int len; - - /* skb headroom large enough to contain all IrDA-headers? */ - if ((skb_headroom(skb) < self->max_header_size) || (skb_shared(skb))) { - struct sk_buff *new_skb = - skb_realloc_headroom(skb, self->max_header_size); - - /* We have to free the original skb anyway */ - dev_kfree_skb(skb); - - /* Did the realloc succeed? */ - if (new_skb == NULL) - return NETDEV_TX_OK; - - /* Use the new skb instead */ - skb = new_skb; - } - - netif_trans_update(dev); - - len = skb->len; - /* Now queue the packet in the transport layer */ - if (self->use_udata) - ret = irttp_udata_request(self->tsap_data, skb); - else - ret = irttp_data_request(self->tsap_data, skb); - - if (ret < 0) { - /* - * IrTTPs tx queue is full, so we just have to - * drop the frame! You might think that we should - * just return -1 and don't deallocate the frame, - * but that is dangerous since it's possible that - * we have replaced the original skb with a new - * one with larger headroom, and that would really - * confuse do_dev_queue_xmit() in dev.c! I have - * tried :-) DB - */ - /* irttp_data_request already free the packet */ - dev->stats.tx_dropped++; - } else { - dev->stats.tx_packets++; - dev->stats.tx_bytes += len; - } - - return NETDEV_TX_OK; -} - -/* - * Function irlan_eth_receive (handle, skb) - * - * This function gets the data that is received on the data channel - * - */ -int irlan_eth_receive(void *instance, void *sap, struct sk_buff *skb) -{ - struct irlan_cb *self = instance; - struct net_device *dev = self->dev; - - if (skb == NULL) { - dev->stats.rx_dropped++; - return 0; - } - if (skb->len < ETH_HLEN) { - pr_debug("%s() : IrLAN frame too short (%d)\n", - __func__, skb->len); - dev->stats.rx_dropped++; - dev_kfree_skb(skb); - return 0; - } - - /* - * Adopt this frame! Important to set all these fields since they - * might have been previously set by the low level IrDA network - * device driver - */ - skb->protocol = eth_type_trans(skb, dev); /* Remove eth header */ - - dev->stats.rx_packets++; - dev->stats.rx_bytes += skb->len; - - netif_rx(skb); /* Eat it! */ - - return 0; -} - -/* - * Function irlan_eth_flow (status) - * - * Do flow control between IP/Ethernet and IrLAN/IrTTP. This is done by - * controlling the queue stop/start. - * - * The IrDA link layer has the advantage to have flow control, and - * IrTTP now properly handles that. Flow controlling the higher layers - * prevent us to drop Tx packets in here (up to 15% for a TCP socket, - * more for UDP socket). - * Also, this allow us to reduce the overall transmit queue, which means - * less latency in case of mixed traffic. - * Jean II - */ -void irlan_eth_flow_indication(void *instance, void *sap, LOCAL_FLOW flow) -{ - struct irlan_cb *self; - struct net_device *dev; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - dev = self->dev; - - IRDA_ASSERT(dev != NULL, return;); - - pr_debug("%s() : flow %s ; running %d\n", __func__, - flow == FLOW_STOP ? "FLOW_STOP" : "FLOW_START", - netif_running(dev)); - - switch (flow) { - case FLOW_STOP: - /* IrTTP is full, stop higher layers */ - netif_stop_queue(dev); - break; - case FLOW_START: - default: - /* Tell upper layers that its time to transmit frames again */ - /* Schedule network layer */ - netif_wake_queue(dev); - break; - } -} - -/* - * Function set_multicast_list (dev) - * - * Configure the filtering of the device - * - */ -#define HW_MAX_ADDRS 4 /* Must query to get it! */ -static void irlan_eth_set_multicast_list(struct net_device *dev) -{ - struct irlan_cb *self = netdev_priv(dev); - - /* Check if data channel has been connected yet */ - if (self->client.state != IRLAN_DATA) { - pr_debug("%s(), delaying!\n", __func__); - return; - } - - if (dev->flags & IFF_PROMISC) { - /* Enable promiscuous mode */ - net_warn_ratelimited("Promiscuous mode not implemented by IrLAN!\n"); - } else if ((dev->flags & IFF_ALLMULTI) || - netdev_mc_count(dev) > HW_MAX_ADDRS) { - /* Disable promiscuous mode, use normal mode. */ - pr_debug("%s(), Setting multicast filter\n", __func__); - /* hardware_set_filter(NULL); */ - - irlan_set_multicast_filter(self, TRUE); - } else if (!netdev_mc_empty(dev)) { - pr_debug("%s(), Setting multicast filter\n", __func__); - /* Walk the address list, and load the filter */ - /* hardware_set_filter(dev->mc_list); */ - - irlan_set_multicast_filter(self, TRUE); - } else { - pr_debug("%s(), Clearing multicast filter\n", __func__); - irlan_set_multicast_filter(self, FALSE); - } - - if (dev->flags & IFF_BROADCAST) - irlan_set_broadcast_filter(self, TRUE); - else - irlan_set_broadcast_filter(self, FALSE); -} diff --git a/net/irda/irlan/irlan_event.c b/net/irda/irlan/irlan_event.c deleted file mode 100644 index 9a1cc11c16f6..000000000000 --- a/net/irda/irlan/irlan_event.c +++ /dev/null @@ -1,60 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_event.c - * Version: - * Description: - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Oct 20 09:10:16 1998 - * Modified at: Sat Oct 30 12:59:01 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <net/irda/irlan_event.h> - -const char * const irlan_state[] = { - "IRLAN_IDLE", - "IRLAN_QUERY", - "IRLAN_CONN", - "IRLAN_INFO", - "IRLAN_MEDIA", - "IRLAN_OPEN", - "IRLAN_WAIT", - "IRLAN_ARB", - "IRLAN_DATA", - "IRLAN_CLOSE", - "IRLAN_SYNC", -}; - -void irlan_next_client_state(struct irlan_cb *self, IRLAN_STATE state) -{ - pr_debug("%s(), %s\n", __func__ , irlan_state[state]); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - self->client.state = state; -} - -void irlan_next_provider_state(struct irlan_cb *self, IRLAN_STATE state) -{ - pr_debug("%s(), %s\n", __func__ , irlan_state[state]); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - self->provider.state = state; -} - diff --git a/net/irda/irlan/irlan_filter.c b/net/irda/irlan/irlan_filter.c deleted file mode 100644 index e755e90b2f26..000000000000 --- a/net/irda/irlan/irlan_filter.c +++ /dev/null @@ -1,240 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_filter.c - * Version: - * Description: - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Fri Jan 29 11:16:38 1999 - * Modified at: Sat Oct 30 12:58:45 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> -#include <linux/random.h> -#include <linux/seq_file.h> - -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_filter.h> - -/* - * Function irlan_filter_request (self, skb) - * - * Handle filter request from client peer device - * - */ -void irlan_filter_request(struct irlan_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - if ((self->provider.filter_type == IRLAN_DIRECTED) && - (self->provider.filter_operation == DYNAMIC)) - { - pr_debug("Giving peer a dynamic Ethernet address\n"); - self->provider.mac_address[0] = 0x40; - self->provider.mac_address[1] = 0x00; - self->provider.mac_address[2] = 0x00; - self->provider.mac_address[3] = 0x00; - - /* Use arbitration value to generate MAC address */ - if (self->provider.access_type == ACCESS_PEER) { - self->provider.mac_address[4] = - self->provider.send_arb_val & 0xff; - self->provider.mac_address[5] = - (self->provider.send_arb_val >> 8) & 0xff; - } else { - /* Just generate something for now */ - get_random_bytes(self->provider.mac_address+4, 1); - get_random_bytes(self->provider.mac_address+5, 1); - } - - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x03; - irlan_insert_string_param(skb, "FILTER_MODE", "NONE"); - irlan_insert_short_param(skb, "MAX_ENTRY", 0x0001); - irlan_insert_array_param(skb, "FILTER_ENTRY", - self->provider.mac_address, 6); - return; - } - - if ((self->provider.filter_type == IRLAN_DIRECTED) && - (self->provider.filter_mode == FILTER)) - { - pr_debug("Directed filter on\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - if ((self->provider.filter_type == IRLAN_DIRECTED) && - (self->provider.filter_mode == NONE)) - { - pr_debug("Directed filter off\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - - if ((self->provider.filter_type == IRLAN_BROADCAST) && - (self->provider.filter_mode == FILTER)) - { - pr_debug("Broadcast filter on\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - if ((self->provider.filter_type == IRLAN_BROADCAST) && - (self->provider.filter_mode == NONE)) - { - pr_debug("Broadcast filter off\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - if ((self->provider.filter_type == IRLAN_MULTICAST) && - (self->provider.filter_mode == FILTER)) - { - pr_debug("Multicast filter on\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - if ((self->provider.filter_type == IRLAN_MULTICAST) && - (self->provider.filter_mode == NONE)) - { - pr_debug("Multicast filter off\n"); - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x00; - return; - } - if ((self->provider.filter_type == IRLAN_MULTICAST) && - (self->provider.filter_operation == GET)) - { - pr_debug("Multicast filter get\n"); - skb->data[0] = 0x00; /* Success? */ - skb->data[1] = 0x02; - irlan_insert_string_param(skb, "FILTER_MODE", "NONE"); - irlan_insert_short_param(skb, "MAX_ENTRY", 16); - return; - } - skb->data[0] = 0x00; /* Command not supported */ - skb->data[1] = 0x00; - - pr_debug("Not implemented!\n"); -} - -/* - * Function check_request_param (self, param, value) - * - * Check parameters in request from peer device - * - */ -void irlan_check_command_param(struct irlan_cb *self, char *param, char *value) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - pr_debug("%s, %s\n", param, value); - - /* - * This is experimental!! DB. - */ - if (strcmp(param, "MODE") == 0) { - self->use_udata = TRUE; - return; - } - - /* - * FILTER_TYPE - */ - if (strcmp(param, "FILTER_TYPE") == 0) { - if (strcmp(value, "DIRECTED") == 0) { - self->provider.filter_type = IRLAN_DIRECTED; - return; - } - if (strcmp(value, "MULTICAST") == 0) { - self->provider.filter_type = IRLAN_MULTICAST; - return; - } - if (strcmp(value, "BROADCAST") == 0) { - self->provider.filter_type = IRLAN_BROADCAST; - return; - } - } - /* - * FILTER_MODE - */ - if (strcmp(param, "FILTER_MODE") == 0) { - if (strcmp(value, "ALL") == 0) { - self->provider.filter_mode = ALL; - return; - } - if (strcmp(value, "FILTER") == 0) { - self->provider.filter_mode = FILTER; - return; - } - if (strcmp(value, "NONE") == 0) { - self->provider.filter_mode = FILTER; - return; - } - } - /* - * FILTER_OPERATION - */ - if (strcmp(param, "FILTER_OPERATION") == 0) { - if (strcmp(value, "DYNAMIC") == 0) { - self->provider.filter_operation = DYNAMIC; - return; - } - if (strcmp(value, "GET") == 0) { - self->provider.filter_operation = GET; - return; - } - } -} - -/* - * Function irlan_print_filter (filter_type, buf) - * - * Print status of filter. Used by /proc file system - * - */ -#ifdef CONFIG_PROC_FS -#define MASK2STR(m,s) { .mask = m, .str = s } - -void irlan_print_filter(struct seq_file *seq, int filter_type) -{ - static struct { - int mask; - const char *str; - } filter_mask2str[] = { - MASK2STR(IRLAN_DIRECTED, "DIRECTED"), - MASK2STR(IRLAN_FUNCTIONAL, "FUNCTIONAL"), - MASK2STR(IRLAN_GROUP, "GROUP"), - MASK2STR(IRLAN_MAC_FRAME, "MAC_FRAME"), - MASK2STR(IRLAN_MULTICAST, "MULTICAST"), - MASK2STR(IRLAN_BROADCAST, "BROADCAST"), - MASK2STR(IRLAN_IPX_SOCKET, "IPX_SOCKET"), - MASK2STR(0, NULL) - }, *p; - - for (p = filter_mask2str; p->str; p++) { - if (filter_type & p->mask) - seq_printf(seq, "%s ", p->str); - } - seq_putc(seq, '\n'); -} -#undef MASK2STR -#endif diff --git a/net/irda/irlan/irlan_provider.c b/net/irda/irlan/irlan_provider.c deleted file mode 100644 index 15c292cf2644..000000000000 --- a/net/irda/irlan/irlan_provider.c +++ /dev/null @@ -1,408 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_provider.c - * Version: 0.9 - * Description: IrDA LAN Access Protocol Implementation - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:37 1997 - * Modified at: Sat Oct 30 12:52:10 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Sources: skeleton.c by Donald Becker <becker@CESDIS.gsfc.nasa.gov> - * slip.c by Laurence Culhane, <loz@holmes.demon.co.uk> - * Fred N. van Kempen, <waltje@uwalt.nl.mugnet.org> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/errno.h> -#include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/init.h> -#include <linux/random.h> -#include <linux/bitops.h> -#include <linux/slab.h> - -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/irttp.h> -#include <net/irda/irlmp.h> -#include <net/irda/irias_object.h> -#include <net/irda/iriap.h> -#include <net/irda/timer.h> - -#include <net/irda/irlan_common.h> -#include <net/irda/irlan_eth.h> -#include <net/irda/irlan_event.h> -#include <net/irda/irlan_provider.h> -#include <net/irda/irlan_filter.h> -#include <net/irda/irlan_client.h> - -static void irlan_provider_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb); - -/* - * Function irlan_provider_control_data_indication (handle, skb) - * - * This function gets the data that is received on the control channel - * - */ -static int irlan_provider_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct irlan_cb *self; - __u8 code; - - self = instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - IRDA_ASSERT(skb != NULL, return -1;); - - code = skb->data[0]; - switch(code) { - case CMD_GET_PROVIDER_INFO: - pr_debug("Got GET_PROVIDER_INFO command!\n"); - irlan_do_provider_event(self, IRLAN_GET_INFO_CMD, skb); - break; - - case CMD_GET_MEDIA_CHAR: - pr_debug("Got GET_MEDIA_CHAR command!\n"); - irlan_do_provider_event(self, IRLAN_GET_MEDIA_CMD, skb); - break; - case CMD_OPEN_DATA_CHANNEL: - pr_debug("Got OPEN_DATA_CHANNEL command!\n"); - irlan_do_provider_event(self, IRLAN_OPEN_DATA_CMD, skb); - break; - case CMD_FILTER_OPERATION: - pr_debug("Got FILTER_OPERATION command!\n"); - irlan_do_provider_event(self, IRLAN_FILTER_CONFIG_CMD, skb); - break; - case CMD_RECONNECT_DATA_CHAN: - pr_debug("%s(), Got RECONNECT_DATA_CHAN command\n", __func__); - pr_debug("%s(), NOT IMPLEMENTED\n", __func__); - break; - case CMD_CLOSE_DATA_CHAN: - pr_debug("Got CLOSE_DATA_CHAN command!\n"); - pr_debug("%s(), NOT IMPLEMENTED\n", __func__); - break; - default: - pr_debug("%s(), Unknown command!\n", __func__); - break; - } - return 0; -} - -/* - * Function irlan_provider_connect_indication (handle, skb, priv) - * - * Got connection from peer IrLAN client - * - */ -static void irlan_provider_connect_indication(void *instance, void *sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - struct irlan_cb *self; - struct tsap_cb *tsap; - - self = instance; - tsap = sap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - IRDA_ASSERT(tsap == self->provider.tsap_ctrl,return;); - IRDA_ASSERT(self->provider.state == IRLAN_IDLE, return;); - - self->provider.max_sdu_size = max_sdu_size; - self->provider.max_header_size = max_header_size; - - irlan_do_provider_event(self, IRLAN_CONNECT_INDICATION, NULL); - - /* - * If we are in peer mode, the client may not have got the discovery - * indication it needs to make progress. If the client is still in - * IDLE state, we must kick it. - */ - if ((self->provider.access_type == ACCESS_PEER) && - (self->client.state == IRLAN_IDLE)) - { - irlan_client_wakeup(self, self->saddr, self->daddr); - } -} - -/* - * Function irlan_provider_connect_response (handle) - * - * Accept incoming connection - * - */ -void irlan_provider_connect_response(struct irlan_cb *self, - struct tsap_cb *tsap) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - /* Just accept */ - irttp_connect_response(tsap, IRLAN_MTU, NULL); -} - -static void irlan_provider_disconnect_indication(void *instance, void *sap, - LM_REASON reason, - struct sk_buff *userdata) -{ - struct irlan_cb *self; - struct tsap_cb *tsap; - - pr_debug("%s(), reason=%d\n", __func__ , reason); - - self = instance; - tsap = sap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - IRDA_ASSERT(tsap != NULL, return;); - IRDA_ASSERT(tsap->magic == TTP_TSAP_MAGIC, return;); - - IRDA_ASSERT(tsap == self->provider.tsap_ctrl, return;); - - irlan_do_provider_event(self, IRLAN_LMP_DISCONNECT, NULL); -} - -/* - * Function irlan_parse_open_data_cmd (self, skb) - * - * - * - */ -int irlan_parse_open_data_cmd(struct irlan_cb *self, struct sk_buff *skb) -{ - int ret; - - ret = irlan_provider_parse_command(self, CMD_OPEN_DATA_CHANNEL, skb); - - /* Open data channel */ - irlan_open_data_tsap(self); - - return ret; -} - -/* - * Function parse_command (skb) - * - * Extract all parameters from received buffer, then feed them to - * check_params for parsing - * - */ -int irlan_provider_parse_command(struct irlan_cb *self, int cmd, - struct sk_buff *skb) -{ - __u8 *frame; - __u8 *ptr; - int count; - __u16 val_len; - int i; - char *name; - char *value; - int ret = RSP_SUCCESS; - - IRDA_ASSERT(skb != NULL, return -RSP_PROTOCOL_ERROR;); - - pr_debug("%s(), skb->len=%d\n", __func__ , (int)skb->len); - - IRDA_ASSERT(self != NULL, return -RSP_PROTOCOL_ERROR;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -RSP_PROTOCOL_ERROR;); - - if (!skb) - return -RSP_PROTOCOL_ERROR; - - frame = skb->data; - - name = kmalloc(255, GFP_ATOMIC); - if (!name) - return -RSP_INSUFFICIENT_RESOURCES; - value = kmalloc(1016, GFP_ATOMIC); - if (!value) { - kfree(name); - return -RSP_INSUFFICIENT_RESOURCES; - } - - /* How many parameters? */ - count = frame[1]; - - pr_debug("Got %d parameters\n", count); - - ptr = frame+2; - - /* For all parameters */ - for (i=0; i<count;i++) { - ret = irlan_extract_param(ptr, name, value, &val_len); - if (ret < 0) { - pr_debug("%s(), IrLAN, Error!\n", __func__); - break; - } - ptr+=ret; - ret = RSP_SUCCESS; - irlan_check_command_param(self, name, value); - } - /* Cleanup */ - kfree(name); - kfree(value); - - return ret; -} - -/* - * Function irlan_provider_send_reply (self, info) - * - * Send reply to query to peer IrLAN layer - * - */ -void irlan_provider_send_reply(struct irlan_cb *self, int command, - int ret_code) -{ - struct sk_buff *skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;); - - skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER + - /* Bigger param length comes from CMD_GET_MEDIA_CHAR */ - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "BROADCAST") + - IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "MULTICAST") + - IRLAN_STRING_PARAMETER_LEN("ACCESS_TYPE", "HOSTED"), - GFP_ATOMIC); - - if (!skb) - return; - - /* Reserve space for TTP, LMP, and LAP header */ - skb_reserve(skb, self->provider.max_header_size); - skb_put(skb, 2); - - switch (command) { - case CMD_GET_PROVIDER_INFO: - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x02; /* 2 parameters */ - switch (self->media) { - case MEDIA_802_3: - irlan_insert_string_param(skb, "MEDIA", "802.3"); - break; - case MEDIA_802_5: - irlan_insert_string_param(skb, "MEDIA", "802.5"); - break; - default: - pr_debug("%s(), unknown media type!\n", __func__); - break; - } - irlan_insert_short_param(skb, "IRLAN_VER", 0x0101); - break; - - case CMD_GET_MEDIA_CHAR: - skb->data[0] = 0x00; /* Success */ - skb->data[1] = 0x05; /* 5 parameters */ - irlan_insert_string_param(skb, "FILTER_TYPE", "DIRECTED"); - irlan_insert_string_param(skb, "FILTER_TYPE", "BROADCAST"); - irlan_insert_string_param(skb, "FILTER_TYPE", "MULTICAST"); - - switch (self->provider.access_type) { - case ACCESS_DIRECT: - irlan_insert_string_param(skb, "ACCESS_TYPE", "DIRECT"); - break; - case ACCESS_PEER: - irlan_insert_string_param(skb, "ACCESS_TYPE", "PEER"); - break; - case ACCESS_HOSTED: - irlan_insert_string_param(skb, "ACCESS_TYPE", "HOSTED"); - break; - default: - pr_debug("%s(), Unknown access type\n", __func__); - break; - } - irlan_insert_short_param(skb, "MAX_FRAME", 0x05ee); - break; - case CMD_OPEN_DATA_CHANNEL: - skb->data[0] = 0x00; /* Success */ - if (self->provider.send_arb_val) { - skb->data[1] = 0x03; /* 3 parameters */ - irlan_insert_short_param(skb, "CON_ARB", - self->provider.send_arb_val); - } else - skb->data[1] = 0x02; /* 2 parameters */ - irlan_insert_byte_param(skb, "DATA_CHAN", self->stsap_sel_data); - irlan_insert_string_param(skb, "RECONNECT_KEY", "LINUX RULES!"); - break; - case CMD_FILTER_OPERATION: - irlan_filter_request(self, skb); - break; - default: - pr_debug("%s(), Unknown command!\n", __func__); - break; - } - - irttp_data_request(self->provider.tsap_ctrl, skb); -} - -/* - * Function irlan_provider_register(void) - * - * Register provider support so we can accept incoming connections. - * - */ -int irlan_provider_open_ctrl_tsap(struct irlan_cb *self) -{ - struct tsap_cb *tsap; - notify_t notify; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - /* Check if already open */ - if (self->provider.tsap_ctrl) - return -1; - - /* - * First register well known control TSAP - */ - irda_notify_init(¬ify); - notify.data_indication = irlan_provider_data_indication; - notify.connect_indication = irlan_provider_connect_indication; - notify.disconnect_indication = irlan_provider_disconnect_indication; - notify.instance = self; - strlcpy(notify.name, "IrLAN ctrl (p)", sizeof(notify.name)); - - tsap = irttp_open_tsap(LSAP_ANY, 1, ¬ify); - if (!tsap) { - pr_debug("%s(), Got no tsap!\n", __func__); - return -1; - } - self->provider.tsap_ctrl = tsap; - - /* Register with LM-IAS */ - irlan_ias_register(self, tsap->stsap_sel); - - return 0; -} - diff --git a/net/irda/irlan/irlan_provider_event.c b/net/irda/irlan/irlan_provider_event.c deleted file mode 100644 index 9c4f7f51d6b5..000000000000 --- a/net/irda/irlan/irlan_provider_event.c +++ /dev/null @@ -1,233 +0,0 @@ -/********************************************************************* - * - * Filename: irlan_provider_event.c - * Version: 0.9 - * Description: IrLAN provider state machine) - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:37 1997 - * Modified at: Sat Oct 30 12:52:41 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <net/irda/irda.h> -#include <net/irda/iriap.h> -#include <net/irda/irlmp.h> -#include <net/irda/irttp.h> - -#include <net/irda/irlan_provider.h> -#include <net/irda/irlan_event.h> - -static int irlan_provider_state_idle(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_provider_state_info(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_provider_state_open(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); -static int irlan_provider_state_data(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb); - -static int (*state[])(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) = -{ - irlan_provider_state_idle, - NULL, /* Query */ - NULL, /* Info */ - irlan_provider_state_info, - NULL, /* Media */ - irlan_provider_state_open, - NULL, /* Wait */ - NULL, /* Arb */ - irlan_provider_state_data, - NULL, /* Close */ - NULL, /* Sync */ -}; - -void irlan_do_provider_event(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(*state[ self->provider.state] != NULL, return;); - - (*state[self->provider.state]) (self, event, skb); -} - -/* - * Function irlan_provider_state_idle (event, skb, info) - * - * IDLE, We are waiting for an indication that there is a provider - * available. - */ -static int irlan_provider_state_idle(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_CONNECT_INDICATION: - irlan_provider_connect_response( self, self->provider.tsap_ctrl); - irlan_next_provider_state( self, IRLAN_INFO); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_provider_state_info (self, event, skb, info) - * - * INFO, We have issued a GetInfo command and is awaiting a reply. - */ -static int irlan_provider_state_info(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_GET_INFO_CMD: - /* Be sure to use 802.3 in case of peer mode */ - if (self->provider.access_type == ACCESS_PEER) { - self->media = MEDIA_802_3; - - /* Check if client has started yet */ - if (self->client.state == IRLAN_IDLE) { - /* This should get the client going */ - irlmp_discovery_request(8); - } - } - - irlan_provider_send_reply(self, CMD_GET_PROVIDER_INFO, - RSP_SUCCESS); - /* Keep state */ - break; - case IRLAN_GET_MEDIA_CMD: - irlan_provider_send_reply(self, CMD_GET_MEDIA_CHAR, - RSP_SUCCESS); - /* Keep state */ - break; - case IRLAN_OPEN_DATA_CMD: - ret = irlan_parse_open_data_cmd(self, skb); - if (self->provider.access_type == ACCESS_PEER) { - /* FIXME: make use of random functions! */ - self->provider.send_arb_val = (jiffies & 0xffff); - } - irlan_provider_send_reply(self, CMD_OPEN_DATA_CHANNEL, ret); - - if (ret == RSP_SUCCESS) { - irlan_next_provider_state(self, IRLAN_OPEN); - - /* Signal client that we are now open */ - irlan_do_client_event(self, IRLAN_PROVIDER_SIGNAL, NULL); - } - break; - case IRLAN_LMP_DISCONNECT: /* FALLTHROUGH */ - case IRLAN_LAP_DISCONNECT: - irlan_next_provider_state(self, IRLAN_IDLE); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_provider_state_open (self, event, skb, info) - * - * OPEN, The client has issued a OpenData command and is awaiting a - * reply - * - */ -static int irlan_provider_state_open(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - - switch(event) { - case IRLAN_FILTER_CONFIG_CMD: - irlan_provider_parse_command(self, CMD_FILTER_OPERATION, skb); - irlan_provider_send_reply(self, CMD_FILTER_OPERATION, - RSP_SUCCESS); - /* Keep state */ - break; - case IRLAN_DATA_CONNECT_INDICATION: - irlan_next_provider_state(self, IRLAN_DATA); - irlan_provider_connect_response(self, self->tsap_data); - break; - case IRLAN_LMP_DISCONNECT: /* FALLTHROUGH */ - case IRLAN_LAP_DISCONNECT: - irlan_next_provider_state(self, IRLAN_IDLE); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irlan_provider_state_data (self, event, skb, info) - * - * DATA, The data channel is connected, allowing data transfers between - * the local and remote machines. - * - */ -static int irlan_provider_state_data(struct irlan_cb *self, IRLAN_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == IRLAN_MAGIC, return -1;); - - switch(event) { - case IRLAN_FILTER_CONFIG_CMD: - irlan_provider_parse_command(self, CMD_FILTER_OPERATION, skb); - irlan_provider_send_reply(self, CMD_FILTER_OPERATION, - RSP_SUCCESS); - break; - case IRLAN_LMP_DISCONNECT: /* FALLTHROUGH */ - case IRLAN_LAP_DISCONNECT: - irlan_next_provider_state(self, IRLAN_IDLE); - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__ , event); - break; - } - if (skb) - dev_kfree_skb(skb); - - return 0; -} - - - - - - - - - - diff --git a/net/irda/irlap.c b/net/irda/irlap.c deleted file mode 100644 index 1cde711bcab5..000000000000 --- a/net/irda/irlap.c +++ /dev/null @@ -1,1207 +0,0 @@ -/********************************************************************* - * - * Filename: irlap.c - * Version: 1.0 - * Description: IrLAP implementation for Linux - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Aug 4 20:40:53 1997 - * Modified at: Tue Dec 14 09:26:44 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/skbuff.h> -#include <linux/delay.h> -#include <linux/proc_fs.h> -#include <linux/init.h> -#include <linux/random.h> -#include <linux/module.h> -#include <linux/seq_file.h> - -#include <net/irda/irda.h> -#include <net/irda/irda_device.h> -#include <net/irda/irqueue.h> -#include <net/irda/irlmp.h> -#include <net/irda/irlmp_frame.h> -#include <net/irda/irlap_frame.h> -#include <net/irda/irlap.h> -#include <net/irda/timer.h> -#include <net/irda/qos.h> - -static hashbin_t *irlap = NULL; -int sysctl_slot_timeout = SLOT_TIMEOUT * 1000 / HZ; - -/* This is the delay of missed pf period before generating an event - * to the application. The spec mandate 3 seconds, but in some cases - * it's way too long. - Jean II */ -int sysctl_warn_noreply_time = 3; - -extern void irlap_queue_xmit(struct irlap_cb *self, struct sk_buff *skb); -static void __irlap_close(struct irlap_cb *self); -static void irlap_init_qos_capabilities(struct irlap_cb *self, - struct qos_info *qos_user); - -static const char *const lap_reasons[] __maybe_unused = { - "ERROR, NOT USED", - "LAP_DISC_INDICATION", - "LAP_NO_RESPONSE", - "LAP_RESET_INDICATION", - "LAP_FOUND_NONE", - "LAP_MEDIA_BUSY", - "LAP_PRIMARY_CONFLICT", - "ERROR, NOT USED", -}; - -int __init irlap_init(void) -{ - /* Check if the compiler did its job properly. - * May happen on some ARM configuration, check with Russell King. */ - IRDA_ASSERT(sizeof(struct xid_frame) == 14, ;); - IRDA_ASSERT(sizeof(struct test_frame) == 10, ;); - IRDA_ASSERT(sizeof(struct ua_frame) == 10, ;); - IRDA_ASSERT(sizeof(struct snrm_frame) == 11, ;); - - /* Allocate master array */ - irlap = hashbin_new(HB_LOCK); - if (irlap == NULL) { - net_err_ratelimited("%s: can't allocate irlap hashbin!\n", - __func__); - return -ENOMEM; - } - - return 0; -} - -void irlap_cleanup(void) -{ - IRDA_ASSERT(irlap != NULL, return;); - - hashbin_delete(irlap, (FREE_FUNC) __irlap_close); -} - -/* - * Function irlap_open (driver) - * - * Initialize IrLAP layer - * - */ -struct irlap_cb *irlap_open(struct net_device *dev, struct qos_info *qos, - const char *hw_name) -{ - struct irlap_cb *self; - - /* Initialize the irlap structure. */ - self = kzalloc(sizeof(struct irlap_cb), GFP_KERNEL); - if (self == NULL) - return NULL; - - self->magic = LAP_MAGIC; - - /* Make a binding between the layers */ - self->netdev = dev; - self->qos_dev = qos; - /* Copy hardware name */ - if(hw_name != NULL) { - strlcpy(self->hw_name, hw_name, sizeof(self->hw_name)); - } else { - self->hw_name[0] = '\0'; - } - - /* FIXME: should we get our own field? */ - dev->atalk_ptr = self; - - self->state = LAP_OFFLINE; - - /* Initialize transmit queue */ - skb_queue_head_init(&self->txq); - skb_queue_head_init(&self->txq_ultra); - skb_queue_head_init(&self->wx_list); - - /* My unique IrLAP device address! */ - /* We don't want the broadcast address, neither the NULL address - * (most often used to signify "invalid"), and we don't want an - * address already in use (otherwise connect won't be able - * to select the proper link). - Jean II */ - do { - get_random_bytes(&self->saddr, sizeof(self->saddr)); - } while ((self->saddr == 0x0) || (self->saddr == BROADCAST) || - (hashbin_lock_find(irlap, self->saddr, NULL)) ); - /* Copy to the driver */ - memcpy(dev->dev_addr, &self->saddr, 4); - - init_timer(&self->slot_timer); - init_timer(&self->query_timer); - init_timer(&self->discovery_timer); - init_timer(&self->final_timer); - init_timer(&self->poll_timer); - init_timer(&self->wd_timer); - init_timer(&self->backoff_timer); - init_timer(&self->media_busy_timer); - - irlap_apply_default_connection_parameters(self); - - self->N3 = 3; /* # connections attempts to try before giving up */ - - self->state = LAP_NDM; - - hashbin_insert(irlap, (irda_queue_t *) self, self->saddr, NULL); - - irlmp_register_link(self, self->saddr, &self->notify); - - return self; -} -EXPORT_SYMBOL(irlap_open); - -/* - * Function __irlap_close (self) - * - * Remove IrLAP and all allocated memory. Stop any pending timers. - * - */ -static void __irlap_close(struct irlap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Stop timers */ - del_timer(&self->slot_timer); - del_timer(&self->query_timer); - del_timer(&self->discovery_timer); - del_timer(&self->final_timer); - del_timer(&self->poll_timer); - del_timer(&self->wd_timer); - del_timer(&self->backoff_timer); - del_timer(&self->media_busy_timer); - - irlap_flush_all_queues(self); - - self->magic = 0; - - kfree(self); -} - -/* - * Function irlap_close (self) - * - * Remove IrLAP instance - * - */ -void irlap_close(struct irlap_cb *self) -{ - struct irlap_cb *lap; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* We used to send a LAP_DISC_INDICATION here, but this was - * racy. This has been move within irlmp_unregister_link() - * itself. Jean II */ - - /* Kill the LAP and all LSAPs on top of it */ - irlmp_unregister_link(self->saddr); - self->notify.instance = NULL; - - /* Be sure that we manage to remove ourself from the hash */ - lap = hashbin_remove(irlap, self->saddr, NULL); - if (!lap) { - pr_debug("%s(), Didn't find myself!\n", __func__); - return; - } - __irlap_close(lap); -} -EXPORT_SYMBOL(irlap_close); - -/* - * Function irlap_connect_indication (self, skb) - * - * Another device is attempting to make a connection - * - */ -void irlap_connect_indication(struct irlap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_init_qos_capabilities(self, NULL); /* No user QoS! */ - - irlmp_link_connect_indication(self->notify.instance, self->saddr, - self->daddr, &self->qos_tx, skb); -} - -/* - * Function irlap_connect_response (self, skb) - * - * Service user has accepted incoming connection - * - */ -void irlap_connect_response(struct irlap_cb *self, struct sk_buff *userdata) -{ - irlap_do_event(self, CONNECT_RESPONSE, userdata, NULL); -} - -/* - * Function irlap_connect_request (self, daddr, qos_user, sniff) - * - * Request connection with another device, sniffing is not implemented - * yet. - * - */ -void irlap_connect_request(struct irlap_cb *self, __u32 daddr, - struct qos_info *qos_user, int sniff) -{ - pr_debug("%s(), daddr=0x%08x\n", __func__, daddr); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - self->daddr = daddr; - - /* - * If the service user specifies QoS values for this connection, - * then use them - */ - irlap_init_qos_capabilities(self, qos_user); - - if ((self->state == LAP_NDM) && !self->media_busy) - irlap_do_event(self, CONNECT_REQUEST, NULL, NULL); - else - self->connect_pending = TRUE; -} - -/* - * Function irlap_connect_confirm (self, skb) - * - * Connection request has been accepted - * - */ -void irlap_connect_confirm(struct irlap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlmp_link_connect_confirm(self->notify.instance, &self->qos_tx, skb); -} - -/* - * Function irlap_data_indication (self, skb) - * - * Received data frames from IR-port, so we just pass them up to - * IrLMP for further processing - * - */ -void irlap_data_indication(struct irlap_cb *self, struct sk_buff *skb, - int unreliable) -{ - /* Hide LAP header from IrLMP layer */ - skb_pull(skb, LAP_ADDR_HEADER+LAP_CTRL_HEADER); - - irlmp_link_data_indication(self->notify.instance, skb, unreliable); -} - - -/* - * Function irlap_data_request (self, skb) - * - * Queue data for transmission, must wait until XMIT state - * - */ -void irlap_data_request(struct irlap_cb *self, struct sk_buff *skb, - int unreliable) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - IRDA_ASSERT(skb_headroom(skb) >= (LAP_ADDR_HEADER+LAP_CTRL_HEADER), - return;); - skb_push(skb, LAP_ADDR_HEADER+LAP_CTRL_HEADER); - - /* - * Must set frame format now so that the rest of the code knows - * if its dealing with an I or an UI frame - */ - if (unreliable) - skb->data[1] = UI_FRAME; - else - skb->data[1] = I_FRAME; - - /* Don't forget to refcount it - see irlmp_connect_request(). */ - skb_get(skb); - - /* Add at the end of the queue (keep ordering) - Jean II */ - skb_queue_tail(&self->txq, skb); - - /* - * Send event if this frame only if we are in the right state - * FIXME: udata should be sent first! (skb_queue_head?) - */ - if ((self->state == LAP_XMIT_P) || (self->state == LAP_XMIT_S)) { - /* If we are not already processing the Tx queue, trigger - * transmission immediately - Jean II */ - if((skb_queue_len(&self->txq) <= 1) && (!self->local_busy)) - irlap_do_event(self, DATA_REQUEST, skb, NULL); - /* Otherwise, the packets will be sent normally at the - * next pf-poll - Jean II */ - } -} - -/* - * Function irlap_unitdata_request (self, skb) - * - * Send Ultra data. This is data that must be sent outside any connection - * - */ -#ifdef CONFIG_IRDA_ULTRA -void irlap_unitdata_request(struct irlap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - IRDA_ASSERT(skb_headroom(skb) >= (LAP_ADDR_HEADER+LAP_CTRL_HEADER), - return;); - skb_push(skb, LAP_ADDR_HEADER+LAP_CTRL_HEADER); - - skb->data[0] = CBROADCAST; - skb->data[1] = UI_FRAME; - - /* Don't need to refcount, see irlmp_connless_data_request() */ - - skb_queue_tail(&self->txq_ultra, skb); - - irlap_do_event(self, SEND_UI_FRAME, NULL, NULL); -} -#endif /*CONFIG_IRDA_ULTRA */ - -/* - * Function irlap_udata_indication (self, skb) - * - * Receive Ultra data. This is data that is received outside any connection - * - */ -#ifdef CONFIG_IRDA_ULTRA -void irlap_unitdata_indication(struct irlap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Hide LAP header from IrLMP layer */ - skb_pull(skb, LAP_ADDR_HEADER+LAP_CTRL_HEADER); - - irlmp_link_unitdata_indication(self->notify.instance, skb); -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irlap_disconnect_request (void) - * - * Request to disconnect connection by service user - */ -void irlap_disconnect_request(struct irlap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Don't disconnect until all data frames are successfully sent */ - if (!skb_queue_empty(&self->txq)) { - self->disconnect_pending = TRUE; - return; - } - - /* Check if we are in the right state for disconnecting */ - switch (self->state) { - case LAP_XMIT_P: /* FALLTHROUGH */ - case LAP_XMIT_S: /* FALLTHROUGH */ - case LAP_CONN: /* FALLTHROUGH */ - case LAP_RESET_WAIT: /* FALLTHROUGH */ - case LAP_RESET_CHECK: - irlap_do_event(self, DISCONNECT_REQUEST, NULL, NULL); - break; - default: - pr_debug("%s(), disconnect pending!\n", __func__); - self->disconnect_pending = TRUE; - break; - } -} - -/* - * Function irlap_disconnect_indication (void) - * - * Disconnect request from other device - * - */ -void irlap_disconnect_indication(struct irlap_cb *self, LAP_REASON reason) -{ - pr_debug("%s(), reason=%s\n", __func__, lap_reasons[reason]); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Flush queues */ - irlap_flush_all_queues(self); - - switch (reason) { - case LAP_RESET_INDICATION: - pr_debug("%s(), Sending reset request!\n", __func__); - irlap_do_event(self, RESET_REQUEST, NULL, NULL); - break; - case LAP_NO_RESPONSE: /* FALLTHROUGH */ - case LAP_DISC_INDICATION: /* FALLTHROUGH */ - case LAP_FOUND_NONE: /* FALLTHROUGH */ - case LAP_MEDIA_BUSY: - irlmp_link_disconnect_indication(self->notify.instance, self, - reason, NULL); - break; - default: - net_err_ratelimited("%s: Unknown reason %d\n", - __func__, reason); - } -} - -/* - * Function irlap_discovery_request (gen_addr_bit) - * - * Start one single discovery operation. - * - */ -void irlap_discovery_request(struct irlap_cb *self, discovery_t *discovery) -{ - struct irlap_info info; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(discovery != NULL, return;); - - pr_debug("%s(), nslots = %d\n", __func__, discovery->nslots); - - IRDA_ASSERT((discovery->nslots == 1) || (discovery->nslots == 6) || - (discovery->nslots == 8) || (discovery->nslots == 16), - return;); - - /* Discovery is only possible in NDM mode */ - if (self->state != LAP_NDM) { - pr_debug("%s(), discovery only possible in NDM mode\n", - __func__); - irlap_discovery_confirm(self, NULL); - /* Note : in theory, if we are not in NDM, we could postpone - * the discovery like we do for connection request. - * In practice, it's not worth it. If the media was busy, - * it's likely next time around it won't be busy. If we are - * in REPLY state, we will get passive discovery info & event. - * Jean II */ - return; - } - - /* Check if last discovery request finished in time, or if - * it was aborted due to the media busy flag. */ - if (self->discovery_log != NULL) { - hashbin_delete(self->discovery_log, (FREE_FUNC) kfree); - self->discovery_log = NULL; - } - - /* All operations will occur at predictable time, no need to lock */ - self->discovery_log = hashbin_new(HB_NOLOCK); - - if (self->discovery_log == NULL) { - net_warn_ratelimited("%s(), Unable to allocate discovery log!\n", - __func__); - return; - } - - info.S = discovery->nslots; /* Number of slots */ - info.s = 0; /* Current slot */ - - self->discovery_cmd = discovery; - info.discovery = discovery; - - /* sysctl_slot_timeout bounds are checked in irsysctl.c - Jean II */ - self->slot_timeout = msecs_to_jiffies(sysctl_slot_timeout); - - irlap_do_event(self, DISCOVERY_REQUEST, NULL, &info); -} - -/* - * Function irlap_discovery_confirm (log) - * - * A device has been discovered in front of this station, we - * report directly to LMP. - */ -void irlap_discovery_confirm(struct irlap_cb *self, hashbin_t *discovery_log) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - IRDA_ASSERT(self->notify.instance != NULL, return;); - - /* - * Check for successful discovery, since we are then allowed to clear - * the media busy condition (IrLAP 6.13.4 - p.94). This should allow - * us to make connection attempts much faster and easier (i.e. no - * collisions). - * Setting media busy to false will also generate an event allowing - * to process pending events in NDM state machine. - * Note : the spec doesn't define what's a successful discovery is. - * If we want Ultra to work, it's successful even if there is - * nobody discovered - Jean II - */ - if (discovery_log) - irda_device_set_media_busy(self->netdev, FALSE); - - /* Inform IrLMP */ - irlmp_link_discovery_confirm(self->notify.instance, discovery_log); -} - -/* - * Function irlap_discovery_indication (log) - * - * Somebody is trying to discover us! - * - */ -void irlap_discovery_indication(struct irlap_cb *self, discovery_t *discovery) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(discovery != NULL, return;); - - IRDA_ASSERT(self->notify.instance != NULL, return;); - - /* A device is very likely to connect immediately after it performs - * a successful discovery. This means that in our case, we are much - * more likely to receive a connection request over the medium. - * So, we backoff to avoid collisions. - * IrLAP spec 6.13.4 suggest 100ms... - * Note : this little trick actually make a *BIG* difference. If I set - * my Linux box with discovery enabled and one Ultra frame sent every - * second, my Palm has no trouble connecting to it every time ! - * Jean II */ - irda_device_set_media_busy(self->netdev, SMALL); - - irlmp_link_discovery_indication(self->notify.instance, discovery); -} - -/* - * Function irlap_status_indication (quality_of_link) - */ -void irlap_status_indication(struct irlap_cb *self, int quality_of_link) -{ - switch (quality_of_link) { - case STATUS_NO_ACTIVITY: - net_info_ratelimited("IrLAP, no activity on link!\n"); - break; - case STATUS_NOISY: - net_info_ratelimited("IrLAP, noisy link!\n"); - break; - default: - break; - } - irlmp_status_indication(self->notify.instance, - quality_of_link, LOCK_NO_CHANGE); -} - -/* - * Function irlap_reset_indication (void) - */ -void irlap_reset_indication(struct irlap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - if (self->state == LAP_RESET_WAIT) - irlap_do_event(self, RESET_REQUEST, NULL, NULL); - else - irlap_do_event(self, RESET_RESPONSE, NULL, NULL); -} - -/* - * Function irlap_reset_confirm (void) - */ -void irlap_reset_confirm(void) -{ -} - -/* - * Function irlap_generate_rand_time_slot (S, s) - * - * Generate a random time slot between s and S-1 where - * S = Number of slots (0 -> S-1) - * s = Current slot - */ -int irlap_generate_rand_time_slot(int S, int s) -{ - static int rand; - int slot; - - IRDA_ASSERT((S - s) > 0, return 0;); - - rand += jiffies; - rand ^= (rand << 12); - rand ^= (rand >> 20); - - slot = s + rand % (S-s); - - IRDA_ASSERT((slot >= s) || (slot < S), return 0;); - - return slot; -} - -/* - * Function irlap_update_nr_received (nr) - * - * Remove all acknowledged frames in current window queue. This code is - * not intuitive and you should not try to change it. If you think it - * contains bugs, please mail a patch to the author instead. - */ -void irlap_update_nr_received(struct irlap_cb *self, int nr) -{ - struct sk_buff *skb = NULL; - int count = 0; - - /* - * Remove all the ack-ed frames from the window queue. - */ - - /* - * Optimize for the common case. It is most likely that the receiver - * will acknowledge all the frames we have sent! So in that case we - * delete all frames stored in window. - */ - if (nr == self->vs) { - while ((skb = skb_dequeue(&self->wx_list)) != NULL) { - dev_kfree_skb(skb); - } - /* The last acked frame is the next to send minus one */ - self->va = nr - 1; - } else { - /* Remove all acknowledged frames in current window */ - while ((skb_peek(&self->wx_list) != NULL) && - (((self->va+1) % 8) != nr)) - { - skb = skb_dequeue(&self->wx_list); - dev_kfree_skb(skb); - - self->va = (self->va + 1) % 8; - count++; - } - } - - /* Advance window */ - self->window = self->window_size - skb_queue_len(&self->wx_list); -} - -/* - * Function irlap_validate_ns_received (ns) - * - * Validate the next to send (ns) field from received frame. - */ -int irlap_validate_ns_received(struct irlap_cb *self, int ns) -{ - /* ns as expected? */ - if (ns == self->vr) - return NS_EXPECTED; - /* - * Stations are allowed to treat invalid NS as unexpected NS - * IrLAP, Recv ... with-invalid-Ns. p. 84 - */ - return NS_UNEXPECTED; - - /* return NR_INVALID; */ -} -/* - * Function irlap_validate_nr_received (nr) - * - * Validate the next to receive (nr) field from received frame. - * - */ -int irlap_validate_nr_received(struct irlap_cb *self, int nr) -{ - /* nr as expected? */ - if (nr == self->vs) { - pr_debug("%s(), expected!\n", __func__); - return NR_EXPECTED; - } - - /* - * unexpected nr? (but within current window), first we check if the - * ns numbers of the frames in the current window wrap. - */ - if (self->va < self->vs) { - if ((nr >= self->va) && (nr <= self->vs)) - return NR_UNEXPECTED; - } else { - if ((nr >= self->va) || (nr <= self->vs)) - return NR_UNEXPECTED; - } - - /* Invalid nr! */ - return NR_INVALID; -} - -/* - * Function irlap_initiate_connection_state () - * - * Initialize the connection state parameters - * - */ -void irlap_initiate_connection_state(struct irlap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Next to send and next to receive */ - self->vs = self->vr = 0; - - /* Last frame which got acked (0 - 1) % 8 */ - self->va = 7; - - self->window = 1; - - self->remote_busy = FALSE; - self->retry_count = 0; -} - -/* - * Function irlap_wait_min_turn_around (self, qos) - * - * Wait negotiated minimum turn around time, this function actually sets - * the number of BOS's that must be sent before the next transmitted - * frame in order to delay for the specified amount of time. This is - * done to avoid using timers, and the forbidden udelay! - */ -void irlap_wait_min_turn_around(struct irlap_cb *self, struct qos_info *qos) -{ - __u32 min_turn_time; - __u32 speed; - - /* Get QoS values. */ - speed = qos->baud_rate.value; - min_turn_time = qos->min_turn_time.value; - - /* No need to calculate XBOFs for speeds over 115200 bps */ - if (speed > 115200) { - self->mtt_required = min_turn_time; - return; - } - - /* - * Send additional BOF's for the next frame for the requested - * min turn time, so now we must calculate how many chars (XBOF's) we - * must send for the requested time period (min turn time) - */ - self->xbofs_delay = irlap_min_turn_time_in_bytes(speed, min_turn_time); -} - -/* - * Function irlap_flush_all_queues (void) - * - * Flush all queues - * - */ -void irlap_flush_all_queues(struct irlap_cb *self) -{ - struct sk_buff* skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Free transmission queue */ - while ((skb = skb_dequeue(&self->txq)) != NULL) - dev_kfree_skb(skb); - - while ((skb = skb_dequeue(&self->txq_ultra)) != NULL) - dev_kfree_skb(skb); - - /* Free sliding window buffered packets */ - while ((skb = skb_dequeue(&self->wx_list)) != NULL) - dev_kfree_skb(skb); -} - -/* - * Function irlap_setspeed (self, speed) - * - * Change the speed of the IrDA port - * - */ -static void irlap_change_speed(struct irlap_cb *self, __u32 speed, int now) -{ - struct sk_buff *skb; - - pr_debug("%s(), setting speed to %d\n", __func__, speed); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - self->speed = speed; - - /* Change speed now, or just piggyback speed on frames */ - if (now) { - /* Send down empty frame to trigger speed change */ - skb = alloc_skb(0, GFP_ATOMIC); - if (skb) - irlap_queue_xmit(self, skb); - } -} - -/* - * Function irlap_init_qos_capabilities (self, qos) - * - * Initialize QoS for this IrLAP session, What we do is to compute the - * intersection of the QoS capabilities for the user, driver and for - * IrLAP itself. Normally, IrLAP will not specify any values, but it can - * be used to restrict certain values. - */ -static void irlap_init_qos_capabilities(struct irlap_cb *self, - struct qos_info *qos_user) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(self->netdev != NULL, return;); - - /* Start out with the maximum QoS support possible */ - irda_init_max_qos_capabilies(&self->qos_rx); - - /* Apply drivers QoS capabilities */ - irda_qos_compute_intersection(&self->qos_rx, self->qos_dev); - - /* - * Check for user supplied QoS parameters. The service user is only - * allowed to supply these values. We check each parameter since the - * user may not have set all of them. - */ - if (qos_user) { - pr_debug("%s(), Found user specified QoS!\n", __func__); - - if (qos_user->baud_rate.bits) - self->qos_rx.baud_rate.bits &= qos_user->baud_rate.bits; - - if (qos_user->max_turn_time.bits) - self->qos_rx.max_turn_time.bits &= qos_user->max_turn_time.bits; - if (qos_user->data_size.bits) - self->qos_rx.data_size.bits &= qos_user->data_size.bits; - - if (qos_user->link_disc_time.bits) - self->qos_rx.link_disc_time.bits &= qos_user->link_disc_time.bits; - } - - /* Use 500ms in IrLAP for now */ - self->qos_rx.max_turn_time.bits &= 0x01; - - /* Set data size */ - /*self->qos_rx.data_size.bits &= 0x03;*/ - - irda_qos_bits_to_value(&self->qos_rx); -} - -/* - * Function irlap_apply_default_connection_parameters (void, now) - * - * Use the default connection and transmission parameters - */ -void irlap_apply_default_connection_parameters(struct irlap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* xbofs : Default value in NDM */ - self->next_bofs = 12; - self->bofs_count = 12; - - /* NDM Speed is 9600 */ - irlap_change_speed(self, 9600, TRUE); - - /* Set mbusy when going to NDM state */ - irda_device_set_media_busy(self->netdev, TRUE); - - /* - * Generate random connection address for this session, which must - * be 7 bits wide and different from 0x00 and 0xfe - */ - while ((self->caddr == 0x00) || (self->caddr == 0xfe)) { - get_random_bytes(&self->caddr, sizeof(self->caddr)); - self->caddr &= 0xfe; - } - - /* Use default values until connection has been negitiated */ - self->slot_timeout = sysctl_slot_timeout; - self->final_timeout = FINAL_TIMEOUT; - self->poll_timeout = POLL_TIMEOUT; - self->wd_timeout = WD_TIMEOUT; - - /* Set some default values */ - self->qos_tx.baud_rate.value = 9600; - self->qos_rx.baud_rate.value = 9600; - self->qos_tx.max_turn_time.value = 0; - self->qos_rx.max_turn_time.value = 0; - self->qos_tx.min_turn_time.value = 0; - self->qos_rx.min_turn_time.value = 0; - self->qos_tx.data_size.value = 64; - self->qos_rx.data_size.value = 64; - self->qos_tx.window_size.value = 1; - self->qos_rx.window_size.value = 1; - self->qos_tx.additional_bofs.value = 12; - self->qos_rx.additional_bofs.value = 12; - self->qos_tx.link_disc_time.value = 0; - self->qos_rx.link_disc_time.value = 0; - - irlap_flush_all_queues(self); - - self->disconnect_pending = FALSE; - self->connect_pending = FALSE; -} - -/* - * Function irlap_apply_connection_parameters (qos, now) - * - * Initialize IrLAP with the negotiated QoS values - * - * If 'now' is false, the speed and xbofs will be changed after the next - * frame is sent. - * If 'now' is true, the speed and xbofs is changed immediately - */ -void irlap_apply_connection_parameters(struct irlap_cb *self, int now) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Set the negotiated xbofs value */ - self->next_bofs = self->qos_tx.additional_bofs.value; - if (now) - self->bofs_count = self->next_bofs; - - /* Set the negotiated link speed (may need the new xbofs value) */ - irlap_change_speed(self, self->qos_tx.baud_rate.value, now); - - self->window_size = self->qos_tx.window_size.value; - self->window = self->qos_tx.window_size.value; - -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - /* - * Calculate how many bytes it is possible to transmit before the - * link must be turned around - */ - self->line_capacity = - irlap_max_line_capacity(self->qos_tx.baud_rate.value, - self->qos_tx.max_turn_time.value); - self->bytes_left = self->line_capacity; -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - - - /* - * Initialize timeout values, some of the rules are listed on - * page 92 in IrLAP. - */ - IRDA_ASSERT(self->qos_tx.max_turn_time.value != 0, return;); - IRDA_ASSERT(self->qos_rx.max_turn_time.value != 0, return;); - /* The poll timeout applies only to the primary station. - * It defines the maximum time the primary stay in XMIT mode - * before timeout and turning the link around (sending a RR). - * Or, this is how much we can keep the pf bit in primary mode. - * Therefore, it must be lower or equal than our *OWN* max turn around. - * Jean II */ - self->poll_timeout = msecs_to_jiffies( - self->qos_tx.max_turn_time.value); - /* The Final timeout applies only to the primary station. - * It defines the maximum time the primary wait (mostly in RECV mode) - * for an answer from the secondary station before polling it again. - * Therefore, it must be greater or equal than our *PARTNER* - * max turn around time - Jean II */ - self->final_timeout = msecs_to_jiffies( - self->qos_rx.max_turn_time.value); - /* The Watchdog Bit timeout applies only to the secondary station. - * It defines the maximum time the secondary wait (mostly in RECV mode) - * for poll from the primary station before getting annoyed. - * Therefore, it must be greater or equal than our *PARTNER* - * max turn around time - Jean II */ - self->wd_timeout = self->final_timeout * 2; - - /* - * N1 and N2 are maximum retry count for *both* the final timer - * and the wd timer (with a factor 2) as defined above. - * After N1 retry of a timer, we give a warning to the user. - * After N2 retry, we consider the link dead and disconnect it. - * Jean II - */ - - /* - * Set N1 to 0 if Link Disconnect/Threshold Time = 3 and set it to - * 3 seconds otherwise. See page 71 in IrLAP for more details. - * Actually, it's not always 3 seconds, as we allow to set - * it via sysctl... Max maxtt is 500ms, and N1 need to be multiple - * of 2, so 1 second is minimum we can allow. - Jean II - */ - if (self->qos_tx.link_disc_time.value == sysctl_warn_noreply_time) - /* - * If we set N1 to 0, it will trigger immediately, which is - * not what we want. What we really want is to disable it, - * Jean II - */ - self->N1 = -2; /* Disable - Need to be multiple of 2*/ - else - self->N1 = sysctl_warn_noreply_time * 1000 / - self->qos_rx.max_turn_time.value; - - pr_debug("Setting N1 = %d\n", self->N1); - - /* Set N2 to match our own disconnect time */ - self->N2 = self->qos_tx.link_disc_time.value * 1000 / - self->qos_rx.max_turn_time.value; - pr_debug("Setting N2 = %d\n", self->N2); -} - -#ifdef CONFIG_PROC_FS -struct irlap_iter_state { - int id; -}; - -static void *irlap_seq_start(struct seq_file *seq, loff_t *pos) -{ - struct irlap_iter_state *iter = seq->private; - struct irlap_cb *self; - - /* Protect our access to the tsap list */ - spin_lock_irq(&irlap->hb_spinlock); - iter->id = 0; - - for (self = (struct irlap_cb *) hashbin_get_first(irlap); - self; self = (struct irlap_cb *) hashbin_get_next(irlap)) { - if (iter->id == *pos) - break; - ++iter->id; - } - - return self; -} - -static void *irlap_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - struct irlap_iter_state *iter = seq->private; - - ++*pos; - ++iter->id; - return (void *) hashbin_get_next(irlap); -} - -static void irlap_seq_stop(struct seq_file *seq, void *v) -{ - spin_unlock_irq(&irlap->hb_spinlock); -} - -static int irlap_seq_show(struct seq_file *seq, void *v) -{ - const struct irlap_iter_state *iter = seq->private; - const struct irlap_cb *self = v; - - IRDA_ASSERT(self->magic == LAP_MAGIC, return -EINVAL;); - - seq_printf(seq, "irlap%d ", iter->id); - seq_printf(seq, "state: %s\n", - irlap_state[self->state]); - - seq_printf(seq, " device name: %s, ", - (self->netdev) ? self->netdev->name : "bug"); - seq_printf(seq, "hardware name: %s\n", self->hw_name); - - seq_printf(seq, " caddr: %#02x, ", self->caddr); - seq_printf(seq, "saddr: %#08x, ", self->saddr); - seq_printf(seq, "daddr: %#08x\n", self->daddr); - - seq_printf(seq, " win size: %d, ", - self->window_size); - seq_printf(seq, "win: %d, ", self->window); -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - seq_printf(seq, "line capacity: %d, ", - self->line_capacity); - seq_printf(seq, "bytes left: %d\n", self->bytes_left); -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - seq_printf(seq, " tx queue len: %d ", - skb_queue_len(&self->txq)); - seq_printf(seq, "win queue len: %d ", - skb_queue_len(&self->wx_list)); - seq_printf(seq, "rbusy: %s", self->remote_busy ? - "TRUE" : "FALSE"); - seq_printf(seq, " mbusy: %s\n", self->media_busy ? - "TRUE" : "FALSE"); - - seq_printf(seq, " retrans: %d ", self->retry_count); - seq_printf(seq, "vs: %d ", self->vs); - seq_printf(seq, "vr: %d ", self->vr); - seq_printf(seq, "va: %d\n", self->va); - - seq_printf(seq, " qos\tbps\tmaxtt\tdsize\twinsize\taddbofs\tmintt\tldisc\tcomp\n"); - - seq_printf(seq, " tx\t%d\t", - self->qos_tx.baud_rate.value); - seq_printf(seq, "%d\t", - self->qos_tx.max_turn_time.value); - seq_printf(seq, "%d\t", - self->qos_tx.data_size.value); - seq_printf(seq, "%d\t", - self->qos_tx.window_size.value); - seq_printf(seq, "%d\t", - self->qos_tx.additional_bofs.value); - seq_printf(seq, "%d\t", - self->qos_tx.min_turn_time.value); - seq_printf(seq, "%d\t", - self->qos_tx.link_disc_time.value); - seq_printf(seq, "\n"); - - seq_printf(seq, " rx\t%d\t", - self->qos_rx.baud_rate.value); - seq_printf(seq, "%d\t", - self->qos_rx.max_turn_time.value); - seq_printf(seq, "%d\t", - self->qos_rx.data_size.value); - seq_printf(seq, "%d\t", - self->qos_rx.window_size.value); - seq_printf(seq, "%d\t", - self->qos_rx.additional_bofs.value); - seq_printf(seq, "%d\t", - self->qos_rx.min_turn_time.value); - seq_printf(seq, "%d\n", - self->qos_rx.link_disc_time.value); - - return 0; -} - -static const struct seq_operations irlap_seq_ops = { - .start = irlap_seq_start, - .next = irlap_seq_next, - .stop = irlap_seq_stop, - .show = irlap_seq_show, -}; - -static int irlap_seq_open(struct inode *inode, struct file *file) -{ - if (irlap == NULL) - return -EINVAL; - - return seq_open_private(file, &irlap_seq_ops, - sizeof(struct irlap_iter_state)); -} - -const struct file_operations irlap_seq_fops = { - .owner = THIS_MODULE, - .open = irlap_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, -}; - -#endif /* CONFIG_PROC_FS */ diff --git a/net/irda/irlap_event.c b/net/irda/irlap_event.c deleted file mode 100644 index 0e1b4d79f745..000000000000 --- a/net/irda/irlap_event.c +++ /dev/null @@ -1,2316 +0,0 @@ -/********************************************************************* - * - * Filename: irlap_event.c - * Version: 0.9 - * Description: IrLAP state machine implementation - * Status: Experimental. - * Author: Dag Brattli <dag@brattli.net> - * Created at: Sat Aug 16 00:59:29 1997 - * Modified at: Sat Dec 25 21:07:57 1999 - * Modified by: Dag Brattli <dag@brattli.net> - * - * Copyright (c) 1998-2000 Dag Brattli <dag@brattli.net>, - * Copyright (c) 1998 Thomas Davis <ratbert@radiks.net> - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/string.h> -#include <linux/kernel.h> -#include <linux/delay.h> -#include <linux/skbuff.h> -#include <linux/slab.h> - -#include <net/irda/irda.h> -#include <net/irda/irlap_event.h> - -#include <net/irda/timer.h> -#include <net/irda/irlap.h> -#include <net/irda/irlap_frame.h> -#include <net/irda/qos.h> -#include <net/irda/parameters.h> -#include <net/irda/irlmp.h> /* irlmp_flow_indication(), ... */ - -#include <net/irda/irda_device.h> - -#ifdef CONFIG_IRDA_FAST_RR -int sysctl_fast_poll_increase = 50; -#endif - -static int irlap_state_ndm (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_query (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_reply (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_conn (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_setup (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_offline(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_xmit_p (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_pclose (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_nrm_p (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_reset_wait(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_reset (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_nrm_s (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_xmit_s (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_sclose (struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info); -static int irlap_state_reset_check(struct irlap_cb *, IRLAP_EVENT event, - struct sk_buff *, struct irlap_info *); - -static const char *const irlap_event[] __maybe_unused = { - "DISCOVERY_REQUEST", - "CONNECT_REQUEST", - "CONNECT_RESPONSE", - "DISCONNECT_REQUEST", - "DATA_REQUEST", - "RESET_REQUEST", - "RESET_RESPONSE", - "SEND_I_CMD", - "SEND_UI_FRAME", - "RECV_DISCOVERY_XID_CMD", - "RECV_DISCOVERY_XID_RSP", - "RECV_SNRM_CMD", - "RECV_TEST_CMD", - "RECV_TEST_RSP", - "RECV_UA_RSP", - "RECV_DM_RSP", - "RECV_RD_RSP", - "RECV_I_CMD", - "RECV_I_RSP", - "RECV_UI_FRAME", - "RECV_FRMR_RSP", - "RECV_RR_CMD", - "RECV_RR_RSP", - "RECV_RNR_CMD", - "RECV_RNR_RSP", - "RECV_REJ_CMD", - "RECV_REJ_RSP", - "RECV_SREJ_CMD", - "RECV_SREJ_RSP", - "RECV_DISC_CMD", - "SLOT_TIMER_EXPIRED", - "QUERY_TIMER_EXPIRED", - "FINAL_TIMER_EXPIRED", - "POLL_TIMER_EXPIRED", - "DISCOVERY_TIMER_EXPIRED", - "WD_TIMER_EXPIRED", - "BACKOFF_TIMER_EXPIRED", - "MEDIA_BUSY_TIMER_EXPIRED", -}; - -const char *const irlap_state[] = { - "LAP_NDM", - "LAP_QUERY", - "LAP_REPLY", - "LAP_CONN", - "LAP_SETUP", - "LAP_OFFLINE", - "LAP_XMIT_P", - "LAP_PCLOSE", - "LAP_NRM_P", - "LAP_RESET_WAIT", - "LAP_RESET", - "LAP_NRM_S", - "LAP_XMIT_S", - "LAP_SCLOSE", - "LAP_RESET_CHECK", -}; - -static int (*state[])(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) = -{ - irlap_state_ndm, - irlap_state_query, - irlap_state_reply, - irlap_state_conn, - irlap_state_setup, - irlap_state_offline, - irlap_state_xmit_p, - irlap_state_pclose, - irlap_state_nrm_p, - irlap_state_reset_wait, - irlap_state_reset, - irlap_state_nrm_s, - irlap_state_xmit_s, - irlap_state_sclose, - irlap_state_reset_check, -}; - -/* - * Function irda_poll_timer_expired (data) - * - * Poll timer has expired. Normally we must now send a RR frame to the - * remote device - */ -static void irlap_poll_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, POLL_TIMER_EXPIRED, NULL, NULL); -} - -/* - * Calculate and set time before we will have to send back the pf bit - * to the peer. Use in primary. - * Make sure that state is XMIT_P/XMIT_S when calling this function - * (and that nobody messed up with the state). - Jean II - */ -static void irlap_start_poll_timer(struct irlap_cb *self, int timeout) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - -#ifdef CONFIG_IRDA_FAST_RR - /* - * Send out the RR frames faster if our own transmit queue is empty, or - * if the peer is busy. The effect is a much faster conversation - */ - if (skb_queue_empty(&self->txq) || self->remote_busy) { - if (self->fast_RR == TRUE) { - /* - * Assert that the fast poll timer has not reached the - * normal poll timer yet - */ - if (self->fast_RR_timeout < timeout) { - /* - * FIXME: this should be a more configurable - * function - */ - self->fast_RR_timeout += - (sysctl_fast_poll_increase * HZ/1000); - - /* Use this fast(er) timeout instead */ - timeout = self->fast_RR_timeout; - } - } else { - self->fast_RR = TRUE; - - /* Start with just 0 ms */ - self->fast_RR_timeout = 0; - timeout = 0; - } - } else - self->fast_RR = FALSE; - - pr_debug("%s(), timeout=%d (%ld)\n", __func__, timeout, jiffies); -#endif /* CONFIG_IRDA_FAST_RR */ - - if (timeout == 0) - irlap_do_event(self, POLL_TIMER_EXPIRED, NULL, NULL); - else - irda_start_timer(&self->poll_timer, timeout, self, - irlap_poll_timer_expired); -} - -/* - * Function irlap_do_event (event, skb, info) - * - * Rushes through the state machine without any delay. If state == XMIT - * then send queued data frames. - */ -void irlap_do_event(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret; - - if (!self || self->magic != LAP_MAGIC) - return; - - pr_debug("%s(), event = %s, state = %s\n", __func__, - irlap_event[event], irlap_state[self->state]); - - ret = (*state[self->state])(self, event, skb, info); - - /* - * Check if there are any pending events that needs to be executed - */ - switch (self->state) { - case LAP_XMIT_P: /* FALLTHROUGH */ - case LAP_XMIT_S: - /* - * We just received the pf bit and are at the beginning - * of a new LAP transmit window. - * Check if there are any queued data frames, and do not - * try to disconnect link if we send any data frames, since - * that will change the state away form XMIT - */ - pr_debug("%s() : queue len = %d\n", __func__, - skb_queue_len(&self->txq)); - - if (!skb_queue_empty(&self->txq)) { - /* Prevent race conditions with irlap_data_request() */ - self->local_busy = TRUE; - - /* Theory of operation. - * We send frames up to when we fill the window or - * reach line capacity. Those frames will queue up - * in the device queue, and the driver will slowly - * send them. - * After each frame that we send, we poll the higher - * layer for more data. It's the right time to do - * that because the link layer need to perform the mtt - * and then send the first frame, so we can afford - * to send a bit of time in kernel space. - * The explicit flow indication allow to minimise - * buffers (== lower latency), to avoid higher layer - * polling via timers (== less context switches) and - * to implement a crude scheduler - Jean II */ - - /* Try to send away all queued data frames */ - while ((skb = skb_dequeue(&self->txq)) != NULL) { - /* Send one frame */ - ret = (*state[self->state])(self, SEND_I_CMD, - skb, NULL); - /* Drop reference count. - * It will be increase as needed in - * irlap_send_data_xxx() */ - kfree_skb(skb); - - /* Poll the higher layers for one more frame */ - irlmp_flow_indication(self->notify.instance, - FLOW_START); - - if (ret == -EPROTO) - break; /* Try again later! */ - } - /* Finished transmitting */ - self->local_busy = FALSE; - } else if (self->disconnect_pending) { - self->disconnect_pending = FALSE; - - ret = (*state[self->state])(self, DISCONNECT_REQUEST, - NULL, NULL); - } - break; -/* case LAP_NDM: */ -/* case LAP_CONN: */ -/* case LAP_RESET_WAIT: */ -/* case LAP_RESET_CHECK: */ - default: - break; - } -} - -/* - * Function irlap_state_ndm (event, skb, frame) - * - * NDM (Normal Disconnected Mode) state - * - */ -static int irlap_state_ndm(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - discovery_t *discovery_rsp; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case CONNECT_REQUEST: - IRDA_ASSERT(self->netdev != NULL, return -1;); - - if (self->media_busy) { - /* Note : this will never happen, because we test - * media busy in irlap_connect_request() and - * postpone the event... - Jean II */ - pr_debug("%s(), CONNECT_REQUEST: media busy!\n", - __func__); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_MEDIA_BUSY); - } else { - irlap_send_snrm_frame(self, &self->qos_rx); - - /* Start Final-bit timer */ - irlap_start_final_timer(self, self->final_timeout); - - self->retry_count = 0; - irlap_next_state(self, LAP_SETUP); - } - break; - case RECV_SNRM_CMD: - /* Check if the frame contains and I field */ - if (info) { - self->daddr = info->daddr; - self->caddr = info->caddr; - - irlap_next_state(self, LAP_CONN); - - irlap_connect_indication(self, skb); - } else { - pr_debug("%s(), SNRM frame does not contain an I field!\n", - __func__); - } - break; - case DISCOVERY_REQUEST: - IRDA_ASSERT(info != NULL, return -1;); - - if (self->media_busy) { - pr_debug("%s(), DISCOVERY_REQUEST: media busy!\n", - __func__); - /* irlap->log.condition = MEDIA_BUSY; */ - - /* This will make IrLMP try again */ - irlap_discovery_confirm(self, NULL); - /* Note : the discovery log is not cleaned up here, - * it will be done in irlap_discovery_request() - * Jean II */ - return 0; - } - - self->S = info->S; - self->s = info->s; - irlap_send_discovery_xid_frame(self, info->S, info->s, TRUE, - info->discovery); - self->frame_sent = FALSE; - self->s++; - - irlap_start_slot_timer(self, self->slot_timeout); - irlap_next_state(self, LAP_QUERY); - break; - case RECV_DISCOVERY_XID_CMD: - IRDA_ASSERT(info != NULL, return -1;); - - /* Assert that this is not the final slot */ - if (info->s <= info->S) { - self->slot = irlap_generate_rand_time_slot(info->S, - info->s); - if (self->slot == info->s) { - discovery_rsp = irlmp_get_discovery_response(); - discovery_rsp->data.daddr = info->daddr; - - irlap_send_discovery_xid_frame(self, info->S, - self->slot, - FALSE, - discovery_rsp); - self->frame_sent = TRUE; - } else - self->frame_sent = FALSE; - - /* - * Go to reply state until end of discovery to - * inhibit our own transmissions. Set the timer - * to not stay forever there... Jean II - */ - irlap_start_query_timer(self, info->S, info->s); - irlap_next_state(self, LAP_REPLY); - } else { - /* This is the final slot. How is it possible ? - * This would happen is both discoveries are just slightly - * offset (if they are in sync, all packets are lost). - * Most often, all the discovery requests will be received - * in QUERY state (see my comment there), except for the - * last frame that will come here. - * The big trouble when it happen is that active discovery - * doesn't happen, because nobody answer the discoveries - * frame of the other guy, so the log shows up empty. - * What should we do ? - * Not much. It's too late to answer those discovery frames, - * so we just pass the info to IrLMP who will put it in the - * log (and post an event). - * Another cause would be devices that do discovery much - * slower than us, however the latest fixes should minimise - * those cases... - * Jean II - */ - pr_debug("%s(), Receiving final discovery request, missed the discovery slots :-(\n", - __func__); - - /* Last discovery request -> in the log */ - irlap_discovery_indication(self, info->discovery); - } - break; - case MEDIA_BUSY_TIMER_EXPIRED: - /* A bunch of events may be postponed because the media is - * busy (usually immediately after we close a connection), - * or while we are doing discovery (state query/reply). - * In all those cases, the media busy flag will be cleared - * when it's OK for us to process those postponed events. - * This event is not mentioned in the state machines in the - * IrLAP spec. It's because they didn't consider Ultra and - * postponing connection request is optional. - * Jean II */ -#ifdef CONFIG_IRDA_ULTRA - /* Send any pending Ultra frames if any */ - if (!skb_queue_empty(&self->txq_ultra)) { - /* We don't send the frame, just post an event. - * Also, previously this code was in timer.c... - * Jean II */ - ret = (*state[self->state])(self, SEND_UI_FRAME, - NULL, NULL); - } -#endif /* CONFIG_IRDA_ULTRA */ - /* Check if we should try to connect. - * This code was previously in irlap_do_event() */ - if (self->connect_pending) { - self->connect_pending = FALSE; - - /* This one *should* not pend in this state, except - * if a socket try to connect and immediately - * disconnect. - clear - Jean II */ - if (self->disconnect_pending) - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - else - ret = (*state[self->state])(self, - CONNECT_REQUEST, - NULL, NULL); - self->disconnect_pending = FALSE; - } - /* Note : one way to test if this code works well (including - * media busy and small busy) is to create a user space - * application generating an Ultra packet every 3.05 sec (or - * 2.95 sec) and to see how it interact with discovery. - * It's fairly easy to check that no packet is lost, that the - * packets are postponed during discovery and that after - * discovery indication you have a 100ms "gap". - * As connection request and Ultra are now processed the same - * way, this avoid the tedious job of trying IrLAP connection - * in all those cases... - * Jean II */ - break; -#ifdef CONFIG_IRDA_ULTRA - case SEND_UI_FRAME: - { - int i; - /* Only allowed to repeat an operation twice */ - for (i=0; ((i<2) && (self->media_busy == FALSE)); i++) { - skb = skb_dequeue(&self->txq_ultra); - if (skb) - irlap_send_ui_frame(self, skb, CBROADCAST, - CMD_FRAME); - else - break; - /* irlap_send_ui_frame() won't increase skb reference - * count, so no dev_kfree_skb() - Jean II */ - } - if (i == 2) { - /* Force us to listen 500 ms again */ - irda_device_set_media_busy(self->netdev, TRUE); - } - break; - } - case RECV_UI_FRAME: - /* Only accept broadcast frames in NDM mode */ - if (info->caddr != CBROADCAST) { - pr_debug("%s(), not a broadcast frame!\n", - __func__); - } else - irlap_unitdata_indication(self, skb); - break; -#endif /* CONFIG_IRDA_ULTRA */ - case RECV_TEST_CMD: - /* Remove test frame header */ - skb_pull(skb, sizeof(struct test_frame)); - - /* - * Send response. This skb will not be sent out again, and - * will only be used to send out the same info as the cmd - */ - irlap_send_test_frame(self, CBROADCAST, info->daddr, skb); - break; - case RECV_TEST_RSP: - pr_debug("%s() not implemented!\n", __func__); - break; - default: - pr_debug("%s(), Unknown event %s\n", __func__, - irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_query (event, skb, info) - * - * QUERY state - * - */ -static int irlap_state_query(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case RECV_DISCOVERY_XID_RSP: - IRDA_ASSERT(info != NULL, return -1;); - IRDA_ASSERT(info->discovery != NULL, return -1;); - - pr_debug("%s(), daddr=%08x\n", __func__, - info->discovery->data.daddr); - - if (!self->discovery_log) { - net_warn_ratelimited("%s: discovery log is gone! maybe the discovery timeout has been set too short?\n", - __func__); - break; - } - hashbin_insert(self->discovery_log, - (irda_queue_t *) info->discovery, - info->discovery->data.daddr, NULL); - - /* Keep state */ - /* irlap_next_state(self, LAP_QUERY); */ - - break; - case RECV_DISCOVERY_XID_CMD: - /* Yes, it is possible to receive those frames in this mode. - * Note that most often the last discovery request won't - * occur here but in NDM state (see my comment there). - * What should we do ? - * Not much. We are currently performing our own discovery, - * therefore we can't answer those frames. We don't want - * to change state either. We just pass the info to - * IrLMP who will put it in the log (and post an event). - * Jean II - */ - - IRDA_ASSERT(info != NULL, return -1;); - - pr_debug("%s(), Receiving discovery request (s = %d) while performing discovery :-(\n", - __func__, info->s); - - /* Last discovery request ? */ - if (info->s == 0xff) - irlap_discovery_indication(self, info->discovery); - break; - case SLOT_TIMER_EXPIRED: - /* - * Wait a little longer if we detect an incoming frame. This - * is not mentioned in the spec, but is a good thing to do, - * since we want to work even with devices that violate the - * timing requirements. - */ - if (irda_device_is_receiving(self->netdev) && !self->add_wait) { - pr_debug("%s(), device is slow to answer, waiting some more!\n", - __func__); - irlap_start_slot_timer(self, msecs_to_jiffies(10)); - self->add_wait = TRUE; - return ret; - } - self->add_wait = FALSE; - - if (self->s < self->S) { - irlap_send_discovery_xid_frame(self, self->S, - self->s, TRUE, - self->discovery_cmd); - self->s++; - irlap_start_slot_timer(self, self->slot_timeout); - - /* Keep state */ - irlap_next_state(self, LAP_QUERY); - } else { - /* This is the final slot! */ - irlap_send_discovery_xid_frame(self, self->S, 0xff, - TRUE, - self->discovery_cmd); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - /* - * We are now finished with the discovery procedure, - * so now we must return the results - */ - irlap_discovery_confirm(self, self->discovery_log); - - /* IrLMP should now have taken care of the log */ - self->discovery_log = NULL; - } - break; - default: - pr_debug("%s(), Unknown event %s\n", __func__, - irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_reply (self, event, skb, info) - * - * REPLY, we have received a XID discovery frame from a device and we - * are waiting for the right time slot to send a response XID frame - * - */ -static int irlap_state_reply(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - discovery_t *discovery_rsp; - int ret=0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case QUERY_TIMER_EXPIRED: - pr_debug("%s(), QUERY_TIMER_EXPIRED <%ld>\n", - __func__, jiffies); - irlap_next_state(self, LAP_NDM); - break; - case RECV_DISCOVERY_XID_CMD: - IRDA_ASSERT(info != NULL, return -1;); - /* Last frame? */ - if (info->s == 0xff) { - del_timer(&self->query_timer); - - /* info->log.condition = REMOTE; */ - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_discovery_indication(self, info->discovery); - } else { - /* If it's our slot, send our reply */ - if ((info->s >= self->slot) && (!self->frame_sent)) { - discovery_rsp = irlmp_get_discovery_response(); - discovery_rsp->data.daddr = info->daddr; - - irlap_send_discovery_xid_frame(self, info->S, - self->slot, - FALSE, - discovery_rsp); - - self->frame_sent = TRUE; - } - /* Readjust our timer to accommodate devices - * doing faster or slower discovery than us... - * Jean II */ - irlap_start_query_timer(self, info->S, info->s); - - /* Keep state */ - //irlap_next_state(self, LAP_REPLY); - } - break; - default: - pr_debug("%s(), Unknown event %d, %s\n", __func__, - event, irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_conn (event, skb, info) - * - * CONN, we have received a SNRM command and is waiting for the upper - * layer to accept or refuse connection - * - */ -static int irlap_state_conn(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - pr_debug("%s(), event=%s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case CONNECT_RESPONSE: - skb_pull(skb, sizeof(struct snrm_frame)); - - IRDA_ASSERT(self->netdev != NULL, return -1;); - - irlap_qos_negotiate(self, skb); - - irlap_initiate_connection_state(self); - - /* - * Applying the parameters now will make sure we change speed - * *after* we have sent the next frame - */ - irlap_apply_connection_parameters(self, FALSE); - - /* - * Sending this frame will force a speed change after it has - * been sent (i.e. the frame will be sent at 9600). - */ - irlap_send_ua_response_frame(self, &self->qos_rx); - -#if 0 - /* - * We are allowed to send two frames, but this may increase - * the connect latency, so lets not do it for now. - */ - /* This is full of good intentions, but doesn't work in - * practice. - * After sending the first UA response, we switch the - * dongle to the negotiated speed, which is usually - * different than 9600 kb/s. - * From there, there is two solutions : - * 1) The other end has received the first UA response : - * it will set up the connection, move to state LAP_NRM_P, - * and will ignore and drop the second UA response. - * Actually, it's even worse : the other side will almost - * immediately send a RR that will likely collide with the - * UA response (depending on negotiated turnaround). - * 2) The other end has not received the first UA response, - * will stay at 9600 and will never see the second UA response. - * Jean II */ - irlap_send_ua_response_frame(self, &self->qos_rx); -#endif - - /* - * The WD-timer could be set to the duration of the P-timer - * for this case, but it is recommended to use twice the - * value (note 3 IrLAP p. 60). - */ - irlap_start_wd_timer(self, self->wd_timeout); - irlap_next_state(self, LAP_NRM_S); - - break; - case RECV_DISCOVERY_XID_CMD: - pr_debug("%s(), event RECV_DISCOVER_XID_CMD!\n", - __func__); - irlap_next_state(self, LAP_NDM); - - break; - case DISCONNECT_REQUEST: - pr_debug("%s(), Disconnect request!\n", __func__); - irlap_send_dm_frame(self); - irlap_next_state( self, LAP_NDM); - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - default: - pr_debug("%s(), Unknown event %d, %s\n", __func__, - event, irlap_event[event]); - - ret = -1; - break; - } - - return ret; -} - -/* - * Function irlap_state_setup (event, skb, frame) - * - * SETUP state, The local layer has transmitted a SNRM command frame to - * a remote peer layer and is awaiting a reply . - * - */ -static int irlap_state_setup(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case FINAL_TIMER_EXPIRED: - if (self->retry_count < self->N3) { -/* - * Perform random backoff, Wait a random number of time units, minimum - * duration half the time taken to transmitt a SNRM frame, maximum duration - * 1.5 times the time taken to transmit a SNRM frame. So this time should - * between 15 msecs and 45 msecs. - */ - irlap_start_backoff_timer(self, msecs_to_jiffies(20 + - (jiffies % 30))); - } else { - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_FOUND_NONE); - } - break; - case BACKOFF_TIMER_EXPIRED: - irlap_send_snrm_frame(self, &self->qos_rx); - irlap_start_final_timer(self, self->final_timeout); - self->retry_count++; - break; - case RECV_SNRM_CMD: - pr_debug("%s(), SNRM battle!\n", __func__); - - IRDA_ASSERT(skb != NULL, return 0;); - IRDA_ASSERT(info != NULL, return 0;); - - /* - * The device with the largest device address wins the battle - * (both have sent a SNRM command!) - */ - if (info &&(info->daddr > self->saddr)) { - del_timer(&self->final_timer); - irlap_initiate_connection_state(self); - - IRDA_ASSERT(self->netdev != NULL, return -1;); - - skb_pull(skb, sizeof(struct snrm_frame)); - - irlap_qos_negotiate(self, skb); - - /* Send UA frame and then change link settings */ - irlap_apply_connection_parameters(self, FALSE); - irlap_send_ua_response_frame(self, &self->qos_rx); - - irlap_next_state(self, LAP_NRM_S); - irlap_connect_confirm(self, skb); - - /* - * The WD-timer could be set to the duration of the - * P-timer for this case, but it is recommended - * to use twice the value (note 3 IrLAP p. 60). - */ - irlap_start_wd_timer(self, self->wd_timeout); - } else { - /* We just ignore the other device! */ - irlap_next_state(self, LAP_SETUP); - } - break; - case RECV_UA_RSP: - /* Stop F-timer */ - del_timer(&self->final_timer); - - /* Initiate connection state */ - irlap_initiate_connection_state(self); - - /* Negotiate connection parameters */ - IRDA_ASSERT(skb->len > 10, return -1;); - - skb_pull(skb, sizeof(struct ua_frame)); - - IRDA_ASSERT(self->netdev != NULL, return -1;); - - irlap_qos_negotiate(self, skb); - - /* Set the new link setting *now* (before the rr frame) */ - irlap_apply_connection_parameters(self, TRUE); - self->retry_count = 0; - - /* Wait for turnaround time to give a chance to the other - * device to be ready to receive us. - * Note : the time to switch speed is typically larger - * than the turnaround time, but as we don't have the other - * side speed switch time, that's our best guess... - * Jean II */ - irlap_wait_min_turn_around(self, &self->qos_tx); - - /* This frame will actually be sent at the new speed */ - irlap_send_rr_frame(self, CMD_FRAME); - - /* The timer is set to half the normal timer to quickly - * detect a failure to negotiate the new connection - * parameters. IrLAP 6.11.3.2, note 3. - * Note that currently we don't process this failure - * properly, as we should do a quick disconnect. - * Jean II */ - irlap_start_final_timer(self, self->final_timeout/2); - irlap_next_state(self, LAP_NRM_P); - - irlap_connect_confirm(self, skb); - break; - case RECV_DM_RSP: /* FALLTHROUGH */ - case RECV_DISC_CMD: - del_timer(&self->final_timer); - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - default: - pr_debug("%s(), Unknown event %d, %s\n", __func__, - event, irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_offline (self, event, skb, info) - * - * OFFLINE state, not used for now! - * - */ -static int irlap_state_offline(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - pr_debug("%s(), Unknown event\n", __func__); - - return -1; -} - -/* - * Function irlap_state_xmit_p (self, event, skb, info) - * - * XMIT, Only the primary station has right to transmit, and we - * therefore do not expect to receive any transmissions from other - * stations. - * - */ -static int irlap_state_xmit_p(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - switch (event) { - case SEND_I_CMD: - /* - * Only send frame if send-window > 0. - */ - if ((self->window > 0) && (!self->remote_busy)) { - int nextfit; -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - struct sk_buff *skb_next; - - /* With DYNAMIC_WINDOW, we keep the window size - * maximum, and adapt on the packets we are sending. - * At 115k, we can send only 2 packets of 2048 bytes - * in a 500 ms turnaround. Without this option, we - * would always limit the window to 2. With this - * option, if we send smaller packets, we can send - * up to 7 of them (always depending on QoS). - * Jean II */ - - /* Look at the next skb. This is safe, as we are - * the only consumer of the Tx queue (if we are not, - * we have other problems) - Jean II */ - skb_next = skb_peek(&self->txq); - - /* Check if a subsequent skb exist and would fit in - * the current window (with respect to turnaround - * time). - * This allow us to properly mark the current packet - * with the pf bit, to avoid falling back on the - * second test below, and avoid waiting the - * end of the window and sending a extra RR. - * Note : (skb_next != NULL) <=> (skb_queue_len() > 0) - * Jean II */ - nextfit = ((skb_next != NULL) && - ((skb_next->len + skb->len) <= - self->bytes_left)); - - /* - * The current packet may not fit ! Because of test - * above, this should not happen any more !!! - * Test if we have transmitted more bytes over the - * link than its possible to do with the current - * speed and turn-around-time. - */ - if((!nextfit) && (skb->len > self->bytes_left)) { - pr_debug("%s(), Not allowed to transmit more bytes!\n", - __func__); - /* Requeue the skb */ - skb_queue_head(&self->txq, skb_get(skb)); - /* - * We should switch state to LAP_NRM_P, but - * that is not possible since we must be sure - * that we poll the other side. Since we have - * used up our time, the poll timer should - * trigger anyway now, so we just wait for it - * DB - */ - /* - * Sorry, but that's not totally true. If - * we send 2000B packets, we may wait another - * 1000B until our turnaround expire. That's - * why we need to be proactive in avoiding - * coming here. - Jean II - */ - return -EPROTO; - } - - /* Subtract space used by this skb */ - self->bytes_left -= skb->len; -#else /* CONFIG_IRDA_DYNAMIC_WINDOW */ - /* Window has been adjusted for the max packet - * size, so much simpler... - Jean II */ - nextfit = !skb_queue_empty(&self->txq); -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - /* - * Send data with poll bit cleared only if window > 1 - * and there is more frames after this one to be sent - */ - if ((self->window > 1) && (nextfit)) { - /* More packet to send in current window */ - irlap_send_data_primary(self, skb); - irlap_next_state(self, LAP_XMIT_P); - } else { - /* Final packet of window */ - irlap_send_data_primary_poll(self, skb); - - /* - * Make sure state machine does not try to send - * any more frames - */ - ret = -EPROTO; - } -#ifdef CONFIG_IRDA_FAST_RR - /* Peer may want to reply immediately */ - self->fast_RR = FALSE; -#endif /* CONFIG_IRDA_FAST_RR */ - } else { - pr_debug("%s(), Unable to send! remote busy?\n", - __func__); - skb_queue_head(&self->txq, skb_get(skb)); - - /* - * The next ret is important, because it tells - * irlap_next_state _not_ to deliver more frames - */ - ret = -EPROTO; - } - break; - case POLL_TIMER_EXPIRED: - pr_debug("%s(), POLL_TIMER_EXPIRED <%ld>\n", - __func__, jiffies); - irlap_send_rr_frame(self, CMD_FRAME); - /* Return to NRM properly - Jean II */ - self->window = self->window_size; -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - /* Allowed to transmit a maximum number of bytes again. */ - self->bytes_left = self->line_capacity; -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - irlap_start_final_timer(self, self->final_timeout); - irlap_next_state(self, LAP_NRM_P); - break; - case DISCONNECT_REQUEST: - del_timer(&self->poll_timer); - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_disc_frame(self); - irlap_flush_all_queues(self); - irlap_start_final_timer(self, self->final_timeout); - self->retry_count = 0; - irlap_next_state(self, LAP_PCLOSE); - break; - case DATA_REQUEST: - /* Nothing to do, irlap_do_event() will send the packet - * when we return... - Jean II */ - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlap_event[event]); - - ret = -EINVAL; - break; - } - return ret; -} - -/* - * Function irlap_state_pclose (event, skb, info) - * - * PCLOSE state - */ -static int irlap_state_pclose(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case RECV_UA_RSP: /* FALLTHROUGH */ - case RECV_DM_RSP: - del_timer(&self->final_timer); - - /* Set new link parameters */ - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - case FINAL_TIMER_EXPIRED: - if (self->retry_count < self->N3) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_disc_frame(self); - irlap_start_final_timer(self, self->final_timeout); - self->retry_count++; - /* Keep state */ - } else { - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_NO_RESPONSE); - } - break; - default: - pr_debug("%s(), Unknown event %d\n", __func__, event); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_nrm_p (self, event, skb, info) - * - * NRM_P (Normal Response Mode as Primary), The primary station has given - * permissions to a secondary station to transmit IrLAP resonse frames - * (by sending a frame with the P bit set). The primary station will not - * transmit any frames and is expecting to receive frames only from the - * secondary to which transmission permissions has been given. - */ -static int irlap_state_nrm_p(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - int ns_status; - int nr_status; - - switch (event) { - case RECV_I_RSP: /* Optimize for the common case */ - if (unlikely(skb->len <= LAP_ADDR_HEADER + LAP_CTRL_HEADER)) { - /* - * Input validation check: a stir4200/mcp2150 - * combination sometimes results in an empty i:rsp. - * This makes no sense; we can just ignore the frame - * and send an rr:cmd immediately. This happens before - * changing nr or ns so triggers a retransmit - */ - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, CMD_FRAME); - /* Keep state */ - break; - } - /* FIXME: must check for remote_busy below */ -#ifdef CONFIG_IRDA_FAST_RR - /* - * Reset the fast_RR so we can use the fast RR code with - * full speed the next time since peer may have more frames - * to transmitt - */ - self->fast_RR = FALSE; -#endif /* CONFIG_IRDA_FAST_RR */ - IRDA_ASSERT( info != NULL, return -1;); - - ns_status = irlap_validate_ns_received(self, info->ns); - nr_status = irlap_validate_nr_received(self, info->nr); - - /* - * Check for expected I(nformation) frame - */ - if ((ns_status == NS_EXPECTED) && (nr_status == NR_EXPECTED)) { - - /* Update Vr (next frame for us to receive) */ - self->vr = (self->vr + 1) % 8; - - /* Update Nr received, cleanup our retry queue */ - irlap_update_nr_received(self, info->nr); - - /* - * Got expected NR, so reset the - * retry_count. This is not done by IrLAP spec, - * which is strange! - */ - self->retry_count = 0; - self->ack_required = TRUE; - - /* poll bit cleared? */ - if (!info->pf) { - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_P); - - irlap_data_indication(self, skb, FALSE); - } else { - /* No longer waiting for pf */ - del_timer(&self->final_timer); - - irlap_wait_min_turn_around(self, &self->qos_tx); - - /* Call higher layer *before* changing state - * to give them a chance to send data in the - * next LAP frame. - * Jean II */ - irlap_data_indication(self, skb, FALSE); - - /* XMIT states are the most dangerous state - * to be in, because user requests are - * processed directly and may change state. - * On the other hand, in NDM_P, those - * requests are queued and we will process - * them when we return to irlap_do_event(). - * Jean II - */ - irlap_next_state(self, LAP_XMIT_P); - - /* This is the last frame. - * Make sure it's always called in XMIT state. - * - Jean II */ - irlap_start_poll_timer(self, self->poll_timeout); - } - break; - - } - /* Unexpected next to send (Ns) */ - if ((ns_status == NS_UNEXPECTED) && (nr_status == NR_EXPECTED)) - { - if (!info->pf) { - irlap_update_nr_received(self, info->nr); - - /* - * Wait until the last frame before doing - * anything - */ - - /* Keep state */ - irlap_next_state(self, LAP_NRM_P); - } else { - pr_debug("%s(), missing or duplicate frame!\n", - __func__); - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, CMD_FRAME); - - self->ack_required = FALSE; - - irlap_start_final_timer(self, self->final_timeout); - irlap_next_state(self, LAP_NRM_P); - } - break; - } - /* - * Unexpected next to receive (Nr) - */ - if ((ns_status == NS_EXPECTED) && (nr_status == NR_UNEXPECTED)) - { - if (info->pf) { - self->vr = (self->vr + 1) % 8; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - /* Resend rejected frames */ - irlap_resend_rejected_frames(self, CMD_FRAME); - - self->ack_required = FALSE; - - /* Make sure we account for the time - * to transmit our frames. See comemnts - * in irlap_send_data_primary_poll(). - * Jean II */ - irlap_start_final_timer(self, 2 * self->final_timeout); - - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_P); - - irlap_data_indication(self, skb, FALSE); - } else { - /* - * Do not resend frames until the last - * frame has arrived from the other - * device. This is not documented in - * IrLAP!! - */ - self->vr = (self->vr + 1) % 8; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - self->ack_required = FALSE; - - /* Keep state, do not move this line!*/ - irlap_next_state(self, LAP_NRM_P); - - irlap_data_indication(self, skb, FALSE); - } - break; - } - /* - * Unexpected next to send (Ns) and next to receive (Nr) - * Not documented by IrLAP! - */ - if ((ns_status == NS_UNEXPECTED) && - (nr_status == NR_UNEXPECTED)) - { - pr_debug("%s(), unexpected nr and ns!\n", - __func__); - if (info->pf) { - /* Resend rejected frames */ - irlap_resend_rejected_frames(self, CMD_FRAME); - - /* Give peer some time to retransmit! - * But account for our own Tx. */ - irlap_start_final_timer(self, 2 * self->final_timeout); - - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_P); - } else { - /* Update Nr received */ - /* irlap_update_nr_received( info->nr); */ - - self->ack_required = FALSE; - } - break; - } - - /* - * Invalid NR or NS - */ - if ((nr_status == NR_INVALID) || (ns_status == NS_INVALID)) { - if (info->pf) { - del_timer(&self->final_timer); - - irlap_next_state(self, LAP_RESET_WAIT); - - irlap_disconnect_indication(self, LAP_RESET_INDICATION); - self->xmitflag = TRUE; - } else { - del_timer(&self->final_timer); - - irlap_disconnect_indication(self, LAP_RESET_INDICATION); - - self->xmitflag = FALSE; - } - break; - } - pr_debug("%s(), Not implemented!\n", __func__); - pr_debug("%s(), event=%s, ns_status=%d, nr_status=%d\n", - __func__, irlap_event[event], ns_status, nr_status); - break; - case RECV_UI_FRAME: - /* Poll bit cleared? */ - if (!info->pf) { - irlap_data_indication(self, skb, TRUE); - irlap_next_state(self, LAP_NRM_P); - } else { - del_timer(&self->final_timer); - irlap_data_indication(self, skb, TRUE); - irlap_next_state(self, LAP_XMIT_P); - pr_debug("%s: RECV_UI_FRAME: next state %s\n", - __func__, irlap_state[self->state]); - irlap_start_poll_timer(self, self->poll_timeout); - } - break; - case RECV_RR_RSP: - /* - * If you get a RR, the remote isn't busy anymore, - * no matter what the NR - */ - self->remote_busy = FALSE; - - /* Stop final timer */ - del_timer(&self->final_timer); - - /* - * Nr as expected? - */ - ret = irlap_validate_nr_received(self, info->nr); - if (ret == NR_EXPECTED) { - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - /* - * Got expected NR, so reset the retry_count. This - * is not done by the IrLAP standard , which is - * strange! DB. - */ - self->retry_count = 0; - irlap_wait_min_turn_around(self, &self->qos_tx); - - irlap_next_state(self, LAP_XMIT_P); - - /* Start poll timer */ - irlap_start_poll_timer(self, self->poll_timeout); - } else if (ret == NR_UNEXPECTED) { - IRDA_ASSERT(info != NULL, return -1;); - /* - * Unexpected nr! - */ - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - pr_debug("RECV_RR_FRAME: Retrans:%d, nr=%d, va=%d, vs=%d, vr=%d\n", - self->retry_count, info->nr, self->va, - self->vs, self->vr); - - /* Resend rejected frames */ - irlap_resend_rejected_frames(self, CMD_FRAME); - irlap_start_final_timer(self, self->final_timeout * 2); - - irlap_next_state(self, LAP_NRM_P); - } else if (ret == NR_INVALID) { - pr_debug("%s(), Received RR with invalid nr !\n", - __func__); - - irlap_next_state(self, LAP_RESET_WAIT); - - irlap_disconnect_indication(self, LAP_RESET_INDICATION); - self->xmitflag = TRUE; - } - break; - case RECV_RNR_RSP: - IRDA_ASSERT(info != NULL, return -1;); - - /* Stop final timer */ - del_timer(&self->final_timer); - self->remote_busy = TRUE; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - irlap_next_state(self, LAP_XMIT_P); - - /* Start poll timer */ - irlap_start_poll_timer(self, self->poll_timeout); - break; - case RECV_FRMR_RSP: - del_timer(&self->final_timer); - self->xmitflag = TRUE; - irlap_next_state(self, LAP_RESET_WAIT); - irlap_reset_indication(self); - break; - case FINAL_TIMER_EXPIRED: - /* - * We are allowed to wait for additional 300 ms if - * final timer expires when we are in the middle - * of receiving a frame (page 45, IrLAP). Check that - * we only do this once for each frame. - */ - if (irda_device_is_receiving(self->netdev) && !self->add_wait) { - pr_debug("FINAL_TIMER_EXPIRED when receiving a frame! Waiting a little bit more!\n"); - irlap_start_final_timer(self, msecs_to_jiffies(300)); - - /* - * Don't allow this to happen one more time in a row, - * or else we can get a pretty tight loop here if - * if we only receive half a frame. DB. - */ - self->add_wait = TRUE; - break; - } - self->add_wait = FALSE; - - /* N2 is the disconnect timer. Until we reach it, we retry */ - if (self->retry_count < self->N2) { - if (skb_peek(&self->wx_list) == NULL) { - /* Retry sending the pf bit to the secondary */ - pr_debug("nrm_p: resending rr"); - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, CMD_FRAME); - } else { - pr_debug("nrm_p: resend frames"); - irlap_resend_rejected_frames(self, CMD_FRAME); - } - - irlap_start_final_timer(self, self->final_timeout); - self->retry_count++; - pr_debug("irlap_state_nrm_p: FINAL_TIMER_EXPIRED: retry_count=%d\n", - self->retry_count); - - /* Early warning event. I'm using a pretty liberal - * interpretation of the spec and generate an event - * every time the timer is multiple of N1 (and not - * only the first time). This allow application - * to know precisely if connectivity restart... - * Jean II */ - if((self->retry_count % self->N1) == 0) - irlap_status_indication(self, - STATUS_NO_ACTIVITY); - - /* Keep state */ - } else { - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - irlap_disconnect_indication(self, LAP_NO_RESPONSE); - } - break; - case RECV_REJ_RSP: - irlap_update_nr_received(self, info->nr); - if (self->remote_busy) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, CMD_FRAME); - } else - irlap_resend_rejected_frames(self, CMD_FRAME); - irlap_start_final_timer(self, 2 * self->final_timeout); - break; - case RECV_SREJ_RSP: - irlap_update_nr_received(self, info->nr); - if (self->remote_busy) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, CMD_FRAME); - } else - irlap_resend_rejected_frame(self, CMD_FRAME); - irlap_start_final_timer(self, 2 * self->final_timeout); - break; - case RECV_RD_RSP: - pr_debug("%s(), RECV_RD_RSP\n", __func__); - - irlap_flush_all_queues(self); - irlap_next_state(self, LAP_XMIT_P); - /* Call back the LAP state machine to do a proper disconnect */ - irlap_disconnect_request(self); - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_reset_wait (event, skb, info) - * - * We have informed the service user of a reset condition, and is - * awaiting reset of disconnect request. - * - */ -static int irlap_state_reset_wait(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - pr_debug("%s(), event = %s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case RESET_REQUEST: - if (self->xmitflag) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_snrm_frame(self, NULL); - irlap_start_final_timer(self, self->final_timeout); - irlap_next_state(self, LAP_RESET); - } else { - irlap_start_final_timer(self, self->final_timeout); - irlap_next_state(self, LAP_RESET); - } - break; - case DISCONNECT_REQUEST: - irlap_wait_min_turn_around( self, &self->qos_tx); - irlap_send_disc_frame( self); - irlap_flush_all_queues( self); - irlap_start_final_timer( self, self->final_timeout); - self->retry_count = 0; - irlap_next_state( self, LAP_PCLOSE); - break; - default: - pr_debug("%s(), Unknown event %s\n", __func__, - irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_reset (self, event, skb, info) - * - * We have sent a SNRM reset command to the peer layer, and is awaiting - * reply. - * - */ -static int irlap_state_reset(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - pr_debug("%s(), event = %s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case RECV_DISC_CMD: - del_timer(&self->final_timer); - - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_NO_RESPONSE); - - break; - case RECV_UA_RSP: - del_timer(&self->final_timer); - - /* Initiate connection state */ - irlap_initiate_connection_state(self); - - irlap_reset_confirm(); - - self->remote_busy = FALSE; - - irlap_next_state(self, LAP_XMIT_P); - - irlap_start_poll_timer(self, self->poll_timeout); - - break; - case FINAL_TIMER_EXPIRED: - if (self->retry_count < 3) { - irlap_wait_min_turn_around(self, &self->qos_tx); - - IRDA_ASSERT(self->netdev != NULL, return -1;); - irlap_send_snrm_frame(self, self->qos_dev); - - self->retry_count++; /* Experimental!! */ - - irlap_start_final_timer(self, self->final_timeout); - irlap_next_state(self, LAP_RESET); - } else if (self->retry_count >= self->N3) { - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_disconnect_indication(self, LAP_NO_RESPONSE); - } - break; - case RECV_SNRM_CMD: - /* - * SNRM frame is not allowed to contain an I-field in this - * state - */ - if (!info) { - pr_debug("%s(), RECV_SNRM_CMD\n", __func__); - irlap_initiate_connection_state(self); - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_ua_response_frame(self, &self->qos_rx); - irlap_reset_confirm(); - irlap_start_wd_timer(self, self->wd_timeout); - irlap_next_state(self, LAP_NDM); - } else { - pr_debug("%s(), SNRM frame contained an I field!\n", - __func__); - } - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlap_event[event]); - - ret = -1; - break; - } - return ret; -} - -/* - * Function irlap_state_xmit_s (event, skb, info) - * - * XMIT_S, The secondary station has been given the right to transmit, - * and we therefore do not expect to receive any transmissions from other - * stations. - */ -static int irlap_state_xmit_s(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ret = 0; - - pr_debug("%s(), event=%s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -ENODEV;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -EBADR;); - - switch (event) { - case SEND_I_CMD: - /* - * Send frame only if send window > 0 - */ - if ((self->window > 0) && (!self->remote_busy)) { - int nextfit; -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - struct sk_buff *skb_next; - - /* - * Same deal as in irlap_state_xmit_p(), so see - * the comments at that point. - * We are the secondary, so there are only subtle - * differences. - Jean II - */ - - /* Check if a subsequent skb exist and would fit in - * the current window (with respect to turnaround - * time). - Jean II */ - skb_next = skb_peek(&self->txq); - nextfit = ((skb_next != NULL) && - ((skb_next->len + skb->len) <= - self->bytes_left)); - - /* - * Test if we have transmitted more bytes over the - * link than its possible to do with the current - * speed and turn-around-time. - */ - if((!nextfit) && (skb->len > self->bytes_left)) { - pr_debug("%s(), Not allowed to transmit more bytes!\n", - __func__); - /* Requeue the skb */ - skb_queue_head(&self->txq, skb_get(skb)); - - /* - * Switch to NRM_S, this is only possible - * when we are in secondary mode, since we - * must be sure that we don't miss any RR - * frames - */ - self->window = self->window_size; - self->bytes_left = self->line_capacity; - irlap_start_wd_timer(self, self->wd_timeout); - - irlap_next_state(self, LAP_NRM_S); - /* Slight difference with primary : - * here we would wait for the other side to - * expire the turnaround. - Jean II */ - - return -EPROTO; /* Try again later */ - } - /* Subtract space used by this skb */ - self->bytes_left -= skb->len; -#else /* CONFIG_IRDA_DYNAMIC_WINDOW */ - /* Window has been adjusted for the max packet - * size, so much simpler... - Jean II */ - nextfit = !skb_queue_empty(&self->txq); -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - /* - * Send data with final bit cleared only if window > 1 - * and there is more frames to be sent - */ - if ((self->window > 1) && (nextfit)) { - irlap_send_data_secondary(self, skb); - irlap_next_state(self, LAP_XMIT_S); - } else { - irlap_send_data_secondary_final(self, skb); - irlap_next_state(self, LAP_NRM_S); - - /* - * Make sure state machine does not try to send - * any more frames - */ - ret = -EPROTO; - } - } else { - pr_debug("%s(), Unable to send!\n", __func__); - skb_queue_head(&self->txq, skb_get(skb)); - ret = -EPROTO; - } - break; - case DISCONNECT_REQUEST: - irlap_send_rd_frame(self); - irlap_flush_all_queues(self); - irlap_start_wd_timer(self, self->wd_timeout); - irlap_next_state(self, LAP_SCLOSE); - break; - case DATA_REQUEST: - /* Nothing to do, irlap_do_event() will send the packet - * when we return... - Jean II */ - break; - default: - pr_debug("%s(), Unknown event %s\n", __func__, - irlap_event[event]); - - ret = -EINVAL; - break; - } - return ret; -} - -/* - * Function irlap_state_nrm_s (event, skb, info) - * - * NRM_S (Normal Response Mode as Secondary) state, in this state we are - * expecting to receive frames from the primary station - * - */ -static int irlap_state_nrm_s(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - int ns_status; - int nr_status; - int ret = 0; - - pr_debug("%s(), event=%s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - switch (event) { - case RECV_I_CMD: /* Optimize for the common case */ - /* FIXME: must check for remote_busy below */ - pr_debug("%s(), event=%s nr=%d, vs=%d, ns=%d, vr=%d, pf=%d\n", - __func__, irlap_event[event], info->nr, - self->vs, info->ns, self->vr, info->pf); - - self->retry_count = 0; - - ns_status = irlap_validate_ns_received(self, info->ns); - nr_status = irlap_validate_nr_received(self, info->nr); - /* - * Check for expected I(nformation) frame - */ - if ((ns_status == NS_EXPECTED) && (nr_status == NR_EXPECTED)) { - - /* Update Vr (next frame for us to receive) */ - self->vr = (self->vr + 1) % 8; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - /* - * poll bit cleared? - */ - if (!info->pf) { - - self->ack_required = TRUE; - - /* - * Starting WD-timer here is optional, but - * not recommended. Note 6 IrLAP p. 83 - */ -#if 0 - irda_start_timer(WD_TIMER, self->wd_timeout); -#endif - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_S); - - irlap_data_indication(self, skb, FALSE); - break; - } else { - /* - * We should wait before sending RR, and - * also before changing to XMIT_S - * state. (note 1, IrLAP p. 82) - */ - irlap_wait_min_turn_around(self, &self->qos_tx); - - /* - * Give higher layers a chance to - * immediately reply with some data before - * we decide if we should send a RR frame - * or not - */ - irlap_data_indication(self, skb, FALSE); - - /* Any pending data requests? */ - if (!skb_queue_empty(&self->txq) && - (self->window > 0)) - { - self->ack_required = TRUE; - - del_timer(&self->wd_timer); - - irlap_next_state(self, LAP_XMIT_S); - } else { - irlap_send_rr_frame(self, RSP_FRAME); - irlap_start_wd_timer(self, - self->wd_timeout); - - /* Keep the state */ - irlap_next_state(self, LAP_NRM_S); - } - break; - } - } - /* - * Check for Unexpected next to send (Ns) - */ - if ((ns_status == NS_UNEXPECTED) && (nr_status == NR_EXPECTED)) - { - /* Unexpected next to send, with final bit cleared */ - if (!info->pf) { - irlap_update_nr_received(self, info->nr); - - irlap_start_wd_timer(self, self->wd_timeout); - } else { - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, RSP_FRAME); - - irlap_start_wd_timer(self, self->wd_timeout); - } - break; - } - - /* - * Unexpected Next to Receive(NR) ? - */ - if ((ns_status == NS_EXPECTED) && (nr_status == NR_UNEXPECTED)) - { - if (info->pf) { - pr_debug("RECV_I_RSP: frame(s) lost\n"); - - self->vr = (self->vr + 1) % 8; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - /* Resend rejected frames */ - irlap_resend_rejected_frames(self, RSP_FRAME); - - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_S); - - irlap_data_indication(self, skb, FALSE); - irlap_start_wd_timer(self, self->wd_timeout); - break; - } - /* - * This is not documented in IrLAP!! Unexpected NR - * with poll bit cleared - */ - if (!info->pf) { - self->vr = (self->vr + 1) % 8; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - - /* Keep state, do not move this line */ - irlap_next_state(self, LAP_NRM_S); - - irlap_data_indication(self, skb, FALSE); - irlap_start_wd_timer(self, self->wd_timeout); - } - break; - } - - if (ret == NR_INVALID) { - pr_debug("NRM_S, NR_INVALID not implemented!\n"); - } - if (ret == NS_INVALID) { - pr_debug("NRM_S, NS_INVALID not implemented!\n"); - } - break; - case RECV_UI_FRAME: - /* - * poll bit cleared? - */ - if (!info->pf) { - irlap_data_indication(self, skb, TRUE); - irlap_next_state(self, LAP_NRM_S); /* Keep state */ - } else { - /* - * Any pending data requests? - */ - if (!skb_queue_empty(&self->txq) && - (self->window > 0) && !self->remote_busy) - { - irlap_data_indication(self, skb, TRUE); - - del_timer(&self->wd_timer); - - irlap_next_state(self, LAP_XMIT_S); - } else { - irlap_data_indication(self, skb, TRUE); - - irlap_wait_min_turn_around(self, &self->qos_tx); - - irlap_send_rr_frame(self, RSP_FRAME); - self->ack_required = FALSE; - - irlap_start_wd_timer(self, self->wd_timeout); - - /* Keep the state */ - irlap_next_state(self, LAP_NRM_S); - } - } - break; - case RECV_RR_CMD: - self->retry_count = 0; - - /* - * Nr as expected? - */ - nr_status = irlap_validate_nr_received(self, info->nr); - if (nr_status == NR_EXPECTED) { - if (!skb_queue_empty(&self->txq) && - (self->window > 0)) { - self->remote_busy = FALSE; - - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - del_timer(&self->wd_timer); - - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_next_state(self, LAP_XMIT_S); - } else { - self->remote_busy = FALSE; - /* Update Nr received */ - irlap_update_nr_received(self, info->nr); - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_start_wd_timer(self, self->wd_timeout); - - /* Note : if the link is idle (this case), - * we never go in XMIT_S, so we never get a - * chance to process any DISCONNECT_REQUEST. - * Do it now ! - Jean II */ - if (self->disconnect_pending) { - /* Disconnect */ - irlap_send_rd_frame(self); - irlap_flush_all_queues(self); - - irlap_next_state(self, LAP_SCLOSE); - } else { - /* Just send back pf bit */ - irlap_send_rr_frame(self, RSP_FRAME); - - irlap_next_state(self, LAP_NRM_S); - } - } - } else if (nr_status == NR_UNEXPECTED) { - self->remote_busy = FALSE; - irlap_update_nr_received(self, info->nr); - irlap_resend_rejected_frames(self, RSP_FRAME); - - irlap_start_wd_timer(self, self->wd_timeout); - - /* Keep state */ - irlap_next_state(self, LAP_NRM_S); - } else { - pr_debug("%s(), invalid nr not implemented!\n", - __func__); - } - break; - case RECV_SNRM_CMD: - /* SNRM frame is not allowed to contain an I-field */ - if (!info) { - del_timer(&self->wd_timer); - pr_debug("%s(), received SNRM cmd\n", __func__); - irlap_next_state(self, LAP_RESET_CHECK); - - irlap_reset_indication(self); - } else { - pr_debug("%s(), SNRM frame contained an I-field!\n", - __func__); - - } - break; - case RECV_REJ_CMD: - irlap_update_nr_received(self, info->nr); - if (self->remote_busy) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, RSP_FRAME); - } else - irlap_resend_rejected_frames(self, RSP_FRAME); - irlap_start_wd_timer(self, self->wd_timeout); - break; - case RECV_SREJ_CMD: - irlap_update_nr_received(self, info->nr); - if (self->remote_busy) { - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, RSP_FRAME); - } else - irlap_resend_rejected_frame(self, RSP_FRAME); - irlap_start_wd_timer(self, self->wd_timeout); - break; - case WD_TIMER_EXPIRED: - /* - * Wait until retry_count * n matches negotiated threshold/ - * disconnect time (note 2 in IrLAP p. 82) - * - * Similar to irlap_state_nrm_p() -> FINAL_TIMER_EXPIRED - * Note : self->wd_timeout = (self->final_timeout * 2), - * which explain why we use (self->N2 / 2) here !!! - * Jean II - */ - pr_debug("%s(), retry_count = %d\n", __func__, - self->retry_count); - - if (self->retry_count < (self->N2 / 2)) { - /* No retry, just wait for primary */ - irlap_start_wd_timer(self, self->wd_timeout); - self->retry_count++; - - if((self->retry_count % (self->N1 / 2)) == 0) - irlap_status_indication(self, - STATUS_NO_ACTIVITY); - } else { - irlap_apply_default_connection_parameters(self); - - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - irlap_disconnect_indication(self, LAP_NO_RESPONSE); - } - break; - case RECV_DISC_CMD: - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - /* Send disconnect response */ - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_ua_response_frame(self, NULL); - - del_timer(&self->wd_timer); - irlap_flush_all_queues(self); - /* Set default link parameters */ - irlap_apply_default_connection_parameters(self); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - case RECV_DISCOVERY_XID_CMD: - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rr_frame(self, RSP_FRAME); - self->ack_required = TRUE; - irlap_start_wd_timer(self, self->wd_timeout); - irlap_next_state(self, LAP_NRM_S); - - break; - case RECV_TEST_CMD: - /* Remove test frame header (only LAP header in NRM) */ - skb_pull(skb, LAP_ADDR_HEADER + LAP_CTRL_HEADER); - - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_start_wd_timer(self, self->wd_timeout); - - /* Send response (info will be copied) */ - irlap_send_test_frame(self, self->caddr, info->daddr, skb); - break; - default: - pr_debug("%s(), Unknown event %d, (%s)\n", __func__, - event, irlap_event[event]); - - ret = -EINVAL; - break; - } - return ret; -} - -/* - * Function irlap_state_sclose (self, event, skb, info) - */ -static int irlap_state_sclose(struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, struct irlap_info *info) -{ - IRDA_ASSERT(self != NULL, return -ENODEV;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -EBADR;); - - switch (event) { - case RECV_DISC_CMD: - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - /* Send disconnect response */ - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_ua_response_frame(self, NULL); - - del_timer(&self->wd_timer); - /* Set default link parameters */ - irlap_apply_default_connection_parameters(self); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - case RECV_DM_RSP: - /* IrLAP-1.1 p.82: in SCLOSE, S and I type RSP frames - * shall take us down into default NDM state, like DM_RSP - */ - case RECV_RR_RSP: - case RECV_RNR_RSP: - case RECV_REJ_RSP: - case RECV_SREJ_RSP: - case RECV_I_RSP: - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - del_timer(&self->wd_timer); - irlap_apply_default_connection_parameters(self); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - case WD_TIMER_EXPIRED: - /* Always switch state before calling upper layers */ - irlap_next_state(self, LAP_NDM); - - irlap_apply_default_connection_parameters(self); - - irlap_disconnect_indication(self, LAP_DISC_INDICATION); - break; - default: - /* IrLAP-1.1 p.82: in SCLOSE, basically any received frame - * with pf=1 shall restart the wd-timer and resend the rd:rsp - */ - if (info != NULL && info->pf) { - del_timer(&self->wd_timer); - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rd_frame(self); - irlap_start_wd_timer(self, self->wd_timeout); - break; /* stay in SCLOSE */ - } - - pr_debug("%s(), Unknown event %d, (%s)\n", __func__, - event, irlap_event[event]); - - break; - } - - return -1; -} - -static int irlap_state_reset_check( struct irlap_cb *self, IRLAP_EVENT event, - struct sk_buff *skb, - struct irlap_info *info) -{ - int ret = 0; - - pr_debug("%s(), event=%s\n", __func__, irlap_event[event]); - - IRDA_ASSERT(self != NULL, return -ENODEV;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -EBADR;); - - switch (event) { - case RESET_RESPONSE: - irlap_send_ua_response_frame(self, &self->qos_rx); - irlap_initiate_connection_state(self); - irlap_start_wd_timer(self, WD_TIMEOUT); - irlap_flush_all_queues(self); - - irlap_next_state(self, LAP_NRM_S); - break; - case DISCONNECT_REQUEST: - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_send_rd_frame(self); - irlap_start_wd_timer(self, WD_TIMEOUT); - irlap_next_state(self, LAP_SCLOSE); - break; - default: - pr_debug("%s(), Unknown event %d, (%s)\n", __func__, - event, irlap_event[event]); - - ret = -EINVAL; - break; - } - return ret; -} diff --git a/net/irda/irlap_frame.c b/net/irda/irlap_frame.c deleted file mode 100644 index debda3de4726..000000000000 --- a/net/irda/irlap_frame.c +++ /dev/null @@ -1,1407 +0,0 @@ -/********************************************************************* - * - * Filename: irlap_frame.c - * Version: 1.0 - * Description: Build and transmit IrLAP frames - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Aug 19 10:27:26 1997 - * Modified at: Wed Jan 5 08:59:04 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> -#include <linux/if.h> -#include <linux/if_ether.h> -#include <linux/netdevice.h> -#include <linux/irda.h> -#include <linux/slab.h> - -#include <net/pkt_sched.h> -#include <net/sock.h> - -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/irda_device.h> -#include <net/irda/irlap.h> -#include <net/irda/wrapper.h> -#include <net/irda/timer.h> -#include <net/irda/irlap_frame.h> -#include <net/irda/qos.h> - -static void irlap_send_i_frame(struct irlap_cb *self, struct sk_buff *skb, - int command); - -/* - * Function irlap_insert_info (self, skb) - * - * Insert minimum turnaround time and speed information into the skb. We - * need to do this since it's per packet relevant information. Safe to - * have this function inlined since it's only called from one place - */ -static inline void irlap_insert_info(struct irlap_cb *self, - struct sk_buff *skb) -{ - struct irda_skb_cb *cb = (struct irda_skb_cb *) skb->cb; - - /* - * Insert MTT (min. turn time) and speed into skb, so that the - * device driver knows which settings to use - */ - cb->magic = LAP_MAGIC; - cb->mtt = self->mtt_required; - cb->next_speed = self->speed; - - /* Reset */ - self->mtt_required = 0; - - /* - * Delay equals negotiated BOFs count, plus the number of BOFs to - * force the negotiated minimum turnaround time - */ - cb->xbofs = self->bofs_count; - cb->next_xbofs = self->next_bofs; - cb->xbofs_delay = self->xbofs_delay; - - /* Reset XBOF's delay (used only for getting min turn time) */ - self->xbofs_delay = 0; - /* Put the correct xbofs value for the next packet */ - self->bofs_count = self->next_bofs; -} - -/* - * Function irlap_queue_xmit (self, skb) - * - * A little wrapper for dev_queue_xmit, so we can insert some common - * code into it. - */ -void irlap_queue_xmit(struct irlap_cb *self, struct sk_buff *skb) -{ - /* Some common init stuff */ - skb->dev = self->netdev; - skb_reset_mac_header(skb); - skb_reset_network_header(skb); - skb_reset_transport_header(skb); - skb->protocol = htons(ETH_P_IRDA); - skb->priority = TC_PRIO_BESTEFFORT; - - irlap_insert_info(self, skb); - - if (unlikely(self->mode & IRDA_MODE_MONITOR)) { - pr_debug("%s(): %s is in monitor mode\n", __func__, - self->netdev->name); - dev_kfree_skb(skb); - return; - } - - dev_queue_xmit(skb); -} - -/* - * Function irlap_send_snrm_cmd (void) - * - * Transmits a connect SNRM command frame - */ -void irlap_send_snrm_frame(struct irlap_cb *self, struct qos_info *qos) -{ - struct sk_buff *tx_skb; - struct snrm_frame *frame; - int ret; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Allocate frame */ - tx_skb = alloc_skb(sizeof(struct snrm_frame) + - IRLAP_NEGOCIATION_PARAMS_LEN, - GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 2); - - /* Insert connection address field */ - if (qos) - frame->caddr = CMD_FRAME | CBROADCAST; - else - frame->caddr = CMD_FRAME | self->caddr; - - /* Insert control field */ - frame->control = SNRM_CMD | PF_BIT; - - /* - * If we are establishing a connection then insert QoS parameters - */ - if (qos) { - skb_put(tx_skb, 9); /* 25 left */ - frame->saddr = cpu_to_le32(self->saddr); - frame->daddr = cpu_to_le32(self->daddr); - - frame->ncaddr = self->caddr; - - ret = irlap_insert_qos_negotiation_params(self, tx_skb); - if (ret < 0) { - dev_kfree_skb(tx_skb); - return; - } - } - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_recv_snrm_cmd (skb, info) - * - * Received SNRM (Set Normal Response Mode) command frame - * - */ -static void irlap_recv_snrm_cmd(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info) -{ - struct snrm_frame *frame; - - if (pskb_may_pull(skb,sizeof(struct snrm_frame))) { - frame = (struct snrm_frame *) skb->data; - - /* Copy the new connection address ignoring the C/R bit */ - info->caddr = frame->ncaddr & 0xFE; - - /* Check if the new connection address is valid */ - if ((info->caddr == 0x00) || (info->caddr == 0xfe)) { - pr_debug("%s(), invalid connection address!\n", - __func__); - return; - } - - /* Copy peer device address */ - info->daddr = le32_to_cpu(frame->saddr); - info->saddr = le32_to_cpu(frame->daddr); - - /* Only accept if addressed directly to us */ - if (info->saddr != self->saddr) { - pr_debug("%s(), not addressed to us!\n", - __func__); - return; - } - irlap_do_event(self, RECV_SNRM_CMD, skb, info); - } else { - /* Signal that this SNRM frame does not contain and I-field */ - irlap_do_event(self, RECV_SNRM_CMD, skb, NULL); - } -} - -/* - * Function irlap_send_ua_response_frame (qos) - * - * Send UA (Unnumbered Acknowledgement) frame - * - */ -void irlap_send_ua_response_frame(struct irlap_cb *self, struct qos_info *qos) -{ - struct sk_buff *tx_skb; - struct ua_frame *frame; - int ret; - - pr_debug("%s() <%ld>\n", __func__, jiffies); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Allocate frame */ - tx_skb = alloc_skb(sizeof(struct ua_frame) + - IRLAP_NEGOCIATION_PARAMS_LEN, - GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 10); - - /* Build UA response */ - frame->caddr = self->caddr; - frame->control = UA_RSP | PF_BIT; - - frame->saddr = cpu_to_le32(self->saddr); - frame->daddr = cpu_to_le32(self->daddr); - - /* Should we send QoS negotiation parameters? */ - if (qos) { - ret = irlap_insert_qos_negotiation_params(self, tx_skb); - if (ret < 0) { - dev_kfree_skb(tx_skb); - return; - } - } - - irlap_queue_xmit(self, tx_skb); -} - - -/* - * Function irlap_send_dm_frame (void) - * - * Send disconnected mode (DM) frame - * - */ -void irlap_send_dm_frame( struct irlap_cb *self) -{ - struct sk_buff *tx_skb = NULL; - struct dm_frame *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - tx_skb = alloc_skb(sizeof(struct dm_frame), GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 2); - - if (self->state == LAP_NDM) - frame->caddr = CBROADCAST; - else - frame->caddr = self->caddr; - - frame->control = DM_RSP | PF_BIT; - - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_send_disc_frame (void) - * - * Send disconnect (DISC) frame - * - */ -void irlap_send_disc_frame(struct irlap_cb *self) -{ - struct sk_buff *tx_skb = NULL; - struct disc_frame *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - tx_skb = alloc_skb(sizeof(struct disc_frame), GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 2); - - frame->caddr = self->caddr | CMD_FRAME; - frame->control = DISC_CMD | PF_BIT; - - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_send_discovery_xid_frame (S, s, command) - * - * Build and transmit a XID (eXchange station IDentifier) discovery - * frame. - */ -void irlap_send_discovery_xid_frame(struct irlap_cb *self, int S, __u8 s, - __u8 command, discovery_t *discovery) -{ - struct sk_buff *tx_skb = NULL; - struct xid_frame *frame; - __u32 bcast = BROADCAST; - __u8 *info; - - pr_debug("%s(), s=%d, S=%d, command=%d\n", __func__, - s, S, command); - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(discovery != NULL, return;); - - tx_skb = alloc_skb(sizeof(struct xid_frame) + IRLAP_DISCOVERY_INFO_LEN, - GFP_ATOMIC); - if (!tx_skb) - return; - - skb_put(tx_skb, 14); - frame = (struct xid_frame *) tx_skb->data; - - if (command) { - frame->caddr = CBROADCAST | CMD_FRAME; - frame->control = XID_CMD | PF_BIT; - } else { - frame->caddr = CBROADCAST; - frame->control = XID_RSP | PF_BIT; - } - frame->ident = XID_FORMAT; - - frame->saddr = cpu_to_le32(self->saddr); - - if (command) - frame->daddr = cpu_to_le32(bcast); - else - frame->daddr = cpu_to_le32(discovery->data.daddr); - - switch (S) { - case 1: - frame->flags = 0x00; - break; - case 6: - frame->flags = 0x01; - break; - case 8: - frame->flags = 0x02; - break; - case 16: - frame->flags = 0x03; - break; - default: - frame->flags = 0x02; - break; - } - - frame->slotnr = s; - frame->version = 0x00; - - /* - * Provide info for final slot only in commands, and for all - * responses. Send the second byte of the hint only if the - * EXTENSION bit is set in the first byte. - */ - if (!command || (frame->slotnr == 0xff)) { - int len; - - if (discovery->data.hints[0] & HINT_EXTENSION) { - info = skb_put(tx_skb, 2); - info[0] = discovery->data.hints[0]; - info[1] = discovery->data.hints[1]; - } else { - info = skb_put(tx_skb, 1); - info[0] = discovery->data.hints[0]; - } - info = skb_put(tx_skb, 1); - info[0] = discovery->data.charset; - - len = IRDA_MIN(discovery->name_len, skb_tailroom(tx_skb)); - skb_put_data(tx_skb, discovery->data.info, len); - } - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_recv_discovery_xid_rsp (skb, info) - * - * Received a XID discovery response - * - */ -static void irlap_recv_discovery_xid_rsp(struct irlap_cb *self, - struct sk_buff *skb, - struct irlap_info *info) -{ - struct xid_frame *xid; - discovery_t *discovery = NULL; - __u8 *discovery_info; - char *text; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - if (!pskb_may_pull(skb, sizeof(struct xid_frame))) { - net_err_ratelimited("%s: frame too short!\n", __func__); - return; - } - - xid = (struct xid_frame *) skb->data; - - info->daddr = le32_to_cpu(xid->saddr); - info->saddr = le32_to_cpu(xid->daddr); - - /* Make sure frame is addressed to us */ - if ((info->saddr != self->saddr) && (info->saddr != BROADCAST)) { - pr_debug("%s(), frame is not addressed to us!\n", - __func__); - return; - } - - if ((discovery = kzalloc(sizeof(discovery_t), GFP_ATOMIC)) == NULL) { - net_warn_ratelimited("%s: kmalloc failed!\n", __func__); - return; - } - - discovery->data.daddr = info->daddr; - discovery->data.saddr = self->saddr; - discovery->timestamp = jiffies; - - pr_debug("%s(), daddr=%08x\n", __func__, - discovery->data.daddr); - - discovery_info = skb_pull(skb, sizeof(struct xid_frame)); - - /* Get info returned from peer */ - discovery->data.hints[0] = discovery_info[0]; - if (discovery_info[0] & HINT_EXTENSION) { - pr_debug("EXTENSION\n"); - discovery->data.hints[1] = discovery_info[1]; - discovery->data.charset = discovery_info[2]; - text = (char *) &discovery_info[3]; - } else { - discovery->data.hints[1] = 0; - discovery->data.charset = discovery_info[1]; - text = (char *) &discovery_info[2]; - } - /* - * Terminate info string, should be safe since this is where the - * FCS bytes resides. - */ - skb->data[skb->len] = '\0'; - strncpy(discovery->data.info, text, NICKNAME_MAX_LEN); - discovery->name_len = strlen(discovery->data.info); - - info->discovery = discovery; - - irlap_do_event(self, RECV_DISCOVERY_XID_RSP, skb, info); -} - -/* - * Function irlap_recv_discovery_xid_cmd (skb, info) - * - * Received a XID discovery command - * - */ -static void irlap_recv_discovery_xid_cmd(struct irlap_cb *self, - struct sk_buff *skb, - struct irlap_info *info) -{ - struct xid_frame *xid; - discovery_t *discovery = NULL; - __u8 *discovery_info; - char *text; - - if (!pskb_may_pull(skb, sizeof(struct xid_frame))) { - net_err_ratelimited("%s: frame too short!\n", __func__); - return; - } - - xid = (struct xid_frame *) skb->data; - - info->daddr = le32_to_cpu(xid->saddr); - info->saddr = le32_to_cpu(xid->daddr); - - /* Make sure frame is addressed to us */ - if ((info->saddr != self->saddr) && (info->saddr != BROADCAST)) { - pr_debug("%s(), frame is not addressed to us!\n", - __func__); - return; - } - - switch (xid->flags & 0x03) { - case 0x00: - info->S = 1; - break; - case 0x01: - info->S = 6; - break; - case 0x02: - info->S = 8; - break; - case 0x03: - info->S = 16; - break; - default: - /* Error!! */ - return; - } - info->s = xid->slotnr; - - discovery_info = skb_pull(skb, sizeof(struct xid_frame)); - - /* - * Check if last frame - */ - if (info->s == 0xff) { - /* Check if things are sane at this point... */ - if((discovery_info == NULL) || - !pskb_may_pull(skb, 3)) { - net_err_ratelimited("%s: discovery frame too short!\n", - __func__); - return; - } - - /* - * We now have some discovery info to deliver! - */ - discovery = kzalloc(sizeof(discovery_t), GFP_ATOMIC); - if (!discovery) - return; - - discovery->data.daddr = info->daddr; - discovery->data.saddr = self->saddr; - discovery->timestamp = jiffies; - - discovery->data.hints[0] = discovery_info[0]; - if (discovery_info[0] & HINT_EXTENSION) { - discovery->data.hints[1] = discovery_info[1]; - discovery->data.charset = discovery_info[2]; - text = (char *) &discovery_info[3]; - } else { - discovery->data.hints[1] = 0; - discovery->data.charset = discovery_info[1]; - text = (char *) &discovery_info[2]; - } - /* - * Terminate string, should be safe since this is where the - * FCS bytes resides. - */ - skb->data[skb->len] = '\0'; - strncpy(discovery->data.info, text, NICKNAME_MAX_LEN); - discovery->name_len = strlen(discovery->data.info); - - info->discovery = discovery; - } else - info->discovery = NULL; - - irlap_do_event(self, RECV_DISCOVERY_XID_CMD, skb, info); -} - -/* - * Function irlap_send_rr_frame (self, command) - * - * Build and transmit RR (Receive Ready) frame. Notice that it is currently - * only possible to send RR frames with the poll bit set. - */ -void irlap_send_rr_frame(struct irlap_cb *self, int command) -{ - struct sk_buff *tx_skb; - struct rr_frame *frame; - - tx_skb = alloc_skb(sizeof(struct rr_frame), GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 2); - - frame->caddr = self->caddr; - frame->caddr |= (command) ? CMD_FRAME : 0; - - frame->control = RR | PF_BIT | (self->vr << 5); - - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_send_rd_frame (self) - * - * Request disconnect. Used by a secondary station to request the - * disconnection of the link. - */ -void irlap_send_rd_frame(struct irlap_cb *self) -{ - struct sk_buff *tx_skb; - struct rd_frame *frame; - - tx_skb = alloc_skb(sizeof(struct rd_frame), GFP_ATOMIC); - if (!tx_skb) - return; - - frame = skb_put(tx_skb, 2); - - frame->caddr = self->caddr; - frame->control = RD_RSP | PF_BIT; - - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_recv_rr_frame (skb, info) - * - * Received RR (Receive Ready) frame from peer station, no harm in - * making it inline since its called only from one single place - * (irlap_driver_rcv). - */ -static inline void irlap_recv_rr_frame(struct irlap_cb *self, - struct sk_buff *skb, - struct irlap_info *info, int command) -{ - info->nr = skb->data[1] >> 5; - - /* Check if this is a command or a response frame */ - if (command) - irlap_do_event(self, RECV_RR_CMD, skb, info); - else - irlap_do_event(self, RECV_RR_RSP, skb, info); -} - -/* - * Function irlap_recv_rnr_frame (self, skb, info) - * - * Received RNR (Receive Not Ready) frame from peer station - * - */ -static void irlap_recv_rnr_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info, int command) -{ - info->nr = skb->data[1] >> 5; - - pr_debug("%s(), nr=%d, %ld\n", __func__, info->nr, jiffies); - - if (command) - irlap_do_event(self, RECV_RNR_CMD, skb, info); - else - irlap_do_event(self, RECV_RNR_RSP, skb, info); -} - -static void irlap_recv_rej_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info, int command) -{ - info->nr = skb->data[1] >> 5; - - /* Check if this is a command or a response frame */ - if (command) - irlap_do_event(self, RECV_REJ_CMD, skb, info); - else - irlap_do_event(self, RECV_REJ_RSP, skb, info); -} - -static void irlap_recv_srej_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info, int command) -{ - info->nr = skb->data[1] >> 5; - - /* Check if this is a command or a response frame */ - if (command) - irlap_do_event(self, RECV_SREJ_CMD, skb, info); - else - irlap_do_event(self, RECV_SREJ_RSP, skb, info); -} - -static void irlap_recv_disc_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info, int command) -{ - /* Check if this is a command or a response frame */ - if (command) - irlap_do_event(self, RECV_DISC_CMD, skb, info); - else - irlap_do_event(self, RECV_RD_RSP, skb, info); -} - -/* - * Function irlap_recv_ua_frame (skb, frame) - * - * Received UA (Unnumbered Acknowledgement) frame - * - */ -static inline void irlap_recv_ua_frame(struct irlap_cb *self, - struct sk_buff *skb, - struct irlap_info *info) -{ - irlap_do_event(self, RECV_UA_RSP, skb, info); -} - -/* - * Function irlap_send_data_primary(self, skb) - * - * Send I-frames as the primary station but without the poll bit set - * - */ -void irlap_send_data_primary(struct irlap_cb *self, struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - - if (skb->data[1] == I_FRAME) { - - /* - * Insert frame sequence number (Vs) in control field before - * inserting into transmit window queue. - */ - skb->data[1] = I_FRAME | (self->vs << 1); - - /* - * Insert frame in store, in case of retransmissions - * Increase skb reference count, see irlap_do_event() - */ - skb_get(skb); - skb_queue_tail(&self->wx_list, skb); - - /* Copy buffer */ - tx_skb = skb_clone(skb, GFP_ATOMIC); - if (tx_skb == NULL) { - return; - } - - self->vs = (self->vs + 1) % 8; - self->ack_required = FALSE; - self->window -= 1; - - irlap_send_i_frame( self, tx_skb, CMD_FRAME); - } else { - pr_debug("%s(), sending unreliable frame\n", __func__); - irlap_send_ui_frame(self, skb_get(skb), self->caddr, CMD_FRAME); - self->window -= 1; - } -} -/* - * Function irlap_send_data_primary_poll (self, skb) - * - * Send I(nformation) frame as primary with poll bit set - */ -void irlap_send_data_primary_poll(struct irlap_cb *self, struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - int transmission_time; - - /* Stop P timer */ - del_timer(&self->poll_timer); - - /* Is this reliable or unreliable data? */ - if (skb->data[1] == I_FRAME) { - - /* - * Insert frame sequence number (Vs) in control field before - * inserting into transmit window queue. - */ - skb->data[1] = I_FRAME | (self->vs << 1); - - /* - * Insert frame in store, in case of retransmissions - * Increase skb reference count, see irlap_do_event() - */ - skb_get(skb); - skb_queue_tail(&self->wx_list, skb); - - /* Copy buffer */ - tx_skb = skb_clone(skb, GFP_ATOMIC); - if (tx_skb == NULL) { - return; - } - - /* - * Set poll bit if necessary. We do this to the copied - * skb, since retransmitted need to set or clear the poll - * bit depending on when they are sent. - */ - tx_skb->data[1] |= PF_BIT; - - self->vs = (self->vs + 1) % 8; - self->ack_required = FALSE; - - irlap_next_state(self, LAP_NRM_P); - irlap_send_i_frame(self, tx_skb, CMD_FRAME); - } else { - pr_debug("%s(), sending unreliable frame\n", __func__); - - if (self->ack_required) { - irlap_send_ui_frame(self, skb_get(skb), self->caddr, CMD_FRAME); - irlap_next_state(self, LAP_NRM_P); - irlap_send_rr_frame(self, CMD_FRAME); - self->ack_required = FALSE; - } else { - skb->data[1] |= PF_BIT; - irlap_next_state(self, LAP_NRM_P); - irlap_send_ui_frame(self, skb_get(skb), self->caddr, CMD_FRAME); - } - } - - /* How much time we took for transmission of all frames. - * We don't know, so let assume we used the full window. Jean II */ - transmission_time = self->final_timeout; - - /* Reset parameter so that we can fill next window */ - self->window = self->window_size; - -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - /* Remove what we have not used. Just do a prorata of the - * bytes left in window to window capacity. - * See max_line_capacities[][] in qos.c for details. Jean II */ - transmission_time -= (self->final_timeout * self->bytes_left - / self->line_capacity); - pr_debug("%s() adjusting transmission_time : ft=%d, bl=%d, lc=%d -> tt=%d\n", - __func__, self->final_timeout, self->bytes_left, - self->line_capacity, transmission_time); - - /* We are allowed to transmit a maximum number of bytes again. */ - self->bytes_left = self->line_capacity; -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - - /* - * The network layer has a intermediate buffer between IrLAP - * and the IrDA driver which can contain 8 frames. So, even - * though IrLAP is currently sending the *last* frame of the - * tx-window, the driver most likely has only just started - * sending the *first* frame of the same tx-window. - * I.e. we are always at the very beginning of or Tx window. - * Now, we are supposed to set the final timer from the end - * of our tx-window to let the other peer reply. So, we need - * to add extra time to compensate for the fact that we - * are really at the start of tx-window, otherwise the final timer - * might expire before he can answer... - * Jean II - */ - irlap_start_final_timer(self, self->final_timeout + transmission_time); - - /* - * The clever amongst you might ask why we do this adjustement - * only here, and not in all the other cases in irlap_event.c. - * In all those other case, we only send a very short management - * frame (few bytes), so the adjustement would be lost in the - * noise... - * The exception of course is irlap_resend_rejected_frame(). - * Jean II */ -} - -/* - * Function irlap_send_data_secondary_final (self, skb) - * - * Send I(nformation) frame as secondary with final bit set - * - */ -void irlap_send_data_secondary_final(struct irlap_cb *self, - struct sk_buff *skb) -{ - struct sk_buff *tx_skb = NULL; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Is this reliable or unreliable data? */ - if (skb->data[1] == I_FRAME) { - - /* - * Insert frame sequence number (Vs) in control field before - * inserting into transmit window queue. - */ - skb->data[1] = I_FRAME | (self->vs << 1); - - /* - * Insert frame in store, in case of retransmissions - * Increase skb reference count, see irlap_do_event() - */ - skb_get(skb); - skb_queue_tail(&self->wx_list, skb); - - tx_skb = skb_clone(skb, GFP_ATOMIC); - if (tx_skb == NULL) { - return; - } - - tx_skb->data[1] |= PF_BIT; - - self->vs = (self->vs + 1) % 8; - self->ack_required = FALSE; - - irlap_send_i_frame(self, tx_skb, RSP_FRAME); - } else { - if (self->ack_required) { - irlap_send_ui_frame(self, skb_get(skb), self->caddr, RSP_FRAME); - irlap_send_rr_frame(self, RSP_FRAME); - self->ack_required = FALSE; - } else { - skb->data[1] |= PF_BIT; - irlap_send_ui_frame(self, skb_get(skb), self->caddr, RSP_FRAME); - } - } - - self->window = self->window_size; -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - /* We are allowed to transmit a maximum number of bytes again. */ - self->bytes_left = self->line_capacity; -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - - irlap_start_wd_timer(self, self->wd_timeout); -} - -/* - * Function irlap_send_data_secondary (self, skb) - * - * Send I(nformation) frame as secondary without final bit set - * - */ -void irlap_send_data_secondary(struct irlap_cb *self, struct sk_buff *skb) -{ - struct sk_buff *tx_skb = NULL; - - /* Is this reliable or unreliable data? */ - if (skb->data[1] == I_FRAME) { - - /* - * Insert frame sequence number (Vs) in control field before - * inserting into transmit window queue. - */ - skb->data[1] = I_FRAME | (self->vs << 1); - - /* - * Insert frame in store, in case of retransmissions - * Increase skb reference count, see irlap_do_event() - */ - skb_get(skb); - skb_queue_tail(&self->wx_list, skb); - - tx_skb = skb_clone(skb, GFP_ATOMIC); - if (tx_skb == NULL) { - return; - } - - self->vs = (self->vs + 1) % 8; - self->ack_required = FALSE; - self->window -= 1; - - irlap_send_i_frame(self, tx_skb, RSP_FRAME); - } else { - irlap_send_ui_frame(self, skb_get(skb), self->caddr, RSP_FRAME); - self->window -= 1; - } -} - -/* - * Function irlap_resend_rejected_frames (nr) - * - * Resend frames which has not been acknowledged. Should be safe to - * traverse the list without locking it since this function will only be - * called from interrupt context (BH) - */ -void irlap_resend_rejected_frames(struct irlap_cb *self, int command) -{ - struct sk_buff *tx_skb; - struct sk_buff *skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Resend unacknowledged frame(s) */ - skb_queue_walk(&self->wx_list, skb) { - irlap_wait_min_turn_around(self, &self->qos_tx); - - /* We copy the skb to be retransmitted since we will have to - * modify it. Cloning will confuse packet sniffers - */ - /* tx_skb = skb_clone( skb, GFP_ATOMIC); */ - tx_skb = skb_copy(skb, GFP_ATOMIC); - if (!tx_skb) { - pr_debug("%s(), unable to copy\n", __func__); - return; - } - - /* Clear old Nr field + poll bit */ - tx_skb->data[1] &= 0x0f; - - /* - * Set poll bit on the last frame retransmitted - */ - if (skb_queue_is_last(&self->wx_list, skb)) - tx_skb->data[1] |= PF_BIT; /* Set p/f bit */ - else - tx_skb->data[1] &= ~PF_BIT; /* Clear p/f bit */ - - irlap_send_i_frame(self, tx_skb, command); - } -#if 0 /* Not yet */ - /* - * We can now fill the window with additional data frames - */ - while (!skb_queue_empty(&self->txq)) { - - pr_debug("%s(), sending additional frames!\n", __func__); - if (self->window > 0) { - skb = skb_dequeue( &self->txq); - IRDA_ASSERT(skb != NULL, return;); - - /* - * If send window > 1 then send frame with pf - * bit cleared - */ - if ((self->window > 1) && - !skb_queue_empty(&self->txq)) { - irlap_send_data_primary(self, skb); - } else { - irlap_send_data_primary_poll(self, skb); - } - kfree_skb(skb); - } - } -#endif -} - -void irlap_resend_rejected_frame(struct irlap_cb *self, int command) -{ - struct sk_buff *tx_skb; - struct sk_buff *skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - /* Resend unacknowledged frame(s) */ - skb = skb_peek(&self->wx_list); - if (skb != NULL) { - irlap_wait_min_turn_around(self, &self->qos_tx); - - /* We copy the skb to be retransmitted since we will have to - * modify it. Cloning will confuse packet sniffers - */ - /* tx_skb = skb_clone( skb, GFP_ATOMIC); */ - tx_skb = skb_copy(skb, GFP_ATOMIC); - if (!tx_skb) { - pr_debug("%s(), unable to copy\n", __func__); - return; - } - - /* Clear old Nr field + poll bit */ - tx_skb->data[1] &= 0x0f; - - /* Set poll/final bit */ - tx_skb->data[1] |= PF_BIT; /* Set p/f bit */ - - irlap_send_i_frame(self, tx_skb, command); - } -} - -/* - * Function irlap_send_ui_frame (self, skb, command) - * - * Contruct and transmit an Unnumbered Information (UI) frame - * - */ -void irlap_send_ui_frame(struct irlap_cb *self, struct sk_buff *skb, - __u8 caddr, int command) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Insert connection address */ - skb->data[0] = caddr | ((command) ? CMD_FRAME : 0); - - irlap_queue_xmit(self, skb); -} - -/* - * Function irlap_send_i_frame (skb) - * - * Contruct and transmit Information (I) frame - */ -static void irlap_send_i_frame(struct irlap_cb *self, struct sk_buff *skb, - int command) -{ - /* Insert connection address */ - skb->data[0] = self->caddr; - skb->data[0] |= (command) ? CMD_FRAME : 0; - - /* Insert next to receive (Vr) */ - skb->data[1] |= (self->vr << 5); /* insert nr */ - - irlap_queue_xmit(self, skb); -} - -/* - * Function irlap_recv_i_frame (skb, frame) - * - * Receive and parse an I (Information) frame, no harm in making it inline - * since it's called only from one single place (irlap_driver_rcv). - */ -static inline void irlap_recv_i_frame(struct irlap_cb *self, - struct sk_buff *skb, - struct irlap_info *info, int command) -{ - info->nr = skb->data[1] >> 5; /* Next to receive */ - info->pf = skb->data[1] & PF_BIT; /* Final bit */ - info->ns = (skb->data[1] >> 1) & 0x07; /* Next to send */ - - /* Check if this is a command or a response frame */ - if (command) - irlap_do_event(self, RECV_I_CMD, skb, info); - else - irlap_do_event(self, RECV_I_RSP, skb, info); -} - -/* - * Function irlap_recv_ui_frame (self, skb, info) - * - * Receive and parse an Unnumbered Information (UI) frame - * - */ -static void irlap_recv_ui_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info) -{ - info->pf = skb->data[1] & PF_BIT; /* Final bit */ - - irlap_do_event(self, RECV_UI_FRAME, skb, info); -} - -/* - * Function irlap_recv_frmr_frame (skb, frame) - * - * Received Frame Reject response. - * - */ -static void irlap_recv_frmr_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info) -{ - __u8 *frame; - int w, x, y, z; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(info != NULL, return;); - - if (!pskb_may_pull(skb, 4)) { - net_err_ratelimited("%s: frame too short!\n", __func__); - return; - } - - frame = skb->data; - - info->nr = frame[2] >> 5; /* Next to receive */ - info->pf = frame[2] & PF_BIT; /* Final bit */ - info->ns = (frame[2] >> 1) & 0x07; /* Next to send */ - - w = frame[3] & 0x01; - x = frame[3] & 0x02; - y = frame[3] & 0x04; - z = frame[3] & 0x08; - - if (w) { - pr_debug("Rejected control field is undefined or not implemented\n"); - } - if (x) { - pr_debug("Rejected control field was invalid because it contained a non permitted I field\n"); - } - if (y) { - pr_debug("Received I field exceeded the maximum negotiated for the existing connection or exceeded the maximum this station supports if no connection exists\n"); - } - if (z) { - pr_debug("Rejected control field control field contained an invalid Nr count\n"); - } - irlap_do_event(self, RECV_FRMR_RSP, skb, info); -} - -/* - * Function irlap_send_test_frame (self, daddr) - * - * Send a test frame response - * - */ -void irlap_send_test_frame(struct irlap_cb *self, __u8 caddr, __u32 daddr, - struct sk_buff *cmd) -{ - struct sk_buff *tx_skb; - struct test_frame *frame; - - tx_skb = alloc_skb(cmd->len + sizeof(struct test_frame), GFP_ATOMIC); - if (!tx_skb) - return; - - /* Broadcast frames must include saddr and daddr fields */ - if (caddr == CBROADCAST) { - frame = skb_put(tx_skb, sizeof(struct test_frame)); - - /* Insert the swapped addresses */ - frame->saddr = cpu_to_le32(self->saddr); - frame->daddr = cpu_to_le32(daddr); - } else - frame = skb_put(tx_skb, LAP_ADDR_HEADER + LAP_CTRL_HEADER); - - frame->caddr = caddr; - frame->control = TEST_RSP | PF_BIT; - - /* Copy info */ - skb_put_data(tx_skb, cmd->data, cmd->len); - - /* Return to sender */ - irlap_wait_min_turn_around(self, &self->qos_tx); - irlap_queue_xmit(self, tx_skb); -} - -/* - * Function irlap_recv_test_frame (self, skb) - * - * Receive a test frame - * - */ -static void irlap_recv_test_frame(struct irlap_cb *self, struct sk_buff *skb, - struct irlap_info *info, int command) -{ - struct test_frame *frame; - - if (!pskb_may_pull(skb, sizeof(*frame))) { - net_err_ratelimited("%s: frame too short!\n", __func__); - return; - } - frame = (struct test_frame *) skb->data; - - /* Broadcast frames must carry saddr and daddr fields */ - if (info->caddr == CBROADCAST) { - if (skb->len < sizeof(struct test_frame)) { - pr_debug("%s() test frame too short!\n", - __func__); - return; - } - - /* Read and swap addresses */ - info->daddr = le32_to_cpu(frame->saddr); - info->saddr = le32_to_cpu(frame->daddr); - - /* Make sure frame is addressed to us */ - if ((info->saddr != self->saddr) && - (info->saddr != BROADCAST)) { - return; - } - } - - if (command) - irlap_do_event(self, RECV_TEST_CMD, skb, info); - else - irlap_do_event(self, RECV_TEST_RSP, skb, info); -} - -/* - * Function irlap_driver_rcv (skb, netdev, ptype) - * - * Called when a frame is received. Dispatches the right receive function - * for processing of the frame. - * - * Note on skb management : - * After calling the higher layers of the IrDA stack, we always - * kfree() the skb, which drop the reference count (and potentially - * destroy it). - * If a higher layer of the stack want to keep the skb around (to put - * in a queue or pass it to the higher layer), it will need to use - * skb_get() to keep a reference on it. This is usually done at the - * LMP level in irlmp.c. - * Jean II - */ -int irlap_driver_rcv(struct sk_buff *skb, struct net_device *dev, - struct packet_type *ptype, struct net_device *orig_dev) -{ - struct irlap_info info; - struct irlap_cb *self; - int command; - __u8 control; - int ret = -1; - - if (!net_eq(dev_net(dev), &init_net)) - goto out; - - /* FIXME: should we get our own field? */ - self = (struct irlap_cb *) dev->atalk_ptr; - - /* If the net device is down, then IrLAP is gone! */ - if (!self || self->magic != LAP_MAGIC) - goto err; - - /* We are no longer an "old" protocol, so we need to handle - * share and non linear skbs. This should never happen, so - * we don't need to be clever about it. Jean II */ - if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) { - net_err_ratelimited("%s: can't clone shared skb!\n", __func__); - goto err; - } - - /* Check if frame is large enough for parsing */ - if (!pskb_may_pull(skb, 2)) { - net_err_ratelimited("%s: frame too short!\n", __func__); - goto err; - } - - command = skb->data[0] & CMD_FRAME; - info.caddr = skb->data[0] & CBROADCAST; - - info.pf = skb->data[1] & PF_BIT; - info.control = skb->data[1] & ~PF_BIT; /* Mask away poll/final bit */ - - control = info.control; - - /* First we check if this frame has a valid connection address */ - if ((info.caddr != self->caddr) && (info.caddr != CBROADCAST)) { - pr_debug("%s(), wrong connection address!\n", - __func__); - goto out; - } - /* - * Optimize for the common case and check if the frame is an - * I(nformation) frame. Only I-frames have bit 0 set to 0 - */ - if (~control & 0x01) { - irlap_recv_i_frame(self, skb, &info, command); - goto out; - } - /* - * We now check is the frame is an S(upervisory) frame. Only - * S-frames have bit 0 set to 1 and bit 1 set to 0 - */ - if (~control & 0x02) { - /* - * Received S(upervisory) frame, check which frame type it is - * only the first nibble is of interest - */ - switch (control & 0x0f) { - case RR: - irlap_recv_rr_frame(self, skb, &info, command); - break; - case RNR: - irlap_recv_rnr_frame(self, skb, &info, command); - break; - case REJ: - irlap_recv_rej_frame(self, skb, &info, command); - break; - case SREJ: - irlap_recv_srej_frame(self, skb, &info, command); - break; - default: - net_warn_ratelimited("%s: Unknown S-frame %02x received!\n", - __func__, info.control); - break; - } - goto out; - } - /* - * This must be a C(ontrol) frame - */ - switch (control) { - case XID_RSP: - irlap_recv_discovery_xid_rsp(self, skb, &info); - break; - case XID_CMD: - irlap_recv_discovery_xid_cmd(self, skb, &info); - break; - case SNRM_CMD: - irlap_recv_snrm_cmd(self, skb, &info); - break; - case DM_RSP: - irlap_do_event(self, RECV_DM_RSP, skb, &info); - break; - case DISC_CMD: /* And RD_RSP since they have the same value */ - irlap_recv_disc_frame(self, skb, &info, command); - break; - case TEST_CMD: - irlap_recv_test_frame(self, skb, &info, command); - break; - case UA_RSP: - irlap_recv_ua_frame(self, skb, &info); - break; - case FRMR_RSP: - irlap_recv_frmr_frame(self, skb, &info); - break; - case UI_FRAME: - irlap_recv_ui_frame(self, skb, &info); - break; - default: - net_warn_ratelimited("%s: Unknown frame %02x received!\n", - __func__, info.control); - break; - } -out: - ret = 0; -err: - /* Always drop our reference on the skb */ - dev_kfree_skb(skb); - return ret; -} diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c deleted file mode 100644 index 43964594aa12..000000000000 --- a/net/irda/irlmp.c +++ /dev/null @@ -1,1996 +0,0 @@ -/********************************************************************* - * - * Filename: irlmp.c - * Version: 1.0 - * Description: IrDA Link Management Protocol (LMP) layer - * Status: Stable. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 17 20:54:32 1997 - * Modified at: Wed Jan 5 11:26:03 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/skbuff.h> -#include <linux/types.h> -#include <linux/proc_fs.h> -#include <linux/init.h> -#include <linux/kmod.h> -#include <linux/random.h> -#include <linux/seq_file.h> - -#include <net/irda/irda.h> -#include <net/irda/timer.h> -#include <net/irda/qos.h> -#include <net/irda/irlap.h> -#include <net/irda/iriap.h> -#include <net/irda/irlmp.h> -#include <net/irda/irlmp_frame.h> - -#include <asm/unaligned.h> - -static __u8 irlmp_find_free_slsap(void); -static int irlmp_slsap_inuse(__u8 slsap_sel); - -/* Master structure */ -struct irlmp_cb *irlmp = NULL; - -/* These can be altered by the sysctl interface */ -int sysctl_discovery = 0; -int sysctl_discovery_timeout = 3; /* 3 seconds by default */ -int sysctl_discovery_slots = 6; /* 6 slots by default */ -int sysctl_lap_keepalive_time = LM_IDLE_TIMEOUT * 1000 / HZ; -char sysctl_devname[65]; - -static const char *irlmp_reasons[] = { - "ERROR, NOT USED", - "LM_USER_REQUEST", - "LM_LAP_DISCONNECT", - "LM_CONNECT_FAILURE", - "LM_LAP_RESET", - "LM_INIT_DISCONNECT", - "ERROR, NOT USED", - "UNKNOWN", -}; - -const char *irlmp_reason_str(LM_REASON reason) -{ - reason = min_t(size_t, reason, ARRAY_SIZE(irlmp_reasons) - 1); - return irlmp_reasons[reason]; -} - -/* - * Function irlmp_init (void) - * - * Create (allocate) the main IrLMP structure - * - */ -int __init irlmp_init(void) -{ - /* Initialize the irlmp structure. */ - irlmp = kzalloc( sizeof(struct irlmp_cb), GFP_KERNEL); - if (irlmp == NULL) - return -ENOMEM; - - irlmp->magic = LMP_MAGIC; - - irlmp->clients = hashbin_new(HB_LOCK); - irlmp->services = hashbin_new(HB_LOCK); - irlmp->links = hashbin_new(HB_LOCK); - irlmp->unconnected_lsaps = hashbin_new(HB_LOCK); - irlmp->cachelog = hashbin_new(HB_NOLOCK); - - if ((irlmp->clients == NULL) || - (irlmp->services == NULL) || - (irlmp->links == NULL) || - (irlmp->unconnected_lsaps == NULL) || - (irlmp->cachelog == NULL)) { - return -ENOMEM; - } - - spin_lock_init(&irlmp->cachelog->hb_spinlock); - - irlmp->last_lsap_sel = 0x0f; /* Reserved 0x00-0x0f */ - strcpy(sysctl_devname, "Linux"); - - init_timer(&irlmp->discovery_timer); - - /* Do discovery every 3 seconds, conditionally */ - if (sysctl_discovery) - irlmp_start_discovery_timer(irlmp, - sysctl_discovery_timeout*HZ); - - return 0; -} - -/* - * Function irlmp_cleanup (void) - * - * Remove IrLMP layer - * - */ -void irlmp_cleanup(void) -{ - /* Check for main structure */ - IRDA_ASSERT(irlmp != NULL, return;); - IRDA_ASSERT(irlmp->magic == LMP_MAGIC, return;); - - del_timer(&irlmp->discovery_timer); - - hashbin_delete(irlmp->links, (FREE_FUNC) kfree); - hashbin_delete(irlmp->unconnected_lsaps, (FREE_FUNC) kfree); - hashbin_delete(irlmp->clients, (FREE_FUNC) kfree); - hashbin_delete(irlmp->services, (FREE_FUNC) kfree); - hashbin_delete(irlmp->cachelog, (FREE_FUNC) kfree); - - /* De-allocate main structure */ - kfree(irlmp); - irlmp = NULL; -} - -/* - * Function irlmp_open_lsap (slsap, notify) - * - * Register with IrLMP and create a local LSAP, - * returns handle to LSAP. - */ -struct lsap_cb *irlmp_open_lsap(__u8 slsap_sel, notify_t *notify, __u8 pid) -{ - struct lsap_cb *self; - - IRDA_ASSERT(notify != NULL, return NULL;); - IRDA_ASSERT(irlmp != NULL, return NULL;); - IRDA_ASSERT(irlmp->magic == LMP_MAGIC, return NULL;); - IRDA_ASSERT(notify->instance != NULL, return NULL;); - - /* Does the client care which Source LSAP selector it gets? */ - if (slsap_sel == LSAP_ANY) { - slsap_sel = irlmp_find_free_slsap(); - if (!slsap_sel) - return NULL; - } else if (irlmp_slsap_inuse(slsap_sel)) - return NULL; - - /* Allocate new instance of a LSAP connection */ - self = kzalloc(sizeof(struct lsap_cb), GFP_ATOMIC); - if (self == NULL) - return NULL; - - self->magic = LMP_LSAP_MAGIC; - self->slsap_sel = slsap_sel; - - /* Fix connectionless LSAP's */ - if (slsap_sel == LSAP_CONNLESS) { -#ifdef CONFIG_IRDA_ULTRA - self->dlsap_sel = LSAP_CONNLESS; - self->pid = pid; -#endif /* CONFIG_IRDA_ULTRA */ - } else - self->dlsap_sel = LSAP_ANY; - /* self->connected = FALSE; -> already NULL via memset() */ - - init_timer(&self->watchdog_timer); - - self->notify = *notify; - - self->lsap_state = LSAP_DISCONNECTED; - - /* Insert into queue of unconnected LSAPs */ - hashbin_insert(irlmp->unconnected_lsaps, (irda_queue_t *) self, - (long) self, NULL); - - return self; -} -EXPORT_SYMBOL(irlmp_open_lsap); - -/* - * Function __irlmp_close_lsap (self) - * - * Remove an instance of LSAP - */ -static void __irlmp_close_lsap(struct lsap_cb *self) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - - /* - * Set some of the variables to preset values - */ - self->magic = 0; - del_timer(&self->watchdog_timer); /* Important! */ - - if (self->conn_skb) - dev_kfree_skb(self->conn_skb); - - kfree(self); -} - -/* - * Function irlmp_close_lsap (self) - * - * Close and remove LSAP - * - */ -void irlmp_close_lsap(struct lsap_cb *self) -{ - struct lap_cb *lap; - struct lsap_cb *lsap = NULL; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - - /* - * Find out if we should remove this LSAP from a link or from the - * list of unconnected lsaps (not associated with a link) - */ - lap = self->lap; - if (lap) { - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, return;); - /* We might close a LSAP before it has completed the - * connection setup. In those case, higher layers won't - * send a proper disconnect request. Harmless, except - * that we will forget to close LAP... - Jean II */ - if(self->lsap_state != LSAP_DISCONNECTED) { - self->lsap_state = LSAP_DISCONNECTED; - irlmp_do_lap_event(self->lap, - LM_LAP_DISCONNECT_REQUEST, NULL); - } - /* Now, remove from the link */ - lsap = hashbin_remove(lap->lsaps, (long) self, NULL); -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - lap->cache.valid = FALSE; -#endif - } - self->lap = NULL; - /* Check if we found the LSAP! If not then try the unconnected lsaps */ - if (!lsap) { - lsap = hashbin_remove(irlmp->unconnected_lsaps, (long) self, - NULL); - } - if (!lsap) { - pr_debug("%s(), Looks like somebody has removed me already!\n", - __func__); - return; - } - __irlmp_close_lsap(self); -} -EXPORT_SYMBOL(irlmp_close_lsap); - -/* - * Function irlmp_register_irlap (saddr, notify) - * - * Register IrLAP layer with IrLMP. There is possible to have multiple - * instances of the IrLAP layer, each connected to different IrDA ports - * - */ -void irlmp_register_link(struct irlap_cb *irlap, __u32 saddr, notify_t *notify) -{ - struct lap_cb *lap; - - IRDA_ASSERT(irlmp != NULL, return;); - IRDA_ASSERT(irlmp->magic == LMP_MAGIC, return;); - IRDA_ASSERT(notify != NULL, return;); - - /* - * Allocate new instance of a LSAP connection - */ - lap = kzalloc(sizeof(struct lap_cb), GFP_KERNEL); - if (lap == NULL) - return; - - lap->irlap = irlap; - lap->magic = LMP_LAP_MAGIC; - lap->saddr = saddr; - lap->daddr = DEV_ADDR_ANY; -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - lap->cache.valid = FALSE; -#endif - lap->lsaps = hashbin_new(HB_LOCK); - if (lap->lsaps == NULL) { - net_warn_ratelimited("%s(), unable to kmalloc lsaps\n", - __func__); - kfree(lap); - return; - } - - lap->lap_state = LAP_STANDBY; - - init_timer(&lap->idle_timer); - - /* - * Insert into queue of LMP links - */ - hashbin_insert(irlmp->links, (irda_queue_t *) lap, lap->saddr, NULL); - - /* - * We set only this variable so IrLAP can tell us on which link the - * different events happened on - */ - irda_notify_init(notify); - notify->instance = lap; -} - -/* - * Function irlmp_unregister_irlap (saddr) - * - * IrLAP layer has been removed! - * - */ -void irlmp_unregister_link(__u32 saddr) -{ - struct lap_cb *link; - - /* We must remove ourselves from the hashbin *first*. This ensure - * that no more LSAPs will be open on this link and no discovery - * will be triggered anymore. Jean II */ - link = hashbin_remove(irlmp->links, saddr, NULL); - if (link) { - IRDA_ASSERT(link->magic == LMP_LAP_MAGIC, return;); - - /* Kill all the LSAPs on this link. Jean II */ - link->reason = LAP_DISC_INDICATION; - link->daddr = DEV_ADDR_ANY; - irlmp_do_lap_event(link, LM_LAP_DISCONNECT_INDICATION, NULL); - - /* Remove all discoveries discovered at this link */ - irlmp_expire_discoveries(irlmp->cachelog, link->saddr, TRUE); - - /* Final cleanup */ - del_timer(&link->idle_timer); - link->magic = 0; - hashbin_delete(link->lsaps, (FREE_FUNC) __irlmp_close_lsap); - kfree(link); - } -} - -/* - * Function irlmp_connect_request (handle, dlsap, userdata) - * - * Connect with a peer LSAP - * - */ -int irlmp_connect_request(struct lsap_cb *self, __u8 dlsap_sel, - __u32 saddr, __u32 daddr, - struct qos_info *qos, struct sk_buff *userdata) -{ - struct sk_buff *tx_skb = userdata; - struct lap_cb *lap; - struct lsap_cb *lsap; - int ret; - - IRDA_ASSERT(self != NULL, return -EBADR;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -EBADR;); - - pr_debug("%s(), slsap_sel=%02x, dlsap_sel=%02x, saddr=%08x, daddr=%08x\n", - __func__, self->slsap_sel, dlsap_sel, saddr, daddr); - - if (test_bit(0, &self->connected)) { - ret = -EISCONN; - goto err; - } - - /* Client must supply destination device address */ - if (!daddr) { - ret = -EINVAL; - goto err; - } - - /* Any userdata? */ - if (tx_skb == NULL) { - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - skb_reserve(tx_skb, LMP_MAX_HEADER); - } - - /* Make room for MUX control header (3 bytes) */ - IRDA_ASSERT(skb_headroom(tx_skb) >= LMP_CONTROL_HEADER, return -1;); - skb_push(tx_skb, LMP_CONTROL_HEADER); - - self->dlsap_sel = dlsap_sel; - - /* - * Find the link to where we should try to connect since there may - * be more than one IrDA port on this machine. If the client has - * passed us the saddr (and already knows which link to use), then - * we use that to find the link, if not then we have to look in the - * discovery log and check if any of the links has discovered a - * device with the given daddr - */ - if ((!saddr) || (saddr == DEV_ADDR_ANY)) { - discovery_t *discovery; - unsigned long flags; - - spin_lock_irqsave(&irlmp->cachelog->hb_spinlock, flags); - if (daddr != DEV_ADDR_ANY) - discovery = hashbin_find(irlmp->cachelog, daddr, NULL); - else { - pr_debug("%s(), no daddr\n", __func__); - discovery = (discovery_t *) - hashbin_get_first(irlmp->cachelog); - } - - if (discovery) { - saddr = discovery->data.saddr; - daddr = discovery->data.daddr; - } - spin_unlock_irqrestore(&irlmp->cachelog->hb_spinlock, flags); - } - lap = hashbin_lock_find(irlmp->links, saddr, NULL); - if (lap == NULL) { - pr_debug("%s(), Unable to find a usable link!\n", __func__); - ret = -EHOSTUNREACH; - goto err; - } - - /* Check if LAP is disconnected or already connected */ - if (lap->daddr == DEV_ADDR_ANY) - lap->daddr = daddr; - else if (lap->daddr != daddr) { - /* Check if some LSAPs are active on this LAP */ - if (HASHBIN_GET_SIZE(lap->lsaps) == 0) { - /* No active connection, but LAP hasn't been - * disconnected yet (waiting for timeout in LAP). - * Maybe we could give LAP a bit of help in this case. - */ - pr_debug("%s(), sorry, but I'm waiting for LAP to timeout!\n", - __func__); - ret = -EAGAIN; - goto err; - } - - /* LAP is already connected to a different node, and LAP - * can only talk to one node at a time */ - pr_debug("%s(), sorry, but link is busy!\n", __func__); - ret = -EBUSY; - goto err; - } - - self->lap = lap; - - /* - * Remove LSAP from list of unconnected LSAPs and insert it into the - * list of connected LSAPs for the particular link - */ - lsap = hashbin_remove(irlmp->unconnected_lsaps, (long) self, NULL); - - IRDA_ASSERT(lsap != NULL, return -1;); - IRDA_ASSERT(lsap->magic == LMP_LSAP_MAGIC, return -1;); - IRDA_ASSERT(lsap->lap != NULL, return -1;); - IRDA_ASSERT(lsap->lap->magic == LMP_LAP_MAGIC, return -1;); - - hashbin_insert(self->lap->lsaps, (irda_queue_t *) self, (long) self, - NULL); - - set_bit(0, &self->connected); /* TRUE */ - - /* - * User supplied qos specifications? - */ - if (qos) - self->qos = *qos; - - irlmp_do_lsap_event(self, LM_CONNECT_REQUEST, tx_skb); - - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(tx_skb); - - return 0; - -err: - /* Cleanup */ - if(tx_skb) - dev_kfree_skb(tx_skb); - return ret; -} -EXPORT_SYMBOL(irlmp_connect_request); - -/* - * Function irlmp_connect_indication (self) - * - * Incoming connection - * - */ -void irlmp_connect_indication(struct lsap_cb *self, struct sk_buff *skb) -{ - int max_seg_size; - int lap_header_size; - int max_header_size; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(self->lap != NULL, return;); - - pr_debug("%s(), slsap_sel=%02x, dlsap_sel=%02x\n", - __func__, self->slsap_sel, self->dlsap_sel); - - /* Note : self->lap is set in irlmp_link_data_indication(), - * (case CONNECT_CMD:) because we have no way to set it here. - * Similarly, self->dlsap_sel is usually set in irlmp_find_lsap(). - * Jean II */ - - self->qos = *self->lap->qos; - - max_seg_size = self->lap->qos->data_size.value-LMP_HEADER; - lap_header_size = IRLAP_GET_HEADER_SIZE(self->lap->irlap); - max_header_size = LMP_HEADER + lap_header_size; - - /* Hide LMP_CONTROL_HEADER header from layer above */ - skb_pull(skb, LMP_CONTROL_HEADER); - - if (self->notify.connect_indication) { - /* Don't forget to refcount it - see irlap_driver_rcv(). */ - skb_get(skb); - self->notify.connect_indication(self->notify.instance, self, - &self->qos, max_seg_size, - max_header_size, skb); - } -} - -/* - * Function irlmp_connect_response (handle, userdata) - * - * Service user is accepting connection - * - */ -int irlmp_connect_response(struct lsap_cb *self, struct sk_buff *userdata) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - IRDA_ASSERT(userdata != NULL, return -1;); - - /* We set the connected bit and move the lsap to the connected list - * in the state machine itself. Jean II */ - - pr_debug("%s(), slsap_sel=%02x, dlsap_sel=%02x\n", - __func__, self->slsap_sel, self->dlsap_sel); - - /* Make room for MUX control header (3 bytes) */ - IRDA_ASSERT(skb_headroom(userdata) >= LMP_CONTROL_HEADER, return -1;); - skb_push(userdata, LMP_CONTROL_HEADER); - - irlmp_do_lsap_event(self, LM_CONNECT_RESPONSE, userdata); - - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(userdata); - - return 0; -} -EXPORT_SYMBOL(irlmp_connect_response); - -/* - * Function irlmp_connect_confirm (handle, skb) - * - * LSAP connection confirmed peer device! - */ -void irlmp_connect_confirm(struct lsap_cb *self, struct sk_buff *skb) -{ - int max_header_size; - int lap_header_size; - int max_seg_size; - - IRDA_ASSERT(skb != NULL, return;); - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - IRDA_ASSERT(self->lap != NULL, return;); - - self->qos = *self->lap->qos; - - max_seg_size = self->lap->qos->data_size.value-LMP_HEADER; - lap_header_size = IRLAP_GET_HEADER_SIZE(self->lap->irlap); - max_header_size = LMP_HEADER + lap_header_size; - - pr_debug("%s(), max_header_size=%d\n", - __func__, max_header_size); - - /* Hide LMP_CONTROL_HEADER header from layer above */ - skb_pull(skb, LMP_CONTROL_HEADER); - - if (self->notify.connect_confirm) { - /* Don't forget to refcount it - see irlap_driver_rcv() */ - skb_get(skb); - self->notify.connect_confirm(self->notify.instance, self, - &self->qos, max_seg_size, - max_header_size, skb); - } -} - -/* - * Function irlmp_dup (orig, instance) - * - * Duplicate LSAP, can be used by servers to confirm a connection on a - * new LSAP so it can keep listening on the old one. - * - */ -struct lsap_cb *irlmp_dup(struct lsap_cb *orig, void *instance) -{ - struct lsap_cb *new; - unsigned long flags; - - spin_lock_irqsave(&irlmp->unconnected_lsaps->hb_spinlock, flags); - - /* Only allowed to duplicate unconnected LSAP's, and only LSAPs - * that have received a connect indication. Jean II */ - if ((!hashbin_find(irlmp->unconnected_lsaps, (long) orig, NULL)) || - (orig->lap == NULL)) { - pr_debug("%s(), invalid LSAP (wrong state)\n", - __func__); - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, - flags); - return NULL; - } - - /* Allocate a new instance */ - new = kmemdup(orig, sizeof(*new), GFP_ATOMIC); - if (!new) { - pr_debug("%s(), unable to kmalloc\n", __func__); - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, - flags); - return NULL; - } - /* new->lap = orig->lap; => done in the memcpy() */ - /* new->slsap_sel = orig->slsap_sel; => done in the memcpy() */ - new->conn_skb = NULL; - - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, flags); - - /* Not everything is the same */ - new->notify.instance = instance; - - init_timer(&new->watchdog_timer); - - hashbin_insert(irlmp->unconnected_lsaps, (irda_queue_t *) new, - (long) new, NULL); - -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - /* Make sure that we invalidate the LSAP cache */ - new->lap->cache.valid = FALSE; -#endif /* CONFIG_IRDA_CACHE_LAST_LSAP */ - - return new; -} - -/* - * Function irlmp_disconnect_request (handle, userdata) - * - * The service user is requesting disconnection, this will not remove the - * LSAP, but only mark it as disconnected - */ -int irlmp_disconnect_request(struct lsap_cb *self, struct sk_buff *userdata) -{ - struct lsap_cb *lsap; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - IRDA_ASSERT(userdata != NULL, return -1;); - - /* Already disconnected ? - * There is a race condition between irlmp_disconnect_indication() - * and us that might mess up the hashbins below. This fixes it. - * Jean II */ - if (! test_and_clear_bit(0, &self->connected)) { - pr_debug("%s(), already disconnected!\n", __func__); - dev_kfree_skb(userdata); - return -1; - } - - skb_push(userdata, LMP_CONTROL_HEADER); - - /* - * Do the event before the other stuff since we must know - * which lap layer that the frame should be transmitted on - */ - irlmp_do_lsap_event(self, LM_DISCONNECT_REQUEST, userdata); - - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(userdata); - - /* - * Remove LSAP from list of connected LSAPs for the particular link - * and insert it into the list of unconnected LSAPs - */ - IRDA_ASSERT(self->lap != NULL, return -1;); - IRDA_ASSERT(self->lap->magic == LMP_LAP_MAGIC, return -1;); - IRDA_ASSERT(self->lap->lsaps != NULL, return -1;); - - lsap = hashbin_remove(self->lap->lsaps, (long) self, NULL); -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - self->lap->cache.valid = FALSE; -#endif - - IRDA_ASSERT(lsap != NULL, return -1;); - IRDA_ASSERT(lsap->magic == LMP_LSAP_MAGIC, return -1;); - IRDA_ASSERT(lsap == self, return -1;); - - hashbin_insert(irlmp->unconnected_lsaps, (irda_queue_t *) self, - (long) self, NULL); - - /* Reset some values */ - self->dlsap_sel = LSAP_ANY; - self->lap = NULL; - - return 0; -} -EXPORT_SYMBOL(irlmp_disconnect_request); - -/* - * Function irlmp_disconnect_indication (reason, userdata) - * - * LSAP is being closed! - */ -void irlmp_disconnect_indication(struct lsap_cb *self, LM_REASON reason, - struct sk_buff *skb) -{ - struct lsap_cb *lsap; - - pr_debug("%s(), reason=%s [%d]\n", __func__, - irlmp_reason_str(reason), reason); - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - - pr_debug("%s(), slsap_sel=%02x, dlsap_sel=%02x\n", - __func__, self->slsap_sel, self->dlsap_sel); - - /* Already disconnected ? - * There is a race condition between irlmp_disconnect_request() - * and us that might mess up the hashbins below. This fixes it. - * Jean II */ - if (! test_and_clear_bit(0, &self->connected)) { - pr_debug("%s(), already disconnected!\n", __func__); - return; - } - - /* - * Remove association between this LSAP and the link it used - */ - IRDA_ASSERT(self->lap != NULL, return;); - IRDA_ASSERT(self->lap->lsaps != NULL, return;); - - lsap = hashbin_remove(self->lap->lsaps, (long) self, NULL); -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - self->lap->cache.valid = FALSE; -#endif - - IRDA_ASSERT(lsap != NULL, return;); - IRDA_ASSERT(lsap == self, return;); - hashbin_insert(irlmp->unconnected_lsaps, (irda_queue_t *) lsap, - (long) lsap, NULL); - - self->dlsap_sel = LSAP_ANY; - self->lap = NULL; - - /* - * Inform service user - */ - if (self->notify.disconnect_indication) { - /* Don't forget to refcount it - see irlap_driver_rcv(). */ - if(skb) - skb_get(skb); - self->notify.disconnect_indication(self->notify.instance, - self, reason, skb); - } else { - pr_debug("%s(), no handler\n", __func__); - } -} - -/* - * Function irlmp_do_expiry (void) - * - * Do a cleanup of the discovery log (remove old entries) - * - * Note : separate from irlmp_do_discovery() so that we can handle - * passive discovery properly. - */ -void irlmp_do_expiry(void) -{ - struct lap_cb *lap; - - /* - * Expire discovery on all links which are *not* connected. - * On links which are connected, we can't do discovery - * anymore and can't refresh the log, so we freeze the - * discovery log to keep info about the device we are - * connected to. - * This info is mandatory if we want irlmp_connect_request() - * to work properly. - Jean II - */ - lap = (struct lap_cb *) hashbin_get_first(irlmp->links); - while (lap != NULL) { - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, return;); - - if (lap->lap_state == LAP_STANDBY) { - /* Expire discoveries discovered on this link */ - irlmp_expire_discoveries(irlmp->cachelog, lap->saddr, - FALSE); - } - lap = (struct lap_cb *) hashbin_get_next(irlmp->links); - } -} - -/* - * Function irlmp_do_discovery (nslots) - * - * Do some discovery on all links - * - * Note : log expiry is done above. - */ -void irlmp_do_discovery(int nslots) -{ - struct lap_cb *lap; - __u16 *data_hintsp; - - /* Make sure the value is sane */ - if ((nslots != 1) && (nslots != 6) && (nslots != 8) && (nslots != 16)){ - net_warn_ratelimited("%s: invalid value for number of slots!\n", - __func__); - nslots = sysctl_discovery_slots = 8; - } - - /* Construct new discovery info to be used by IrLAP, */ - data_hintsp = (__u16 *) irlmp->discovery_cmd.data.hints; - put_unaligned(irlmp->hints.word, data_hintsp); - - /* - * Set character set for device name (we use ASCII), and - * copy device name. Remember to make room for a \0 at the - * end - */ - irlmp->discovery_cmd.data.charset = CS_ASCII; - strncpy(irlmp->discovery_cmd.data.info, sysctl_devname, - NICKNAME_MAX_LEN); - irlmp->discovery_cmd.name_len = strlen(irlmp->discovery_cmd.data.info); - irlmp->discovery_cmd.nslots = nslots; - - /* - * Try to send discovery packets on all links - */ - lap = (struct lap_cb *) hashbin_get_first(irlmp->links); - while (lap != NULL) { - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, return;); - - if (lap->lap_state == LAP_STANDBY) { - /* Try to discover */ - irlmp_do_lap_event(lap, LM_LAP_DISCOVERY_REQUEST, - NULL); - } - lap = (struct lap_cb *) hashbin_get_next(irlmp->links); - } -} - -/* - * Function irlmp_discovery_request (nslots) - * - * Do a discovery of devices in front of the computer - * - * If the caller has registered a client discovery callback, this - * allow him to receive the full content of the discovery log through - * this callback (as normally he will receive only new discoveries). - */ -void irlmp_discovery_request(int nslots) -{ - /* Return current cached discovery log (in full) */ - irlmp_discovery_confirm(irlmp->cachelog, DISCOVERY_LOG); - - /* - * Start a single discovery operation if discovery is not already - * running - */ - if (!sysctl_discovery) { - /* Check if user wants to override the default */ - if (nslots == DISCOVERY_DEFAULT_SLOTS) - nslots = sysctl_discovery_slots; - - irlmp_do_discovery(nslots); - /* Note : we never do expiry here. Expiry will run on the - * discovery timer regardless of the state of sysctl_discovery - * Jean II */ - } -} -EXPORT_SYMBOL(irlmp_discovery_request); - -/* - * Function irlmp_get_discoveries (pn, mask, slots) - * - * Return the current discovery log - * - * If discovery is not enabled, you should call this function again - * after 1 or 2 seconds (i.e. after discovery has been done). - */ -struct irda_device_info *irlmp_get_discoveries(int *pn, __u16 mask, int nslots) -{ - /* If discovery is not enabled, it's likely that the discovery log - * will be empty. So, we trigger a single discovery, so that next - * time the user call us there might be some results in the log. - * Jean II - */ - if (!sysctl_discovery) { - /* Check if user wants to override the default */ - if (nslots == DISCOVERY_DEFAULT_SLOTS) - nslots = sysctl_discovery_slots; - - /* Start discovery - will complete sometime later */ - irlmp_do_discovery(nslots); - /* Note : we never do expiry here. Expiry will run on the - * discovery timer regardless of the state of sysctl_discovery - * Jean II */ - } - - /* Return current cached discovery log */ - return irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE); -} -EXPORT_SYMBOL(irlmp_get_discoveries); - -/* - * Function irlmp_notify_client (log) - * - * Notify all about discovered devices - * - * Clients registered with IrLMP are : - * o IrComm - * o IrLAN - * o Any socket (in any state - ouch, that may be a lot !) - * The client may have defined a callback to be notified in case of - * partial/selective discovery based on the hints that it passed to IrLMP. - */ -static inline void -irlmp_notify_client(irlmp_client_t *client, - hashbin_t *log, DISCOVERY_MODE mode) -{ - discinfo_t *discoveries; /* Copy of the discovery log */ - int number; /* Number of nodes in the log */ - int i; - - /* Check if client wants or not partial/selective log (optimisation) */ - if (!client->disco_callback) - return; - - /* - * Locking notes : - * the old code was manipulating the log directly, which was - * very racy. Now, we use copy_discoveries, that protects - * itself while dumping the log for us. - * The overhead of the copy is compensated by the fact that - * we only pass new discoveries in normal mode and don't - * pass the same old entry every 3s to the caller as we used - * to do (virtual function calling is expensive). - * Jean II - */ - - /* - * Now, check all discovered devices (if any), and notify client - * only about the services that the client is interested in - * We also notify only about the new devices unless the caller - * explicitly request a dump of the log. Jean II - */ - discoveries = irlmp_copy_discoveries(log, &number, - client->hint_mask.word, - (mode == DISCOVERY_LOG)); - /* Check if the we got some results */ - if (discoveries == NULL) - return; /* No nodes discovered */ - - /* Pass all entries to the listener */ - for(i = 0; i < number; i++) - client->disco_callback(&(discoveries[i]), mode, client->priv); - - /* Free up our buffer */ - kfree(discoveries); -} - -/* - * Function irlmp_discovery_confirm ( self, log) - * - * Some device(s) answered to our discovery request! Check to see which - * device it is, and give indication to the client(s) - * - */ -void irlmp_discovery_confirm(hashbin_t *log, DISCOVERY_MODE mode) -{ - irlmp_client_t *client; - irlmp_client_t *client_next; - - IRDA_ASSERT(log != NULL, return;); - - if (!(HASHBIN_GET_SIZE(log))) - return; - - /* For each client - notify callback may touch client list */ - client = (irlmp_client_t *) hashbin_get_first(irlmp->clients); - while (NULL != hashbin_find_next(irlmp->clients, (long) client, NULL, - (void *) &client_next) ) { - /* Check if we should notify client */ - irlmp_notify_client(client, log, mode); - - client = client_next; - } -} - -/* - * Function irlmp_discovery_expiry (expiry) - * - * This device is no longer been discovered, and therefore it is being - * purged from the discovery log. Inform all clients who have - * registered for this event... - * - * Note : called exclusively from discovery.c - * Note : this is no longer called under discovery spinlock, so the - * client can do whatever he wants in the callback. - */ -void irlmp_discovery_expiry(discinfo_t *expiries, int number) -{ - irlmp_client_t *client; - irlmp_client_t *client_next; - int i; - - IRDA_ASSERT(expiries != NULL, return;); - - /* For each client - notify callback may touch client list */ - client = (irlmp_client_t *) hashbin_get_first(irlmp->clients); - while (NULL != hashbin_find_next(irlmp->clients, (long) client, NULL, - (void *) &client_next) ) { - - /* Pass all entries to the listener */ - for(i = 0; i < number; i++) { - /* Check if we should notify client */ - if ((client->expir_callback) && - (client->hint_mask.word & - get_unaligned((__u16 *)expiries[i].hints) - & 0x7f7f) ) - client->expir_callback(&(expiries[i]), - EXPIRY_TIMEOUT, - client->priv); - } - - /* Next client */ - client = client_next; - } -} - -/* - * Function irlmp_get_discovery_response () - * - * Used by IrLAP to get the discovery info it needs when answering - * discovery requests by other devices. - */ -discovery_t *irlmp_get_discovery_response(void) -{ - IRDA_ASSERT(irlmp != NULL, return NULL;); - - put_unaligned(irlmp->hints.word, (__u16 *)irlmp->discovery_rsp.data.hints); - - /* - * Set character set for device name (we use ASCII), and - * copy device name. Remember to make room for a \0 at the - * end - */ - irlmp->discovery_rsp.data.charset = CS_ASCII; - - strncpy(irlmp->discovery_rsp.data.info, sysctl_devname, - NICKNAME_MAX_LEN); - irlmp->discovery_rsp.name_len = strlen(irlmp->discovery_rsp.data.info); - - return &irlmp->discovery_rsp; -} - -/* - * Function irlmp_data_request (self, skb) - * - * Send some data to peer device - * - * Note on skb management : - * After calling the lower layers of the IrDA stack, we always - * kfree() the skb, which drop the reference count (and potentially - * destroy it). - * IrLMP and IrLAP may queue the packet, and in those cases will need - * to use skb_get() to keep it around. - * Jean II - */ -int irlmp_data_request(struct lsap_cb *self, struct sk_buff *userdata) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - /* Make room for MUX header */ - IRDA_ASSERT(skb_headroom(userdata) >= LMP_HEADER, return -1;); - skb_push(userdata, LMP_HEADER); - - ret = irlmp_do_lsap_event(self, LM_DATA_REQUEST, userdata); - - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(userdata); - - return ret; -} -EXPORT_SYMBOL(irlmp_data_request); - -/* - * Function irlmp_data_indication (handle, skb) - * - * Got data from LAP layer so pass it up to upper layer - * - */ -void irlmp_data_indication(struct lsap_cb *self, struct sk_buff *skb) -{ - /* Hide LMP header from layer above */ - skb_pull(skb, LMP_HEADER); - - if (self->notify.data_indication) { - /* Don't forget to refcount it - see irlap_driver_rcv(). */ - skb_get(skb); - self->notify.data_indication(self->notify.instance, self, skb); - } -} - -/* - * Function irlmp_udata_request (self, skb) - */ -int irlmp_udata_request(struct lsap_cb *self, struct sk_buff *userdata) -{ - int ret; - - IRDA_ASSERT(userdata != NULL, return -1;); - - /* Make room for MUX header */ - IRDA_ASSERT(skb_headroom(userdata) >= LMP_HEADER, return -1;); - skb_push(userdata, LMP_HEADER); - - ret = irlmp_do_lsap_event(self, LM_UDATA_REQUEST, userdata); - - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(userdata); - - return ret; -} - -/* - * Function irlmp_udata_indication (self, skb) - * - * Send unreliable data (but still within the connection) - * - */ -void irlmp_udata_indication(struct lsap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Hide LMP header from layer above */ - skb_pull(skb, LMP_HEADER); - - if (self->notify.udata_indication) { - /* Don't forget to refcount it - see irlap_driver_rcv(). */ - skb_get(skb); - self->notify.udata_indication(self->notify.instance, self, - skb); - } -} - -/* - * Function irlmp_connless_data_request (self, skb) - */ -#ifdef CONFIG_IRDA_ULTRA -int irlmp_connless_data_request(struct lsap_cb *self, struct sk_buff *userdata, - __u8 pid) -{ - struct sk_buff *clone_skb; - struct lap_cb *lap; - - IRDA_ASSERT(userdata != NULL, return -1;); - - /* Make room for MUX and PID header */ - IRDA_ASSERT(skb_headroom(userdata) >= LMP_HEADER+LMP_PID_HEADER, - return -1;); - - /* Insert protocol identifier */ - skb_push(userdata, LMP_PID_HEADER); - if(self != NULL) - userdata->data[0] = self->pid; - else - userdata->data[0] = pid; - - /* Connectionless sockets must use 0x70 */ - skb_push(userdata, LMP_HEADER); - userdata->data[0] = userdata->data[1] = LSAP_CONNLESS; - - /* Try to send Connectionless packets out on all links */ - lap = (struct lap_cb *) hashbin_get_first(irlmp->links); - while (lap != NULL) { - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, return -1;); - - clone_skb = skb_clone(userdata, GFP_ATOMIC); - if (!clone_skb) { - dev_kfree_skb(userdata); - return -ENOMEM; - } - - irlap_unitdata_request(lap->irlap, clone_skb); - /* irlap_unitdata_request() don't increase refcount, - * so no dev_kfree_skb() - Jean II */ - - lap = (struct lap_cb *) hashbin_get_next(irlmp->links); - } - dev_kfree_skb(userdata); - - return 0; -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irlmp_connless_data_indication (self, skb) - * - * Receive unreliable data outside any connection. Mostly used by Ultra - * - */ -#ifdef CONFIG_IRDA_ULTRA -void irlmp_connless_data_indication(struct lsap_cb *self, struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* Hide LMP and PID header from layer above */ - skb_pull(skb, LMP_HEADER+LMP_PID_HEADER); - - if (self->notify.udata_indication) { - /* Don't forget to refcount it - see irlap_driver_rcv(). */ - skb_get(skb); - self->notify.udata_indication(self->notify.instance, self, - skb); - } -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Propagate status indication from LAP to LSAPs (via LMP) - * This don't trigger any change of state in lap_cb, lmp_cb or lsap_cb, - * and the event is stateless, therefore we can bypass both state machines - * and send the event direct to the LSAP user. - * Jean II - */ -void irlmp_status_indication(struct lap_cb *self, - LINK_STATUS link, LOCK_STATUS lock) -{ - struct lsap_cb *next; - struct lsap_cb *curr; - - /* Send status_indication to all LSAPs using this link */ - curr = (struct lsap_cb *) hashbin_get_first( self->lsaps); - while (NULL != hashbin_find_next(self->lsaps, (long) curr, NULL, - (void *) &next) ) { - IRDA_ASSERT(curr->magic == LMP_LSAP_MAGIC, return;); - /* - * Inform service user if he has requested it - */ - if (curr->notify.status_indication != NULL) - curr->notify.status_indication(curr->notify.instance, - link, lock); - else - pr_debug("%s(), no handler\n", __func__); - - curr = next; - } -} - -/* - * Receive flow control indication from LAP. - * LAP want us to send it one more frame. We implement a simple round - * robin scheduler between the active sockets so that we get a bit of - * fairness. Note that the round robin is far from perfect, but it's - * better than nothing. - * We then poll the selected socket so that we can do synchronous - * refilling of IrLAP (which allow to minimise the number of buffers). - * Jean II - */ -void irlmp_flow_indication(struct lap_cb *self, LOCAL_FLOW flow) -{ - struct lsap_cb *next; - struct lsap_cb *curr; - int lsap_todo; - - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - IRDA_ASSERT(flow == FLOW_START, return;); - - /* Get the number of lsap. That's the only safe way to know - * that we have looped around... - Jean II */ - lsap_todo = HASHBIN_GET_SIZE(self->lsaps); - pr_debug("%s() : %d lsaps to scan\n", __func__, lsap_todo); - - /* Poll lsap in order until the queue is full or until we - * tried them all. - * Most often, the current LSAP will have something to send, - * so we will go through this loop only once. - Jean II */ - while((lsap_todo--) && - (IRLAP_GET_TX_QUEUE_LEN(self->irlap) < LAP_HIGH_THRESHOLD)) { - /* Try to find the next lsap we should poll. */ - next = self->flow_next; - /* If we have no lsap, restart from first one */ - if(next == NULL) - next = (struct lsap_cb *) hashbin_get_first(self->lsaps); - /* Verify current one and find the next one */ - curr = hashbin_find_next(self->lsaps, (long) next, NULL, - (void *) &self->flow_next); - /* Uh-oh... Paranoia */ - if(curr == NULL) - break; - pr_debug("%s() : curr is %p, next was %p and is now %p, still %d to go - queue len = %d\n", - __func__, curr, next, self->flow_next, lsap_todo, - IRLAP_GET_TX_QUEUE_LEN(self->irlap)); - - /* Inform lsap user that it can send one more packet. */ - if (curr->notify.flow_indication != NULL) - curr->notify.flow_indication(curr->notify.instance, - curr, flow); - else - pr_debug("%s(), no handler\n", __func__); - } -} - -#if 0 -/* - * Function irlmp_hint_to_service (hint) - * - * Returns a list of all servics contained in the given hint bits. This - * function assumes that the hint bits have the size of two bytes only - */ -__u8 *irlmp_hint_to_service(__u8 *hint) -{ - __u8 *service; - int i = 0; - - /* - * Allocate array to store services in. 16 entries should be safe - * since we currently only support 2 hint bytes - */ - service = kmalloc(16, GFP_ATOMIC); - if (!service) - return NULL; - - if (!hint[0]) { - pr_debug("<None>\n"); - kfree(service); - return NULL; - } - if (hint[0] & HINT_PNP) - pr_debug("PnP Compatible "); - if (hint[0] & HINT_PDA) - pr_debug("PDA/Palmtop "); - if (hint[0] & HINT_COMPUTER) - pr_debug("Computer "); - if (hint[0] & HINT_PRINTER) { - pr_debug("Printer "); - service[i++] = S_PRINTER; - } - if (hint[0] & HINT_MODEM) - pr_debug("Modem "); - if (hint[0] & HINT_FAX) - pr_debug("Fax "); - if (hint[0] & HINT_LAN) { - pr_debug("LAN Access "); - service[i++] = S_LAN; - } - /* - * Test if extension byte exists. This byte will usually be - * there, but this is not really required by the standard. - * (IrLMP p. 29) - */ - if (hint[0] & HINT_EXTENSION) { - if (hint[1] & HINT_TELEPHONY) { - pr_debug("Telephony "); - service[i++] = S_TELEPHONY; - } - if (hint[1] & HINT_FILE_SERVER) - pr_debug("File Server "); - - if (hint[1] & HINT_COMM) { - pr_debug("IrCOMM "); - service[i++] = S_COMM; - } - if (hint[1] & HINT_OBEX) { - pr_debug("IrOBEX "); - service[i++] = S_OBEX; - } - } - pr_debug("\n"); - - /* So that client can be notified about any discovery */ - service[i++] = S_ANY; - - service[i] = S_END; - - return service; -} -#endif - -static const __u16 service_hint_mapping[S_END][2] = { - { HINT_PNP, 0 }, /* S_PNP */ - { HINT_PDA, 0 }, /* S_PDA */ - { HINT_COMPUTER, 0 }, /* S_COMPUTER */ - { HINT_PRINTER, 0 }, /* S_PRINTER */ - { HINT_MODEM, 0 }, /* S_MODEM */ - { HINT_FAX, 0 }, /* S_FAX */ - { HINT_LAN, 0 }, /* S_LAN */ - { HINT_EXTENSION, HINT_TELEPHONY }, /* S_TELEPHONY */ - { HINT_EXTENSION, HINT_COMM }, /* S_COMM */ - { HINT_EXTENSION, HINT_OBEX }, /* S_OBEX */ - { 0xFF, 0xFF }, /* S_ANY */ -}; - -/* - * Function irlmp_service_to_hint (service) - * - * Converts a service type, to a hint bit - * - * Returns: a 16 bit hint value, with the service bit set - */ -__u16 irlmp_service_to_hint(int service) -{ - __u16_host_order hint; - - hint.byte[0] = service_hint_mapping[service][0]; - hint.byte[1] = service_hint_mapping[service][1]; - - return hint.word; -} -EXPORT_SYMBOL(irlmp_service_to_hint); - -/* - * Function irlmp_register_service (service) - * - * Register local service with IrLMP - * - */ -void *irlmp_register_service(__u16 hints) -{ - irlmp_service_t *service; - - pr_debug("%s(), hints = %04x\n", __func__, hints); - - /* Make a new registration */ - service = kmalloc(sizeof(irlmp_service_t), GFP_ATOMIC); - if (!service) - return NULL; - - service->hints.word = hints; - hashbin_insert(irlmp->services, (irda_queue_t *) service, - (long) service, NULL); - - irlmp->hints.word |= hints; - - return (void *)service; -} -EXPORT_SYMBOL(irlmp_register_service); - -/* - * Function irlmp_unregister_service (handle) - * - * Unregister service with IrLMP. - * - * Returns: 0 on success, -1 on error - */ -int irlmp_unregister_service(void *handle) -{ - irlmp_service_t *service; - unsigned long flags; - - if (!handle) - return -1; - - /* Caller may call with invalid handle (it's legal) - Jean II */ - service = hashbin_lock_find(irlmp->services, (long) handle, NULL); - if (!service) { - pr_debug("%s(), Unknown service!\n", __func__); - return -1; - } - - hashbin_remove_this(irlmp->services, (irda_queue_t *) service); - kfree(service); - - /* Remove old hint bits */ - irlmp->hints.word = 0; - - /* Refresh current hint bits */ - spin_lock_irqsave(&irlmp->services->hb_spinlock, flags); - service = (irlmp_service_t *) hashbin_get_first(irlmp->services); - while (service) { - irlmp->hints.word |= service->hints.word; - - service = (irlmp_service_t *)hashbin_get_next(irlmp->services); - } - spin_unlock_irqrestore(&irlmp->services->hb_spinlock, flags); - return 0; -} -EXPORT_SYMBOL(irlmp_unregister_service); - -/* - * Function irlmp_register_client (hint_mask, callback1, callback2) - * - * Register a local client with IrLMP - * First callback is selective discovery (based on hints) - * Second callback is for selective discovery expiries - * - * Returns: handle > 0 on success, 0 on error - */ -void *irlmp_register_client(__u16 hint_mask, DISCOVERY_CALLBACK1 disco_clb, - DISCOVERY_CALLBACK2 expir_clb, void *priv) -{ - irlmp_client_t *client; - - IRDA_ASSERT(irlmp != NULL, return NULL;); - - /* Make a new registration */ - client = kmalloc(sizeof(irlmp_client_t), GFP_ATOMIC); - if (!client) - return NULL; - - /* Register the details */ - client->hint_mask.word = hint_mask; - client->disco_callback = disco_clb; - client->expir_callback = expir_clb; - client->priv = priv; - - hashbin_insert(irlmp->clients, (irda_queue_t *) client, - (long) client, NULL); - - return (void *) client; -} -EXPORT_SYMBOL(irlmp_register_client); - -/* - * Function irlmp_update_client (handle, hint_mask, callback1, callback2) - * - * Updates specified client (handle) with possibly new hint_mask and - * callback - * - * Returns: 0 on success, -1 on error - */ -int irlmp_update_client(void *handle, __u16 hint_mask, - DISCOVERY_CALLBACK1 disco_clb, - DISCOVERY_CALLBACK2 expir_clb, void *priv) -{ - irlmp_client_t *client; - - if (!handle) - return -1; - - client = hashbin_lock_find(irlmp->clients, (long) handle, NULL); - if (!client) { - pr_debug("%s(), Unknown client!\n", __func__); - return -1; - } - - client->hint_mask.word = hint_mask; - client->disco_callback = disco_clb; - client->expir_callback = expir_clb; - client->priv = priv; - - return 0; -} -EXPORT_SYMBOL(irlmp_update_client); - -/* - * Function irlmp_unregister_client (handle) - * - * Returns: 0 on success, -1 on error - * - */ -int irlmp_unregister_client(void *handle) -{ - struct irlmp_client *client; - - if (!handle) - return -1; - - /* Caller may call with invalid handle (it's legal) - Jean II */ - client = hashbin_lock_find(irlmp->clients, (long) handle, NULL); - if (!client) { - pr_debug("%s(), Unknown client!\n", __func__); - return -1; - } - - pr_debug("%s(), removing client!\n", __func__); - hashbin_remove_this(irlmp->clients, (irda_queue_t *) client); - kfree(client); - - return 0; -} -EXPORT_SYMBOL(irlmp_unregister_client); - -/* - * Function irlmp_slsap_inuse (slsap) - * - * Check if the given source LSAP selector is in use - * - * This function is clearly not very efficient. On the mitigating side, the - * stack make sure that in 99% of the cases, we are called only once - * for each socket allocation. We could probably keep a bitmap - * of the allocated LSAP, but I'm not sure the complexity is worth it. - * Jean II - */ -static int irlmp_slsap_inuse(__u8 slsap_sel) -{ - struct lsap_cb *self; - struct lap_cb *lap; - unsigned long flags; - - IRDA_ASSERT(irlmp != NULL, return TRUE;); - IRDA_ASSERT(irlmp->magic == LMP_MAGIC, return TRUE;); - IRDA_ASSERT(slsap_sel != LSAP_ANY, return TRUE;); - -#ifdef CONFIG_IRDA_ULTRA - /* Accept all bindings to the connectionless LSAP */ - if (slsap_sel == LSAP_CONNLESS) - return FALSE; -#endif /* CONFIG_IRDA_ULTRA */ - - /* Valid values are between 0 and 127 (0x0-0x6F) */ - if (slsap_sel > LSAP_MAX) - return TRUE; - - /* - * Check if slsap is already in use. To do this we have to loop over - * every IrLAP connection and check every LSAP associated with each - * the connection. - */ - spin_lock_irqsave_nested(&irlmp->links->hb_spinlock, flags, - SINGLE_DEPTH_NESTING); - lap = (struct lap_cb *) hashbin_get_first(irlmp->links); - while (lap != NULL) { - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, goto errlap;); - - /* Careful for priority inversions here ! - * irlmp->links is never taken while another IrDA - * spinlock is held, so we are safe. Jean II */ - spin_lock(&lap->lsaps->hb_spinlock); - - /* For this IrLAP, check all the LSAPs */ - self = (struct lsap_cb *) hashbin_get_first(lap->lsaps); - while (self != NULL) { - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, - goto errlsap;); - - if ((self->slsap_sel == slsap_sel)) { - pr_debug("Source LSAP selector=%02x in use\n", - self->slsap_sel); - goto errlsap; - } - self = (struct lsap_cb*) hashbin_get_next(lap->lsaps); - } - spin_unlock(&lap->lsaps->hb_spinlock); - - /* Next LAP */ - lap = (struct lap_cb *) hashbin_get_next(irlmp->links); - } - spin_unlock_irqrestore(&irlmp->links->hb_spinlock, flags); - - /* - * Server sockets are typically waiting for connections and - * therefore reside in the unconnected list. We don't want - * to give out their LSAPs for obvious reasons... - * Jean II - */ - spin_lock_irqsave(&irlmp->unconnected_lsaps->hb_spinlock, flags); - - self = (struct lsap_cb *) hashbin_get_first(irlmp->unconnected_lsaps); - while (self != NULL) { - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, goto erruncon;); - if ((self->slsap_sel == slsap_sel)) { - pr_debug("Source LSAP selector=%02x in use (unconnected)\n", - self->slsap_sel); - goto erruncon; - } - self = (struct lsap_cb*) hashbin_get_next(irlmp->unconnected_lsaps); - } - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, flags); - - return FALSE; - - /* Error exit from within one of the two nested loops. - * Make sure we release the right spinlock in the righ order. - * Jean II */ -errlsap: - spin_unlock(&lap->lsaps->hb_spinlock); -IRDA_ASSERT_LABEL(errlap:) - spin_unlock_irqrestore(&irlmp->links->hb_spinlock, flags); - return TRUE; - - /* Error exit from within the unconnected loop. - * Just one spinlock to release... Jean II */ -erruncon: - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, flags); - return TRUE; -} - -/* - * Function irlmp_find_free_slsap () - * - * Find a free source LSAP to use. This function is called if the service - * user has requested a source LSAP equal to LM_ANY - */ -static __u8 irlmp_find_free_slsap(void) -{ - __u8 lsap_sel; - int wrapped = 0; - - IRDA_ASSERT(irlmp != NULL, return -1;); - IRDA_ASSERT(irlmp->magic == LMP_MAGIC, return -1;); - - /* Most users don't really care which LSAPs they are given, - * and therefore we automatically give them a free LSAP. - * This function try to find a suitable LSAP, i.e. which is - * not in use and is within the acceptable range. Jean II */ - - do { - /* Always increment to LSAP number before using it. - * In theory, we could reuse the last LSAP number, as long - * as it is no longer in use. Some IrDA stack do that. - * However, the previous socket may be half closed, i.e. - * we closed it, we think it's no longer in use, but the - * other side did not receive our close and think it's - * active and still send data on it. - * This is similar to what is done with PIDs and TCP ports. - * Also, this reduce the number of calls to irlmp_slsap_inuse() - * which is an expensive function to call. - * Jean II */ - irlmp->last_lsap_sel++; - - /* Check if we need to wraparound (0x70-0x7f are reserved) */ - if (irlmp->last_lsap_sel > LSAP_MAX) { - /* 0x00-0x10 are also reserved for well know ports */ - irlmp->last_lsap_sel = 0x10; - - /* Make sure we terminate the loop */ - if (wrapped++) { - net_err_ratelimited("%s: no more free LSAPs !\n", - __func__); - return 0; - } - } - - /* If the LSAP is in use, try the next one. - * Despite the autoincrement, we need to check if the lsap - * is really in use or not, first because LSAP may be - * directly allocated in irlmp_open_lsap(), and also because - * we may wraparound on old sockets. Jean II */ - } while (irlmp_slsap_inuse(irlmp->last_lsap_sel)); - - /* Got it ! */ - lsap_sel = irlmp->last_lsap_sel; - pr_debug("%s(), found free lsap_sel=%02x\n", - __func__, lsap_sel); - - return lsap_sel; -} - -/* - * Function irlmp_convert_lap_reason (lap_reason) - * - * Converts IrLAP disconnect reason codes to IrLMP disconnect reason - * codes - * - */ -LM_REASON irlmp_convert_lap_reason( LAP_REASON lap_reason) -{ - int reason = LM_LAP_DISCONNECT; - - switch (lap_reason) { - case LAP_DISC_INDICATION: /* Received a disconnect request from peer */ - pr_debug("%s(), LAP_DISC_INDICATION\n", __func__); - reason = LM_USER_REQUEST; - break; - case LAP_NO_RESPONSE: /* To many retransmits without response */ - pr_debug("%s(), LAP_NO_RESPONSE\n", __func__); - reason = LM_LAP_DISCONNECT; - break; - case LAP_RESET_INDICATION: - pr_debug("%s(), LAP_RESET_INDICATION\n", __func__); - reason = LM_LAP_RESET; - break; - case LAP_FOUND_NONE: - case LAP_MEDIA_BUSY: - case LAP_PRIMARY_CONFLICT: - pr_debug("%s(), LAP_FOUND_NONE, LAP_MEDIA_BUSY or LAP_PRIMARY_CONFLICT\n", - __func__); - reason = LM_CONNECT_FAILURE; - break; - default: - pr_debug("%s(), Unknown IrLAP disconnect reason %d!\n", - __func__, lap_reason); - reason = LM_LAP_DISCONNECT; - break; - } - - return reason; -} - -#ifdef CONFIG_PROC_FS - -struct irlmp_iter_state { - hashbin_t *hashbin; -}; - -#define LSAP_START_TOKEN ((void *)1) -#define LINK_START_TOKEN ((void *)2) - -static void *irlmp_seq_hb_idx(struct irlmp_iter_state *iter, loff_t *off) -{ - void *element; - - spin_lock_irq(&iter->hashbin->hb_spinlock); - for (element = hashbin_get_first(iter->hashbin); - element != NULL; - element = hashbin_get_next(iter->hashbin)) { - if (!off || (*off)-- == 0) { - /* NB: hashbin left locked */ - return element; - } - } - spin_unlock_irq(&iter->hashbin->hb_spinlock); - iter->hashbin = NULL; - return NULL; -} - - -static void *irlmp_seq_start(struct seq_file *seq, loff_t *pos) -{ - struct irlmp_iter_state *iter = seq->private; - void *v; - loff_t off = *pos; - - iter->hashbin = NULL; - if (off-- == 0) - return LSAP_START_TOKEN; - - iter->hashbin = irlmp->unconnected_lsaps; - v = irlmp_seq_hb_idx(iter, &off); - if (v) - return v; - - if (off-- == 0) - return LINK_START_TOKEN; - - iter->hashbin = irlmp->links; - return irlmp_seq_hb_idx(iter, &off); -} - -static void *irlmp_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - struct irlmp_iter_state *iter = seq->private; - - ++*pos; - - if (v == LSAP_START_TOKEN) { /* start of list of lsaps */ - iter->hashbin = irlmp->unconnected_lsaps; - v = irlmp_seq_hb_idx(iter, NULL); - return v ? v : LINK_START_TOKEN; - } - - if (v == LINK_START_TOKEN) { /* start of list of links */ - iter->hashbin = irlmp->links; - return irlmp_seq_hb_idx(iter, NULL); - } - - v = hashbin_get_next(iter->hashbin); - - if (v == NULL) { /* no more in this hash bin */ - spin_unlock_irq(&iter->hashbin->hb_spinlock); - - if (iter->hashbin == irlmp->unconnected_lsaps) - v = LINK_START_TOKEN; - - iter->hashbin = NULL; - } - return v; -} - -static void irlmp_seq_stop(struct seq_file *seq, void *v) -{ - struct irlmp_iter_state *iter = seq->private; - - if (iter->hashbin) - spin_unlock_irq(&iter->hashbin->hb_spinlock); -} - -static int irlmp_seq_show(struct seq_file *seq, void *v) -{ - const struct irlmp_iter_state *iter = seq->private; - struct lsap_cb *self = v; - - if (v == LSAP_START_TOKEN) - seq_puts(seq, "Unconnected LSAPs:\n"); - else if (v == LINK_START_TOKEN) - seq_puts(seq, "\nRegistered Link Layers:\n"); - else if (iter->hashbin == irlmp->unconnected_lsaps) { - self = v; - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -EINVAL; ); - seq_printf(seq, "lsap state: %s, ", - irlsap_state[ self->lsap_state]); - seq_printf(seq, - "slsap_sel: %#02x, dlsap_sel: %#02x, ", - self->slsap_sel, self->dlsap_sel); - seq_printf(seq, "(%s)", self->notify.name); - seq_printf(seq, "\n"); - } else if (iter->hashbin == irlmp->links) { - struct lap_cb *lap = v; - - seq_printf(seq, "lap state: %s, ", - irlmp_state[lap->lap_state]); - - seq_printf(seq, "saddr: %#08x, daddr: %#08x, ", - lap->saddr, lap->daddr); - seq_printf(seq, "num lsaps: %d", - HASHBIN_GET_SIZE(lap->lsaps)); - seq_printf(seq, "\n"); - - /* Careful for priority inversions here ! - * All other uses of attrib spinlock are independent of - * the object spinlock, so we are safe. Jean II */ - spin_lock(&lap->lsaps->hb_spinlock); - - seq_printf(seq, "\n Connected LSAPs:\n"); - for (self = (struct lsap_cb *) hashbin_get_first(lap->lsaps); - self != NULL; - self = (struct lsap_cb *)hashbin_get_next(lap->lsaps)) { - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, - goto outloop;); - seq_printf(seq, " lsap state: %s, ", - irlsap_state[ self->lsap_state]); - seq_printf(seq, - "slsap_sel: %#02x, dlsap_sel: %#02x, ", - self->slsap_sel, self->dlsap_sel); - seq_printf(seq, "(%s)", self->notify.name); - seq_putc(seq, '\n'); - - } - IRDA_ASSERT_LABEL(outloop:) - spin_unlock(&lap->lsaps->hb_spinlock); - seq_putc(seq, '\n'); - } else - return -EINVAL; - - return 0; -} - -static const struct seq_operations irlmp_seq_ops = { - .start = irlmp_seq_start, - .next = irlmp_seq_next, - .stop = irlmp_seq_stop, - .show = irlmp_seq_show, -}; - -static int irlmp_seq_open(struct inode *inode, struct file *file) -{ - IRDA_ASSERT(irlmp != NULL, return -EINVAL;); - - return seq_open_private(file, &irlmp_seq_ops, - sizeof(struct irlmp_iter_state)); -} - -const struct file_operations irlmp_seq_fops = { - .owner = THIS_MODULE, - .open = irlmp_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, -}; - -#endif /* PROC_FS */ diff --git a/net/irda/irlmp_event.c b/net/irda/irlmp_event.c deleted file mode 100644 index e306cf2c1e04..000000000000 --- a/net/irda/irlmp_event.c +++ /dev/null @@ -1,886 +0,0 @@ -/********************************************************************* - * - * Filename: irlmp_event.c - * Version: 0.8 - * Description: An IrDA LMP event driver for Linux - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Aug 4 20:40:53 1997 - * Modified at: Tue Dec 14 23:04:16 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/kernel.h> - -#include <net/irda/irda.h> -#include <net/irda/timer.h> -#include <net/irda/irlap.h> -#include <net/irda/irlmp.h> -#include <net/irda/irlmp_frame.h> -#include <net/irda/irlmp_event.h> - -const char *const irlmp_state[] = { - "LAP_STANDBY", - "LAP_U_CONNECT", - "LAP_ACTIVE", -}; - -const char *const irlsap_state[] = { - "LSAP_DISCONNECTED", - "LSAP_CONNECT", - "LSAP_CONNECT_PEND", - "LSAP_DATA_TRANSFER_READY", - "LSAP_SETUP", - "LSAP_SETUP_PEND", -}; - -static const char *const irlmp_event[] __maybe_unused = { - "LM_CONNECT_REQUEST", - "LM_CONNECT_CONFIRM", - "LM_CONNECT_RESPONSE", - "LM_CONNECT_INDICATION", - - "LM_DISCONNECT_INDICATION", - "LM_DISCONNECT_REQUEST", - - "LM_DATA_REQUEST", - "LM_UDATA_REQUEST", - "LM_DATA_INDICATION", - "LM_UDATA_INDICATION", - - "LM_WATCHDOG_TIMEOUT", - - /* IrLAP events */ - "LM_LAP_CONNECT_REQUEST", - "LM_LAP_CONNECT_INDICATION", - "LM_LAP_CONNECT_CONFIRM", - "LM_LAP_DISCONNECT_INDICATION", - "LM_LAP_DISCONNECT_REQUEST", - "LM_LAP_DISCOVERY_REQUEST", - "LM_LAP_DISCOVERY_CONFIRM", - "LM_LAP_IDLE_TIMEOUT", -}; - -/* LAP Connection control proto declarations */ -static void irlmp_state_standby (struct lap_cb *, IRLMP_EVENT, - struct sk_buff *); -static void irlmp_state_u_connect(struct lap_cb *, IRLMP_EVENT, - struct sk_buff *); -static void irlmp_state_active (struct lap_cb *, IRLMP_EVENT, - struct sk_buff *); - -/* LSAP Connection control proto declarations */ -static int irlmp_state_disconnected(struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); -static int irlmp_state_connect (struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); -static int irlmp_state_connect_pend(struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); -static int irlmp_state_dtr (struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); -static int irlmp_state_setup (struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); -static int irlmp_state_setup_pend (struct lsap_cb *, IRLMP_EVENT, - struct sk_buff *); - -static void (*lap_state[]) (struct lap_cb *, IRLMP_EVENT, struct sk_buff *) = -{ - irlmp_state_standby, - irlmp_state_u_connect, - irlmp_state_active, -}; - -static int (*lsap_state[])( struct lsap_cb *, IRLMP_EVENT, struct sk_buff *) = -{ - irlmp_state_disconnected, - irlmp_state_connect, - irlmp_state_connect_pend, - irlmp_state_dtr, - irlmp_state_setup, - irlmp_state_setup_pend -}; - -static inline void irlmp_next_lap_state(struct lap_cb *self, - IRLMP_STATE state) -{ - /* - pr_debug("%s(), LMP LAP = %s\n", __func__, irlmp_state[state]); - */ - self->lap_state = state; -} - -static inline void irlmp_next_lsap_state(struct lsap_cb *self, - LSAP_STATE state) -{ - /* - IRDA_ASSERT(self != NULL, return;); - pr_debug("%s(), LMP LSAP = %s\n", __func__, irlsap_state[state]); - */ - self->lsap_state = state; -} - -/* Do connection control events */ -int irlmp_do_lsap_event(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - pr_debug("%s(), EVENT = %s, STATE = %s\n", - __func__, irlmp_event[event], irlsap_state[self->lsap_state]); - - return (*lsap_state[self->lsap_state]) (self, event, skb); -} - -/* - * Function do_lap_event (event, skb, info) - * - * Do IrLAP control events - * - */ -void irlmp_do_lap_event(struct lap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - - pr_debug("%s(), EVENT = %s, STATE = %s\n", __func__, - irlmp_event[event], - irlmp_state[self->lap_state]); - - (*lap_state[self->lap_state]) (self, event, skb); -} - -void irlmp_discovery_timer_expired(void *data) -{ - /* We always cleanup the log (active & passive discovery) */ - irlmp_do_expiry(); - - irlmp_do_discovery(sysctl_discovery_slots); - - /* Restart timer */ - irlmp_start_discovery_timer(irlmp, sysctl_discovery_timeout * HZ); -} - -void irlmp_watchdog_timer_expired(void *data) -{ - struct lsap_cb *self = (struct lsap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return;); - - irlmp_do_lsap_event(self, LM_WATCHDOG_TIMEOUT, NULL); -} - -void irlmp_idle_timer_expired(void *data) -{ - struct lap_cb *self = (struct lap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - - irlmp_do_lap_event(self, LM_LAP_IDLE_TIMEOUT, NULL); -} - -/* - * Send an event on all LSAPs attached to this LAP. - */ -static inline void -irlmp_do_all_lsap_event(hashbin_t * lsap_hashbin, - IRLMP_EVENT event) -{ - struct lsap_cb *lsap; - struct lsap_cb *lsap_next; - - /* Note : this function use the new hashbin_find_next() - * function, instead of the old hashbin_get_next(). - * This make sure that we are always pointing one lsap - * ahead, so that if the current lsap is removed as the - * result of sending the event, we don't care. - * Also, as we store the context ourselves, if an enumeration - * of the same lsap hashbin happens as the result of sending the - * event, we don't care. - * The only problem is if the next lsap is removed. In that case, - * hashbin_find_next() will return NULL and we will abort the - * enumeration. - Jean II */ - - /* Also : we don't accept any skb in input. We can *NOT* pass - * the same skb to multiple clients safely, we would need to - * skb_clone() it. - Jean II */ - - lsap = (struct lsap_cb *) hashbin_get_first(lsap_hashbin); - - while (NULL != hashbin_find_next(lsap_hashbin, - (long) lsap, - NULL, - (void *) &lsap_next) ) { - irlmp_do_lsap_event(lsap, event, NULL); - lsap = lsap_next; - } -} - -/********************************************************************* - * - * LAP connection control states - * - ********************************************************************/ - -/* - * Function irlmp_state_standby (event, skb, info) - * - * STANDBY, The IrLAP connection does not exist. - * - */ -static void irlmp_state_standby(struct lap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - IRDA_ASSERT(self->irlap != NULL, return;); - - switch (event) { - case LM_LAP_DISCOVERY_REQUEST: - /* irlmp_next_station_state( LMP_DISCOVER); */ - - irlap_discovery_request(self->irlap, &irlmp->discovery_cmd); - break; - case LM_LAP_CONNECT_INDICATION: - /* It's important to switch state first, to avoid IrLMP to - * think that the link is free since IrLMP may then start - * discovery before the connection is properly set up. DB. - */ - irlmp_next_lap_state(self, LAP_ACTIVE); - - /* Just accept connection TODO, this should be fixed */ - irlap_connect_response(self->irlap, skb); - break; - case LM_LAP_CONNECT_REQUEST: - pr_debug("%s() LS_CONNECT_REQUEST\n", __func__); - - irlmp_next_lap_state(self, LAP_U_CONNECT); - - /* FIXME: need to set users requested QoS */ - irlap_connect_request(self->irlap, self->daddr, NULL, 0); - break; - case LM_LAP_DISCONNECT_INDICATION: - pr_debug("%s(), Error LM_LAP_DISCONNECT_INDICATION\n", - __func__); - - irlmp_next_lap_state(self, LAP_STANDBY); - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlmp_event[event]); - break; - } -} - -/* - * Function irlmp_state_u_connect (event, skb, info) - * - * U_CONNECT, The layer above has tried to open an LSAP connection but - * since the IrLAP connection does not exist, we must first start an - * IrLAP connection. We are now waiting response from IrLAP. - * */ -static void irlmp_state_u_connect(struct lap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - pr_debug("%s(), event=%s\n", __func__, irlmp_event[event]); - - switch (event) { - case LM_LAP_CONNECT_INDICATION: - /* It's important to switch state first, to avoid IrLMP to - * think that the link is free since IrLMP may then start - * discovery before the connection is properly set up. DB. - */ - irlmp_next_lap_state(self, LAP_ACTIVE); - - /* Just accept connection TODO, this should be fixed */ - irlap_connect_response(self->irlap, skb); - - /* Tell LSAPs that they can start sending data */ - irlmp_do_all_lsap_event(self->lsaps, LM_LAP_CONNECT_CONFIRM); - - /* Note : by the time we get there (LAP retries and co), - * the lsaps may already have gone. This avoid getting stuck - * forever in LAP_ACTIVE state - Jean II */ - if (HASHBIN_GET_SIZE(self->lsaps) == 0) { - pr_debug("%s() NO LSAPs !\n", __func__); - irlmp_start_idle_timer(self, LM_IDLE_TIMEOUT); - } - break; - case LM_LAP_CONNECT_REQUEST: - /* Already trying to connect */ - break; - case LM_LAP_CONNECT_CONFIRM: - /* For all lsap_ce E Associated do LS_Connect_confirm */ - irlmp_next_lap_state(self, LAP_ACTIVE); - - /* Tell LSAPs that they can start sending data */ - irlmp_do_all_lsap_event(self->lsaps, LM_LAP_CONNECT_CONFIRM); - - /* Note : by the time we get there (LAP retries and co), - * the lsaps may already have gone. This avoid getting stuck - * forever in LAP_ACTIVE state - Jean II */ - if (HASHBIN_GET_SIZE(self->lsaps) == 0) { - pr_debug("%s() NO LSAPs !\n", __func__); - irlmp_start_idle_timer(self, LM_IDLE_TIMEOUT); - } - break; - case LM_LAP_DISCONNECT_INDICATION: - pr_debug("%s(), LM_LAP_DISCONNECT_INDICATION\n", __func__); - irlmp_next_lap_state(self, LAP_STANDBY); - - /* Send disconnect event to all LSAPs using this link */ - irlmp_do_all_lsap_event(self->lsaps, - LM_LAP_DISCONNECT_INDICATION); - break; - case LM_LAP_DISCONNECT_REQUEST: - pr_debug("%s(), LM_LAP_DISCONNECT_REQUEST\n", __func__); - - /* One of the LSAP did timeout or was closed, if it was - * the last one, try to get out of here - Jean II */ - if (HASHBIN_GET_SIZE(self->lsaps) <= 1) { - irlap_disconnect_request(self->irlap); - } - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlmp_event[event]); - break; - } -} - -/* - * Function irlmp_state_active (event, skb, info) - * - * ACTIVE, IrLAP connection is active - * - */ -static void irlmp_state_active(struct lap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - switch (event) { - case LM_LAP_CONNECT_REQUEST: - pr_debug("%s(), LS_CONNECT_REQUEST\n", __func__); - - /* - * IrLAP may have a pending disconnect. We tried to close - * IrLAP, but it was postponed because the link was - * busy or we were still sending packets. As we now - * need it, make sure it stays on. Jean II - */ - irlap_clear_disconnect(self->irlap); - - /* - * LAP connection already active, just bounce back! Since we - * don't know which LSAP that tried to do this, we have to - * notify all LSAPs using this LAP, but that should be safe to - * do anyway. - */ - irlmp_do_all_lsap_event(self->lsaps, LM_LAP_CONNECT_CONFIRM); - - /* Needed by connect indication */ - irlmp_do_all_lsap_event(irlmp->unconnected_lsaps, - LM_LAP_CONNECT_CONFIRM); - /* Keep state */ - break; - case LM_LAP_DISCONNECT_REQUEST: - /* - * Need to find out if we should close IrLAP or not. If there - * is only one LSAP connection left on this link, that LSAP - * must be the one that tries to close IrLAP. It will be - * removed later and moved to the list of unconnected LSAPs - */ - if (HASHBIN_GET_SIZE(self->lsaps) > 0) { - /* Timer value is checked in irsysctl - Jean II */ - irlmp_start_idle_timer(self, sysctl_lap_keepalive_time * HZ / 1000); - } else { - /* No more connections, so close IrLAP */ - - /* We don't want to change state just yet, because - * we want to reflect accurately the real state of - * the LAP, not the state we wish it was in, - * so that we don't lose LM_LAP_CONNECT_REQUEST. - * In some cases, IrLAP won't close the LAP - * immediately. For example, it might still be - * retrying packets or waiting for the pf bit. - * As the LAP always send a DISCONNECT_INDICATION - * in PCLOSE or SCLOSE, just change state on that. - * Jean II */ - irlap_disconnect_request(self->irlap); - } - break; - case LM_LAP_IDLE_TIMEOUT: - if (HASHBIN_GET_SIZE(self->lsaps) == 0) { - /* Same reasoning as above - keep state */ - irlap_disconnect_request(self->irlap); - } - break; - case LM_LAP_DISCONNECT_INDICATION: - irlmp_next_lap_state(self, LAP_STANDBY); - - /* In some case, at this point our side has already closed - * all lsaps, and we are waiting for the idle_timer to - * expire. If another device reconnect immediately, the - * idle timer will expire in the midle of the connection - * initialisation, screwing up things a lot... - * Therefore, we must stop the timer... */ - irlmp_stop_idle_timer(self); - - /* - * Inform all connected LSAP's using this link - */ - irlmp_do_all_lsap_event(self->lsaps, - LM_LAP_DISCONNECT_INDICATION); - - /* Force an expiry of the discovery log. - * Now that the LAP is free, the system may attempt to - * connect to another device. Unfortunately, our entries - * are stale. There is a small window (<3s) before the - * normal discovery will run and where irlmp_connect_request() - * can get the wrong info, so make sure things get - * cleaned *NOW* ;-) - Jean II */ - irlmp_do_expiry(); - break; - default: - pr_debug("%s(), Unknown event %s\n", - __func__, irlmp_event[event]); - break; - } -} - -/********************************************************************* - * - * LSAP connection control states - * - ********************************************************************/ - -/* - * Function irlmp_state_disconnected (event, skb, info) - * - * DISCONNECTED - * - */ -static int irlmp_state_disconnected(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - switch (event) { -#ifdef CONFIG_IRDA_ULTRA - case LM_UDATA_INDICATION: - /* This is most bizarre. Those packets are aka unreliable - * connected, aka IrLPT or SOCK_DGRAM/IRDAPROTO_UNITDATA. - * Why do we pass them as Ultra ??? Jean II */ - irlmp_connless_data_indication(self, skb); - break; -#endif /* CONFIG_IRDA_ULTRA */ - case LM_CONNECT_REQUEST: - pr_debug("%s(), LM_CONNECT_REQUEST\n", __func__); - - if (self->conn_skb) { - net_warn_ratelimited("%s: busy with another request!\n", - __func__); - return -EBUSY; - } - /* Don't forget to refcount it (see irlmp_connect_request()) */ - skb_get(skb); - self->conn_skb = skb; - - irlmp_next_lsap_state(self, LSAP_SETUP_PEND); - - /* Start watchdog timer (5 secs for now) */ - irlmp_start_watchdog_timer(self, 5*HZ); - - irlmp_do_lap_event(self->lap, LM_LAP_CONNECT_REQUEST, NULL); - break; - case LM_CONNECT_INDICATION: - if (self->conn_skb) { - net_warn_ratelimited("%s: busy with another request!\n", - __func__); - return -EBUSY; - } - /* Don't forget to refcount it (see irlap_driver_rcv()) */ - skb_get(skb); - self->conn_skb = skb; - - irlmp_next_lsap_state(self, LSAP_CONNECT_PEND); - - /* Start watchdog timer - * This is not mentionned in the spec, but there is a rare - * race condition that can get the socket stuck. - * If we receive this event while our LAP is closing down, - * the LM_LAP_CONNECT_REQUEST get lost and we get stuck in - * CONNECT_PEND state forever. - * The other cause of getting stuck down there is if the - * higher layer never reply to the CONNECT_INDICATION. - * Anyway, it make sense to make sure that we always have - * a backup plan. 1 second is plenty (should be immediate). - * Jean II */ - irlmp_start_watchdog_timer(self, 1*HZ); - - irlmp_do_lap_event(self->lap, LM_LAP_CONNECT_REQUEST, NULL); - break; - default: - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} - -/* - * Function irlmp_state_connect (self, event, skb) - * - * CONNECT - * - */ -static int irlmp_state_connect(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - struct lsap_cb *lsap; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - switch (event) { - case LM_CONNECT_RESPONSE: - /* - * Bind this LSAP to the IrLAP link where the connect was - * received - */ - lsap = hashbin_remove(irlmp->unconnected_lsaps, (long) self, - NULL); - - IRDA_ASSERT(lsap == self, return -1;); - IRDA_ASSERT(self->lap != NULL, return -1;); - IRDA_ASSERT(self->lap->lsaps != NULL, return -1;); - - hashbin_insert(self->lap->lsaps, (irda_queue_t *) self, - (long) self, NULL); - - set_bit(0, &self->connected); /* TRUE */ - - irlmp_send_lcf_pdu(self->lap, self->dlsap_sel, - self->slsap_sel, CONNECT_CNF, skb); - - del_timer(&self->watchdog_timer); - - irlmp_next_lsap_state(self, LSAP_DATA_TRANSFER_READY); - break; - case LM_WATCHDOG_TIMEOUT: - /* May happen, who knows... - * Jean II */ - pr_debug("%s() WATCHDOG_TIMEOUT!\n", __func__); - - /* Disconnect, get out... - Jean II */ - self->lap = NULL; - self->dlsap_sel = LSAP_ANY; - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - break; - default: - /* LM_LAP_DISCONNECT_INDICATION : Should never happen, we - * are *not* yet bound to the IrLAP link. Jean II */ - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} - -/* - * Function irlmp_state_connect_pend (event, skb, info) - * - * CONNECT_PEND - * - */ -static int irlmp_state_connect_pend(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - switch (event) { - case LM_CONNECT_REQUEST: - /* Keep state */ - break; - case LM_CONNECT_RESPONSE: - pr_debug("%s(), LM_CONNECT_RESPONSE, no indication issued yet\n", - __func__); - /* Keep state */ - break; - case LM_DISCONNECT_REQUEST: - pr_debug("%s(), LM_DISCONNECT_REQUEST, not yet bound to IrLAP connection\n", - __func__); - /* Keep state */ - break; - case LM_LAP_CONNECT_CONFIRM: - pr_debug("%s(), LS_CONNECT_CONFIRM\n", __func__); - irlmp_next_lsap_state(self, LSAP_CONNECT); - - tx_skb = self->conn_skb; - self->conn_skb = NULL; - - irlmp_connect_indication(self, tx_skb); - /* Drop reference count - see irlmp_connect_indication(). */ - dev_kfree_skb(tx_skb); - break; - case LM_WATCHDOG_TIMEOUT: - /* Will happen in some rare cases because of a race condition. - * Just make sure we don't stay there forever... - * Jean II */ - pr_debug("%s() WATCHDOG_TIMEOUT!\n", __func__); - - /* Go back to disconnected mode, keep the socket waiting */ - self->lap = NULL; - self->dlsap_sel = LSAP_ANY; - if(self->conn_skb) - dev_kfree_skb(self->conn_skb); - self->conn_skb = NULL; - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - break; - default: - /* LM_LAP_DISCONNECT_INDICATION : Should never happen, we - * are *not* yet bound to the IrLAP link. Jean II */ - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} - -/* - * Function irlmp_state_dtr (self, event, skb) - * - * DATA_TRANSFER_READY - * - */ -static int irlmp_state_dtr(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - LM_REASON reason; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - IRDA_ASSERT(self->lap != NULL, return -1;); - - switch (event) { - case LM_DATA_REQUEST: /* Optimize for the common case */ - irlmp_send_data_pdu(self->lap, self->dlsap_sel, - self->slsap_sel, FALSE, skb); - break; - case LM_DATA_INDICATION: /* Optimize for the common case */ - irlmp_data_indication(self, skb); - break; - case LM_UDATA_REQUEST: - IRDA_ASSERT(skb != NULL, return -1;); - irlmp_send_data_pdu(self->lap, self->dlsap_sel, - self->slsap_sel, TRUE, skb); - break; - case LM_UDATA_INDICATION: - irlmp_udata_indication(self, skb); - break; - case LM_CONNECT_REQUEST: - pr_debug("%s(), LM_CONNECT_REQUEST, error, LSAP already connected\n", - __func__); - /* Keep state */ - break; - case LM_CONNECT_RESPONSE: - pr_debug("%s(), LM_CONNECT_RESPONSE, error, LSAP already connected\n", - __func__); - /* Keep state */ - break; - case LM_DISCONNECT_REQUEST: - irlmp_send_lcf_pdu(self->lap, self->dlsap_sel, self->slsap_sel, - DISCONNECT, skb); - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - /* Called only from irlmp_disconnect_request(), will - * unbind from LAP over there. Jean II */ - - /* Try to close the LAP connection if its still there */ - if (self->lap) { - pr_debug("%s(), trying to close IrLAP\n", - __func__); - irlmp_do_lap_event(self->lap, - LM_LAP_DISCONNECT_REQUEST, - NULL); - } - break; - case LM_LAP_DISCONNECT_INDICATION: - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - reason = irlmp_convert_lap_reason(self->lap->reason); - - irlmp_disconnect_indication(self, reason, NULL); - break; - case LM_DISCONNECT_INDICATION: - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - IRDA_ASSERT(self->lap != NULL, return -1;); - IRDA_ASSERT(self->lap->magic == LMP_LAP_MAGIC, return -1;); - - IRDA_ASSERT(skb != NULL, return -1;); - IRDA_ASSERT(skb->len > 3, return -1;); - reason = skb->data[3]; - - /* Try to close the LAP connection */ - pr_debug("%s(), trying to close IrLAP\n", __func__); - irlmp_do_lap_event(self->lap, LM_LAP_DISCONNECT_REQUEST, NULL); - - irlmp_disconnect_indication(self, reason, skb); - break; - default: - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} - -/* - * Function irlmp_state_setup (event, skb, info) - * - * SETUP, Station Control has set up the underlying IrLAP connection. - * An LSAP connection request has been transmitted to the peer - * LSAP-Connection Control FSM and we are awaiting reply. - */ -static int irlmp_state_setup(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - LM_REASON reason; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LMP_LSAP_MAGIC, return -1;); - - switch (event) { - case LM_CONNECT_CONFIRM: - irlmp_next_lsap_state(self, LSAP_DATA_TRANSFER_READY); - - del_timer(&self->watchdog_timer); - - irlmp_connect_confirm(self, skb); - break; - case LM_DISCONNECT_INDICATION: - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - IRDA_ASSERT(self->lap != NULL, return -1;); - IRDA_ASSERT(self->lap->magic == LMP_LAP_MAGIC, return -1;); - - IRDA_ASSERT(skb != NULL, return -1;); - IRDA_ASSERT(skb->len > 3, return -1;); - reason = skb->data[3]; - - /* Try to close the LAP connection */ - pr_debug("%s(), trying to close IrLAP\n", __func__); - irlmp_do_lap_event(self->lap, LM_LAP_DISCONNECT_REQUEST, NULL); - - irlmp_disconnect_indication(self, reason, skb); - break; - case LM_LAP_DISCONNECT_INDICATION: - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - del_timer(&self->watchdog_timer); - - IRDA_ASSERT(self->lap != NULL, return -1;); - IRDA_ASSERT(self->lap->magic == LMP_LAP_MAGIC, return -1;); - - reason = irlmp_convert_lap_reason(self->lap->reason); - - irlmp_disconnect_indication(self, reason, skb); - break; - case LM_WATCHDOG_TIMEOUT: - pr_debug("%s() WATCHDOG_TIMEOUT!\n", __func__); - - IRDA_ASSERT(self->lap != NULL, return -1;); - irlmp_do_lap_event(self->lap, LM_LAP_DISCONNECT_REQUEST, NULL); - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - irlmp_disconnect_indication(self, LM_CONNECT_FAILURE, NULL); - break; - default: - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} - -/* - * Function irlmp_state_setup_pend (event, skb, info) - * - * SETUP_PEND, An LM_CONNECT_REQUEST has been received from the service - * user to set up an LSAP connection. A request has been sent to the - * LAP FSM to set up the underlying IrLAP connection, and we - * are awaiting confirm. - */ -static int irlmp_state_setup_pend(struct lsap_cb *self, IRLMP_EVENT event, - struct sk_buff *skb) -{ - struct sk_buff *tx_skb; - LM_REASON reason; - int ret = 0; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(irlmp != NULL, return -1;); - - switch (event) { - case LM_LAP_CONNECT_CONFIRM: - IRDA_ASSERT(self->conn_skb != NULL, return -1;); - - tx_skb = self->conn_skb; - self->conn_skb = NULL; - - irlmp_send_lcf_pdu(self->lap, self->dlsap_sel, - self->slsap_sel, CONNECT_CMD, tx_skb); - /* Drop reference count - see irlap_data_request(). */ - dev_kfree_skb(tx_skb); - - irlmp_next_lsap_state(self, LSAP_SETUP); - break; - case LM_WATCHDOG_TIMEOUT: - pr_debug("%s() : WATCHDOG_TIMEOUT !\n", __func__); - - IRDA_ASSERT(self->lap != NULL, return -1;); - irlmp_do_lap_event(self->lap, LM_LAP_DISCONNECT_REQUEST, NULL); - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - irlmp_disconnect_indication(self, LM_CONNECT_FAILURE, NULL); - break; - case LM_LAP_DISCONNECT_INDICATION: /* LS_Disconnect.indication */ - del_timer( &self->watchdog_timer); - - irlmp_next_lsap_state(self, LSAP_DISCONNECTED); - - reason = irlmp_convert_lap_reason(self->lap->reason); - - irlmp_disconnect_indication(self, reason, NULL); - break; - default: - pr_debug("%s(), Unknown event %s on LSAP %#02x\n", - __func__, irlmp_event[event], self->slsap_sel); - break; - } - return ret; -} diff --git a/net/irda/irlmp_frame.c b/net/irda/irlmp_frame.c deleted file mode 100644 index 38b0f994bc7b..000000000000 --- a/net/irda/irlmp_frame.c +++ /dev/null @@ -1,476 +0,0 @@ -/********************************************************************* - * - * Filename: irlmp_frame.c - * Version: 0.9 - * Description: IrLMP frame implementation - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Aug 19 02:09:59 1997 - * Modified at: Mon Dec 13 13:41:12 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999 Dag Brattli <dagb@cs.uit.no> - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> -#include <linux/kernel.h> - -#include <net/irda/irda.h> -#include <net/irda/irlap.h> -#include <net/irda/timer.h> -#include <net/irda/irlmp.h> -#include <net/irda/irlmp_frame.h> -#include <net/irda/discovery.h> - -static struct lsap_cb *irlmp_find_lsap(struct lap_cb *self, __u8 dlsap, - __u8 slsap, int status, hashbin_t *); - -inline void irlmp_send_data_pdu(struct lap_cb *self, __u8 dlsap, __u8 slsap, - int expedited, struct sk_buff *skb) -{ - skb->data[0] = dlsap; - skb->data[1] = slsap; - - if (expedited) { - pr_debug("%s(), sending expedited data\n", __func__); - irlap_data_request(self->irlap, skb, TRUE); - } else - irlap_data_request(self->irlap, skb, FALSE); -} - -/* - * Function irlmp_send_lcf_pdu (dlsap, slsap, opcode,skb) - * - * Send Link Control Frame to IrLAP - */ -void irlmp_send_lcf_pdu(struct lap_cb *self, __u8 dlsap, __u8 slsap, - __u8 opcode, struct sk_buff *skb) -{ - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - frame = skb->data; - - frame[0] = dlsap | CONTROL_BIT; - frame[1] = slsap; - - frame[2] = opcode; - - if (opcode == DISCONNECT) - frame[3] = 0x01; /* Service user request */ - else - frame[3] = 0x00; /* rsvd */ - - irlap_data_request(self->irlap, skb, FALSE); -} - -/* - * Function irlmp_input (skb) - * - * Used by IrLAP to pass received data frames to IrLMP layer - * - */ -void irlmp_link_data_indication(struct lap_cb *self, struct sk_buff *skb, - int unreliable) -{ - struct lsap_cb *lsap; - __u8 slsap_sel; /* Source (this) LSAP address */ - __u8 dlsap_sel; /* Destination LSAP address */ - __u8 *fp; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - IRDA_ASSERT(skb->len > 2, return;); - - fp = skb->data; - - /* - * The next statements may be confusing, but we do this so that - * destination LSAP of received frame is source LSAP in our view - */ - slsap_sel = fp[0] & LSAP_MASK; - dlsap_sel = fp[1]; - - /* - * Check if this is an incoming connection, since we must deal with - * it in a different way than other established connections. - */ - if ((fp[0] & CONTROL_BIT) && (fp[2] == CONNECT_CMD)) { - pr_debug("%s(), incoming connection, source LSAP=%d, dest LSAP=%d\n", - __func__, slsap_sel, dlsap_sel); - - /* Try to find LSAP among the unconnected LSAPs */ - lsap = irlmp_find_lsap(self, dlsap_sel, slsap_sel, CONNECT_CMD, - irlmp->unconnected_lsaps); - - /* Maybe LSAP was already connected, so try one more time */ - if (!lsap) { - pr_debug("%s(), incoming connection for LSAP already connected\n", - __func__); - lsap = irlmp_find_lsap(self, dlsap_sel, slsap_sel, 0, - self->lsaps); - } - } else - lsap = irlmp_find_lsap(self, dlsap_sel, slsap_sel, 0, - self->lsaps); - - if (lsap == NULL) { - pr_debug("IrLMP, Sorry, no LSAP for received frame!\n"); - pr_debug("%s(), slsap_sel = %02x, dlsap_sel = %02x\n", - __func__, slsap_sel, dlsap_sel); - if (fp[0] & CONTROL_BIT) { - pr_debug("%s(), received control frame %02x\n", - __func__, fp[2]); - } else { - pr_debug("%s(), received data frame\n", __func__); - } - return; - } - - /* - * Check if we received a control frame? - */ - if (fp[0] & CONTROL_BIT) { - switch (fp[2]) { - case CONNECT_CMD: - lsap->lap = self; - irlmp_do_lsap_event(lsap, LM_CONNECT_INDICATION, skb); - break; - case CONNECT_CNF: - irlmp_do_lsap_event(lsap, LM_CONNECT_CONFIRM, skb); - break; - case DISCONNECT: - pr_debug("%s(), Disconnect indication!\n", - __func__); - irlmp_do_lsap_event(lsap, LM_DISCONNECT_INDICATION, - skb); - break; - case ACCESSMODE_CMD: - pr_debug("Access mode cmd not implemented!\n"); - break; - case ACCESSMODE_CNF: - pr_debug("Access mode cnf not implemented!\n"); - break; - default: - pr_debug("%s(), Unknown control frame %02x\n", - __func__, fp[2]); - break; - } - } else if (unreliable) { - /* Optimize and bypass the state machine if possible */ - if (lsap->lsap_state == LSAP_DATA_TRANSFER_READY) - irlmp_udata_indication(lsap, skb); - else - irlmp_do_lsap_event(lsap, LM_UDATA_INDICATION, skb); - } else { - /* Optimize and bypass the state machine if possible */ - if (lsap->lsap_state == LSAP_DATA_TRANSFER_READY) - irlmp_data_indication(lsap, skb); - else - irlmp_do_lsap_event(lsap, LM_DATA_INDICATION, skb); - } -} - -/* - * Function irlmp_link_unitdata_indication (self, skb) - * - * - * - */ -#ifdef CONFIG_IRDA_ULTRA -void irlmp_link_unitdata_indication(struct lap_cb *self, struct sk_buff *skb) -{ - struct lsap_cb *lsap; - __u8 slsap_sel; /* Source (this) LSAP address */ - __u8 dlsap_sel; /* Destination LSAP address */ - __u8 pid; /* Protocol identifier */ - __u8 *fp; - unsigned long flags; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - IRDA_ASSERT(skb->len > 2, return;); - - fp = skb->data; - - /* - * The next statements may be confusing, but we do this so that - * destination LSAP of received frame is source LSAP in our view - */ - slsap_sel = fp[0] & LSAP_MASK; - dlsap_sel = fp[1]; - pid = fp[2]; - - if (pid & 0x80) { - pr_debug("%s(), extension in PID not supp!\n", - __func__); - return; - } - - /* Check if frame is addressed to the connectionless LSAP */ - if ((slsap_sel != LSAP_CONNLESS) || (dlsap_sel != LSAP_CONNLESS)) { - pr_debug("%s(), dropping frame!\n", __func__); - return; - } - - /* Search the connectionless LSAP */ - spin_lock_irqsave(&irlmp->unconnected_lsaps->hb_spinlock, flags); - lsap = (struct lsap_cb *) hashbin_get_first(irlmp->unconnected_lsaps); - while (lsap != NULL) { - /* - * Check if source LSAP and dest LSAP selectors and PID match. - */ - if ((lsap->slsap_sel == slsap_sel) && - (lsap->dlsap_sel == dlsap_sel) && - (lsap->pid == pid)) - { - break; - } - lsap = (struct lsap_cb *) hashbin_get_next(irlmp->unconnected_lsaps); - } - spin_unlock_irqrestore(&irlmp->unconnected_lsaps->hb_spinlock, flags); - - if (lsap) - irlmp_connless_data_indication(lsap, skb); - else { - pr_debug("%s(), found no matching LSAP!\n", __func__); - } -} -#endif /* CONFIG_IRDA_ULTRA */ - -/* - * Function irlmp_link_disconnect_indication (reason, userdata) - * - * IrLAP has disconnected - * - */ -void irlmp_link_disconnect_indication(struct lap_cb *lap, - struct irlap_cb *irlap, - LAP_REASON reason, - struct sk_buff *skb) -{ - IRDA_ASSERT(lap != NULL, return;); - IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, return;); - - lap->reason = reason; - lap->daddr = DEV_ADDR_ANY; - - /* FIXME: must do something with the skb if any */ - - /* - * Inform station state machine - */ - irlmp_do_lap_event(lap, LM_LAP_DISCONNECT_INDICATION, NULL); -} - -/* - * Function irlmp_link_connect_indication (qos) - * - * Incoming LAP connection! - * - */ -void irlmp_link_connect_indication(struct lap_cb *self, __u32 saddr, - __u32 daddr, struct qos_info *qos, - struct sk_buff *skb) -{ - /* Copy QoS settings for this session */ - self->qos = qos; - - /* Update destination device address */ - self->daddr = daddr; - IRDA_ASSERT(self->saddr == saddr, return;); - - irlmp_do_lap_event(self, LM_LAP_CONNECT_INDICATION, skb); -} - -/* - * Function irlmp_link_connect_confirm (qos) - * - * LAP connection confirmed! - * - */ -void irlmp_link_connect_confirm(struct lap_cb *self, struct qos_info *qos, - struct sk_buff *skb) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - IRDA_ASSERT(qos != NULL, return;); - - /* Don't need use the skb for now */ - - /* Copy QoS settings for this session */ - self->qos = qos; - - irlmp_do_lap_event(self, LM_LAP_CONNECT_CONFIRM, NULL); -} - -/* - * Function irlmp_link_discovery_indication (self, log) - * - * Device is discovering us - * - * It's not an answer to our own discoveries, just another device trying - * to perform discovery, but we don't want to miss the opportunity - * to exploit this information, because : - * o We may not actively perform discovery (just passive discovery) - * o This type of discovery is much more reliable. In some cases, it - * seem that less than 50% of our discoveries get an answer, while - * we always get ~100% of these. - * o Make faster discovery, statistically divide time of discovery - * events by 2 (important for the latency aspect and user feel) - * o Even is we do active discovery, the other node might not - * answer our discoveries (ex: Palm). The Palm will just perform - * one active discovery and connect directly to us. - * - * However, when both devices discover each other, they might attempt to - * connect to each other following the discovery event, and it would create - * collisions on the medium (SNRM battle). - * The "fix" for that is to disable all connection requests in IrLAP - * for 100ms after a discovery indication by setting the media_busy flag. - * Previously, we used to postpone the event which was quite ugly. Now - * that IrLAP takes care of this problem, just pass the event up... - * - * Jean II - */ -void irlmp_link_discovery_indication(struct lap_cb *self, - discovery_t *discovery) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - - /* Add to main log, cleanup */ - irlmp_add_discovery(irlmp->cachelog, discovery); - - /* Just handle it the same way as a discovery confirm, - * bypass the LM_LAP state machine (see below) */ - irlmp_discovery_confirm(irlmp->cachelog, DISCOVERY_PASSIVE); -} - -/* - * Function irlmp_link_discovery_confirm (self, log) - * - * Called by IrLAP with a list of discoveries after the discovery - * request has been carried out. A NULL log is received if IrLAP - * was unable to carry out the discovery request - * - */ -void irlmp_link_discovery_confirm(struct lap_cb *self, hashbin_t *log) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LMP_LAP_MAGIC, return;); - - /* Add to main log, cleanup */ - irlmp_add_discovery_log(irlmp->cachelog, log); - - /* Propagate event to various LSAPs registered for it. - * We bypass the LM_LAP state machine because - * 1) We do it regardless of the LM_LAP state - * 2) It doesn't affect the LM_LAP state - * 3) Faster, slimer, simpler, ... - * Jean II */ - irlmp_discovery_confirm(irlmp->cachelog, DISCOVERY_ACTIVE); -} - -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP -static inline void irlmp_update_cache(struct lap_cb *lap, - struct lsap_cb *lsap) -{ - /* Prevent concurrent read to get garbage */ - lap->cache.valid = FALSE; - /* Update cache entry */ - lap->cache.dlsap_sel = lsap->dlsap_sel; - lap->cache.slsap_sel = lsap->slsap_sel; - lap->cache.lsap = lsap; - lap->cache.valid = TRUE; -} -#endif - -/* - * Function irlmp_find_handle (self, dlsap_sel, slsap_sel, status, queue) - * - * Find handle associated with destination and source LSAP - * - * Any IrDA connection (LSAP/TSAP) is uniquely identified by - * 3 parameters, the local lsap, the remote lsap and the remote address. - * We may initiate multiple connections to the same remote service - * (they will have different local lsap), a remote device may initiate - * multiple connections to the same local service (they will have - * different remote lsap), or multiple devices may connect to the same - * service and may use the same remote lsap (and they will have - * different remote address). - * So, where is the remote address ? Each LAP connection is made with - * a single remote device, so imply a specific remote address. - * Jean II - */ -static struct lsap_cb *irlmp_find_lsap(struct lap_cb *self, __u8 dlsap_sel, - __u8 slsap_sel, int status, - hashbin_t *queue) -{ - struct lsap_cb *lsap; - unsigned long flags; - - /* - * Optimize for the common case. We assume that the last frame - * received is in the same connection as the last one, so check in - * cache first to avoid the linear search - */ -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - if ((self->cache.valid) && - (self->cache.slsap_sel == slsap_sel) && - (self->cache.dlsap_sel == dlsap_sel)) - { - return self->cache.lsap; - } -#endif - - spin_lock_irqsave(&queue->hb_spinlock, flags); - - lsap = (struct lsap_cb *) hashbin_get_first(queue); - while (lsap != NULL) { - /* - * If this is an incoming connection, then the destination - * LSAP selector may have been specified as LM_ANY so that - * any client can connect. In that case we only need to check - * if the source LSAP (in our view!) match! - */ - if ((status == CONNECT_CMD) && - (lsap->slsap_sel == slsap_sel) && - (lsap->dlsap_sel == LSAP_ANY)) { - /* This is where the dest lsap sel is set on incoming - * lsaps */ - lsap->dlsap_sel = dlsap_sel; - break; - } - /* - * Check if source LSAP and dest LSAP selectors match. - */ - if ((lsap->slsap_sel == slsap_sel) && - (lsap->dlsap_sel == dlsap_sel)) - break; - - lsap = (struct lsap_cb *) hashbin_get_next(queue); - } -#ifdef CONFIG_IRDA_CACHE_LAST_LSAP - if(lsap) - irlmp_update_cache(self, lsap); -#endif - spin_unlock_irqrestore(&queue->hb_spinlock, flags); - - /* Return what we've found or NULL */ - return lsap; -} diff --git a/net/irda/irmod.c b/net/irda/irmod.c deleted file mode 100644 index c5e35b85c477..000000000000 --- a/net/irda/irmod.c +++ /dev/null @@ -1,199 +0,0 @@ -/********************************************************************* - * - * Filename: irmod.c - * Version: 0.9 - * Description: IrDA stack main entry points - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Dec 15 13:55:39 1997 - * Modified at: Wed Jan 5 15:12:41 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1997, 1999-2000 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2004 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -/* - * This file contains the main entry points of the IrDA stack. - * They are in this file and not af_irda.c because some developpers - * are using the IrDA stack without the socket API (compiling out - * af_irda.c). - * Jean II - */ - -#include <linux/module.h> -#include <linux/moduleparam.h> - -#include <net/irda/irda.h> -#include <net/irda/irmod.h> /* notify_t */ -#include <net/irda/irlap.h> /* irlap_init */ -#include <net/irda/irlmp.h> /* irlmp_init */ -#include <net/irda/iriap.h> /* iriap_init */ -#include <net/irda/irttp.h> /* irttp_init */ -#include <net/irda/irda_device.h> /* irda_device_init */ - -/* Packet type handler. - * Tell the kernel how IrDA packets should be handled. - */ -static struct packet_type irda_packet_type __read_mostly = { - .type = cpu_to_be16(ETH_P_IRDA), - .func = irlap_driver_rcv, /* Packet type handler irlap_frame.c */ -}; - -/* - * Function irda_notify_init (notify) - * - * Used for initializing the notify structure - * - */ -void irda_notify_init(notify_t *notify) -{ - notify->data_indication = NULL; - notify->udata_indication = NULL; - notify->connect_confirm = NULL; - notify->connect_indication = NULL; - notify->disconnect_indication = NULL; - notify->flow_indication = NULL; - notify->status_indication = NULL; - notify->instance = NULL; - strlcpy(notify->name, "Unknown", sizeof(notify->name)); -} -EXPORT_SYMBOL(irda_notify_init); - -/* - * Function irda_init (void) - * - * Protocol stack initialisation entry point. - * Initialise the various components of the IrDA stack - */ -static int __init irda_init(void) -{ - int ret = 0; - - /* Lower layer of the stack */ - irlmp_init(); - irlap_init(); - - /* Driver/dongle support */ - irda_device_init(); - - /* Higher layers of the stack */ - iriap_init(); - irttp_init(); - ret = irsock_init(); - if (ret < 0) - goto out_err_1; - - /* Add IrDA packet type (Start receiving packets) */ - dev_add_pack(&irda_packet_type); - - /* External APIs */ -#ifdef CONFIG_PROC_FS - irda_proc_register(); -#endif -#ifdef CONFIG_SYSCTL - ret = irda_sysctl_register(); - if (ret < 0) - goto out_err_2; -#endif - - ret = irda_nl_register(); - if (ret < 0) - goto out_err_3; - - return 0; - - out_err_3: -#ifdef CONFIG_SYSCTL - irda_sysctl_unregister(); - out_err_2: -#endif -#ifdef CONFIG_PROC_FS - irda_proc_unregister(); -#endif - - /* Remove IrDA packet type (stop receiving packets) */ - dev_remove_pack(&irda_packet_type); - - /* Remove higher layers */ - irsock_cleanup(); - out_err_1: - irttp_cleanup(); - iriap_cleanup(); - - /* Remove lower layers */ - irda_device_cleanup(); - irlap_cleanup(); /* Must be done before irlmp_cleanup()! DB */ - - /* Remove middle layer */ - irlmp_cleanup(); - - - return ret; -} - -/* - * Function irda_cleanup (void) - * - * Protocol stack cleanup/removal entry point. - * Cleanup the various components of the IrDA stack - */ -static void __exit irda_cleanup(void) -{ - /* Remove External APIs */ - irda_nl_unregister(); - -#ifdef CONFIG_SYSCTL - irda_sysctl_unregister(); -#endif -#ifdef CONFIG_PROC_FS - irda_proc_unregister(); -#endif - - /* Remove IrDA packet type (stop receiving packets) */ - dev_remove_pack(&irda_packet_type); - - /* Remove higher layers */ - irsock_cleanup(); - irttp_cleanup(); - iriap_cleanup(); - - /* Remove lower layers */ - irda_device_cleanup(); - irlap_cleanup(); /* Must be done before irlmp_cleanup()! DB */ - - /* Remove middle layer */ - irlmp_cleanup(); -} - -/* - * The IrDA stack must be initialised *before* drivers get initialised, - * and *before* higher protocols (IrLAN/IrCOMM/IrNET) get initialised, - * otherwise bad things will happen (hashbins will be NULL for example). - * Those modules are at module_init()/device_initcall() level. - * - * On the other hand, it needs to be initialised *after* the basic - * networking, the /proc/net filesystem and sysctl module. Those are - * currently initialised in .../init/main.c (before initcalls). - * Also, IrDA drivers needs to be initialised *after* the random number - * generator (main stack and higher layer init don't need it anymore). - * - * Jean II - */ -subsys_initcall(irda_init); -module_exit(irda_cleanup); - -MODULE_AUTHOR("Dag Brattli <dagb@cs.uit.no> & Jean Tourrilhes <jt@hpl.hp.com>"); -MODULE_DESCRIPTION("The Linux IrDA Protocol Stack"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_NETPROTO(PF_IRDA); diff --git a/net/irda/irnet/Kconfig b/net/irda/irnet/Kconfig deleted file mode 100644 index 28c557f0fdd2..000000000000 --- a/net/irda/irnet/Kconfig +++ /dev/null @@ -1,13 +0,0 @@ -config IRNET - tristate "IrNET protocol" - depends on IRDA && PPP - help - Say Y here if you want to build support for the IrNET protocol. - To compile it as a module, choose M here: the module will be - called irnet. IrNET is a PPP driver, so you will also need a - working PPP subsystem (driver, daemon and config)... - - IrNET is an alternate way to transfer TCP/IP traffic over IrDA. It - uses synchronous PPP over a set of point to point IrDA sockets. You - can use it between Linux machine or with W2k. - diff --git a/net/irda/irnet/Makefile b/net/irda/irnet/Makefile deleted file mode 100644 index 61c365c8a2a0..000000000000 --- a/net/irda/irnet/Makefile +++ /dev/null @@ -1,7 +0,0 @@ -# -# Makefile for the Linux IrDA IrNET protocol layer. -# - -obj-$(CONFIG_IRNET) += irnet.o - -irnet-y := irnet_ppp.o irnet_irda.o diff --git a/net/irda/irnet/irnet.h b/net/irda/irnet/irnet.h deleted file mode 100644 index 9d451f8ed47a..000000000000 --- a/net/irda/irnet/irnet.h +++ /dev/null @@ -1,522 +0,0 @@ -/* - * IrNET protocol module : Synchronous PPP over an IrDA socket. - * - * Jean II - HPL `00 - <jt@hpl.hp.com> - * - * This file contains definitions and declarations global to the IrNET module, - * all grouped in one place... - * This file is a *private* header, so other modules don't want to know - * what's in there... - * - * Note : as most part of the Linux kernel, this module is available - * under the GNU General Public License (GPL). - */ - -#ifndef IRNET_H -#define IRNET_H - -/************************** DOCUMENTATION ***************************/ -/* - * What is IrNET - * ------------- - * IrNET is a protocol allowing to carry TCP/IP traffic between two - * IrDA peers in an efficient fashion. It is a thin layer, passing PPP - * packets to IrTTP and vice versa. It uses PPP in synchronous mode, - * because IrTTP offer a reliable sequenced packet service (as opposed - * to a byte stream). In fact, you could see IrNET as carrying TCP/IP - * in a IrDA socket, using PPP to provide the glue. - * - * The main difference with traditional PPP over IrCOMM is that we - * avoid the framing and serial emulation which are a performance - * bottleneck. It also allows multipoint communications in a sensible - * fashion. - * - * The main difference with IrLAN is that we use PPP for the link - * management, which is more standard, interoperable and flexible than - * the IrLAN protocol. For example, PPP adds authentication, - * encryption, compression, header compression and automated routing - * setup. And, as IrNET let PPP do the hard work, the implementation - * is much simpler than IrLAN. - * - * The Linux implementation - * ------------------------ - * IrNET is written on top of the Linux-IrDA stack, and interface with - * the generic Linux PPP driver. Because IrNET depend on recent - * changes of the PPP driver interface, IrNET will work only with very - * recent kernel (2.3.99-pre6 and up). - * - * The present implementation offer the following features : - * o simple user interface using pppd - * o efficient implementation (interface directly to PPP and IrTTP) - * o addressing (you can specify the name of the IrNET recipient) - * o multipoint operation (limited by IrLAP specification) - * o information in /proc/net/irda/irnet - * o IrNET events on /dev/irnet (for user space daemon) - * o IrNET daemon (irnetd) to automatically handle incoming requests - * o Windows 2000 compatibility (tested, but need more work) - * Currently missing : - * o Lot's of testing (that's your job) - * o Connection retries (may be too hard to do) - * o Check pppd persist mode - * o User space daemon (to automatically handle incoming requests) - * - * The setup is not currently the most easy, but this should get much - * better when everything will get integrated... - * - * Acknowledgements - * ---------------- - * This module is based on : - * o The PPP driver (ppp_synctty/ppp_generic) by Paul Mackerras - * o The IrLAN protocol (irlan_common/XXX) by Dag Brattli - * o The IrSock interface (af_irda) by Dag Brattli - * o Some other bits from the kernel and my drivers... - * Infinite thanks to those brave souls for providing the infrastructure - * upon which IrNET is built. - * - * Thanks to all my colleagues in HP for helping me. In particular, - * thanks to Salil Pradhan and Bill Serra for W2k testing... - * Thanks to Luiz Magalhaes for irnetd and much testing... - * - * Thanks to Alan Cox for answering lot's of my stupid questions, and - * to Paul Mackerras answering my questions on how to best integrate - * IrNET and pppd. - * - * Jean II - * - * Note on some implementations choices... - * ------------------------------------ - * 1) Direct interface vs tty/socket - * I could have used a tty interface to hook to ppp and use the full - * socket API to connect to IrDA. The code would have been easier to - * maintain, and maybe the code would have been smaller... - * Instead, we hook directly to ppp_generic and to IrTTP, which make - * things more complicated... - * - * The first reason is flexibility : this allow us to create IrNET - * instances on demand (no /dev/ircommX crap) and to allow linkname - * specification on pppd command line... - * - * Second reason is speed optimisation. If you look closely at the - * transmit and receive paths, you will notice that they are "super lean" - * (that's why they look ugly), with no function calls and as little data - * copy and modification as I could... - * - * 2) irnetd in user space - * irnetd is implemented in user space, which is necessary to call pppd. - * This also give maximum benefits in term of flexibility and customability, - * and allow to offer the event channel, useful for other stuff like debug. - * - * On the other hand, this require a loose coordination between the - * present module and irnetd. One critical area is how incoming request - * are handled. - * When irnet receive an incoming request, it send an event to irnetd and - * drop the incoming IrNET socket. - * irnetd start a pppd instance, which create a new IrNET socket. This new - * socket is then connected in the originating node to the pppd instance. - * At this point, in the originating node, the first socket is closed. - * - * I admit, this is a bit messy and waste some resources. The alternative - * is caching incoming socket, and that's also quite messy and waste - * resources. - * We also make connection time slower. For example, on a 115 kb/s link it - * adds 60ms to the connection time (770 ms). However, this is slower than - * the time it takes to fire up pppd on my P133... - * - * - * History : - * ------- - * - * v1 - 15.5.00 - Jean II - * o Basic IrNET (hook to ppp_generic & IrTTP - incl. multipoint) - * o control channel on /dev/irnet (set name/address) - * o event channel on /dev/irnet (for user space daemon) - * - * v2 - 5.6.00 - Jean II - * o Enable DROP_NOT_READY to avoid PPP timeouts & other weirdness... - * o Add DISCONNECT_TO event and rename DISCONNECT_FROM. - * o Set official device number alloaction on /dev/irnet - * - * v3 - 30.8.00 - Jean II - * o Update to latest Linux-IrDA changes : - * - queue_t => irda_queue_t - * o Update to ppp-2.4.0 : - * - move irda_irnet_connect from PPPIOCATTACH to TIOCSETD - * o Add EXPIRE event (depend on new IrDA-Linux patch) - * o Switch from `hashbin_remove' to `hashbin_remove_this' to fix - * a multilink bug... (depend on new IrDA-Linux patch) - * o fix a self->daddr to self->raddr in irda_irnet_connect to fix - * another multilink bug (darn !) - * o Remove LINKNAME_IOCTL cruft - * - * v3b - 31.8.00 - Jean II - * o Dump discovery log at event channel startup - * - * v4 - 28.9.00 - Jean II - * o Fix interaction between poll/select and dump discovery log - * o Add IRNET_BLOCKED_LINK event (depend on new IrDA-Linux patch) - * o Add IRNET_NOANSWER_FROM event (mostly to help support) - * o Release flow control in disconnect_indication - * o Block packets while connecting (speed up connections) - * - * v5 - 11.01.01 - Jean II - * o Init self->max_header_size, just in case... - * o Set up ap->chan.hdrlen, to get zero copy on tx side working. - * o avoid tx->ttp->flow->ppp->tx->... loop, by checking flow state - * Thanks to Christian Gennerat for finding this bug ! - * --- - * o Declare the proper MTU/MRU that we can support - * (but PPP doesn't read the MTU value :-() - * o Declare hashbin HB_NOLOCK instead of HB_LOCAL to avoid - * disabling and enabling irq twice - * - * v6 - 31.05.01 - Jean II - * o Print source address in Found, Discovery, Expiry & Request events - * o Print requested source address in /proc/net/irnet - * o Change control channel input. Allow multiple commands in one line. - * o Add saddr command to change ap->rsaddr (and use that in IrDA) - * --- - * o Make the IrDA connection procedure totally asynchronous. - * Heavy rewrite of the IAS query code and the whole connection - * procedure. Now, irnet_connect() no longer need to be called from - * a process context... - * o Enable IrDA connect retries in ppp_irnet_send(). The good thing - * is that IrDA connect retries are directly driven by PPP LCP - * retries (we retry for each LCP packet), so that everything - * is transparently controlled from pppd lcp-max-configure. - * o Add ttp_connect flag to prevent rentry on the connect procedure - * o Test and fixups to eliminate side effects of retries - * - * v7 - 22.08.01 - Jean II - * o Cleanup : Change "saddr = 0x0" to "saddr = DEV_ADDR_ANY" - * o Fix bug in BLOCK_WHEN_CONNECT introduced in v6 : due to the - * asynchronous IAS query, self->tsap is NULL when PPP send the - * first packet. This was preventing "connect-delay 0" to work. - * Change the test in ppp_irnet_send() to self->ttp_connect. - * - * v8 - 1.11.01 - Jean II - * o Tighten the use of self->ttp_connect and self->ttp_open to - * prevent various race conditions. - * o Avoid leaking discovery log and skb - * o Replace "self" with "server" in irnet_connect_indication() to - * better detect cut'n'paste error ;-) - * - * v9 - 29.11.01 - Jean II - * o Fix event generation in disconnect indication that I broke in v8 - * It was always generation "No-Answer" because I was testing ttp_open - * just after clearing it. *blush*. - * o Use newly created irttp_listen() to fix potential crash when LAP - * destroyed before irnet module removed. - * - * v10 - 4.3.2 - Jean II - * o When receiving a disconnect indication, don't reenable the - * PPP Tx queue, this will trigger a reconnect. Instead, close - * the channel, which will kill pppd... - * - * v11 - 20.3.02 - Jean II - * o Oops ! v10 fix disabled IrNET retries and passive behaviour. - * Better fix in irnet_disconnect_indication() : - * - if connected, kill pppd via hangup. - * - if not connected, reenable ppp Tx, which trigger IrNET retry. - * - * v12 - 10.4.02 - Jean II - * o Fix race condition in irnet_connect_indication(). - * If the socket was already trying to connect, drop old connection - * and use new one only if acting as primary. See comments. - * - * v13 - 30.5.02 - Jean II - * o Update module init code - * - * v14 - 20.2.03 - Jean II - * o Add discovery hint bits in the control channel. - * o Remove obsolete MOD_INC/DEC_USE_COUNT in favor of .owner - * - * v15 - 7.4.03 - Jean II - * o Replace spin_lock_irqsave() with spin_lock_bh() so that we can - * use ppp_unit_number(). It's probably also better overall... - * o Disable call to ppp_unregister_channel(), because we can't do it. - */ - -/***************************** INCLUDES *****************************/ - -#include <linux/module.h> - -#include <linux/kernel.h> -#include <linux/skbuff.h> -#include <linux/tty.h> -#include <linux/proc_fs.h> -#include <linux/netdevice.h> -#include <linux/poll.h> -#include <linux/capability.h> -#include <linux/ctype.h> /* isspace() */ -#include <linux/string.h> /* skip_spaces() */ -#include <linux/uaccess.h> -#include <linux/init.h> - -#include <linux/ppp_defs.h> -#include <linux/ppp-ioctl.h> -#include <linux/ppp_channel.h> - -#include <net/irda/irda.h> -#include <net/irda/iriap.h> -#include <net/irda/irias_object.h> -#include <net/irda/irlmp.h> -#include <net/irda/irttp.h> -#include <net/irda/discovery.h> - -/***************************** OPTIONS *****************************/ -/* - * Define or undefine to compile or not some optional part of the - * IrNET driver... - * Note : the present defaults make sense, play with that at your - * own risk... - */ -/* IrDA side of the business... */ -#define DISCOVERY_NOMASK /* To enable W2k compatibility... */ -#define ADVERTISE_HINT /* Advertise IrLAN hint bit */ -#define ALLOW_SIMULT_CONNECT /* This seem to work, cross fingers... */ -#define DISCOVERY_EVENTS /* Query the discovery log to post events */ -#define INITIAL_DISCOVERY /* Dump current discovery log as events */ -#undef STREAM_COMPAT /* Not needed - potentially messy */ -#undef CONNECT_INDIC_KICK /* Might mess IrDA, not needed */ -#undef FAIL_SEND_DISCONNECT /* Might mess IrDA, not needed */ -#undef PASS_CONNECT_PACKETS /* Not needed ? Safe */ -#undef MISSING_PPP_API /* Stuff I wish I could do */ - -/* PPP side of the business */ -#define BLOCK_WHEN_CONNECT /* Block packets when connecting */ -#define CONNECT_IN_SEND /* Retry IrDA connection procedure */ -#undef FLUSH_TO_PPP /* Not sure about this one, let's play safe */ -#undef SECURE_DEVIRNET /* Bah... */ - -/****************************** DEBUG ******************************/ - -/* - * This set of flags enable and disable all the various warning, - * error and debug message of this driver. - * Each section can be enabled and disabled independently - */ -/* In the PPP part */ -#define DEBUG_CTRL_TRACE 0 /* Control channel */ -#define DEBUG_CTRL_INFO 0 /* various info */ -#define DEBUG_CTRL_ERROR 1 /* problems */ -#define DEBUG_FS_TRACE 0 /* filesystem callbacks */ -#define DEBUG_FS_INFO 0 /* various info */ -#define DEBUG_FS_ERROR 1 /* problems */ -#define DEBUG_PPP_TRACE 0 /* PPP related functions */ -#define DEBUG_PPP_INFO 0 /* various info */ -#define DEBUG_PPP_ERROR 1 /* problems */ -#define DEBUG_MODULE_TRACE 0 /* module insertion/removal */ -#define DEBUG_MODULE_ERROR 1 /* problems */ - -/* In the IrDA part */ -#define DEBUG_IRDA_SR_TRACE 0 /* IRDA subroutines */ -#define DEBUG_IRDA_SR_INFO 0 /* various info */ -#define DEBUG_IRDA_SR_ERROR 1 /* problems */ -#define DEBUG_IRDA_SOCK_TRACE 0 /* IRDA main socket functions */ -#define DEBUG_IRDA_SOCK_INFO 0 /* various info */ -#define DEBUG_IRDA_SOCK_ERROR 1 /* problems */ -#define DEBUG_IRDA_SERV_TRACE 0 /* The IrNET server */ -#define DEBUG_IRDA_SERV_INFO 0 /* various info */ -#define DEBUG_IRDA_SERV_ERROR 1 /* problems */ -#define DEBUG_IRDA_TCB_TRACE 0 /* IRDA IrTTP callbacks */ -#define DEBUG_IRDA_CB_INFO 0 /* various info */ -#define DEBUG_IRDA_CB_ERROR 1 /* problems */ -#define DEBUG_IRDA_OCB_TRACE 0 /* IRDA other callbacks */ -#define DEBUG_IRDA_OCB_INFO 0 /* various info */ -#define DEBUG_IRDA_OCB_ERROR 1 /* problems */ - -#define DEBUG_ASSERT 0 /* Verify all assertions */ - -/* - * These are the macros we are using to actually print the debug - * statements. Don't look at it, it's ugly... - * - * One of the trick is that, as the DEBUG_XXX are constant, the - * compiler will optimise away the if() in all cases. - */ -/* All error messages (will show up in the normal logs) */ -#define DERROR(dbg, format, args...) \ - {if(DEBUG_##dbg) \ - printk(KERN_INFO "irnet: %s(): " format, __func__ , ##args);} - -/* Normal debug message (will show up in /var/log/debug) */ -#define DEBUG(dbg, format, args...) \ - {if(DEBUG_##dbg) \ - printk(KERN_DEBUG "irnet: %s(): " format, __func__ , ##args);} - -/* Entering a function (trace) */ -#define DENTER(dbg, format, args...) \ - {if(DEBUG_##dbg) \ - printk(KERN_DEBUG "irnet: -> %s" format, __func__ , ##args);} - -/* Entering and exiting a function in one go (trace) */ -#define DPASS(dbg, format, args...) \ - {if(DEBUG_##dbg) \ - printk(KERN_DEBUG "irnet: <>%s" format, __func__ , ##args);} - -/* Exiting a function (trace) */ -#define DEXIT(dbg, format, args...) \ - {if(DEBUG_##dbg) \ - printk(KERN_DEBUG "irnet: <-%s()" format, __func__ , ##args);} - -/* Exit a function with debug */ -#define DRETURN(ret, dbg, args...) \ - {DEXIT(dbg, ": " args);\ - return ret; } - -/* Exit a function on failed condition */ -#define DABORT(cond, ret, dbg, args...) \ - {if(cond) {\ - DERROR(dbg, args);\ - return ret; }} - -/* Invalid assertion, print out an error and exit... */ -#define DASSERT(cond, ret, dbg, args...) \ - {if((DEBUG_ASSERT) && !(cond)) {\ - DERROR(dbg, "Invalid assertion: " args);\ - return ret; }} - -/************************ CONSTANTS & MACROS ************************/ - -/* Paranoia */ -#define IRNET_MAGIC 0xB00754 - -/* Number of control events in the control channel buffer... */ -#define IRNET_MAX_EVENTS 8 /* Should be more than enough... */ - -/****************************** TYPES ******************************/ - -/* - * This is the main structure where we store all the data pertaining to - * one instance of irnet. - * Note : in irnet functions, a pointer this structure is usually called - * "ap" or "self". If the code is borrowed from the IrDA stack, it tend - * to be called "self", and if it is borrowed from the PPP driver it is - * "ap". Apart from that, it's exactly the same structure ;-) - */ -typedef struct irnet_socket -{ - /* ------------------- Instance management ------------------- */ - /* We manage a linked list of IrNET socket instances */ - irda_queue_t q; /* Must be first - for hasbin */ - int magic; /* Paranoia */ - - /* --------------------- FileSystem part --------------------- */ - /* "pppd" interact directly with us on a /dev/ file */ - struct file * file; /* File descriptor of this instance */ - /* TTY stuff - to keep "pppd" happy */ - struct ktermios termios; /* Various tty flags */ - /* Stuff for the control channel */ - int event_index; /* Last read in the event log */ - - /* ------------------------- PPP part ------------------------- */ - /* We interface directly to the ppp_generic driver in the kernel */ - int ppp_open; /* registered with ppp_generic */ - struct ppp_channel chan; /* Interface to generic ppp layer */ - - int mru; /* Max size of PPP payload */ - u32 xaccm[8]; /* Asynchronous character map (just */ - u32 raccm; /* to please pppd - dummy) */ - unsigned int flags; /* PPP flags (compression, ...) */ - unsigned int rbits; /* Unused receive flags ??? */ - struct work_struct disconnect_work; /* Process context disconnection */ - /* ------------------------ IrTTP part ------------------------ */ - /* We create a pseudo "socket" over the IrDA tranport */ - unsigned long ttp_open; /* Set when IrTTP is ready */ - unsigned long ttp_connect; /* Set when IrTTP is connecting */ - struct tsap_cb * tsap; /* IrTTP instance (the connection) */ - - char rname[NICKNAME_MAX_LEN + 1]; - /* IrDA nickname of destination */ - __u32 rdaddr; /* Requested peer IrDA address */ - __u32 rsaddr; /* Requested local IrDA address */ - __u32 daddr; /* actual peer IrDA address */ - __u32 saddr; /* my local IrDA address */ - __u8 dtsap_sel; /* Remote TSAP selector */ - __u8 stsap_sel; /* Local TSAP selector */ - - __u32 max_sdu_size_rx;/* Socket parameters used for IrTTP */ - __u32 max_sdu_size_tx; - __u32 max_data_size; - __u8 max_header_size; - LOCAL_FLOW tx_flow; /* State of the Tx path in IrTTP */ - - /* ------------------- IrLMP and IrIAS part ------------------- */ - /* Used for IrDA Discovery and socket name resolution */ - void * ckey; /* IrLMP client handle */ - __u16 mask; /* Hint bits mask (filter discov.)*/ - int nslots; /* Number of slots for discovery */ - - struct iriap_cb * iriap; /* Used to query remote IAS */ - int errno; /* status of the IAS query */ - - /* -------------------- Discovery log part -------------------- */ - /* Used by initial discovery on the control channel - * and by irnet_discover_daddr_and_lsap_sel() */ - struct irda_device_info *discoveries; /* Copy of the discovery log */ - int disco_index; /* Last read in the discovery log */ - int disco_number; /* Size of the discovery log */ - - struct mutex lock; - -} irnet_socket; - -/* - * This is the various event that we will generate on the control channel - */ -typedef enum irnet_event -{ - IRNET_DISCOVER, /* New IrNET node discovered */ - IRNET_EXPIRE, /* IrNET node expired */ - IRNET_CONNECT_TO, /* IrNET socket has connected to other node */ - IRNET_CONNECT_FROM, /* Other node has connected to IrNET socket */ - IRNET_REQUEST_FROM, /* Non satisfied connection request */ - IRNET_NOANSWER_FROM, /* Failed connection request */ - IRNET_BLOCKED_LINK, /* Link (IrLAP) is blocked for > 3s */ - IRNET_DISCONNECT_FROM, /* IrNET socket has disconnected */ - IRNET_DISCONNECT_TO /* Closing IrNET socket */ -} irnet_event; - -/* - * This is the storage for an event and its arguments - */ -typedef struct irnet_log -{ - irnet_event event; - int unit; - __u32 saddr; - __u32 daddr; - char name[NICKNAME_MAX_LEN + 1]; /* 21 + 1 */ - __u16_host_order hints; /* Discovery hint bits */ -} irnet_log; - -/* - * This is the storage for all events and related stuff... - */ -typedef struct irnet_ctrl_channel -{ - irnet_log log[IRNET_MAX_EVENTS]; /* Event log */ - int index; /* Current index in log */ - spinlock_t spinlock; /* Serialize access to the event log */ - wait_queue_head_t rwait; /* processes blocked on read (or poll) */ -} irnet_ctrl_channel; - -/**************************** PROTOTYPES ****************************/ -/* - * Global functions of the IrNET module - * Note : we list here also functions called from one file to the other. - */ - -/* -------------------------- IRDA PART -------------------------- */ -int irda_irnet_create(irnet_socket *); /* Initialise an IrNET socket */ -int irda_irnet_connect(irnet_socket *); /* Try to connect over IrDA */ -void irda_irnet_destroy(irnet_socket *); /* Teardown an IrNET socket */ -int irda_irnet_init(void); /* Initialise IrDA part of IrNET */ -void irda_irnet_cleanup(void); /* Teardown IrDA part of IrNET */ - -/**************************** VARIABLES ****************************/ - -/* Control channel stuff - allocated in irnet_irda.h */ -extern struct irnet_ctrl_channel irnet_events; - -#endif /* IRNET_H */ diff --git a/net/irda/irnet/irnet_irda.c b/net/irda/irnet/irnet_irda.c deleted file mode 100644 index e390bceeb2f8..000000000000 --- a/net/irda/irnet/irnet_irda.c +++ /dev/null @@ -1,1885 +0,0 @@ -/* - * IrNET protocol module : Synchronous PPP over an IrDA socket. - * - * Jean II - HPL `00 - <jt@hpl.hp.com> - * - * This file implement the IRDA interface of IrNET. - * Basically, we sit on top of IrTTP. We set up IrTTP, IrIAS properly, - * and exchange frames with IrTTP. - */ - -#include "irnet_irda.h" /* Private header */ -#include <linux/sched.h> -#include <linux/seq_file.h> -#include <linux/slab.h> -#include <asm/unaligned.h> - -/* - * PPP disconnect work: we need to make sure we're in - * process context when calling ppp_unregister_channel(). - */ -static void irnet_ppp_disconnect(struct work_struct *work) -{ - irnet_socket * self = - container_of(work, irnet_socket, disconnect_work); - - if (self == NULL) - return; - /* - * If we were connected, cleanup & close the PPP - * channel, which will kill pppd (hangup) and the rest. - */ - if (self->ppp_open && !self->ttp_open && !self->ttp_connect) { - ppp_unregister_channel(&self->chan); - self->ppp_open = 0; - } -} - -/************************* CONTROL CHANNEL *************************/ -/* - * When ppp is not active, /dev/irnet act as a control channel. - * Writing allow to set up the IrDA destination of the IrNET channel, - * and any application may be read events happening on IrNET... - */ - -/*------------------------------------------------------------------*/ -/* - * Post an event to the control channel... - * Put the event in the log, and then wait all process blocked on read - * so they can read the log... - */ -static void -irnet_post_event(irnet_socket * ap, - irnet_event event, - __u32 saddr, - __u32 daddr, - char * name, - __u16 hints) -{ - int index; /* In the log */ - - DENTER(CTRL_TRACE, "(ap=0x%p, event=%d, daddr=%08x, name=``%s'')\n", - ap, event, daddr, name); - - /* Protect this section via spinlock. - * Note : as we are the only event producer, we only need to exclude - * ourself when touching the log, which is nice and easy. - */ - spin_lock_bh(&irnet_events.spinlock); - - /* Copy the event in the log */ - index = irnet_events.index; - irnet_events.log[index].event = event; - irnet_events.log[index].daddr = daddr; - irnet_events.log[index].saddr = saddr; - /* Try to copy IrDA nickname */ - if(name) - strcpy(irnet_events.log[index].name, name); - else - irnet_events.log[index].name[0] = '\0'; - /* Copy hints */ - irnet_events.log[index].hints.word = hints; - /* Try to get ppp unit number */ - if((ap != (irnet_socket *) NULL) && (ap->ppp_open)) - irnet_events.log[index].unit = ppp_unit_number(&ap->chan); - else - irnet_events.log[index].unit = -1; - - /* Increment the index - * Note that we increment the index only after the event is written, - * to make sure that the readers don't get garbage... */ - irnet_events.index = (index + 1) % IRNET_MAX_EVENTS; - - DEBUG(CTRL_INFO, "New event index is %d\n", irnet_events.index); - - /* Spin lock end */ - spin_unlock_bh(&irnet_events.spinlock); - - /* Now : wake up everybody waiting for events... */ - wake_up_interruptible_all(&irnet_events.rwait); - - DEXIT(CTRL_TRACE, "\n"); -} - -/************************* IRDA SUBROUTINES *************************/ -/* - * These are a bunch of subroutines called from other functions - * down there, mostly common code or to improve readability... - * - * Note : we duplicate quite heavily some routines of af_irda.c, - * because our input structure (self) is quite different - * (struct irnet instead of struct irda_sock), which make sharing - * the same code impossible (at least, without templates). - */ - -/*------------------------------------------------------------------*/ -/* - * Function irda_open_tsap (self) - * - * Open local Transport Service Access Point (TSAP) - * - * Create a IrTTP instance for us and set all the IrTTP callbacks. - */ -static inline int -irnet_open_tsap(irnet_socket * self) -{ - notify_t notify; /* Callback structure */ - - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - DABORT(self->tsap != NULL, -EBUSY, IRDA_SR_ERROR, "Already busy !\n"); - - /* Initialize IrTTP callbacks to be used by the IrDA stack */ - irda_notify_init(¬ify); - notify.connect_confirm = irnet_connect_confirm; - notify.connect_indication = irnet_connect_indication; - notify.disconnect_indication = irnet_disconnect_indication; - notify.data_indication = irnet_data_indication; - /*notify.udata_indication = NULL;*/ - notify.flow_indication = irnet_flow_indication; - notify.status_indication = irnet_status_indication; - notify.instance = self; - strlcpy(notify.name, IRNET_NOTIFY_NAME, sizeof(notify.name)); - - /* Open an IrTTP instance */ - self->tsap = irttp_open_tsap(LSAP_ANY, DEFAULT_INITIAL_CREDIT, - ¬ify); - DABORT(self->tsap == NULL, -ENOMEM, - IRDA_SR_ERROR, "Unable to allocate TSAP !\n"); - - /* Remember which TSAP selector we actually got */ - self->stsap_sel = self->tsap->stsap_sel; - - DEXIT(IRDA_SR_TRACE, " - tsap=0x%p, sel=0x%X\n", - self->tsap, self->stsap_sel); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_ias_to_tsap (self, result, value) - * - * Examine an IAS object and extract TSAP - * - * We do an IAP query to find the TSAP associated with the IrNET service. - * When IrIAP pass us the result of the query, this function look at - * the return values to check for failures and extract the TSAP if - * possible. - * Also deallocate value - * The failure is in self->errno - * Return TSAP or -1 - */ -static inline __u8 -irnet_ias_to_tsap(irnet_socket * self, - int result, - struct ias_value * value) -{ - __u8 dtsap_sel = 0; /* TSAP we are looking for */ - - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - /* By default, no error */ - self->errno = 0; - - /* Check if request succeeded */ - switch(result) - { - /* Standard errors : service not available */ - case IAS_CLASS_UNKNOWN: - case IAS_ATTRIB_UNKNOWN: - DEBUG(IRDA_SR_INFO, "IAS object doesn't exist ! (%d)\n", result); - self->errno = -EADDRNOTAVAIL; - break; - - /* Other errors, most likely IrDA stack failure */ - default : - DEBUG(IRDA_SR_INFO, "IAS query failed ! (%d)\n", result); - self->errno = -EHOSTUNREACH; - break; - - /* Success : we got what we wanted */ - case IAS_SUCCESS: - break; - } - - /* Check what was returned to us */ - if(value != NULL) - { - /* What type of argument have we got ? */ - switch(value->type) - { - case IAS_INTEGER: - DEBUG(IRDA_SR_INFO, "result=%d\n", value->t.integer); - if(value->t.integer != -1) - /* Get the remote TSAP selector */ - dtsap_sel = value->t.integer; - else - self->errno = -EADDRNOTAVAIL; - break; - default: - self->errno = -EADDRNOTAVAIL; - DERROR(IRDA_SR_ERROR, "bad type ! (0x%X)\n", value->type); - break; - } - - /* Cleanup */ - irias_delete_value(value); - } - else /* value == NULL */ - { - /* Nothing returned to us - usually result != SUCCESS */ - if(!(self->errno)) - { - DERROR(IRDA_SR_ERROR, - "IrDA bug : result == SUCCESS && value == NULL\n"); - self->errno = -EHOSTUNREACH; - } - } - DEXIT(IRDA_SR_TRACE, "\n"); - - /* Return the TSAP */ - return dtsap_sel; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_find_lsap_sel (self) - * - * Try to lookup LSAP selector in remote LM-IAS - * - * Basically, we start a IAP query, and then go to sleep. When the query - * return, irnet_getvalue_confirm will wake us up, and we can examine the - * result of the query... - * Note that in some case, the query fail even before we go to sleep, - * creating some races... - */ -static inline int -irnet_find_lsap_sel(irnet_socket * self) -{ - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - /* This should not happen */ - DABORT(self->iriap, -EBUSY, IRDA_SR_ERROR, "busy with a previous query.\n"); - - /* Create an IAP instance, will be closed in irnet_getvalue_confirm() */ - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - irnet_getvalue_confirm); - - /* Treat unexpected signals as disconnect */ - self->errno = -EHOSTUNREACH; - - /* Query remote LM-IAS */ - iriap_getvaluebyclass_request(self->iriap, self->rsaddr, self->daddr, - IRNET_SERVICE_NAME, IRNET_IAS_VALUE); - - /* The above request is non-blocking. - * After a while, IrDA will call us back in irnet_getvalue_confirm() - * We will then call irnet_ias_to_tsap() and finish the - * connection procedure */ - - DEXIT(IRDA_SR_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_connect_tsap (self) - * - * Initialise the TTP socket and initiate TTP connection - * - */ -static inline int -irnet_connect_tsap(irnet_socket * self) -{ - int err; - - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - /* Open a local TSAP (an IrTTP instance) */ - err = irnet_open_tsap(self); - if(err != 0) - { - clear_bit(0, &self->ttp_connect); - DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return err; - } - - /* Connect to remote device */ - err = irttp_connect_request(self->tsap, self->dtsap_sel, - self->rsaddr, self->daddr, NULL, - self->max_sdu_size_rx, NULL); - if(err != 0) - { - clear_bit(0, &self->ttp_connect); - DERROR(IRDA_SR_ERROR, "connect aborted!\n"); - return err; - } - - /* The above call is non-blocking. - * After a while, the IrDA stack will either call us back in - * irnet_connect_confirm() or irnet_disconnect_indication() - * See you there ;-) */ - - DEXIT(IRDA_SR_TRACE, "\n"); - return err; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_discover_next_daddr (self) - * - * Query the IrNET TSAP of the next device in the log. - * - * Used in the TSAP discovery procedure. - */ -static inline int -irnet_discover_next_daddr(irnet_socket * self) -{ - /* Close the last instance of IrIAP, and open a new one. - * We can't reuse the IrIAP instance in the IrIAP callback */ - if(self->iriap) - { - iriap_close(self->iriap); - self->iriap = NULL; - } - /* Create a new IAP instance */ - self->iriap = iriap_open(LSAP_ANY, IAS_CLIENT, self, - irnet_discovervalue_confirm); - if(self->iriap == NULL) - return -ENOMEM; - - /* Next discovery - before the call to avoid races */ - self->disco_index++; - - /* Check if we have one more address to try */ - if(self->disco_index < self->disco_number) - { - /* Query remote LM-IAS */ - iriap_getvaluebyclass_request(self->iriap, - self->discoveries[self->disco_index].saddr, - self->discoveries[self->disco_index].daddr, - IRNET_SERVICE_NAME, IRNET_IAS_VALUE); - /* The above request is non-blocking. - * After a while, IrDA will call us back in irnet_discovervalue_confirm() - * We will then call irnet_ias_to_tsap() and come back here again... */ - return 0; - } - else - return 1; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_discover_daddr_and_lsap_sel (self) - * - * This try to find a device with the requested service. - * - * Initiate a TSAP discovery procedure. - * It basically look into the discovery log. For each address in the list, - * it queries the LM-IAS of the device to find if this device offer - * the requested service. - * If there is more than one node supporting the service, we complain - * to the user (it should move devices around). - * If we find one node which have the requested TSAP, we connect to it. - * - * This function just start the whole procedure. It request the discovery - * log and submit the first IAS query. - * The bulk of the job is handled in irnet_discovervalue_confirm() - * - * Note : this procedure fails if there is more than one device in range - * on the same dongle, because IrLMP doesn't disconnect the LAP when the - * last LSAP is closed. Moreover, we would need to wait the LAP - * disconnection... - */ -static inline int -irnet_discover_daddr_and_lsap_sel(irnet_socket * self) -{ - int ret; - - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - /* Ask lmp for the current discovery log */ - self->discoveries = irlmp_get_discoveries(&self->disco_number, self->mask, - DISCOVERY_DEFAULT_SLOTS); - - /* Check if the we got some results */ - if(self->discoveries == NULL) - { - self->disco_number = -1; - clear_bit(0, &self->ttp_connect); - DRETURN(-ENETUNREACH, IRDA_SR_INFO, "No Cachelog...\n"); - } - DEBUG(IRDA_SR_INFO, "Got the log (0x%p), size is %d\n", - self->discoveries, self->disco_number); - - /* Start with the first discovery */ - self->disco_index = -1; - self->daddr = DEV_ADDR_ANY; - - /* This will fail if the log is empty - this is non-blocking */ - ret = irnet_discover_next_daddr(self); - if(ret) - { - /* Close IAP */ - if(self->iriap) - iriap_close(self->iriap); - self->iriap = NULL; - - /* Cleanup our copy of the discovery log */ - kfree(self->discoveries); - self->discoveries = NULL; - - clear_bit(0, &self->ttp_connect); - DRETURN(-ENETUNREACH, IRDA_SR_INFO, "Cachelog empty...\n"); - } - - /* Follow me in irnet_discovervalue_confirm() */ - - DEXIT(IRDA_SR_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_dname_to_daddr (self) - * - * Convert an IrDA nickname to a valid IrDA address - * - * It basically look into the discovery log until there is a match. - */ -static inline int -irnet_dname_to_daddr(irnet_socket * self) -{ - struct irda_device_info *discoveries; /* Copy of the discovery log */ - int number; /* Number of nodes in the log */ - int i; - - DENTER(IRDA_SR_TRACE, "(self=0x%p)\n", self); - - /* Ask lmp for the current discovery log */ - discoveries = irlmp_get_discoveries(&number, 0xffff, - DISCOVERY_DEFAULT_SLOTS); - /* Check if the we got some results */ - if(discoveries == NULL) - DRETURN(-ENETUNREACH, IRDA_SR_INFO, "Cachelog empty...\n"); - - /* - * Now, check all discovered devices (if any), and connect - * client only about the services that the client is - * interested in... - */ - for(i = 0; i < number; i++) - { - /* Does the name match ? */ - if(!strncmp(discoveries[i].info, self->rname, NICKNAME_MAX_LEN)) - { - /* Yes !!! Get it.. */ - self->daddr = discoveries[i].daddr; - DEBUG(IRDA_SR_INFO, "discovered device ``%s'' at address 0x%08x.\n", - self->rname, self->daddr); - kfree(discoveries); - DEXIT(IRDA_SR_TRACE, "\n"); - return 0; - } - } - /* No luck ! */ - DEBUG(IRDA_SR_INFO, "cannot discover device ``%s'' !!!\n", self->rname); - kfree(discoveries); - return -EADDRNOTAVAIL; -} - - -/************************* SOCKET ROUTINES *************************/ -/* - * This are the main operations on IrNET sockets, basically to create - * and destroy IrNET sockets. These are called from the PPP part... - */ - -/*------------------------------------------------------------------*/ -/* - * Create a IrNET instance : just initialise some parameters... - */ -int -irda_irnet_create(irnet_socket * self) -{ - DENTER(IRDA_SOCK_TRACE, "(self=0x%p)\n", self); - - self->magic = IRNET_MAGIC; /* Paranoia */ - - self->ttp_open = 0; /* Prevent higher layer from accessing IrTTP */ - self->ttp_connect = 0; /* Not connecting yet */ - self->rname[0] = '\0'; /* May be set via control channel */ - self->rdaddr = DEV_ADDR_ANY; /* May be set via control channel */ - self->rsaddr = DEV_ADDR_ANY; /* May be set via control channel */ - self->daddr = DEV_ADDR_ANY; /* Until we get connected */ - self->saddr = DEV_ADDR_ANY; /* Until we get connected */ - self->max_sdu_size_rx = TTP_SAR_UNBOUND; - - /* Register as a client with IrLMP */ - self->ckey = irlmp_register_client(0, NULL, NULL, NULL); -#ifdef DISCOVERY_NOMASK - self->mask = 0xffff; /* For W2k compatibility */ -#else /* DISCOVERY_NOMASK */ - self->mask = irlmp_service_to_hint(S_LAN); -#endif /* DISCOVERY_NOMASK */ - self->tx_flow = FLOW_START; /* Flow control from IrTTP */ - - INIT_WORK(&self->disconnect_work, irnet_ppp_disconnect); - - DEXIT(IRDA_SOCK_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Connect to the other side : - * o convert device name to an address - * o find the socket number (dlsap) - * o Establish the connection - * - * Note : We no longer mimic af_irda. The IAS query for finding the TSAP - * is done asynchronously, like the TTP connection. This allow us to - * call this function from any context (not only process). - * The downside is that following what's happening in there is tricky - * because it involve various functions all over the place... - */ -int -irda_irnet_connect(irnet_socket * self) -{ - int err; - - DENTER(IRDA_SOCK_TRACE, "(self=0x%p)\n", self); - - /* Check if we are already trying to connect. - * Because irda_irnet_connect() can be called directly by pppd plus - * packet retries in ppp_generic and connect may take time, plus we may - * race with irnet_connect_indication(), we need to be careful there... */ - if(test_and_set_bit(0, &self->ttp_connect)) - DRETURN(-EBUSY, IRDA_SOCK_INFO, "Already connecting...\n"); - if((self->iriap != NULL) || (self->tsap != NULL)) - DERROR(IRDA_SOCK_ERROR, "Socket not cleaned up...\n"); - - /* Insert ourselves in the hashbin so that the IrNET server can find us. - * Notes : 4th arg is string of 32 char max and must be null terminated - * When 4th arg is used (string), 3rd arg isn't (int) - * Can't re-insert (MUST remove first) so check for that... */ - if((irnet_server.running) && (self->q.q_next == NULL)) - { - spin_lock_bh(&irnet_server.spinlock); - hashbin_insert(irnet_server.list, (irda_queue_t *) self, 0, self->rname); - spin_unlock_bh(&irnet_server.spinlock); - DEBUG(IRDA_SOCK_INFO, "Inserted ``%s'' in hashbin...\n", self->rname); - } - - /* If we don't have anything (no address, no name) */ - if((self->rdaddr == DEV_ADDR_ANY) && (self->rname[0] == '\0')) - { - /* Try to find a suitable address */ - if((err = irnet_discover_daddr_and_lsap_sel(self)) != 0) - DRETURN(err, IRDA_SOCK_INFO, "auto-connect failed!\n"); - /* In most cases, the call above is non-blocking */ - } - else - { - /* If we have only the name (no address), try to get an address */ - if(self->rdaddr == DEV_ADDR_ANY) - { - if((err = irnet_dname_to_daddr(self)) != 0) - DRETURN(err, IRDA_SOCK_INFO, "name connect failed!\n"); - } - else - /* Use the requested destination address */ - self->daddr = self->rdaddr; - - /* Query remote LM-IAS to find LSAP selector */ - irnet_find_lsap_sel(self); - /* The above call is non blocking */ - } - - /* At this point, we are waiting for the IrDA stack to call us back, - * or we have already failed. - * We will finish the connection procedure in irnet_connect_tsap(). - */ - DEXIT(IRDA_SOCK_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_irnet_destroy(self) - * - * Destroy irnet instance - * - * Note : this need to be called from a process context. - */ -void -irda_irnet_destroy(irnet_socket * self) -{ - DENTER(IRDA_SOCK_TRACE, "(self=0x%p)\n", self); - if(self == NULL) - return; - - /* Remove ourselves from hashbin (if we are queued in hashbin) - * Note : `irnet_server.running' protect us from calls in hashbin_delete() */ - if((irnet_server.running) && (self->q.q_next != NULL)) - { - struct irnet_socket * entry; - DEBUG(IRDA_SOCK_INFO, "Removing from hash..\n"); - spin_lock_bh(&irnet_server.spinlock); - entry = hashbin_remove_this(irnet_server.list, (irda_queue_t *) self); - self->q.q_next = NULL; - spin_unlock_bh(&irnet_server.spinlock); - DASSERT(entry == self, , IRDA_SOCK_ERROR, "Can't remove from hash.\n"); - } - - /* If we were connected, post a message */ - if(test_bit(0, &self->ttp_open)) - { - /* Note : as the disconnect comes from ppp_generic, the unit number - * doesn't exist anymore when we post the event, so we need to pass - * NULL as the first arg... */ - irnet_post_event(NULL, IRNET_DISCONNECT_TO, - self->saddr, self->daddr, self->rname, 0); - } - - /* Prevent various IrDA callbacks from messing up things - * Need to be first */ - clear_bit(0, &self->ttp_connect); - - /* Prevent higher layer from accessing IrTTP */ - clear_bit(0, &self->ttp_open); - - /* Unregister with IrLMP */ - irlmp_unregister_client(self->ckey); - - /* Unregister with LM-IAS */ - if(self->iriap) - { - iriap_close(self->iriap); - self->iriap = NULL; - } - - /* Cleanup eventual discoveries from connection attempt or control channel */ - if(self->discoveries != NULL) - { - /* Cleanup our copy of the discovery log */ - kfree(self->discoveries); - self->discoveries = NULL; - } - - /* Close our IrTTP connection */ - if(self->tsap) - { - DEBUG(IRDA_SOCK_INFO, "Closing our TTP connection.\n"); - irttp_disconnect_request(self->tsap, NULL, P_NORMAL); - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } - self->stsap_sel = 0; - - DEXIT(IRDA_SOCK_TRACE, "\n"); -} - - -/************************** SERVER SOCKET **************************/ -/* - * The IrNET service is composed of one server socket and a variable - * number of regular IrNET sockets. The server socket is supposed to - * handle incoming connections and redirect them to one IrNET sockets. - * It's a superset of the regular IrNET socket, but has a very distinct - * behaviour... - */ - -/*------------------------------------------------------------------*/ -/* - * Function irnet_daddr_to_dname (self) - * - * Convert an IrDA address to a IrDA nickname - * - * It basically look into the discovery log until there is a match. - */ -static inline int -irnet_daddr_to_dname(irnet_socket * self) -{ - struct irda_device_info *discoveries; /* Copy of the discovery log */ - int number; /* Number of nodes in the log */ - int i; - - DENTER(IRDA_SERV_TRACE, "(self=0x%p)\n", self); - - /* Ask lmp for the current discovery log */ - discoveries = irlmp_get_discoveries(&number, 0xffff, - DISCOVERY_DEFAULT_SLOTS); - /* Check if the we got some results */ - if (discoveries == NULL) - DRETURN(-ENETUNREACH, IRDA_SERV_INFO, "Cachelog empty...\n"); - - /* Now, check all discovered devices (if any) */ - for(i = 0; i < number; i++) - { - /* Does the name match ? */ - if(discoveries[i].daddr == self->daddr) - { - /* Yes !!! Get it.. */ - strlcpy(self->rname, discoveries[i].info, sizeof(self->rname)); - self->rname[sizeof(self->rname) - 1] = '\0'; - DEBUG(IRDA_SERV_INFO, "Device 0x%08x is in fact ``%s''.\n", - self->daddr, self->rname); - kfree(discoveries); - DEXIT(IRDA_SERV_TRACE, "\n"); - return 0; - } - } - /* No luck ! */ - DEXIT(IRDA_SERV_INFO, ": cannot discover device 0x%08x !!!\n", self->daddr); - kfree(discoveries); - return -EADDRNOTAVAIL; -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_find_socket (self) - * - * Find the correct IrNET socket - * - * Look into the list of IrNET sockets and finds one with the right - * properties... - */ -static inline irnet_socket * -irnet_find_socket(irnet_socket * self) -{ - irnet_socket * new = (irnet_socket *) NULL; - int err; - - DENTER(IRDA_SERV_TRACE, "(self=0x%p)\n", self); - - /* Get the addresses of the requester */ - self->daddr = irttp_get_daddr(self->tsap); - self->saddr = irttp_get_saddr(self->tsap); - - /* Try to get the IrDA nickname of the requester */ - err = irnet_daddr_to_dname(self); - - /* Protect access to the instance list */ - spin_lock_bh(&irnet_server.spinlock); - - /* So now, try to get an socket having specifically - * requested that nickname */ - if(err == 0) - { - new = (irnet_socket *) hashbin_find(irnet_server.list, - 0, self->rname); - if(new) - DEBUG(IRDA_SERV_INFO, "Socket 0x%p matches rname ``%s''.\n", - new, new->rname); - } - - /* If no name matches, try to find an socket by the destination address */ - /* It can be either the requested destination address (set via the - * control channel), or the current destination address if the - * socket is in the middle of a connection request */ - if(new == (irnet_socket *) NULL) - { - new = (irnet_socket *) hashbin_get_first(irnet_server.list); - while(new !=(irnet_socket *) NULL) - { - /* Does it have the same address ? */ - if((new->rdaddr == self->daddr) || (new->daddr == self->daddr)) - { - /* Yes !!! Get it.. */ - DEBUG(IRDA_SERV_INFO, "Socket 0x%p matches daddr %#08x.\n", - new, self->daddr); - break; - } - new = (irnet_socket *) hashbin_get_next(irnet_server.list); - } - } - - /* If we don't have any socket, get the first unconnected socket */ - if(new == (irnet_socket *) NULL) - { - new = (irnet_socket *) hashbin_get_first(irnet_server.list); - while(new !=(irnet_socket *) NULL) - { - /* Is it available ? */ - if(!(test_bit(0, &new->ttp_open)) && (new->rdaddr == DEV_ADDR_ANY) && - (new->rname[0] == '\0') && (new->ppp_open)) - { - /* Yes !!! Get it.. */ - DEBUG(IRDA_SERV_INFO, "Socket 0x%p is free.\n", - new); - break; - } - new = (irnet_socket *) hashbin_get_next(irnet_server.list); - } - } - - /* Spin lock end */ - spin_unlock_bh(&irnet_server.spinlock); - - DEXIT(IRDA_SERV_TRACE, " - new = 0x%p\n", new); - return new; -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_connect_socket (self) - * - * Connect an incoming connection to the socket - * - */ -static inline int -irnet_connect_socket(irnet_socket * server, - irnet_socket * new, - struct qos_info * qos, - __u32 max_sdu_size, - __u8 max_header_size) -{ - DENTER(IRDA_SERV_TRACE, "(server=0x%p, new=0x%p)\n", - server, new); - - /* Now attach up the new socket */ - new->tsap = irttp_dup(server->tsap, new); - DABORT(new->tsap == NULL, -1, IRDA_SERV_ERROR, "dup failed!\n"); - - /* Set up all the relevant parameters on the new socket */ - new->stsap_sel = new->tsap->stsap_sel; - new->dtsap_sel = new->tsap->dtsap_sel; - new->saddr = irttp_get_saddr(new->tsap); - new->daddr = irttp_get_daddr(new->tsap); - - new->max_header_size = max_header_size; - new->max_sdu_size_tx = max_sdu_size; - new->max_data_size = max_sdu_size; -#ifdef STREAM_COMPAT - /* If we want to receive "stream sockets" */ - if(max_sdu_size == 0) - new->max_data_size = irttp_get_max_seg_size(new->tsap); -#endif /* STREAM_COMPAT */ - - /* Clean up the original one to keep it in listen state */ - irttp_listen(server->tsap); - - /* Send a connection response on the new socket */ - irttp_connect_response(new->tsap, new->max_sdu_size_rx, NULL); - - /* Allow PPP to send its junk over the new socket... */ - set_bit(0, &new->ttp_open); - - /* Not connecting anymore, and clean up last possible remains - * of connection attempts on the socket */ - clear_bit(0, &new->ttp_connect); - if(new->iriap) - { - iriap_close(new->iriap); - new->iriap = NULL; - } - if(new->discoveries != NULL) - { - kfree(new->discoveries); - new->discoveries = NULL; - } - -#ifdef CONNECT_INDIC_KICK - /* As currently we don't block packets in ppp_irnet_send() while passive, - * this is not really needed... - * Also, not doing it give IrDA a chance to finish the setup properly - * before being swamped with packets... */ - ppp_output_wakeup(&new->chan); -#endif /* CONNECT_INDIC_KICK */ - - /* Notify the control channel */ - irnet_post_event(new, IRNET_CONNECT_FROM, - new->saddr, new->daddr, server->rname, 0); - - DEXIT(IRDA_SERV_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_disconnect_server (self) - * - * Cleanup the server socket when the incoming connection abort - * - */ -static inline void -irnet_disconnect_server(irnet_socket * self, - struct sk_buff *skb) -{ - DENTER(IRDA_SERV_TRACE, "(self=0x%p)\n", self); - - /* Put the received packet in the black hole */ - kfree_skb(skb); - -#ifdef FAIL_SEND_DISCONNECT - /* Tell the other party we don't want to be connected */ - /* Hum... Is it the right thing to do ? And do we need to send - * a connect response before ? It looks ok without this... */ - irttp_disconnect_request(self->tsap, NULL, P_NORMAL); -#endif /* FAIL_SEND_DISCONNECT */ - - /* Notify the control channel (see irnet_find_socket()) */ - irnet_post_event(NULL, IRNET_REQUEST_FROM, - self->saddr, self->daddr, self->rname, 0); - - /* Clean up the server to keep it in listen state */ - irttp_listen(self->tsap); - - DEXIT(IRDA_SERV_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_setup_server (self) - * - * Create a IrTTP server and set it up... - * - * Register the IrLAN hint bit, create a IrTTP instance for us, - * set all the IrTTP callbacks and create an IrIAS entry... - */ -static inline int -irnet_setup_server(void) -{ - __u16 hints; - - DENTER(IRDA_SERV_TRACE, "()\n"); - - /* Initialise the regular socket part of the server */ - irda_irnet_create(&irnet_server.s); - - /* Open a local TSAP (an IrTTP instance) for the server */ - irnet_open_tsap(&irnet_server.s); - - /* PPP part setup */ - irnet_server.s.ppp_open = 0; - irnet_server.s.chan.private = NULL; - irnet_server.s.file = NULL; - - /* Get the hint bit corresponding to IrLAN */ - /* Note : we overload the IrLAN hint bit. As it is only a "hint", and as - * we provide roughly the same functionality as IrLAN, this is ok. - * In fact, the situation is similar as JetSend overloading the Obex hint - */ - hints = irlmp_service_to_hint(S_LAN); - -#ifdef ADVERTISE_HINT - /* Register with IrLMP as a service (advertise our hint bit) */ - irnet_server.skey = irlmp_register_service(hints); -#endif /* ADVERTISE_HINT */ - - /* Register with LM-IAS (so that people can connect to us) */ - irnet_server.ias_obj = irias_new_object(IRNET_SERVICE_NAME, jiffies); - irias_add_integer_attrib(irnet_server.ias_obj, IRNET_IAS_VALUE, - irnet_server.s.stsap_sel, IAS_KERNEL_ATTR); - irias_insert_object(irnet_server.ias_obj); - -#ifdef DISCOVERY_EVENTS - /* Tell IrLMP we want to be notified of newly discovered nodes */ - irlmp_update_client(irnet_server.s.ckey, hints, - irnet_discovery_indication, irnet_expiry_indication, - (void *) &irnet_server.s); -#endif - - DEXIT(IRDA_SERV_TRACE, " - self=0x%p\n", &irnet_server.s); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Function irda_destroy_server (self) - * - * Destroy the IrTTP server... - * - * Reverse of the previous function... - */ -static inline void -irnet_destroy_server(void) -{ - DENTER(IRDA_SERV_TRACE, "()\n"); - -#ifdef ADVERTISE_HINT - /* Unregister with IrLMP */ - irlmp_unregister_service(irnet_server.skey); -#endif /* ADVERTISE_HINT */ - - /* Unregister with LM-IAS */ - if(irnet_server.ias_obj) - irias_delete_object(irnet_server.ias_obj); - - /* Cleanup the socket part */ - irda_irnet_destroy(&irnet_server.s); - - DEXIT(IRDA_SERV_TRACE, "\n"); -} - - -/************************ IRDA-TTP CALLBACKS ************************/ -/* - * When we create a IrTTP instance, we pass to it a set of callbacks - * that IrTTP will call in case of various events. - * We take care of those events here. - */ - -/*------------------------------------------------------------------*/ -/* - * Function irnet_data_indication (instance, sap, skb) - * - * Received some data from TinyTP. Just queue it on the receive queue - * - */ -static int -irnet_data_indication(void * instance, - void * sap, - struct sk_buff *skb) -{ - irnet_socket * ap = (irnet_socket *) instance; - unsigned char * p; - int code = 0; - - DENTER(IRDA_TCB_TRACE, "(self/ap=0x%p, skb=0x%p)\n", - ap, skb); - DASSERT(skb != NULL, 0, IRDA_CB_ERROR, "skb is NULL !!!\n"); - - /* Check is ppp is ready to receive our packet */ - if(!ap->ppp_open) - { - DERROR(IRDA_CB_ERROR, "PPP not ready, dropping packet...\n"); - /* When we return error, TTP will need to requeue the skb and - * will stop the sender. IrTTP will stall until we send it a - * flow control request... */ - return -ENOMEM; - } - - /* strip address/control field if present */ - p = skb->data; - if((p[0] == PPP_ALLSTATIONS) && (p[1] == PPP_UI)) - { - /* chop off address/control */ - if(skb->len < 3) - goto err_exit; - p = skb_pull(skb, 2); - } - - /* decompress protocol field if compressed */ - if(p[0] & 1) - { - /* protocol is compressed */ - *(u8 *)skb_push(skb, 1) = 0; - } - else - if(skb->len < 2) - goto err_exit; - - /* pass to generic ppp layer */ - /* Note : how do I know if ppp can accept or not the packet ? This is - * essential if I want to manage flow control smoothly... */ - ppp_input(&ap->chan, skb); - - DEXIT(IRDA_TCB_TRACE, "\n"); - return 0; - - err_exit: - DERROR(IRDA_CB_ERROR, "Packet too small, dropping...\n"); - kfree_skb(skb); - ppp_input_error(&ap->chan, code); - return 0; /* Don't return an error code, only for flow control... */ -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_disconnect_indication (instance, sap, reason, skb) - * - * Connection has been closed. Chech reason to find out why - * - * Note : there are many cases where we come here : - * o attempted to connect, timeout - * o connected, link is broken, LAP has timeout - * o connected, other side close the link - * o connection request on the server not handled - */ -static void -irnet_disconnect_indication(void * instance, - void * sap, - LM_REASON reason, - struct sk_buff *skb) -{ - irnet_socket * self = (irnet_socket *) instance; - int test_open; - int test_connect; - - DENTER(IRDA_TCB_TRACE, "(self=0x%p)\n", self); - DASSERT(self != NULL, , IRDA_CB_ERROR, "Self is NULL !!!\n"); - - /* Don't care about it, but let's not leak it */ - if(skb) - dev_kfree_skb(skb); - - /* Prevent higher layer from accessing IrTTP */ - test_open = test_and_clear_bit(0, &self->ttp_open); - /* Not connecting anymore... - * (note : TSAP is open, so IAP callbacks are no longer pending...) */ - test_connect = test_and_clear_bit(0, &self->ttp_connect); - - /* If both self->ttp_open and self->ttp_connect are NULL, it mean that we - * have a race condition with irda_irnet_destroy() or - * irnet_connect_indication(), so don't mess up tsap... - */ - if(!(test_open || test_connect)) - { - DERROR(IRDA_CB_ERROR, "Race condition detected...\n"); - return; - } - - /* If we were active, notify the control channel */ - if(test_open) - irnet_post_event(self, IRNET_DISCONNECT_FROM, - self->saddr, self->daddr, self->rname, 0); - else - /* If we were trying to connect, notify the control channel */ - if((self->tsap) && (self != &irnet_server.s)) - irnet_post_event(self, IRNET_NOANSWER_FROM, - self->saddr, self->daddr, self->rname, 0); - - /* Close our IrTTP connection, cleanup tsap */ - if((self->tsap) && (self != &irnet_server.s)) - { - DEBUG(IRDA_CB_INFO, "Closing our TTP connection.\n"); - irttp_close_tsap(self->tsap); - self->tsap = NULL; - } - /* Cleanup the socket in case we want to reconnect in ppp_output_wakeup() */ - self->stsap_sel = 0; - self->daddr = DEV_ADDR_ANY; - self->tx_flow = FLOW_START; - - /* Deal with the ppp instance if it's still alive */ - if(self->ppp_open) - { - if(test_open) - { - /* ppp_unregister_channel() wants a user context. */ - schedule_work(&self->disconnect_work); - } - else - { - /* If we were trying to connect, flush (drain) ppp_generic - * Tx queue (most often we have blocked it), which will - * trigger an other attempt to connect. If we are passive, - * this will empty the Tx queue after last try. */ - ppp_output_wakeup(&self->chan); - } - } - - DEXIT(IRDA_TCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_connect_confirm (instance, sap, qos, max_sdu_size, skb) - * - * Connections has been confirmed by the remote device - * - */ -static void -irnet_connect_confirm(void * instance, - void * sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - irnet_socket * self = (irnet_socket *) instance; - - DENTER(IRDA_TCB_TRACE, "(self=0x%p)\n", self); - - /* Check if socket is closing down (via irda_irnet_destroy()) */ - if(! test_bit(0, &self->ttp_connect)) - { - DERROR(IRDA_CB_ERROR, "Socket no longer connecting. Ouch !\n"); - return; - } - - /* How much header space do we need to reserve */ - self->max_header_size = max_header_size; - - /* IrTTP max SDU size in transmit direction */ - self->max_sdu_size_tx = max_sdu_size; - self->max_data_size = max_sdu_size; -#ifdef STREAM_COMPAT - if(max_sdu_size == 0) - self->max_data_size = irttp_get_max_seg_size(self->tsap); -#endif /* STREAM_COMPAT */ - - /* At this point, IrLMP has assigned our source address */ - self->saddr = irttp_get_saddr(self->tsap); - - /* Allow higher layer to access IrTTP */ - set_bit(0, &self->ttp_open); - clear_bit(0, &self->ttp_connect); /* Not racy, IrDA traffic is serial */ - /* Give a kick in the ass of ppp_generic so that he sends us some data */ - ppp_output_wakeup(&self->chan); - - /* Check size of received packet */ - if(skb->len > 0) - { -#ifdef PASS_CONNECT_PACKETS - DEBUG(IRDA_CB_INFO, "Passing connect packet to PPP.\n"); - /* Try to pass it to PPP */ - irnet_data_indication(instance, sap, skb); -#else /* PASS_CONNECT_PACKETS */ - DERROR(IRDA_CB_ERROR, "Dropping non empty packet.\n"); - kfree_skb(skb); /* Note : will be optimised with other kfree... */ -#endif /* PASS_CONNECT_PACKETS */ - } - else - kfree_skb(skb); - - /* Notify the control channel */ - irnet_post_event(self, IRNET_CONNECT_TO, - self->saddr, self->daddr, self->rname, 0); - - DEXIT(IRDA_TCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_flow_indication (instance, sap, flow) - * - * Used by TinyTP to tell us if it can accept more data or not - * - */ -static void -irnet_flow_indication(void * instance, - void * sap, - LOCAL_FLOW flow) -{ - irnet_socket * self = (irnet_socket *) instance; - LOCAL_FLOW oldflow = self->tx_flow; - - DENTER(IRDA_TCB_TRACE, "(self=0x%p, flow=%d)\n", self, flow); - - /* Update our state */ - self->tx_flow = flow; - - /* Check what IrTTP want us to do... */ - switch(flow) - { - case FLOW_START: - DEBUG(IRDA_CB_INFO, "IrTTP wants us to start again\n"); - /* Check if we really need to wake up PPP */ - if(oldflow == FLOW_STOP) - ppp_output_wakeup(&self->chan); - else - DEBUG(IRDA_CB_INFO, "But we were already transmitting !!!\n"); - break; - case FLOW_STOP: - DEBUG(IRDA_CB_INFO, "IrTTP wants us to slow down\n"); - break; - default: - DEBUG(IRDA_CB_INFO, "Unknown flow command!\n"); - break; - } - - DEXIT(IRDA_TCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_status_indication (instance, sap, reason, skb) - * - * Link (IrLAP) status report. - * - */ -static void -irnet_status_indication(void * instance, - LINK_STATUS link, - LOCK_STATUS lock) -{ - irnet_socket * self = (irnet_socket *) instance; - - DENTER(IRDA_TCB_TRACE, "(self=0x%p)\n", self); - DASSERT(self != NULL, , IRDA_CB_ERROR, "Self is NULL !!!\n"); - - /* We can only get this event if we are connected */ - switch(link) - { - case STATUS_NO_ACTIVITY: - irnet_post_event(self, IRNET_BLOCKED_LINK, - self->saddr, self->daddr, self->rname, 0); - break; - default: - DEBUG(IRDA_CB_INFO, "Unknown status...\n"); - } - - DEXIT(IRDA_TCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_connect_indication(instance, sap, qos, max_sdu_size, userdata) - * - * Incoming connection - * - * In theory, this function is called only on the server socket. - * Some other node is attempting to connect to the IrNET service, and has - * sent a connection request on our server socket. - * We just redirect the connection to the relevant IrNET socket. - * - * Note : we also make sure that between 2 irnet nodes, there can - * exist only one irnet connection. - */ -static void -irnet_connect_indication(void * instance, - void * sap, - struct qos_info *qos, - __u32 max_sdu_size, - __u8 max_header_size, - struct sk_buff *skb) -{ - irnet_socket * server = &irnet_server.s; - irnet_socket * new = (irnet_socket *) NULL; - - DENTER(IRDA_TCB_TRACE, "(server=0x%p)\n", server); - DASSERT(instance == &irnet_server, , IRDA_CB_ERROR, - "Invalid instance (0x%p) !!!\n", instance); - DASSERT(sap == irnet_server.s.tsap, , IRDA_CB_ERROR, "Invalid sap !!!\n"); - - /* Try to find the most appropriate IrNET socket */ - new = irnet_find_socket(server); - - /* After all this hard work, do we have an socket ? */ - if(new == (irnet_socket *) NULL) - { - DEXIT(IRDA_CB_INFO, ": No socket waiting for this connection.\n"); - irnet_disconnect_server(server, skb); - return; - } - - /* Is the socket already busy ? */ - if(test_bit(0, &new->ttp_open)) - { - DEXIT(IRDA_CB_INFO, ": Socket already connected.\n"); - irnet_disconnect_server(server, skb); - return; - } - - /* The following code is a bit tricky, so need comments ;-) - */ - /* If ttp_connect is set, the socket is trying to connect to the other - * end and may have sent a IrTTP connection request and is waiting for - * a connection response (that may never come). - * Now, the pain is that the socket may have opened a tsap and is - * waiting on it, while the other end is trying to connect to it on - * another tsap. - * Because IrNET can be peer to peer, we need to workaround this. - * Furthermore, the way the irnetd script is implemented, the - * target will create a second IrNET connection back to the - * originator and expect the originator to bind this new connection - * to the original PPPD instance. - * And of course, if we don't use irnetd, we can have a race when - * both side try to connect simultaneously, which could leave both - * connections half closed (yuck). - * Conclusions : - * 1) The "originator" must accept the new connection and get rid - * of the old one so that irnetd works - * 2) One side must deny the new connection to avoid races, - * but both side must agree on which side it is... - * Most often, the originator is primary at the LAP layer. - * Jean II - */ - /* Now, let's look at the way I wrote the test... - * We need to clear up the ttp_connect flag atomically to prevent - * irnet_disconnect_indication() to mess up the tsap we are going to close. - * We want to clear the ttp_connect flag only if we close the tsap, - * otherwise we will never close it, so we need to check for primary - * *before* doing the test on the flag. - * And of course, ALLOW_SIMULT_CONNECT can disable this entirely... - * Jean II - */ - - /* Socket already connecting ? On primary ? */ - if(0 -#ifdef ALLOW_SIMULT_CONNECT - || ((irttp_is_primary(server->tsap) == 1) && /* primary */ - (test_and_clear_bit(0, &new->ttp_connect))) -#endif /* ALLOW_SIMULT_CONNECT */ - ) - { - DERROR(IRDA_CB_ERROR, "Socket already connecting, but going to reuse it !\n"); - - /* Cleanup the old TSAP if necessary - IrIAP will be cleaned up later */ - if(new->tsap != NULL) - { - /* Close the old connection the new socket was attempting, - * so that we can hook it up to the new connection. - * It's now safe to do it... */ - irttp_close_tsap(new->tsap); - new->tsap = NULL; - } - } - else - { - /* Three options : - * 1) socket was not connecting or connected : ttp_connect should be 0. - * 2) we don't want to connect the socket because we are secondary or - * ALLOW_SIMULT_CONNECT is undefined. ttp_connect should be 1. - * 3) we are half way in irnet_disconnect_indication(), and it's a - * nice race condition... Fortunately, we can detect that by checking - * if tsap is still alive. On the other hand, we can't be in - * irda_irnet_destroy() otherwise we would not have found this - * socket in the hashbin. - * Jean II */ - if((test_bit(0, &new->ttp_connect)) || (new->tsap != NULL)) - { - /* Don't mess this socket, somebody else in in charge... */ - DERROR(IRDA_CB_ERROR, "Race condition detected, socket in use, abort connect...\n"); - irnet_disconnect_server(server, skb); - return; - } - } - - /* So : at this point, we have a socket, and it is idle. Good ! */ - irnet_connect_socket(server, new, qos, max_sdu_size, max_header_size); - - /* Check size of received packet */ - if(skb->len > 0) - { -#ifdef PASS_CONNECT_PACKETS - DEBUG(IRDA_CB_INFO, "Passing connect packet to PPP.\n"); - /* Try to pass it to PPP */ - irnet_data_indication(new, new->tsap, skb); -#else /* PASS_CONNECT_PACKETS */ - DERROR(IRDA_CB_ERROR, "Dropping non empty packet.\n"); - kfree_skb(skb); /* Note : will be optimised with other kfree... */ -#endif /* PASS_CONNECT_PACKETS */ - } - else - kfree_skb(skb); - - DEXIT(IRDA_TCB_TRACE, "\n"); -} - - -/********************** IRDA-IAS/LMP CALLBACKS **********************/ -/* - * These are the callbacks called by other layers of the IrDA stack, - * mainly LMP for discovery and IAS for name queries. - */ - -/*------------------------------------------------------------------*/ -/* - * Function irnet_getvalue_confirm (result, obj_id, value, priv) - * - * Got answer from remote LM-IAS, just connect - * - * This is the reply to a IAS query we were doing to find the TSAP of - * the device we want to connect to. - * If we have found a valid TSAP, just initiate the TTP connection - * on this TSAP. - */ -static void -irnet_getvalue_confirm(int result, - __u16 obj_id, - struct ias_value *value, - void * priv) -{ - irnet_socket * self = (irnet_socket *) priv; - - DENTER(IRDA_OCB_TRACE, "(self=0x%p)\n", self); - DASSERT(self != NULL, , IRDA_OCB_ERROR, "Self is NULL !!!\n"); - - /* Check if already connected (via irnet_connect_socket()) - * or socket is closing down (via irda_irnet_destroy()) */ - if(! test_bit(0, &self->ttp_connect)) - { - DERROR(IRDA_OCB_ERROR, "Socket no longer connecting. Ouch !\n"); - return; - } - - /* We probably don't need to make any more queries */ - iriap_close(self->iriap); - self->iriap = NULL; - - /* Post process the IAS reply */ - self->dtsap_sel = irnet_ias_to_tsap(self, result, value); - - /* If error, just go out */ - if(self->errno) - { - clear_bit(0, &self->ttp_connect); - DERROR(IRDA_OCB_ERROR, "IAS connect failed ! (0x%X)\n", self->errno); - return; - } - - DEBUG(IRDA_OCB_INFO, "daddr = %08x, lsap = %d, starting IrTTP connection\n", - self->daddr, self->dtsap_sel); - - /* Start up TTP - non blocking */ - irnet_connect_tsap(self); - - DEXIT(IRDA_OCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_discovervalue_confirm (result, obj_id, value, priv) - * - * Handle the TSAP discovery procedure state machine. - * Got answer from remote LM-IAS, try next device - * - * We are doing a TSAP discovery procedure, and we got an answer to - * a IAS query we were doing to find the TSAP on one of the address - * in the discovery log. - * - * If we have found a valid TSAP for the first time, save it. If it's - * not the first time we found one, complain. - * - * If we have more addresses in the log, just initiate a new query. - * Note that those query may fail (see irnet_discover_daddr_and_lsap_sel()) - * - * Otherwise, wrap up the procedure (cleanup), check if we have found - * any device and connect to it. - */ -static void -irnet_discovervalue_confirm(int result, - __u16 obj_id, - struct ias_value *value, - void * priv) -{ - irnet_socket * self = (irnet_socket *) priv; - __u8 dtsap_sel; /* TSAP we are looking for */ - - DENTER(IRDA_OCB_TRACE, "(self=0x%p)\n", self); - DASSERT(self != NULL, , IRDA_OCB_ERROR, "Self is NULL !!!\n"); - - /* Check if already connected (via irnet_connect_socket()) - * or socket is closing down (via irda_irnet_destroy()) */ - if(! test_bit(0, &self->ttp_connect)) - { - DERROR(IRDA_OCB_ERROR, "Socket no longer connecting. Ouch !\n"); - return; - } - - /* Post process the IAS reply */ - dtsap_sel = irnet_ias_to_tsap(self, result, value); - - /* Have we got something ? */ - if(self->errno == 0) - { - /* We found the requested service */ - if(self->daddr != DEV_ADDR_ANY) - { - DERROR(IRDA_OCB_ERROR, "More than one device in range supports IrNET...\n"); - } - else - { - /* First time we found that one, save it ! */ - self->daddr = self->discoveries[self->disco_index].daddr; - self->dtsap_sel = dtsap_sel; - } - } - - /* If no failure */ - if((self->errno == -EADDRNOTAVAIL) || (self->errno == 0)) - { - int ret; - - /* Search the next node */ - ret = irnet_discover_next_daddr(self); - if(!ret) - { - /* In this case, the above request was non-blocking. - * We will return here after a while... */ - return; - } - /* In this case, we have processed the last discovery item */ - } - - /* No more queries to be done (failure or last one) */ - - /* We probably don't need to make any more queries */ - iriap_close(self->iriap); - self->iriap = NULL; - - /* No more items : remove the log and signal termination */ - DEBUG(IRDA_OCB_INFO, "Cleaning up log (0x%p)\n", - self->discoveries); - if(self->discoveries != NULL) - { - /* Cleanup our copy of the discovery log */ - kfree(self->discoveries); - self->discoveries = NULL; - } - self->disco_number = -1; - - /* Check out what we found */ - if(self->daddr == DEV_ADDR_ANY) - { - self->daddr = DEV_ADDR_ANY; - clear_bit(0, &self->ttp_connect); - DEXIT(IRDA_OCB_TRACE, ": cannot discover IrNET in any device !!!\n"); - return; - } - - /* We have a valid address - just connect */ - - DEBUG(IRDA_OCB_INFO, "daddr = %08x, lsap = %d, starting IrTTP connection\n", - self->daddr, self->dtsap_sel); - - /* Start up TTP - non blocking */ - irnet_connect_tsap(self); - - DEXIT(IRDA_OCB_TRACE, "\n"); -} - -#ifdef DISCOVERY_EVENTS -/*------------------------------------------------------------------*/ -/* - * Function irnet_discovery_indication (discovery) - * - * Got a discovery indication from IrLMP, post an event - * - * Note : IrLMP take care of matching the hint mask for us, and also - * check if it is a "new" node for us... - * - * As IrLMP filter on the IrLAN hint bit, we get both IrLAN and IrNET - * nodes, so it's only at connection time that we will know if the - * node support IrNET, IrLAN or both. The other solution is to check - * in IAS the PNP ids and service name. - * Note : even if a node support IrNET (or IrLAN), it's no guarantee - * that we will be able to connect to it, the node might already be - * busy... - * - * One last thing : in some case, this function will trigger duplicate - * discovery events. On the other hand, we should catch all - * discoveries properly (i.e. not miss one). Filtering duplicate here - * is to messy, so we leave that to user space... - */ -static void -irnet_discovery_indication(discinfo_t * discovery, - DISCOVERY_MODE mode, - void * priv) -{ - irnet_socket * self = &irnet_server.s; - - DENTER(IRDA_OCB_TRACE, "(self=0x%p)\n", self); - DASSERT(priv == &irnet_server, , IRDA_OCB_ERROR, - "Invalid instance (0x%p) !!!\n", priv); - - DEBUG(IRDA_OCB_INFO, "Discovered new IrNET/IrLAN node %s...\n", - discovery->info); - - /* Notify the control channel */ - irnet_post_event(NULL, IRNET_DISCOVER, - discovery->saddr, discovery->daddr, discovery->info, - get_unaligned((__u16 *)discovery->hints)); - - DEXIT(IRDA_OCB_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_expiry_indication (expiry) - * - * Got a expiry indication from IrLMP, post an event - * - * Note : IrLMP take care of matching the hint mask for us, we only - * check if it is a "new" node... - */ -static void -irnet_expiry_indication(discinfo_t * expiry, - DISCOVERY_MODE mode, - void * priv) -{ - irnet_socket * self = &irnet_server.s; - - DENTER(IRDA_OCB_TRACE, "(self=0x%p)\n", self); - DASSERT(priv == &irnet_server, , IRDA_OCB_ERROR, - "Invalid instance (0x%p) !!!\n", priv); - - DEBUG(IRDA_OCB_INFO, "IrNET/IrLAN node %s expired...\n", - expiry->info); - - /* Notify the control channel */ - irnet_post_event(NULL, IRNET_EXPIRE, - expiry->saddr, expiry->daddr, expiry->info, - get_unaligned((__u16 *)expiry->hints)); - - DEXIT(IRDA_OCB_TRACE, "\n"); -} -#endif /* DISCOVERY_EVENTS */ - - -/*********************** PROC ENTRY CALLBACKS ***********************/ -/* - * We create a instance in the /proc filesystem, and here we take care - * of that... - */ - -#ifdef CONFIG_PROC_FS -static int -irnet_proc_show(struct seq_file *m, void *v) -{ - irnet_socket * self; - char * state; - int i = 0; - - /* Get the IrNET server information... */ - seq_printf(m, "IrNET server - "); - seq_printf(m, "IrDA state: %s, ", - (irnet_server.running ? "running" : "dead")); - seq_printf(m, "stsap_sel: %02x, ", irnet_server.s.stsap_sel); - seq_printf(m, "dtsap_sel: %02x\n", irnet_server.s.dtsap_sel); - - /* Do we need to continue ? */ - if(!irnet_server.running) - return 0; - - /* Protect access to the instance list */ - spin_lock_bh(&irnet_server.spinlock); - - /* Get the sockets one by one... */ - self = (irnet_socket *) hashbin_get_first(irnet_server.list); - while(self != NULL) - { - /* Start printing info about the socket. */ - seq_printf(m, "\nIrNET socket %d - ", i++); - - /* First, get the requested configuration */ - seq_printf(m, "Requested IrDA name: \"%s\", ", self->rname); - seq_printf(m, "daddr: %08x, ", self->rdaddr); - seq_printf(m, "saddr: %08x\n", self->rsaddr); - - /* Second, get all the PPP info */ - seq_printf(m, " PPP state: %s", - (self->ppp_open ? "registered" : "unregistered")); - if(self->ppp_open) - { - seq_printf(m, ", unit: ppp%d", - ppp_unit_number(&self->chan)); - seq_printf(m, ", channel: %d", - ppp_channel_index(&self->chan)); - seq_printf(m, ", mru: %d", - self->mru); - /* Maybe add self->flags ? Later... */ - } - - /* Then, get all the IrDA specific info... */ - if(self->ttp_open) - state = "connected"; - else - if(self->tsap != NULL) - state = "connecting"; - else - if(self->iriap != NULL) - state = "searching"; - else - if(self->ttp_connect) - state = "weird"; - else - state = "idle"; - seq_printf(m, "\n IrDA state: %s, ", state); - seq_printf(m, "daddr: %08x, ", self->daddr); - seq_printf(m, "stsap_sel: %02x, ", self->stsap_sel); - seq_printf(m, "dtsap_sel: %02x\n", self->dtsap_sel); - - /* Next socket, please... */ - self = (irnet_socket *) hashbin_get_next(irnet_server.list); - } - - /* Spin lock end */ - spin_unlock_bh(&irnet_server.spinlock); - - return 0; -} - -static int irnet_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, irnet_proc_show, NULL); -} - -static const struct file_operations irnet_proc_fops = { - .owner = THIS_MODULE, - .open = irnet_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; -#endif /* PROC_FS */ - - -/********************** CONFIGURATION/CLEANUP **********************/ -/* - * Initialisation and teardown of the IrDA part, called at module - * insertion and removal... - */ - -/*------------------------------------------------------------------*/ -/* - * Prepare the IrNET layer for operation... - */ -int __init -irda_irnet_init(void) -{ - int err = 0; - - DENTER(MODULE_TRACE, "()\n"); - - /* Pure paranoia - should be redundant */ - memset(&irnet_server, 0, sizeof(struct irnet_root)); - - /* Setup start of irnet instance list */ - irnet_server.list = hashbin_new(HB_NOLOCK); - DABORT(irnet_server.list == NULL, -ENOMEM, - MODULE_ERROR, "Can't allocate hashbin!\n"); - /* Init spinlock for instance list */ - spin_lock_init(&irnet_server.spinlock); - - /* Initialise control channel */ - init_waitqueue_head(&irnet_events.rwait); - irnet_events.index = 0; - /* Init spinlock for event logging */ - spin_lock_init(&irnet_events.spinlock); - -#ifdef CONFIG_PROC_FS - /* Add a /proc file for irnet infos */ - proc_create("irnet", 0, proc_irda, &irnet_proc_fops); -#endif /* CONFIG_PROC_FS */ - - /* Setup the IrNET server */ - err = irnet_setup_server(); - - if(!err) - /* We are no longer functional... */ - irnet_server.running = 1; - - DEXIT(MODULE_TRACE, "\n"); - return err; -} - -/*------------------------------------------------------------------*/ -/* - * Cleanup at exit... - */ -void __exit -irda_irnet_cleanup(void) -{ - DENTER(MODULE_TRACE, "()\n"); - - /* We are no longer there... */ - irnet_server.running = 0; - -#ifdef CONFIG_PROC_FS - /* Remove our /proc file */ - remove_proc_entry("irnet", proc_irda); -#endif /* CONFIG_PROC_FS */ - - /* Remove our IrNET server from existence */ - irnet_destroy_server(); - - /* Remove all instances of IrNET socket still present */ - hashbin_delete(irnet_server.list, (FREE_FUNC) irda_irnet_destroy); - - DEXIT(MODULE_TRACE, "\n"); -} diff --git a/net/irda/irnet/irnet_irda.h b/net/irda/irnet/irnet_irda.h deleted file mode 100644 index 3e408952a3f1..000000000000 --- a/net/irda/irnet/irnet_irda.h +++ /dev/null @@ -1,178 +0,0 @@ -/* - * IrNET protocol module : Synchronous PPP over an IrDA socket. - * - * Jean II - HPL `00 - <jt@hpl.hp.com> - * - * This file contains all definitions and declarations necessary for the - * IRDA part of the IrNET module (dealing with IrTTP, IrIAS and co). - * This file is a private header, so other modules don't want to know - * what's in there... - */ - -#ifndef IRNET_IRDA_H -#define IRNET_IRDA_H - -/***************************** INCLUDES *****************************/ -/* Please add other headers in irnet.h */ - -#include "irnet.h" /* Module global include */ - -/************************ CONSTANTS & MACROS ************************/ - -/* - * Name of the service (socket name) used by IrNET - */ -/* IAS object name (or part of it) */ -#define IRNET_SERVICE_NAME "IrNetv1" -/* IAS attribute */ -#define IRNET_IAS_VALUE "IrDA:TinyTP:LsapSel" -/* LMP notify name for client (only for /proc/net/irda/irlmp) */ -#define IRNET_NOTIFY_NAME "IrNET socket" -/* LMP notify name for server (only for /proc/net/irda/irlmp) */ -#define IRNET_NOTIFY_NAME_SERV "IrNET server" - -/****************************** TYPES ******************************/ - -/* - * This is the main structure where we store all the data pertaining to - * the IrNET server (listen for connection requests) and the root - * of the IrNET socket list - */ -typedef struct irnet_root -{ - irnet_socket s; /* To pretend we are a client... */ - - /* Generic stuff */ - int magic; /* Paranoia */ - int running; /* Are we operational ? */ - - /* Link list of all IrNET instances opened */ - hashbin_t * list; - spinlock_t spinlock; /* Serialize access to the list */ - /* Note : the way hashbin has been designed is absolutely not - * reentrant, beware... So, we blindly protect all with spinlock */ - - /* Handle for the hint bit advertised in IrLMP */ - void * skey; - - /* Server socket part */ - struct ias_object * ias_obj; /* Our service name + lsap in IAS */ - -} irnet_root; - - -/**************************** PROTOTYPES ****************************/ - -/* ----------------------- CONTROL CHANNEL ----------------------- */ -static void - irnet_post_event(irnet_socket *, - irnet_event, - __u32, - __u32, - char *, - __u16); -/* ----------------------- IRDA SUBROUTINES ----------------------- */ -static inline int - irnet_open_tsap(irnet_socket *); -static inline __u8 - irnet_ias_to_tsap(irnet_socket *, - int, - struct ias_value *); -static inline int - irnet_find_lsap_sel(irnet_socket *); -static inline int - irnet_connect_tsap(irnet_socket *); -static inline int - irnet_discover_next_daddr(irnet_socket *); -static inline int - irnet_discover_daddr_and_lsap_sel(irnet_socket *); -static inline int - irnet_dname_to_daddr(irnet_socket *); -/* ------------------------ SERVER SOCKET ------------------------ */ -static inline int - irnet_daddr_to_dname(irnet_socket *); -static inline irnet_socket * - irnet_find_socket(irnet_socket *); -static inline int - irnet_connect_socket(irnet_socket *, - irnet_socket *, - struct qos_info *, - __u32, - __u8); -static inline void - irnet_disconnect_server(irnet_socket *, - struct sk_buff *); -static inline int - irnet_setup_server(void); -static inline void - irnet_destroy_server(void); -/* ---------------------- IRDA-TTP CALLBACKS ---------------------- */ -static int - irnet_data_indication(void *, /* instance */ - void *, /* sap */ - struct sk_buff *); -static void - irnet_disconnect_indication(void *, - void *, - LM_REASON, - struct sk_buff *); -static void - irnet_connect_confirm(void *, - void *, - struct qos_info *, - __u32, - __u8, - struct sk_buff *); -static void - irnet_flow_indication(void *, - void *, - LOCAL_FLOW); -static void - irnet_status_indication(void *, - LINK_STATUS, - LOCK_STATUS); -static void - irnet_connect_indication(void *, - void *, - struct qos_info *, - __u32, - __u8, - struct sk_buff *); -/* -------------------- IRDA-IAS/LMP CALLBACKS -------------------- */ -static void - irnet_getvalue_confirm(int, - __u16, - struct ias_value *, - void *); -static void - irnet_discovervalue_confirm(int, - __u16, - struct ias_value *, - void *); -#ifdef DISCOVERY_EVENTS -static void - irnet_discovery_indication(discinfo_t *, - DISCOVERY_MODE, - void *); -static void - irnet_expiry_indication(discinfo_t *, - DISCOVERY_MODE, - void *); -#endif - -/**************************** VARIABLES ****************************/ - -/* - * The IrNET server. Listen to connection requests and co... - */ -static struct irnet_root irnet_server; - -/* Control channel stuff (note : extern) */ -struct irnet_ctrl_channel irnet_events; - -/* The /proc/net/irda directory, defined elsewhere... */ -#ifdef CONFIG_PROC_FS -extern struct proc_dir_entry *proc_irda; -#endif /* CONFIG_PROC_FS */ - -#endif /* IRNET_IRDA_H */ diff --git a/net/irda/irnet/irnet_ppp.c b/net/irda/irnet/irnet_ppp.c deleted file mode 100644 index 7025dcb853d0..000000000000 --- a/net/irda/irnet/irnet_ppp.c +++ /dev/null @@ -1,1189 +0,0 @@ -/* - * IrNET protocol module : Synchronous PPP over an IrDA socket. - * - * Jean II - HPL `00 - <jt@hpl.hp.com> - * - * This file implement the PPP interface and /dev/irnet character device. - * The PPP interface hook to the ppp_generic module, handle all our - * relationship to the PPP code in the kernel (and by extension to pppd), - * and exchange PPP frames with this module (send/receive). - * The /dev/irnet device is used primarily for 2 functions : - * 1) as a stub for pppd (the ppp daemon), so that we can appropriately - * generate PPP sessions (we pretend we are a tty). - * 2) as a control channel (write commands, read events) - */ - -#include <linux/sched/signal.h> -#include <linux/slab.h> - -#include "irnet_ppp.h" /* Private header */ -/* Please put other headers in irnet.h - Thanks */ - -/* Generic PPP callbacks (to call us) */ -static const struct ppp_channel_ops irnet_ppp_ops = { - .start_xmit = ppp_irnet_send, - .ioctl = ppp_irnet_ioctl -}; - -/************************* CONTROL CHANNEL *************************/ -/* - * When a pppd instance is not active on /dev/irnet, it acts as a control - * channel. - * Writing allow to set up the IrDA destination of the IrNET channel, - * and any application may be read events happening in IrNET... - */ - -/*------------------------------------------------------------------*/ -/* - * Write is used to send a command to configure a IrNET channel - * before it is open by pppd. The syntax is : "command argument" - * Currently there is only two defined commands : - * o name : set the requested IrDA nickname of the IrNET peer. - * o addr : set the requested IrDA address of the IrNET peer. - * Note : the code is crude, but effective... - */ -static inline ssize_t -irnet_ctrl_write(irnet_socket * ap, - const char __user *buf, - size_t count) -{ - char command[IRNET_MAX_COMMAND]; - char * start; /* Current command being processed */ - char * next; /* Next command to process */ - int length; /* Length of current command */ - - DENTER(CTRL_TRACE, "(ap=0x%p, count=%zd)\n", ap, count); - - /* Check for overflow... */ - DABORT(count >= IRNET_MAX_COMMAND, -ENOMEM, - CTRL_ERROR, "Too much data !!!\n"); - - /* Get the data in the driver */ - if(copy_from_user(command, buf, count)) - { - DERROR(CTRL_ERROR, "Invalid user space pointer.\n"); - return -EFAULT; - } - - /* Safe terminate the string */ - command[count] = '\0'; - DEBUG(CTRL_INFO, "Command line received is ``%s'' (%zd).\n", - command, count); - - /* Check every commands in the command line */ - next = command; - while(next != NULL) - { - /* Look at the next command */ - start = next; - - /* Scrap whitespaces before the command */ - start = skip_spaces(start); - - /* ',' is our command separator */ - next = strchr(start, ','); - if(next) - { - *next = '\0'; /* Terminate command */ - length = next - start; /* Length */ - next++; /* Skip the '\0' */ - } - else - length = strlen(start); - - DEBUG(CTRL_INFO, "Found command ``%s'' (%d).\n", start, length); - - /* Check if we recognised one of the known command - * We can't use "switch" with strings, so hack with "continue" */ - - /* First command : name -> Requested IrDA nickname */ - if(!strncmp(start, "name", 4)) - { - /* Copy the name only if is included and not "any" */ - if((length > 5) && (strcmp(start + 5, "any"))) - { - /* Strip out trailing whitespaces */ - while(isspace(start[length - 1])) - length--; - - DABORT(length < 5 || length > NICKNAME_MAX_LEN + 5, - -EINVAL, CTRL_ERROR, "Invalid nickname.\n"); - - /* Copy the name for later reuse */ - memcpy(ap->rname, start + 5, length - 5); - ap->rname[length - 5] = '\0'; - } - else - ap->rname[0] = '\0'; - DEBUG(CTRL_INFO, "Got rname = ``%s''\n", ap->rname); - - /* Restart the loop */ - continue; - } - - /* Second command : addr, daddr -> Requested IrDA destination address - * Also process : saddr -> Requested IrDA source address */ - if((!strncmp(start, "addr", 4)) || - (!strncmp(start, "daddr", 5)) || - (!strncmp(start, "saddr", 5))) - { - __u32 addr = DEV_ADDR_ANY; - - /* Copy the address only if is included and not "any" */ - if((length > 5) && (strcmp(start + 5, "any"))) - { - char * begp = start + 5; - char * endp; - - /* Scrap whitespaces before the command */ - begp = skip_spaces(begp); - - /* Convert argument to a number (last arg is the base) */ - addr = simple_strtoul(begp, &endp, 16); - /* Has it worked ? (endp should be start + length) */ - DABORT(endp <= (start + 5), -EINVAL, - CTRL_ERROR, "Invalid address.\n"); - } - /* Which type of address ? */ - if(start[0] == 's') - { - /* Save it */ - ap->rsaddr = addr; - DEBUG(CTRL_INFO, "Got rsaddr = %08x\n", ap->rsaddr); - } - else - { - /* Save it */ - ap->rdaddr = addr; - DEBUG(CTRL_INFO, "Got rdaddr = %08x\n", ap->rdaddr); - } - - /* Restart the loop */ - continue; - } - - /* Other possible command : connect N (number of retries) */ - - /* No command matched -> Failed... */ - DABORT(1, -EINVAL, CTRL_ERROR, "Not a recognised IrNET command.\n"); - } - - /* Success : we have parsed all commands successfully */ - return count; -} - -#ifdef INITIAL_DISCOVERY -/*------------------------------------------------------------------*/ -/* - * Function irnet_get_discovery_log (self) - * - * Query the content on the discovery log if not done - * - * This function query the current content of the discovery log - * at the startup of the event channel and save it in the internal struct. - */ -static void -irnet_get_discovery_log(irnet_socket * ap) -{ - __u16 mask = irlmp_service_to_hint(S_LAN); - - /* Ask IrLMP for the current discovery log */ - ap->discoveries = irlmp_get_discoveries(&ap->disco_number, mask, - DISCOVERY_DEFAULT_SLOTS); - - /* Check if the we got some results */ - if(ap->discoveries == NULL) - ap->disco_number = -1; - - DEBUG(CTRL_INFO, "Got the log (0x%p), size is %d\n", - ap->discoveries, ap->disco_number); -} - -/*------------------------------------------------------------------*/ -/* - * Function irnet_read_discovery_log (self, event) - * - * Read the content on the discovery log - * - * This function dump the current content of the discovery log - * at the startup of the event channel. - * Return 1 if wrote an event on the control channel... - * - * State of the ap->disco_XXX variables : - * Socket creation : discoveries = NULL ; disco_index = 0 ; disco_number = 0 - * While reading : discoveries = ptr ; disco_index = X ; disco_number = Y - * After reading : discoveries = NULL ; disco_index = Y ; disco_number = -1 - */ -static inline int -irnet_read_discovery_log(irnet_socket *ap, char *event, int buf_size) -{ - int done_event = 0; - - DENTER(CTRL_TRACE, "(ap=0x%p, event=0x%p)\n", - ap, event); - - /* Test if we have some work to do or we have already finished */ - if(ap->disco_number == -1) - { - DEBUG(CTRL_INFO, "Already done\n"); - return 0; - } - - /* Test if it's the first time and therefore we need to get the log */ - if(ap->discoveries == NULL) - irnet_get_discovery_log(ap); - - /* Check if we have more item to dump */ - if(ap->disco_index < ap->disco_number) - { - /* Write an event */ - snprintf(event, buf_size, - "Found %08x (%s) behind %08x {hints %02X-%02X}\n", - ap->discoveries[ap->disco_index].daddr, - ap->discoveries[ap->disco_index].info, - ap->discoveries[ap->disco_index].saddr, - ap->discoveries[ap->disco_index].hints[0], - ap->discoveries[ap->disco_index].hints[1]); - DEBUG(CTRL_INFO, "Writing discovery %d : %s\n", - ap->disco_index, ap->discoveries[ap->disco_index].info); - - /* We have an event */ - done_event = 1; - /* Next discovery */ - ap->disco_index++; - } - - /* Check if we have done the last item */ - if(ap->disco_index >= ap->disco_number) - { - /* No more items : remove the log and signal termination */ - DEBUG(CTRL_INFO, "Cleaning up log (0x%p)\n", - ap->discoveries); - if(ap->discoveries != NULL) - { - /* Cleanup our copy of the discovery log */ - kfree(ap->discoveries); - ap->discoveries = NULL; - } - ap->disco_number = -1; - } - - return done_event; -} -#endif /* INITIAL_DISCOVERY */ - -/*------------------------------------------------------------------*/ -/* - * Read is used to get IrNET events - */ -static inline ssize_t -irnet_ctrl_read(irnet_socket * ap, - struct file * file, - char __user * buf, - size_t count) -{ - DECLARE_WAITQUEUE(wait, current); - char event[75]; - ssize_t ret = 0; - - DENTER(CTRL_TRACE, "(ap=0x%p, count=%zd)\n", ap, count); - -#ifdef INITIAL_DISCOVERY - /* Check if we have read the log */ - if (irnet_read_discovery_log(ap, event, sizeof(event))) - { - count = min(strlen(event), count); - if (copy_to_user(buf, event, count)) - { - DERROR(CTRL_ERROR, "Invalid user space pointer.\n"); - return -EFAULT; - } - - DEXIT(CTRL_TRACE, "\n"); - return count; - } -#endif /* INITIAL_DISCOVERY */ - - /* Put ourselves on the wait queue to be woken up */ - add_wait_queue(&irnet_events.rwait, &wait); - set_current_state(TASK_INTERRUPTIBLE); - for(;;) - { - /* If there is unread events */ - ret = 0; - if(ap->event_index != irnet_events.index) - break; - ret = -EAGAIN; - if(file->f_flags & O_NONBLOCK) - break; - ret = -ERESTARTSYS; - if(signal_pending(current)) - break; - /* Yield and wait to be woken up */ - schedule(); - } - __set_current_state(TASK_RUNNING); - remove_wait_queue(&irnet_events.rwait, &wait); - - /* Did we got it ? */ - if(ret != 0) - { - /* No, return the error code */ - DEXIT(CTRL_TRACE, " - ret %zd\n", ret); - return ret; - } - - /* Which event is it ? */ - switch(irnet_events.log[ap->event_index].event) - { - case IRNET_DISCOVER: - snprintf(event, sizeof(event), - "Discovered %08x (%s) behind %08x {hints %02X-%02X}\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].saddr, - irnet_events.log[ap->event_index].hints.byte[0], - irnet_events.log[ap->event_index].hints.byte[1]); - break; - case IRNET_EXPIRE: - snprintf(event, sizeof(event), - "Expired %08x (%s) behind %08x {hints %02X-%02X}\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].saddr, - irnet_events.log[ap->event_index].hints.byte[0], - irnet_events.log[ap->event_index].hints.byte[1]); - break; - case IRNET_CONNECT_TO: - snprintf(event, sizeof(event), "Connected to %08x (%s) on ppp%d\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].unit); - break; - case IRNET_CONNECT_FROM: - snprintf(event, sizeof(event), "Connection from %08x (%s) on ppp%d\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].unit); - break; - case IRNET_REQUEST_FROM: - snprintf(event, sizeof(event), "Request from %08x (%s) behind %08x\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].saddr); - break; - case IRNET_NOANSWER_FROM: - snprintf(event, sizeof(event), "No-answer from %08x (%s) on ppp%d\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].unit); - break; - case IRNET_BLOCKED_LINK: - snprintf(event, sizeof(event), "Blocked link with %08x (%s) on ppp%d\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].unit); - break; - case IRNET_DISCONNECT_FROM: - snprintf(event, sizeof(event), "Disconnection from %08x (%s) on ppp%d\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name, - irnet_events.log[ap->event_index].unit); - break; - case IRNET_DISCONNECT_TO: - snprintf(event, sizeof(event), "Disconnected to %08x (%s)\n", - irnet_events.log[ap->event_index].daddr, - irnet_events.log[ap->event_index].name); - break; - default: - snprintf(event, sizeof(event), "Bug\n"); - } - /* Increment our event index */ - ap->event_index = (ap->event_index + 1) % IRNET_MAX_EVENTS; - - DEBUG(CTRL_INFO, "Event is :%s", event); - - count = min(strlen(event), count); - if (copy_to_user(buf, event, count)) - { - DERROR(CTRL_ERROR, "Invalid user space pointer.\n"); - return -EFAULT; - } - - DEXIT(CTRL_TRACE, "\n"); - return count; -} - -/*------------------------------------------------------------------*/ -/* - * Poll : called when someone do a select on /dev/irnet. - * Just check if there are new events... - */ -static inline unsigned int -irnet_ctrl_poll(irnet_socket * ap, - struct file * file, - poll_table * wait) -{ - unsigned int mask; - - DENTER(CTRL_TRACE, "(ap=0x%p)\n", ap); - - poll_wait(file, &irnet_events.rwait, wait); - mask = POLLOUT | POLLWRNORM; - /* If there is unread events */ - if(ap->event_index != irnet_events.index) - mask |= POLLIN | POLLRDNORM; -#ifdef INITIAL_DISCOVERY - if(ap->disco_number != -1) - { - /* Test if it's the first time and therefore we need to get the log */ - if(ap->discoveries == NULL) - irnet_get_discovery_log(ap); - /* Recheck */ - if(ap->disco_number != -1) - mask |= POLLIN | POLLRDNORM; - } -#endif /* INITIAL_DISCOVERY */ - - DEXIT(CTRL_TRACE, " - mask=0x%X\n", mask); - return mask; -} - - -/*********************** FILESYSTEM CALLBACKS ***********************/ -/* - * Implement the usual open, read, write functions that will be called - * by the file system when some action is performed on /dev/irnet. - * Most of those actions will in fact be performed by "pppd" or - * the control channel, we just act as a redirector... - */ - -/*------------------------------------------------------------------*/ -/* - * Open : when somebody open /dev/irnet - * We basically create a new instance of irnet and initialise it. - */ -static int -dev_irnet_open(struct inode * inode, - struct file * file) -{ - struct irnet_socket * ap; - int err; - - DENTER(FS_TRACE, "(file=0x%p)\n", file); - -#ifdef SECURE_DEVIRNET - /* This could (should?) be enforced by the permissions on /dev/irnet. */ - if(!capable(CAP_NET_ADMIN)) - return -EPERM; -#endif /* SECURE_DEVIRNET */ - - /* Allocate a private structure for this IrNET instance */ - ap = kzalloc(sizeof(*ap), GFP_KERNEL); - DABORT(ap == NULL, -ENOMEM, FS_ERROR, "Can't allocate struct irnet...\n"); - - /* initialize the irnet structure */ - ap->file = file; - - /* PPP channel setup */ - ap->ppp_open = 0; - ap->chan.private = ap; - ap->chan.ops = &irnet_ppp_ops; - ap->chan.mtu = (2048 - TTP_MAX_HEADER - 2 - PPP_HDRLEN); - ap->chan.hdrlen = 2 + TTP_MAX_HEADER; /* for A/C + Max IrDA hdr */ - /* PPP parameters */ - ap->mru = (2048 - TTP_MAX_HEADER - 2 - PPP_HDRLEN); - ap->xaccm[0] = ~0U; - ap->xaccm[3] = 0x60000000U; - ap->raccm = ~0U; - - /* Setup the IrDA part... */ - err = irda_irnet_create(ap); - if(err) - { - DERROR(FS_ERROR, "Can't setup IrDA link...\n"); - kfree(ap); - - return err; - } - - /* For the control channel */ - ap->event_index = irnet_events.index; /* Cancel all past events */ - - mutex_init(&ap->lock); - - /* Put our stuff where we will be able to find it later */ - file->private_data = ap; - - DEXIT(FS_TRACE, " - ap=0x%p\n", ap); - - return 0; -} - - -/*------------------------------------------------------------------*/ -/* - * Close : when somebody close /dev/irnet - * Destroy the instance of /dev/irnet - */ -static int -dev_irnet_close(struct inode * inode, - struct file * file) -{ - irnet_socket * ap = file->private_data; - - DENTER(FS_TRACE, "(file=0x%p, ap=0x%p)\n", - file, ap); - DABORT(ap == NULL, 0, FS_ERROR, "ap is NULL !!!\n"); - - /* Detach ourselves */ - file->private_data = NULL; - - /* Close IrDA stuff */ - irda_irnet_destroy(ap); - - /* Disconnect from the generic PPP layer if not already done */ - if(ap->ppp_open) - { - DERROR(FS_ERROR, "Channel still registered - deregistering !\n"); - ap->ppp_open = 0; - ppp_unregister_channel(&ap->chan); - } - - kfree(ap); - - DEXIT(FS_TRACE, "\n"); - return 0; -} - -/*------------------------------------------------------------------*/ -/* - * Write does nothing. - * (we receive packet from ppp_generic through ppp_irnet_send()) - */ -static ssize_t -dev_irnet_write(struct file * file, - const char __user *buf, - size_t count, - loff_t * ppos) -{ - irnet_socket * ap = file->private_data; - - DPASS(FS_TRACE, "(file=0x%p, ap=0x%p, count=%zd)\n", - file, ap, count); - DABORT(ap == NULL, -ENXIO, FS_ERROR, "ap is NULL !!!\n"); - - /* If we are connected to ppp_generic, let it handle the job */ - if(ap->ppp_open) - return -EAGAIN; - else - return irnet_ctrl_write(ap, buf, count); -} - -/*------------------------------------------------------------------*/ -/* - * Read doesn't do much either. - * (pppd poll us, but ultimately reads through /dev/ppp) - */ -static ssize_t -dev_irnet_read(struct file * file, - char __user * buf, - size_t count, - loff_t * ppos) -{ - irnet_socket * ap = file->private_data; - - DPASS(FS_TRACE, "(file=0x%p, ap=0x%p, count=%zd)\n", - file, ap, count); - DABORT(ap == NULL, -ENXIO, FS_ERROR, "ap is NULL !!!\n"); - - /* If we are connected to ppp_generic, let it handle the job */ - if(ap->ppp_open) - return -EAGAIN; - else - return irnet_ctrl_read(ap, file, buf, count); -} - -/*------------------------------------------------------------------*/ -/* - * Poll : called when someone do a select on /dev/irnet - */ -static unsigned int -dev_irnet_poll(struct file * file, - poll_table * wait) -{ - irnet_socket * ap = file->private_data; - unsigned int mask; - - DENTER(FS_TRACE, "(file=0x%p, ap=0x%p)\n", - file, ap); - - mask = POLLOUT | POLLWRNORM; - DABORT(ap == NULL, mask, FS_ERROR, "ap is NULL !!!\n"); - - /* If we are connected to ppp_generic, let it handle the job */ - if(!ap->ppp_open) - mask |= irnet_ctrl_poll(ap, file, wait); - - DEXIT(FS_TRACE, " - mask=0x%X\n", mask); - return mask; -} - -/*------------------------------------------------------------------*/ -/* - * IOCtl : Called when someone does some ioctls on /dev/irnet - * This is the way pppd configure us and control us while the PPP - * instance is active. - */ -static long -dev_irnet_ioctl( - struct file * file, - unsigned int cmd, - unsigned long arg) -{ - irnet_socket * ap = file->private_data; - int err; - int val; - void __user *argp = (void __user *)arg; - - DENTER(FS_TRACE, "(file=0x%p, ap=0x%p, cmd=0x%X)\n", - file, ap, cmd); - - /* Basic checks... */ - DASSERT(ap != NULL, -ENXIO, PPP_ERROR, "ap is NULL...\n"); -#ifdef SECURE_DEVIRNET - if(!capable(CAP_NET_ADMIN)) - return -EPERM; -#endif /* SECURE_DEVIRNET */ - - err = -EFAULT; - switch(cmd) - { - /* Set discipline (should be N_SYNC_PPP or N_TTY) */ - case TIOCSETD: - if(get_user(val, (int __user *)argp)) - break; - if((val == N_SYNC_PPP) || (val == N_PPP)) - { - DEBUG(FS_INFO, "Entering PPP discipline.\n"); - /* PPP channel setup (ap->chan in configured in dev_irnet_open())*/ - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - - err = ppp_register_channel(&ap->chan); - if(err == 0) - { - /* Our ppp side is active */ - ap->ppp_open = 1; - - DEBUG(FS_INFO, "Trying to establish a connection.\n"); - /* Setup the IrDA link now - may fail... */ - irda_irnet_connect(ap); - } - else - DERROR(FS_ERROR, "Can't setup PPP channel...\n"); - - mutex_unlock(&ap->lock); - } - else - { - /* In theory, should be N_TTY */ - DEBUG(FS_INFO, "Exiting PPP discipline.\n"); - /* Disconnect from the generic PPP layer */ - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - - if(ap->ppp_open) - { - ap->ppp_open = 0; - ppp_unregister_channel(&ap->chan); - } - else - DERROR(FS_ERROR, "Channel not registered !\n"); - err = 0; - - mutex_unlock(&ap->lock); - } - break; - - /* Query PPP channel and unit number */ - case PPPIOCGCHAN: - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - - if(ap->ppp_open && !put_user(ppp_channel_index(&ap->chan), - (int __user *)argp)) - err = 0; - - mutex_unlock(&ap->lock); - break; - case PPPIOCGUNIT: - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - - if(ap->ppp_open && !put_user(ppp_unit_number(&ap->chan), - (int __user *)argp)) - err = 0; - - mutex_unlock(&ap->lock); - break; - - /* All these ioctls can be passed both directly and from ppp_generic, - * so we just deal with them in one place... - */ - case PPPIOCGFLAGS: - case PPPIOCSFLAGS: - case PPPIOCGASYNCMAP: - case PPPIOCSASYNCMAP: - case PPPIOCGRASYNCMAP: - case PPPIOCSRASYNCMAP: - case PPPIOCGXASYNCMAP: - case PPPIOCSXASYNCMAP: - case PPPIOCGMRU: - case PPPIOCSMRU: - DEBUG(FS_INFO, "Standard PPP ioctl.\n"); - if(!capable(CAP_NET_ADMIN)) - err = -EPERM; - else { - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - - err = ppp_irnet_ioctl(&ap->chan, cmd, arg); - - mutex_unlock(&ap->lock); - } - break; - - /* TTY IOCTLs : Pretend that we are a tty, to keep pppd happy */ - /* Get termios */ - case TCGETS: - DEBUG(FS_INFO, "Get termios.\n"); - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - -#ifndef TCGETS2 - if(!kernel_termios_to_user_termios((struct termios __user *)argp, &ap->termios)) - err = 0; -#else - if(kernel_termios_to_user_termios_1((struct termios __user *)argp, &ap->termios)) - err = 0; -#endif - - mutex_unlock(&ap->lock); - break; - /* Set termios */ - case TCSETSF: - DEBUG(FS_INFO, "Set termios.\n"); - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - -#ifndef TCGETS2 - if(!user_termios_to_kernel_termios(&ap->termios, (struct termios __user *)argp)) - err = 0; -#else - if(!user_termios_to_kernel_termios_1(&ap->termios, (struct termios __user *)argp)) - err = 0; -#endif - - mutex_unlock(&ap->lock); - break; - - /* Set DTR/RTS */ - case TIOCMBIS: - case TIOCMBIC: - /* Set exclusive/non-exclusive mode */ - case TIOCEXCL: - case TIOCNXCL: - DEBUG(FS_INFO, "TTY compatibility.\n"); - err = 0; - break; - - case TCGETA: - DEBUG(FS_INFO, "TCGETA\n"); - break; - - case TCFLSH: - DEBUG(FS_INFO, "TCFLSH\n"); - /* Note : this will flush buffers in PPP, so it *must* be done - * We should also worry that we don't accept junk here and that - * we get rid of our own buffers */ -#ifdef FLUSH_TO_PPP - if (mutex_lock_interruptible(&ap->lock)) - return -EINTR; - ppp_output_wakeup(&ap->chan); - mutex_unlock(&ap->lock); -#endif /* FLUSH_TO_PPP */ - err = 0; - break; - - case FIONREAD: - DEBUG(FS_INFO, "FIONREAD\n"); - val = 0; - if(put_user(val, (int __user *)argp)) - break; - err = 0; - break; - - default: - DERROR(FS_ERROR, "Unsupported ioctl (0x%X)\n", cmd); - err = -ENOTTY; - } - - DEXIT(FS_TRACE, " - err = 0x%X\n", err); - return err; -} - -/************************** PPP CALLBACKS **************************/ -/* - * This are the functions that the generic PPP driver in the kernel - * will call to communicate to us. - */ - -/*------------------------------------------------------------------*/ -/* - * Prepare the ppp frame for transmission over the IrDA socket. - * We make sure that the header space is enough, and we change ppp header - * according to flags passed by pppd. - * This is not a callback, but just a helper function used in ppp_irnet_send() - */ -static inline struct sk_buff * -irnet_prepare_skb(irnet_socket * ap, - struct sk_buff * skb) -{ - unsigned char * data; - int proto; /* PPP protocol */ - int islcp; /* Protocol == LCP */ - int needaddr; /* Need PPP address */ - - DENTER(PPP_TRACE, "(ap=0x%p, skb=0x%p)\n", - ap, skb); - - /* Extract PPP protocol from the frame */ - data = skb->data; - proto = (data[0] << 8) + data[1]; - - /* LCP packets with codes between 1 (configure-request) - * and 7 (code-reject) must be sent as though no options - * have been negotiated. */ - islcp = (proto == PPP_LCP) && (1 <= data[2]) && (data[2] <= 7); - - /* compress protocol field if option enabled */ - if((data[0] == 0) && (ap->flags & SC_COMP_PROT) && (!islcp)) - skb_pull(skb,1); - - /* Check if we need address/control fields */ - needaddr = 2*((ap->flags & SC_COMP_AC) == 0 || islcp); - - /* Is the skb headroom large enough to contain all IrDA-headers? */ - if((skb_headroom(skb) < (ap->max_header_size + needaddr)) || - (skb_shared(skb))) - { - struct sk_buff * new_skb; - - DEBUG(PPP_INFO, "Reallocating skb\n"); - - /* Create a new skb */ - new_skb = skb_realloc_headroom(skb, ap->max_header_size + needaddr); - - /* We have to free the original skb anyway */ - dev_kfree_skb(skb); - - /* Did the realloc succeed ? */ - DABORT(new_skb == NULL, NULL, PPP_ERROR, "Could not realloc skb\n"); - - /* Use the new skb instead */ - skb = new_skb; - } - - /* prepend address/control fields if necessary */ - if(needaddr) - { - skb_push(skb, 2); - skb->data[0] = PPP_ALLSTATIONS; - skb->data[1] = PPP_UI; - } - - DEXIT(PPP_TRACE, "\n"); - - return skb; -} - -/*------------------------------------------------------------------*/ -/* - * Send a packet to the peer over the IrTTP connection. - * Returns 1 iff the packet was accepted. - * Returns 0 iff packet was not consumed. - * If the packet was not accepted, we will call ppp_output_wakeup - * at some later time to reactivate flow control in ppp_generic. - */ -static int -ppp_irnet_send(struct ppp_channel * chan, - struct sk_buff * skb) -{ - irnet_socket * self = (struct irnet_socket *) chan->private; - int ret; - - DENTER(PPP_TRACE, "(channel=0x%p, ap/self=0x%p)\n", - chan, self); - - /* Check if things are somewhat valid... */ - DASSERT(self != NULL, 0, PPP_ERROR, "Self is NULL !!!\n"); - - /* Check if we are connected */ - if(!(test_bit(0, &self->ttp_open))) - { -#ifdef CONNECT_IN_SEND - /* Let's try to connect one more time... */ - /* Note : we won't be connected after this call, but we should be - * ready for next packet... */ - /* If we are already connecting, this will fail */ - irda_irnet_connect(self); -#endif /* CONNECT_IN_SEND */ - - DEBUG(PPP_INFO, "IrTTP not ready ! (%ld-%ld)\n", - self->ttp_open, self->ttp_connect); - - /* Note : we can either drop the packet or block the packet. - * - * Blocking the packet allow us a better connection time, - * because by calling ppp_output_wakeup() we can have - * ppp_generic resending the LCP request immediately to us, - * rather than waiting for one of pppd periodic transmission of - * LCP request. - * - * On the other hand, if we block all packet, all those periodic - * transmissions of pppd accumulate in ppp_generic, creating a - * backlog of LCP request. When we eventually connect later on, - * we have to transmit all this backlog before we can connect - * proper (if we don't timeout before). - * - * The current strategy is as follow : - * While we are attempting to connect, we block packets to get - * a better connection time. - * If we fail to connect, we drain the queue and start dropping packets - */ -#ifdef BLOCK_WHEN_CONNECT - /* If we are attempting to connect */ - if(test_bit(0, &self->ttp_connect)) - { - /* Blocking packet, ppp_generic will retry later */ - return 0; - } -#endif /* BLOCK_WHEN_CONNECT */ - - /* Dropping packet, pppd will retry later */ - dev_kfree_skb(skb); - return 1; - } - - /* Check if the queue can accept any packet, otherwise block */ - if(self->tx_flow != FLOW_START) - DRETURN(0, PPP_INFO, "IrTTP queue full (%d skbs)...\n", - skb_queue_len(&self->tsap->tx_queue)); - - /* Prepare ppp frame for transmission */ - skb = irnet_prepare_skb(self, skb); - DABORT(skb == NULL, 1, PPP_ERROR, "Prepare skb for Tx failed.\n"); - - /* Send the packet to IrTTP */ - ret = irttp_data_request(self->tsap, skb); - if(ret < 0) - { - /* - * > IrTTPs tx queue is full, so we just have to - * > drop the frame! You might think that we should - * > just return -1 and don't deallocate the frame, - * > but that is dangerous since it's possible that - * > we have replaced the original skb with a new - * > one with larger headroom, and that would really - * > confuse do_dev_queue_xmit() in dev.c! I have - * > tried :-) DB - * Correction : we verify the flow control above (self->tx_flow), - * so we come here only if IrTTP doesn't like the packet (empty, - * too large, IrTTP not connected). In those rare cases, it's ok - * to drop it, we don't want to see it here again... - * Jean II - */ - DERROR(PPP_ERROR, "IrTTP doesn't like this packet !!! (0x%X)\n", ret); - /* irttp_data_request already free the packet */ - } - - DEXIT(PPP_TRACE, "\n"); - return 1; /* Packet has been consumed */ -} - -/*------------------------------------------------------------------*/ -/* - * Take care of the ioctls that ppp_generic doesn't want to deal with... - * Note : we are also called from dev_irnet_ioctl(). - */ -static int -ppp_irnet_ioctl(struct ppp_channel * chan, - unsigned int cmd, - unsigned long arg) -{ - irnet_socket * ap = (struct irnet_socket *) chan->private; - int err; - int val; - u32 accm[8]; - void __user *argp = (void __user *)arg; - - DENTER(PPP_TRACE, "(channel=0x%p, ap=0x%p, cmd=0x%X)\n", - chan, ap, cmd); - - /* Basic checks... */ - DASSERT(ap != NULL, -ENXIO, PPP_ERROR, "ap is NULL...\n"); - - err = -EFAULT; - switch(cmd) - { - /* PPP flags */ - case PPPIOCGFLAGS: - val = ap->flags | ap->rbits; - if(put_user(val, (int __user *) argp)) - break; - err = 0; - break; - case PPPIOCSFLAGS: - if(get_user(val, (int __user *) argp)) - break; - ap->flags = val & ~SC_RCV_BITS; - ap->rbits = val & SC_RCV_BITS; - err = 0; - break; - - /* Async map stuff - all dummy to please pppd */ - case PPPIOCGASYNCMAP: - if(put_user(ap->xaccm[0], (u32 __user *) argp)) - break; - err = 0; - break; - case PPPIOCSASYNCMAP: - if(get_user(ap->xaccm[0], (u32 __user *) argp)) - break; - err = 0; - break; - case PPPIOCGRASYNCMAP: - if(put_user(ap->raccm, (u32 __user *) argp)) - break; - err = 0; - break; - case PPPIOCSRASYNCMAP: - if(get_user(ap->raccm, (u32 __user *) argp)) - break; - err = 0; - break; - case PPPIOCGXASYNCMAP: - if(copy_to_user(argp, ap->xaccm, sizeof(ap->xaccm))) - break; - err = 0; - break; - case PPPIOCSXASYNCMAP: - if(copy_from_user(accm, argp, sizeof(accm))) - break; - accm[2] &= ~0x40000000U; /* can't escape 0x5e */ - accm[3] |= 0x60000000U; /* must escape 0x7d, 0x7e */ - memcpy(ap->xaccm, accm, sizeof(ap->xaccm)); - err = 0; - break; - - /* Max PPP frame size */ - case PPPIOCGMRU: - if(put_user(ap->mru, (int __user *) argp)) - break; - err = 0; - break; - case PPPIOCSMRU: - if(get_user(val, (int __user *) argp)) - break; - if(val < PPP_MRU) - val = PPP_MRU; - ap->mru = val; - err = 0; - break; - - default: - DEBUG(PPP_INFO, "Unsupported ioctl (0x%X)\n", cmd); - err = -ENOIOCTLCMD; - } - - DEXIT(PPP_TRACE, " - err = 0x%X\n", err); - return err; -} - -/************************** INITIALISATION **************************/ -/* - * Module initialisation and all that jazz... - */ - -/*------------------------------------------------------------------*/ -/* - * Hook our device callbacks in the filesystem, to connect our code - * to /dev/irnet - */ -static inline int __init -ppp_irnet_init(void) -{ - int err = 0; - - DENTER(MODULE_TRACE, "()\n"); - - /* Allocate ourselves as a minor in the misc range */ - err = misc_register(&irnet_misc_device); - - DEXIT(MODULE_TRACE, "\n"); - return err; -} - -/*------------------------------------------------------------------*/ -/* - * Cleanup at exit... - */ -static inline void __exit -ppp_irnet_cleanup(void) -{ - DENTER(MODULE_TRACE, "()\n"); - - /* De-allocate /dev/irnet minor in misc range */ - misc_deregister(&irnet_misc_device); - - DEXIT(MODULE_TRACE, "\n"); -} - -/*------------------------------------------------------------------*/ -/* - * Module main entry point - */ -static int __init -irnet_init(void) -{ - int err; - - /* Initialise both parts... */ - err = irda_irnet_init(); - if(!err) - err = ppp_irnet_init(); - return err; -} - -/*------------------------------------------------------------------*/ -/* - * Module exit - */ -static void __exit -irnet_cleanup(void) -{ - irda_irnet_cleanup(); - ppp_irnet_cleanup(); -} - -/*------------------------------------------------------------------*/ -/* - * Module magic - */ -module_init(irnet_init); -module_exit(irnet_cleanup); -MODULE_AUTHOR("Jean Tourrilhes <jt@hpl.hp.com>"); -MODULE_DESCRIPTION("IrNET : Synchronous PPP over IrDA"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_CHARDEV(10, 187); diff --git a/net/irda/irnet/irnet_ppp.h b/net/irda/irnet/irnet_ppp.h deleted file mode 100644 index 32061442cc8e..000000000000 --- a/net/irda/irnet/irnet_ppp.h +++ /dev/null @@ -1,116 +0,0 @@ -/* - * IrNET protocol module : Synchronous PPP over an IrDA socket. - * - * Jean II - HPL `00 - <jt@hpl.hp.com> - * - * This file contains all definitions and declarations necessary for the - * PPP part of the IrNET module. - * This file is a private header, so other modules don't want to know - * what's in there... - */ - -#ifndef IRNET_PPP_H -#define IRNET_PPP_H - -/***************************** INCLUDES *****************************/ - -#include "irnet.h" /* Module global include */ -#include <linux/miscdevice.h> - -/************************ CONSTANTS & MACROS ************************/ - -/* IrNET control channel stuff */ -#define IRNET_MAX_COMMAND 256 /* Max length of a command line */ - -/* PPP hardcore stuff */ - -/* Bits in rbits (PPP flags in irnet struct) */ -#define SC_RCV_BITS (SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP) - -/* Bit numbers in busy */ -#define XMIT_BUSY 0 -#define RECV_BUSY 1 -#define XMIT_WAKEUP 2 -#define XMIT_FULL 3 - -/* Queue management */ -#define PPPSYNC_MAX_RQLEN 32 /* arbitrary */ - -/****************************** TYPES ******************************/ - - -/**************************** PROTOTYPES ****************************/ - -/* ----------------------- CONTROL CHANNEL ----------------------- */ -static inline ssize_t - irnet_ctrl_write(irnet_socket *, - const char *, - size_t); -static inline ssize_t - irnet_ctrl_read(irnet_socket *, - struct file *, - char *, - size_t); -static inline unsigned int - irnet_ctrl_poll(irnet_socket *, - struct file *, - poll_table *); -/* ----------------------- CHARACTER DEVICE ----------------------- */ -static int - dev_irnet_open(struct inode *, /* fs callback : open */ - struct file *), - dev_irnet_close(struct inode *, - struct file *); -static ssize_t - dev_irnet_write(struct file *, - const char __user *, - size_t, - loff_t *), - dev_irnet_read(struct file *, - char __user *, - size_t, - loff_t *); -static unsigned int - dev_irnet_poll(struct file *, - poll_table *); -static long - dev_irnet_ioctl(struct file *, - unsigned int, - unsigned long); -/* ------------------------ PPP INTERFACE ------------------------ */ -static inline struct sk_buff * - irnet_prepare_skb(irnet_socket *, - struct sk_buff *); -static int - ppp_irnet_send(struct ppp_channel *, - struct sk_buff *); -static int - ppp_irnet_ioctl(struct ppp_channel *, - unsigned int, - unsigned long); - -/**************************** VARIABLES ****************************/ - -/* Filesystem callbacks (to call us) */ -static const struct file_operations irnet_device_fops = -{ - .owner = THIS_MODULE, - .read = dev_irnet_read, - .write = dev_irnet_write, - .poll = dev_irnet_poll, - .unlocked_ioctl = dev_irnet_ioctl, - .open = dev_irnet_open, - .release = dev_irnet_close, - .llseek = noop_llseek, - /* Also : llseek, readdir, mmap, flush, fsync, fasync, lock, readv, writev */ -}; - -/* Structure so that the misc major (drivers/char/misc.c) take care of us... */ -static struct miscdevice irnet_misc_device = -{ - .minor = IRNET_MINOR, - .name = "irnet", - .fops = &irnet_device_fops -}; - -#endif /* IRNET_PPP_H */ diff --git a/net/irda/irnetlink.c b/net/irda/irnetlink.c deleted file mode 100644 index 7fc340e574cf..000000000000 --- a/net/irda/irnetlink.c +++ /dev/null @@ -1,162 +0,0 @@ -/* - * IrDA netlink layer, for stack configuration. - * - * Copyright (c) 2007 Samuel Ortiz <samuel@sortiz.org> - * - * Partly based on the 802.11 nelink implementation - * (see net/wireless/nl80211.c) which is: - * Copyright 2006 Johannes Berg <johannes@sipsolutions.net> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - */ - -#include <linux/socket.h> -#include <linux/irda.h> -#include <linux/gfp.h> -#include <net/net_namespace.h> -#include <net/sock.h> -#include <net/irda/irda.h> -#include <net/irda/irlap.h> -#include <net/genetlink.h> - - - -static struct genl_family irda_nl_family; - -static struct net_device * ifname_to_netdev(struct net *net, struct genl_info *info) -{ - char * ifname; - - if (!info->attrs[IRDA_NL_ATTR_IFNAME]) - return NULL; - - ifname = nla_data(info->attrs[IRDA_NL_ATTR_IFNAME]); - - pr_debug("%s(): Looking for %s\n", __func__, ifname); - - return dev_get_by_name(net, ifname); -} - -static int irda_nl_set_mode(struct sk_buff *skb, struct genl_info *info) -{ - struct net_device * dev; - struct irlap_cb * irlap; - u32 mode; - - if (!info->attrs[IRDA_NL_ATTR_MODE]) - return -EINVAL; - - mode = nla_get_u32(info->attrs[IRDA_NL_ATTR_MODE]); - - pr_debug("%s(): Switching to mode: %d\n", __func__, mode); - - dev = ifname_to_netdev(&init_net, info); - if (!dev) - return -ENODEV; - - irlap = (struct irlap_cb *)dev->atalk_ptr; - if (!irlap) { - dev_put(dev); - return -ENODEV; - } - - irlap->mode = mode; - - dev_put(dev); - - return 0; -} - -static int irda_nl_get_mode(struct sk_buff *skb, struct genl_info *info) -{ - struct net_device * dev; - struct irlap_cb * irlap; - struct sk_buff *msg; - void *hdr; - int ret = -ENOBUFS; - - dev = ifname_to_netdev(&init_net, info); - if (!dev) - return -ENODEV; - - msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!msg) { - dev_put(dev); - return -ENOMEM; - } - - irlap = (struct irlap_cb *)dev->atalk_ptr; - if (!irlap) { - ret = -ENODEV; - goto err_out; - } - - hdr = genlmsg_put(msg, info->snd_portid, info->snd_seq, - &irda_nl_family, 0, IRDA_NL_CMD_GET_MODE); - if (hdr == NULL) { - ret = -EMSGSIZE; - goto err_out; - } - - if(nla_put_string(msg, IRDA_NL_ATTR_IFNAME, - dev->name)) - goto err_out; - - if(nla_put_u32(msg, IRDA_NL_ATTR_MODE, irlap->mode)) - goto err_out; - - genlmsg_end(msg, hdr); - - return genlmsg_reply(msg, info); - - err_out: - nlmsg_free(msg); - dev_put(dev); - - return ret; -} - -static const struct nla_policy irda_nl_policy[IRDA_NL_ATTR_MAX + 1] = { - [IRDA_NL_ATTR_IFNAME] = { .type = NLA_NUL_STRING, - .len = IFNAMSIZ-1 }, - [IRDA_NL_ATTR_MODE] = { .type = NLA_U32 }, -}; - -static const struct genl_ops irda_nl_ops[] = { - { - .cmd = IRDA_NL_CMD_SET_MODE, - .doit = irda_nl_set_mode, - .policy = irda_nl_policy, - .flags = GENL_ADMIN_PERM, - }, - { - .cmd = IRDA_NL_CMD_GET_MODE, - .doit = irda_nl_get_mode, - .policy = irda_nl_policy, - /* can be retrieved by unprivileged users */ - }, - -}; - -static struct genl_family irda_nl_family __ro_after_init = { - .name = IRDA_NL_NAME, - .hdrsize = 0, - .version = IRDA_NL_VERSION, - .maxattr = IRDA_NL_CMD_MAX, - .module = THIS_MODULE, - .ops = irda_nl_ops, - .n_ops = ARRAY_SIZE(irda_nl_ops), -}; - -int __init irda_nl_register(void) -{ - return genl_register_family(&irda_nl_family); -} - -void irda_nl_unregister(void) -{ - genl_unregister_family(&irda_nl_family); -} diff --git a/net/irda/irproc.c b/net/irda/irproc.c deleted file mode 100644 index 77cfdde9d82f..000000000000 --- a/net/irda/irproc.c +++ /dev/null @@ -1,96 +0,0 @@ -/********************************************************************* - * - * Filename: irproc.c - * Version: 1.0 - * Description: Various entries in the /proc file system - * Status: Experimental. - * Author: Thomas Davis, <ratbert@radiks.net> - * Created at: Sat Feb 21 21:33:24 1998 - * Modified at: Sun Nov 14 08:54:54 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-1999, Dag Brattli <dagb@cs.uit.no> - * Copyright (c) 1998, Thomas Davis, <ratbert@radiks.net>, - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * I, Thomas Davis, provide no warranty for any of this software. - * This material is provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/module.h> -#include <linux/init.h> -#include <net/net_namespace.h> - -#include <net/irda/irda.h> -#include <net/irda/irlap.h> -#include <net/irda/irlmp.h> - -extern const struct file_operations discovery_seq_fops; -extern const struct file_operations irlap_seq_fops; -extern const struct file_operations irlmp_seq_fops; -extern const struct file_operations irttp_seq_fops; -extern const struct file_operations irias_seq_fops; - -struct irda_entry { - const char *name; - const struct file_operations *fops; -}; - -struct proc_dir_entry *proc_irda; -EXPORT_SYMBOL(proc_irda); - -static const struct irda_entry irda_dirs[] = { - {"discovery", &discovery_seq_fops}, - {"irttp", &irttp_seq_fops}, - {"irlmp", &irlmp_seq_fops}, - {"irlap", &irlap_seq_fops}, - {"irias", &irias_seq_fops}, -}; - -/* - * Function irda_proc_register (void) - * - * Register irda entry in /proc file system - * - */ -void __init irda_proc_register(void) -{ - int i; - - proc_irda = proc_mkdir("irda", init_net.proc_net); - if (proc_irda == NULL) - return; - - for (i = 0; i < ARRAY_SIZE(irda_dirs); i++) - (void) proc_create(irda_dirs[i].name, 0, proc_irda, - irda_dirs[i].fops); -} - -/* - * Function irda_proc_unregister (void) - * - * Unregister irda entry in /proc file system - * - */ -void irda_proc_unregister(void) -{ - int i; - - if (proc_irda) { - for (i=0; i<ARRAY_SIZE(irda_dirs); i++) - remove_proc_entry(irda_dirs[i].name, proc_irda); - - remove_proc_entry("irda", init_net.proc_net); - proc_irda = NULL; - } -} - - diff --git a/net/irda/irqueue.c b/net/irda/irqueue.c deleted file mode 100644 index 160dc89335e2..000000000000 --- a/net/irda/irqueue.c +++ /dev/null @@ -1,911 +0,0 @@ -/********************************************************************* - * - * Filename: irqueue.c - * Version: 0.3 - * Description: General queue implementation - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Jun 9 13:29:31 1998 - * Modified at: Sun Dec 12 13:48:22 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Modified at: Thu Jan 4 14:29:10 CET 2001 - * Modified by: Marc Zyngier <mzyngier@freesurf.fr> - * - * Copyright (C) 1998-1999, Aage Kvalnes <aage@cs.uit.no> - * Copyright (C) 1998, Dag Brattli, - * All Rights Reserved. - * - * This code is taken from the Vortex Operating System written by Aage - * Kvalnes. Aage has agreed that this code can use the GPL licence, - * although he does not use that licence in his own code. - * - * This copyright does however _not_ include the ELF hash() function - * which I currently don't know which licence or copyright it - * has. Please inform me if you know. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -/* - * NOTE : - * There are various problems with this package : - * o the hash function for ints is pathetic (but could be changed) - * o locking is sometime suspicious (especially during enumeration) - * o most users have only a few elements (== overhead) - * o most users never use search, so don't benefit from hashing - * Problem already fixed : - * o not 64 bit compliant (most users do hashv = (int) self) - * o hashbin_remove() is broken => use hashbin_remove_this() - * I think most users would be better served by a simple linked list - * (like include/linux/list.h) with a global spinlock per list. - * Jean II - */ - -/* - * Notes on the concurrent access to hashbin and other SMP issues - * ------------------------------------------------------------- - * Hashbins are very often in the IrDA stack a global repository of - * information, and therefore used in a very asynchronous manner following - * various events (driver calls, timers, user calls...). - * Therefore, very often it is highly important to consider the - * management of concurrent access to the hashbin and how to guarantee the - * consistency of the operations on it. - * - * First, we need to define the objective of locking : - * 1) Protect user data (content pointed by the hashbin) - * 2) Protect hashbin structure itself (linked list in each bin) - * - * OLD LOCKING - * ----------- - * - * The previous locking strategy, either HB_LOCAL or HB_GLOBAL were - * both inadequate in *both* aspect. - * o HB_GLOBAL was using a spinlock for each bin (local locking). - * o HB_LOCAL was disabling irq on *all* CPUs, so use a single - * global semaphore. - * The problems were : - * A) Global irq disabling is no longer supported by the kernel - * B) No protection for the hashbin struct global data - * o hashbin_delete() - * o hb_current - * C) No protection for user data in some cases - * - * A) HB_LOCAL use global irq disabling, so doesn't work on kernel - * 2.5.X. Even when it is supported (kernel 2.4.X and earlier), its - * performance is not satisfactory on SMP setups. Most hashbins were - * HB_LOCAL, so (A) definitely need fixing. - * B) HB_LOCAL could be modified to fix (B). However, because HB_GLOBAL - * lock only the individual bins, it will never be able to lock the - * global data, so can't do (B). - * C) Some functions return pointer to data that is still in the - * hashbin : - * o hashbin_find() - * o hashbin_get_first() - * o hashbin_get_next() - * As the data is still in the hashbin, it may be changed or free'd - * while the caller is examinimg the data. In those case, locking can't - * be done within the hashbin, but must include use of the data within - * the caller. - * The caller can easily do this with HB_LOCAL (just disable irqs). - * However, this is impossible with HB_GLOBAL because the caller has no - * way to know the proper bin, so don't know which spinlock to use. - * - * Quick summary : can no longer use HB_LOCAL, and HB_GLOBAL is - * fundamentally broken and will never work. - * - * NEW LOCKING - * ----------- - * - * To fix those problems, I've introduce a few changes in the - * hashbin locking : - * 1) New HB_LOCK scheme - * 2) hashbin->hb_spinlock - * 3) New hashbin usage policy - * - * HB_LOCK : - * ------- - * HB_LOCK is a locking scheme intermediate between the old HB_LOCAL - * and HB_GLOBAL. It uses a single spinlock to protect the whole content - * of the hashbin. As it is a single spinlock, it can protect the global - * data of the hashbin and not only the bins themselves. - * HB_LOCK can only protect some of the hashbin calls, so it only lock - * call that can be made 100% safe and leave other call unprotected. - * HB_LOCK in theory is slower than HB_GLOBAL, but as the hashbin - * content is always small contention is not high, so it doesn't matter - * much. HB_LOCK is probably faster than HB_LOCAL. - * - * hashbin->hb_spinlock : - * -------------------- - * The spinlock that HB_LOCK uses is available for caller, so that - * the caller can protect unprotected calls (see below). - * If the caller want to do entirely its own locking (HB_NOLOCK), he - * can do so and may use safely this spinlock. - * Locking is done like this : - * spin_lock_irqsave(&hashbin->hb_spinlock, flags); - * Releasing the lock : - * spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - * - * Safe & Protected calls : - * ---------------------- - * The following calls are safe or protected via HB_LOCK : - * o hashbin_new() -> safe - * o hashbin_delete() - * o hashbin_insert() - * o hashbin_remove_first() - * o hashbin_remove() - * o hashbin_remove_this() - * o HASHBIN_GET_SIZE() -> atomic - * - * The following calls only protect the hashbin itself : - * o hashbin_lock_find() - * o hashbin_find_next() - * - * Unprotected calls : - * ----------------- - * The following calls need to be protected by the caller : - * o hashbin_find() - * o hashbin_get_first() - * o hashbin_get_next() - * - * Locking Policy : - * -------------- - * If the hashbin is used only in a single thread of execution - * (explicitly or implicitely), you can use HB_NOLOCK - * If the calling module already provide concurrent access protection, - * you may use HB_NOLOCK. - * - * In all other cases, you need to use HB_LOCK and lock the hashbin - * every time before calling one of the unprotected calls. You also must - * use the pointer returned by the unprotected call within the locked - * region. - * - * Extra care for enumeration : - * -------------------------- - * hashbin_get_first() and hashbin_get_next() use the hashbin to - * store the current position, in hb_current. - * As long as the hashbin remains locked, this is safe. If you unlock - * the hashbin, the current position may change if anybody else modify - * or enumerate the hashbin. - * Summary : do the full enumeration while locked. - * - * Alternatively, you may use hashbin_find_next(). But, this will - * be slower, is more complex to use and doesn't protect the hashbin - * content. So, care is needed here as well. - * - * Other issues : - * ------------ - * I believe that we are overdoing it by using spin_lock_irqsave() - * and we should use only spin_lock_bh() or similar. But, I don't have - * the balls to try it out. - * Don't believe that because hashbin are now (somewhat) SMP safe - * that the rest of the code is. Higher layers tend to be safest, - * but LAP and LMP would need some serious dedicated love. - * - * Jean II - */ -#include <linux/module.h> -#include <linux/slab.h> - -#include <net/irda/irda.h> -#include <net/irda/irqueue.h> - -/************************ QUEUE SUBROUTINES ************************/ - -/* - * Hashbin - */ -#define GET_HASHBIN(x) ( x & HASHBIN_MASK ) - -/* - * Function hash (name) - * - * This function hash the input string 'name' using the ELF hash - * function for strings. - */ -static __u32 hash( const char* name) -{ - __u32 h = 0; - __u32 g; - - while(*name) { - h = (h<<4) + *name++; - if ((g = (h & 0xf0000000))) - h ^=g>>24; - h &=~g; - } - return h; -} - -/* - * Function enqueue_first (queue, proc) - * - * Insert item first in queue. - * - */ -static void enqueue_first(irda_queue_t **queue, irda_queue_t* element) -{ - - /* - * Check if queue is empty. - */ - if ( *queue == NULL ) { - /* - * Queue is empty. Insert one element into the queue. - */ - element->q_next = element->q_prev = *queue = element; - - } else { - /* - * Queue is not empty. Insert element into front of queue. - */ - element->q_next = (*queue); - (*queue)->q_prev->q_next = element; - element->q_prev = (*queue)->q_prev; - (*queue)->q_prev = element; - (*queue) = element; - } -} - - -/* - * Function dequeue (queue) - * - * Remove first entry in queue - * - */ -static irda_queue_t *dequeue_first(irda_queue_t **queue) -{ - irda_queue_t *ret; - - pr_debug("dequeue_first()\n"); - - /* - * Set return value - */ - ret = *queue; - - if ( *queue == NULL ) { - /* - * Queue was empty. - */ - } else if ( (*queue)->q_next == *queue ) { - /* - * Queue only contained a single element. It will now be - * empty. - */ - *queue = NULL; - } else { - /* - * Queue contained several element. Remove the first one. - */ - (*queue)->q_prev->q_next = (*queue)->q_next; - (*queue)->q_next->q_prev = (*queue)->q_prev; - *queue = (*queue)->q_next; - } - - /* - * Return the removed entry (or NULL of queue was empty). - */ - return ret; -} - -/* - * Function dequeue_general (queue, element) - * - * - */ -static irda_queue_t *dequeue_general(irda_queue_t **queue, irda_queue_t* element) -{ - irda_queue_t *ret; - - pr_debug("dequeue_general()\n"); - - /* - * Set return value - */ - ret = *queue; - - if ( *queue == NULL ) { - /* - * Queue was empty. - */ - } else if ( (*queue)->q_next == *queue ) { - /* - * Queue only contained a single element. It will now be - * empty. - */ - *queue = NULL; - - } else { - /* - * Remove specific element. - */ - element->q_prev->q_next = element->q_next; - element->q_next->q_prev = element->q_prev; - if ( (*queue) == element) - (*queue) = element->q_next; - } - - /* - * Return the removed entry (or NULL of queue was empty). - */ - return ret; -} - -/************************ HASHBIN MANAGEMENT ************************/ - -/* - * Function hashbin_create ( type, name ) - * - * Create hashbin! - * - */ -hashbin_t *hashbin_new(int type) -{ - hashbin_t* hashbin; - - /* - * Allocate new hashbin - */ - hashbin = kzalloc(sizeof(*hashbin), GFP_ATOMIC); - if (!hashbin) - return NULL; - - /* - * Initialize structure - */ - hashbin->hb_type = type; - hashbin->magic = HB_MAGIC; - //hashbin->hb_current = NULL; - - /* Make sure all spinlock's are unlocked */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_lock_init(&hashbin->hb_spinlock); - } - - return hashbin; -} -EXPORT_SYMBOL(hashbin_new); - - -/* - * Function hashbin_delete (hashbin, free_func) - * - * Destroy hashbin, the free_func can be a user supplied special routine - * for deallocating this structure if it's complex. If not the user can - * just supply kfree, which should take care of the job. - */ -int hashbin_delete( hashbin_t* hashbin, FREE_FUNC free_func) -{ - irda_queue_t* queue; - unsigned long flags = 0; - int i; - - IRDA_ASSERT(hashbin != NULL, return -1;); - IRDA_ASSERT(hashbin->magic == HB_MAGIC, return -1;); - - /* Synchronize */ - if (hashbin->hb_type & HB_LOCK) - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - - /* - * Free the entries in the hashbin, TODO: use hashbin_clear when - * it has been shown to work - */ - for (i = 0; i < HASHBIN_SIZE; i ++ ) { - while (1) { - queue = dequeue_first((irda_queue_t**) &hashbin->hb_queue[i]); - - if (!queue) - break; - - if (free_func) { - if (hashbin->hb_type & HB_LOCK) - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - free_func(queue); - if (hashbin->hb_type & HB_LOCK) - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - } - } - } - - /* Cleanup local data */ - hashbin->hb_current = NULL; - hashbin->magic = ~HB_MAGIC; - - /* Release lock */ - if (hashbin->hb_type & HB_LOCK) - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - - /* - * Free the hashbin structure - */ - kfree(hashbin); - - return 0; -} -EXPORT_SYMBOL(hashbin_delete); - -/********************* HASHBIN LIST OPERATIONS *********************/ - -/* - * Function hashbin_insert (hashbin, entry, name) - * - * Insert an entry into the hashbin - * - */ -void hashbin_insert(hashbin_t* hashbin, irda_queue_t* entry, long hashv, - const char* name) -{ - unsigned long flags = 0; - int bin; - - IRDA_ASSERT( hashbin != NULL, return;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return;); - - /* - * Locate hashbin - */ - if ( name ) - hashv = hash( name ); - bin = GET_HASHBIN( hashv ); - - /* Synchronize */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - /* - * Store name and key - */ - entry->q_hash = hashv; - if ( name ) - strlcpy( entry->q_name, name, sizeof(entry->q_name)); - - /* - * Insert new entry first - */ - enqueue_first( (irda_queue_t**) &hashbin->hb_queue[ bin ], - entry); - hashbin->hb_size++; - - /* Release lock */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ -} -EXPORT_SYMBOL(hashbin_insert); - -/* - * Function hashbin_remove_first (hashbin) - * - * Remove first entry of the hashbin - * - * Note : this function no longer use hashbin_remove(), but does things - * similar to hashbin_remove_this(), so can be considered safe. - * Jean II - */ -void *hashbin_remove_first( hashbin_t *hashbin) -{ - unsigned long flags = 0; - irda_queue_t *entry = NULL; - - /* Synchronize */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - entry = hashbin_get_first( hashbin); - if ( entry != NULL) { - int bin; - long hashv; - /* - * Locate hashbin - */ - hashv = entry->q_hash; - bin = GET_HASHBIN( hashv ); - - /* - * Dequeue the entry... - */ - dequeue_general( (irda_queue_t**) &hashbin->hb_queue[ bin ], - entry); - hashbin->hb_size--; - entry->q_next = NULL; - entry->q_prev = NULL; - - /* - * Check if this item is the currently selected item, and in - * that case we must reset hb_current - */ - if ( entry == hashbin->hb_current) - hashbin->hb_current = NULL; - } - - /* Release lock */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - return entry; -} - - -/* - * Function hashbin_remove (hashbin, hashv, name) - * - * Remove entry with the given name - * - * The use of this function is highly discouraged, because the whole - * concept behind hashbin_remove() is broken. In many cases, it's not - * possible to guarantee the unicity of the index (either hashv or name), - * leading to removing the WRONG entry. - * The only simple safe use is : - * hashbin_remove(hasbin, (int) self, NULL); - * In other case, you must think hard to guarantee unicity of the index. - * Jean II - */ -void* hashbin_remove( hashbin_t* hashbin, long hashv, const char* name) -{ - int bin, found = FALSE; - unsigned long flags = 0; - irda_queue_t* entry; - - IRDA_ASSERT( hashbin != NULL, return NULL;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return NULL;); - - /* - * Locate hashbin - */ - if ( name ) - hashv = hash( name ); - bin = GET_HASHBIN( hashv ); - - /* Synchronize */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - /* - * Search for entry - */ - entry = hashbin->hb_queue[ bin ]; - if ( entry ) { - do { - /* - * Check for key - */ - if ( entry->q_hash == hashv ) { - /* - * Name compare too? - */ - if ( name ) { - if ( strcmp( entry->q_name, name) == 0) - { - found = TRUE; - break; - } - } else { - found = TRUE; - break; - } - } - entry = entry->q_next; - } while ( entry != hashbin->hb_queue[ bin ] ); - } - - /* - * If entry was found, dequeue it - */ - if ( found ) { - dequeue_general( (irda_queue_t**) &hashbin->hb_queue[ bin ], - entry); - hashbin->hb_size--; - - /* - * Check if this item is the currently selected item, and in - * that case we must reset hb_current - */ - if ( entry == hashbin->hb_current) - hashbin->hb_current = NULL; - } - - /* Release lock */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - - /* Return */ - if ( found ) - return entry; - else - return NULL; - -} -EXPORT_SYMBOL(hashbin_remove); - -/* - * Function hashbin_remove_this (hashbin, entry) - * - * Remove entry with the given name - * - * In some cases, the user of hashbin can't guarantee the unicity - * of either the hashv or name. - * In those cases, using the above function is guaranteed to cause troubles, - * so we use this one instead... - * And by the way, it's also faster, because we skip the search phase ;-) - */ -void* hashbin_remove_this( hashbin_t* hashbin, irda_queue_t* entry) -{ - unsigned long flags = 0; - int bin; - long hashv; - - IRDA_ASSERT( hashbin != NULL, return NULL;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return NULL;); - IRDA_ASSERT( entry != NULL, return NULL;); - - /* Synchronize */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - /* Check if valid and not already removed... */ - if((entry->q_next == NULL) || (entry->q_prev == NULL)) { - entry = NULL; - goto out; - } - - /* - * Locate hashbin - */ - hashv = entry->q_hash; - bin = GET_HASHBIN( hashv ); - - /* - * Dequeue the entry... - */ - dequeue_general( (irda_queue_t**) &hashbin->hb_queue[ bin ], - entry); - hashbin->hb_size--; - entry->q_next = NULL; - entry->q_prev = NULL; - - /* - * Check if this item is the currently selected item, and in - * that case we must reset hb_current - */ - if ( entry == hashbin->hb_current) - hashbin->hb_current = NULL; -out: - /* Release lock */ - if ( hashbin->hb_type & HB_LOCK ) { - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - } /* Default is no-lock */ - - return entry; -} -EXPORT_SYMBOL(hashbin_remove_this); - -/*********************** HASHBIN ENUMERATION ***********************/ - -/* - * Function hashbin_common_find (hashbin, hashv, name) - * - * Find item with the given hashv or name - * - */ -void* hashbin_find( hashbin_t* hashbin, long hashv, const char* name ) -{ - int bin; - irda_queue_t* entry; - - pr_debug("hashbin_find()\n"); - - IRDA_ASSERT( hashbin != NULL, return NULL;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return NULL;); - - /* - * Locate hashbin - */ - if ( name ) - hashv = hash( name ); - bin = GET_HASHBIN( hashv ); - - /* - * Search for entry - */ - entry = hashbin->hb_queue[ bin]; - if ( entry ) { - do { - /* - * Check for key - */ - if ( entry->q_hash == hashv ) { - /* - * Name compare too? - */ - if ( name ) { - if ( strcmp( entry->q_name, name ) == 0 ) { - return entry; - } - } else { - return entry; - } - } - entry = entry->q_next; - } while ( entry != hashbin->hb_queue[ bin ] ); - } - - return NULL; -} -EXPORT_SYMBOL(hashbin_find); - -/* - * Function hashbin_lock_find (hashbin, hashv, name) - * - * Find item with the given hashv or name - * - * Same, but with spinlock protection... - * I call it safe, but it's only safe with respect to the hashbin, not its - * content. - Jean II - */ -void* hashbin_lock_find( hashbin_t* hashbin, long hashv, const char* name ) -{ - unsigned long flags = 0; - irda_queue_t* entry; - - /* Synchronize */ - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - - /* - * Search for entry - */ - entry = hashbin_find(hashbin, hashv, name); - - /* Release lock */ - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - - return entry; -} -EXPORT_SYMBOL(hashbin_lock_find); - -/* - * Function hashbin_find (hashbin, hashv, name, pnext) - * - * Find an item with the given hashv or name, and its successor - * - * This function allow to do concurrent enumerations without the - * need to lock over the whole session, because the caller keep the - * context of the search. On the other hand, it might fail and return - * NULL if the entry is removed. - Jean II - */ -void* hashbin_find_next( hashbin_t* hashbin, long hashv, const char* name, - void ** pnext) -{ - unsigned long flags = 0; - irda_queue_t* entry; - - /* Synchronize */ - spin_lock_irqsave(&hashbin->hb_spinlock, flags); - - /* - * Search for current entry - * This allow to check if the current item is still in the - * hashbin or has been removed. - */ - entry = hashbin_find(hashbin, hashv, name); - - /* - * Trick hashbin_get_next() to return what we want - */ - if(entry) { - hashbin->hb_current = entry; - *pnext = hashbin_get_next( hashbin ); - } else - *pnext = NULL; - - /* Release lock */ - spin_unlock_irqrestore(&hashbin->hb_spinlock, flags); - - return entry; -} - -/* - * Function hashbin_get_first (hashbin) - * - * Get a pointer to first element in hashbin, this function must be - * called before any calls to hashbin_get_next()! - * - */ -irda_queue_t *hashbin_get_first( hashbin_t* hashbin) -{ - irda_queue_t *entry; - int i; - - IRDA_ASSERT( hashbin != NULL, return NULL;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return NULL;); - - if ( hashbin == NULL) - return NULL; - - for ( i = 0; i < HASHBIN_SIZE; i ++ ) { - entry = hashbin->hb_queue[ i]; - if ( entry) { - hashbin->hb_current = entry; - return entry; - } - } - /* - * Did not find any item in hashbin - */ - return NULL; -} -EXPORT_SYMBOL(hashbin_get_first); - -/* - * Function hashbin_get_next (hashbin) - * - * Get next item in hashbin. A series of hashbin_get_next() calls must - * be started by a call to hashbin_get_first(). The function returns - * NULL when all items have been traversed - * - * The context of the search is stored within the hashbin, so you must - * protect yourself from concurrent enumerations. - Jean II - */ -irda_queue_t *hashbin_get_next( hashbin_t *hashbin) -{ - irda_queue_t* entry; - int bin; - int i; - - IRDA_ASSERT( hashbin != NULL, return NULL;); - IRDA_ASSERT( hashbin->magic == HB_MAGIC, return NULL;); - - if ( hashbin->hb_current == NULL) { - IRDA_ASSERT( hashbin->hb_current != NULL, return NULL;); - return NULL; - } - entry = hashbin->hb_current->q_next; - bin = GET_HASHBIN( entry->q_hash); - - /* - * Make sure that we are not back at the beginning of the queue - * again - */ - if ( entry != hashbin->hb_queue[ bin ]) { - hashbin->hb_current = entry; - - return entry; - } - - /* - * Check that this is not the last queue in hashbin - */ - if ( bin >= HASHBIN_SIZE) - return NULL; - - /* - * Move to next queue in hashbin - */ - bin++; - for ( i = bin; i < HASHBIN_SIZE; i++ ) { - entry = hashbin->hb_queue[ i]; - if ( entry) { - hashbin->hb_current = entry; - - return entry; - } - } - return NULL; -} -EXPORT_SYMBOL(hashbin_get_next); diff --git a/net/irda/irsysctl.c b/net/irda/irsysctl.c deleted file mode 100644 index 873da5e7d428..000000000000 --- a/net/irda/irsysctl.c +++ /dev/null @@ -1,258 +0,0 @@ -/********************************************************************* - * - * Filename: irsysctl.c - * Version: 1.0 - * Description: Sysctl interface for IrDA - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun May 24 22:12:06 1998 - * Modified at: Fri Jun 4 02:50:15 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1997, 1999 Dag Brattli, All Rights Reserved. - * Copyright (c) 2000-2001 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/mm.h> -#include <linux/ctype.h> -#include <linux/sysctl.h> -#include <linux/init.h> - -#include <net/irda/irda.h> /* irda_debug */ -#include <net/irda/irlmp.h> -#include <net/irda/timer.h> -#include <net/irda/irias_object.h> - -extern int sysctl_discovery; -extern int sysctl_discovery_slots; -extern int sysctl_discovery_timeout; -extern int sysctl_slot_timeout; -extern int sysctl_fast_poll_increase; -extern char sysctl_devname[]; -extern int sysctl_max_baud_rate; -extern unsigned int sysctl_min_tx_turn_time; -extern unsigned int sysctl_max_tx_data_size; -extern unsigned int sysctl_max_tx_window; -extern int sysctl_max_noreply_time; -extern int sysctl_warn_noreply_time; -extern int sysctl_lap_keepalive_time; - -extern struct irlmp_cb *irlmp; - -/* this is needed for the proc_dointvec_minmax - Jean II */ -static int max_discovery_slots = 16; /* ??? */ -static int min_discovery_slots = 1; -/* IrLAP 6.13.2 says 25ms to 10+70ms - allow higher since some devices - * seems to require it. (from Dag's comment) */ -static int max_slot_timeout = 160; -static int min_slot_timeout = 20; -static int max_max_baud_rate = 16000000; /* See qos.c - IrLAP spec */ -static int min_max_baud_rate = 2400; -static int max_min_tx_turn_time = 10000; /* See qos.c - IrLAP spec */ -static int min_min_tx_turn_time; -static int max_max_tx_data_size = 2048; /* See qos.c - IrLAP spec */ -static int min_max_tx_data_size = 64; -static int max_max_tx_window = 7; /* See qos.c - IrLAP spec */ -static int min_max_tx_window = 1; -static int max_max_noreply_time = 40; /* See qos.c - IrLAP spec */ -static int min_max_noreply_time = 3; -static int max_warn_noreply_time = 3; /* 3s == standard */ -static int min_warn_noreply_time = 1; /* 1s == min WD_TIMER */ -static int max_lap_keepalive_time = 10000; /* 10s */ -static int min_lap_keepalive_time = 100; /* 100us */ -/* For other sysctl, I've no idea of the range. Maybe Dag could help - * us on that - Jean II */ - -static int do_devname(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - int ret; - - ret = proc_dostring(table, write, buffer, lenp, ppos); - if (ret == 0 && write) { - struct ias_value *val; - - val = irias_new_string_value(sysctl_devname); - if (val) - irias_object_change_attribute("Device", "DeviceName", val); - } - return ret; -} - - -static int do_discovery(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - int ret; - - ret = proc_dointvec(table, write, buffer, lenp, ppos); - if (ret) - return ret; - - if (irlmp == NULL) - return -ENODEV; - - if (sysctl_discovery) - irlmp_start_discovery_timer(irlmp, sysctl_discovery_timeout*HZ); - else - del_timer_sync(&irlmp->discovery_timer); - - return ret; -} - -/* One file */ -static struct ctl_table irda_table[] = { - { - .procname = "discovery", - .data = &sysctl_discovery, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = do_discovery, - }, - { - .procname = "devname", - .data = sysctl_devname, - .maxlen = 65, - .mode = 0644, - .proc_handler = do_devname, - }, -#ifdef CONFIG_IRDA_FAST_RR - { - .procname = "fast_poll_increase", - .data = &sysctl_fast_poll_increase, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, -#endif - { - .procname = "discovery_slots", - .data = &sysctl_discovery_slots, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_discovery_slots, - .extra2 = &max_discovery_slots - }, - { - .procname = "discovery_timeout", - .data = &sysctl_discovery_timeout, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, - { - .procname = "slot_timeout", - .data = &sysctl_slot_timeout, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_slot_timeout, - .extra2 = &max_slot_timeout - }, - { - .procname = "max_baud_rate", - .data = &sysctl_max_baud_rate, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_max_baud_rate, - .extra2 = &max_max_baud_rate - }, - { - .procname = "min_tx_turn_time", - .data = &sysctl_min_tx_turn_time, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_min_tx_turn_time, - .extra2 = &max_min_tx_turn_time - }, - { - .procname = "max_tx_data_size", - .data = &sysctl_max_tx_data_size, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_max_tx_data_size, - .extra2 = &max_max_tx_data_size - }, - { - .procname = "max_tx_window", - .data = &sysctl_max_tx_window, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_max_tx_window, - .extra2 = &max_max_tx_window - }, - { - .procname = "max_noreply_time", - .data = &sysctl_max_noreply_time, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_max_noreply_time, - .extra2 = &max_max_noreply_time - }, - { - .procname = "warn_noreply_time", - .data = &sysctl_warn_noreply_time, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_warn_noreply_time, - .extra2 = &max_warn_noreply_time - }, - { - .procname = "lap_keepalive_time", - .data = &sysctl_lap_keepalive_time, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &min_lap_keepalive_time, - .extra2 = &max_lap_keepalive_time - }, - { } -}; - -static struct ctl_table_header *irda_table_header; - -/* - * Function irda_sysctl_register (void) - * - * Register our sysctl interface - * - */ -int __init irda_sysctl_register(void) -{ - irda_table_header = register_net_sysctl(&init_net, "net/irda", irda_table); - if (!irda_table_header) - return -ENOMEM; - - return 0; -} - -/* - * Function irda_sysctl_unregister (void) - * - * Unregister our sysctl interface - * - */ -void irda_sysctl_unregister(void) -{ - unregister_net_sysctl_table(irda_table_header); -} - - - diff --git a/net/irda/irttp.c b/net/irda/irttp.c deleted file mode 100644 index b6ab41d5b3a3..000000000000 --- a/net/irda/irttp.c +++ /dev/null @@ -1,1891 +0,0 @@ -/********************************************************************* - * - * Filename: irttp.c - * Version: 1.2 - * Description: Tiny Transport Protocol (TTP) implementation - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sun Aug 31 20:14:31 1997 - * Modified at: Wed Jan 5 11:31:27 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2003 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> -#include <linux/init.h> -#include <linux/fs.h> -#include <linux/seq_file.h> -#include <linux/slab.h> -#include <linux/export.h> - -#include <asm/byteorder.h> -#include <asm/unaligned.h> - -#include <net/irda/irda.h> -#include <net/irda/irlap.h> -#include <net/irda/irlmp.h> -#include <net/irda/parameters.h> -#include <net/irda/irttp.h> - -static struct irttp_cb *irttp; - -static void __irttp_close_tsap(struct tsap_cb *self); - -static int irttp_data_indication(void *instance, void *sap, - struct sk_buff *skb); -static int irttp_udata_indication(void *instance, void *sap, - struct sk_buff *skb); -static void irttp_disconnect_indication(void *instance, void *sap, - LM_REASON reason, struct sk_buff *); -static void irttp_connect_indication(void *instance, void *sap, - struct qos_info *qos, __u32 max_sdu_size, - __u8 header_size, struct sk_buff *skb); -static void irttp_connect_confirm(void *instance, void *sap, - struct qos_info *qos, __u32 max_sdu_size, - __u8 header_size, struct sk_buff *skb); -static void irttp_run_tx_queue(struct tsap_cb *self); -static void irttp_run_rx_queue(struct tsap_cb *self); - -static void irttp_flush_queues(struct tsap_cb *self); -static void irttp_fragment_skb(struct tsap_cb *self, struct sk_buff *skb); -static struct sk_buff *irttp_reassemble_skb(struct tsap_cb *self); -static void irttp_todo_expired(unsigned long data); -static int irttp_param_max_sdu_size(void *instance, irda_param_t *param, - int get); - -static void irttp_flow_indication(void *instance, void *sap, LOCAL_FLOW flow); -static void irttp_status_indication(void *instance, - LINK_STATUS link, LOCK_STATUS lock); - -/* Information for parsing parameters in IrTTP */ -static const pi_minor_info_t pi_minor_call_table[] = { - { NULL, 0 }, /* 0x00 */ - { irttp_param_max_sdu_size, PV_INTEGER | PV_BIG_ENDIAN } /* 0x01 */ -}; -static const pi_major_info_t pi_major_call_table[] = { - { pi_minor_call_table, 2 } -}; -static pi_param_info_t param_info = { pi_major_call_table, 1, 0x0f, 4 }; - -/************************ GLOBAL PROCEDURES ************************/ - -/* - * Function irttp_init (void) - * - * Initialize the IrTTP layer. Called by module initialization code - * - */ -int __init irttp_init(void) -{ - irttp = kzalloc(sizeof(struct irttp_cb), GFP_KERNEL); - if (irttp == NULL) - return -ENOMEM; - - irttp->magic = TTP_MAGIC; - - irttp->tsaps = hashbin_new(HB_LOCK); - if (!irttp->tsaps) { - net_err_ratelimited("%s: can't allocate IrTTP hashbin!\n", - __func__); - kfree(irttp); - return -ENOMEM; - } - - return 0; -} - -/* - * Function irttp_cleanup (void) - * - * Called by module destruction/cleanup code - * - */ -void irttp_cleanup(void) -{ - /* Check for main structure */ - IRDA_ASSERT(irttp->magic == TTP_MAGIC, return;); - - /* - * Delete hashbin and close all TSAP instances in it - */ - hashbin_delete(irttp->tsaps, (FREE_FUNC) __irttp_close_tsap); - - irttp->magic = 0; - - /* De-allocate main structure */ - kfree(irttp); - - irttp = NULL; -} - -/*************************** SUBROUTINES ***************************/ - -/* - * Function irttp_start_todo_timer (self, timeout) - * - * Start todo timer. - * - * Made it more effient and unsensitive to race conditions - Jean II - */ -static inline void irttp_start_todo_timer(struct tsap_cb *self, int timeout) -{ - /* Set new value for timer */ - mod_timer(&self->todo_timer, jiffies + timeout); -} - -/* - * Function irttp_todo_expired (data) - * - * Todo timer has expired! - * - * One of the restriction of the timer is that it is run only on the timer - * interrupt which run every 10ms. This mean that even if you set the timer - * with a delay of 0, it may take up to 10ms before it's run. - * So, to minimise latency and keep cache fresh, we try to avoid using - * it as much as possible. - * Note : we can't use tasklets, because they can't be asynchronously - * killed (need user context), and we can't guarantee that here... - * Jean II - */ -static void irttp_todo_expired(unsigned long data) -{ - struct tsap_cb *self = (struct tsap_cb *) data; - - /* Check that we still exist */ - if (!self || self->magic != TTP_TSAP_MAGIC) - return; - - pr_debug("%s(instance=%p)\n", __func__, self); - - /* Try to make some progress, especially on Tx side - Jean II */ - irttp_run_rx_queue(self); - irttp_run_tx_queue(self); - - /* Check if time for disconnect */ - if (test_bit(0, &self->disconnect_pend)) { - /* Check if it's possible to disconnect yet */ - if (skb_queue_empty(&self->tx_queue)) { - /* Make sure disconnect is not pending anymore */ - clear_bit(0, &self->disconnect_pend); /* FALSE */ - - /* Note : self->disconnect_skb may be NULL */ - irttp_disconnect_request(self, self->disconnect_skb, - P_NORMAL); - self->disconnect_skb = NULL; - } else { - /* Try again later */ - irttp_start_todo_timer(self, HZ/10); - - /* No reason to try and close now */ - return; - } - } - - /* Check if it's closing time */ - if (self->close_pend) - /* Finish cleanup */ - irttp_close_tsap(self); -} - -/* - * Function irttp_flush_queues (self) - * - * Flushes (removes all frames) in transitt-buffer (tx_list) - */ -static void irttp_flush_queues(struct tsap_cb *self) -{ - struct sk_buff *skb; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - /* Deallocate frames waiting to be sent */ - while ((skb = skb_dequeue(&self->tx_queue)) != NULL) - dev_kfree_skb(skb); - - /* Deallocate received frames */ - while ((skb = skb_dequeue(&self->rx_queue)) != NULL) - dev_kfree_skb(skb); - - /* Deallocate received fragments */ - while ((skb = skb_dequeue(&self->rx_fragments)) != NULL) - dev_kfree_skb(skb); -} - -/* - * Function irttp_reassemble (self) - * - * Makes a new (continuous) skb of all the fragments in the fragment - * queue - * - */ -static struct sk_buff *irttp_reassemble_skb(struct tsap_cb *self) -{ - struct sk_buff *skb, *frag; - int n = 0; /* Fragment index */ - - IRDA_ASSERT(self != NULL, return NULL;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return NULL;); - - pr_debug("%s(), self->rx_sdu_size=%d\n", __func__, - self->rx_sdu_size); - - skb = dev_alloc_skb(TTP_HEADER + self->rx_sdu_size); - if (!skb) - return NULL; - - /* - * Need to reserve space for TTP header in case this skb needs to - * be requeued in case delivery failes - */ - skb_reserve(skb, TTP_HEADER); - skb_put(skb, self->rx_sdu_size); - - /* - * Copy all fragments to a new buffer - */ - while ((frag = skb_dequeue(&self->rx_fragments)) != NULL) { - skb_copy_to_linear_data_offset(skb, n, frag->data, frag->len); - n += frag->len; - - dev_kfree_skb(frag); - } - - pr_debug("%s(), frame len=%d, rx_sdu_size=%d, rx_max_sdu_size=%d\n", - __func__, n, self->rx_sdu_size, self->rx_max_sdu_size); - /* Note : irttp_run_rx_queue() calculate self->rx_sdu_size - * by summing the size of all fragments, so we should always - * have n == self->rx_sdu_size, except in cases where we - * droped the last fragment (when self->rx_sdu_size exceed - * self->rx_max_sdu_size), where n < self->rx_sdu_size. - * Jean II */ - IRDA_ASSERT(n <= self->rx_sdu_size, n = self->rx_sdu_size;); - - /* Set the new length */ - skb_trim(skb, n); - - self->rx_sdu_size = 0; - - return skb; -} - -/* - * Function irttp_fragment_skb (skb) - * - * Fragments a frame and queues all the fragments for transmission - * - */ -static inline void irttp_fragment_skb(struct tsap_cb *self, - struct sk_buff *skb) -{ - struct sk_buff *frag; - __u8 *frame; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - /* - * Split frame into a number of segments - */ - while (skb->len > self->max_seg_size) { - pr_debug("%s(), fragmenting ...\n", __func__); - - /* Make new segment */ - frag = alloc_skb(self->max_seg_size+self->max_header_size, - GFP_ATOMIC); - if (!frag) - return; - - skb_reserve(frag, self->max_header_size); - - /* Copy data from the original skb into this fragment. */ - skb_copy_from_linear_data(skb, skb_put(frag, self->max_seg_size), - self->max_seg_size); - - /* Insert TTP header, with the more bit set */ - frame = skb_push(frag, TTP_HEADER); - frame[0] = TTP_MORE; - - /* Hide the copied data from the original skb */ - skb_pull(skb, self->max_seg_size); - - /* Queue fragment */ - skb_queue_tail(&self->tx_queue, frag); - } - /* Queue what is left of the original skb */ - pr_debug("%s(), queuing last segment\n", __func__); - - frame = skb_push(skb, TTP_HEADER); - frame[0] = 0x00; /* Clear more bit */ - - /* Queue fragment */ - skb_queue_tail(&self->tx_queue, skb); -} - -/* - * Function irttp_param_max_sdu_size (self, param) - * - * Handle the MaxSduSize parameter in the connect frames, this function - * will be called both when this parameter needs to be inserted into, and - * extracted from the connect frames - */ -static int irttp_param_max_sdu_size(void *instance, irda_param_t *param, - int get) -{ - struct tsap_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->tx_max_sdu_size; - else - self->tx_max_sdu_size = param->pv.i; - - pr_debug("%s(), MaxSduSize=%d\n", __func__, param->pv.i); - - return 0; -} - -/*************************** CLIENT CALLS ***************************/ -/************************** LMP CALLBACKS **************************/ -/* Everything is happily mixed up. Waiting for next clean up - Jean II */ - -/* - * Initialization, that has to be done on new tsap - * instance allocation and on duplication - */ -static void irttp_init_tsap(struct tsap_cb *tsap) -{ - spin_lock_init(&tsap->lock); - init_timer(&tsap->todo_timer); - - skb_queue_head_init(&tsap->rx_queue); - skb_queue_head_init(&tsap->tx_queue); - skb_queue_head_init(&tsap->rx_fragments); -} - -/* - * Function irttp_open_tsap (stsap, notify) - * - * Create TSAP connection endpoint, - */ -struct tsap_cb *irttp_open_tsap(__u8 stsap_sel, int credit, notify_t *notify) -{ - struct tsap_cb *self; - struct lsap_cb *lsap; - notify_t ttp_notify; - - IRDA_ASSERT(irttp->magic == TTP_MAGIC, return NULL;); - - /* The IrLMP spec (IrLMP 1.1 p10) says that we have the right to - * use only 0x01-0x6F. Of course, we can use LSAP_ANY as well. - * JeanII */ - if ((stsap_sel != LSAP_ANY) && - ((stsap_sel < 0x01) || (stsap_sel >= 0x70))) { - pr_debug("%s(), invalid tsap!\n", __func__); - return NULL; - } - - self = kzalloc(sizeof(struct tsap_cb), GFP_ATOMIC); - if (self == NULL) - return NULL; - - /* Initialize internal objects */ - irttp_init_tsap(self); - - /* Initialise todo timer */ - self->todo_timer.data = (unsigned long) self; - self->todo_timer.function = &irttp_todo_expired; - - /* Initialize callbacks for IrLMP to use */ - irda_notify_init(&ttp_notify); - ttp_notify.connect_confirm = irttp_connect_confirm; - ttp_notify.connect_indication = irttp_connect_indication; - ttp_notify.disconnect_indication = irttp_disconnect_indication; - ttp_notify.data_indication = irttp_data_indication; - ttp_notify.udata_indication = irttp_udata_indication; - ttp_notify.flow_indication = irttp_flow_indication; - if (notify->status_indication != NULL) - ttp_notify.status_indication = irttp_status_indication; - ttp_notify.instance = self; - strncpy(ttp_notify.name, notify->name, NOTIFY_MAX_NAME); - - self->magic = TTP_TSAP_MAGIC; - self->connected = FALSE; - - /* - * Create LSAP at IrLMP layer - */ - lsap = irlmp_open_lsap(stsap_sel, &ttp_notify, 0); - if (lsap == NULL) { - pr_debug("%s: unable to allocate LSAP!!\n", __func__); - __irttp_close_tsap(self); - return NULL; - } - - /* - * If user specified LSAP_ANY as source TSAP selector, then IrLMP - * will replace it with whatever source selector which is free, so - * the stsap_sel we have might not be valid anymore - */ - self->stsap_sel = lsap->slsap_sel; - pr_debug("%s(), stsap_sel=%02x\n", __func__, self->stsap_sel); - - self->notify = *notify; - self->lsap = lsap; - - hashbin_insert(irttp->tsaps, (irda_queue_t *) self, (long) self, NULL); - - if (credit > TTP_RX_MAX_CREDIT) - self->initial_credit = TTP_RX_MAX_CREDIT; - else - self->initial_credit = credit; - - return self; -} -EXPORT_SYMBOL(irttp_open_tsap); - -/* - * Function irttp_close (handle) - * - * Remove an instance of a TSAP. This function should only deal with the - * deallocation of the TSAP, and resetting of the TSAPs values; - * - */ -static void __irttp_close_tsap(struct tsap_cb *self) -{ - /* First make sure we're connected. */ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - irttp_flush_queues(self); - - del_timer(&self->todo_timer); - - /* This one won't be cleaned up if we are disconnect_pend + close_pend - * and we receive a disconnect_indication */ - if (self->disconnect_skb) - dev_kfree_skb(self->disconnect_skb); - - self->connected = FALSE; - self->magic = ~TTP_TSAP_MAGIC; - - kfree(self); -} - -/* - * Function irttp_close (self) - * - * Remove TSAP from list of all TSAPs and then deallocate all resources - * associated with this TSAP - * - * Note : because we *free* the tsap structure, it is the responsibility - * of the caller to make sure we are called only once and to deal with - * possible race conditions. - Jean II - */ -int irttp_close_tsap(struct tsap_cb *self) -{ - struct tsap_cb *tsap; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - - /* Make sure tsap has been disconnected */ - if (self->connected) { - /* Check if disconnect is not pending */ - if (!test_bit(0, &self->disconnect_pend)) { - net_warn_ratelimited("%s: TSAP still connected!\n", - __func__); - irttp_disconnect_request(self, NULL, P_NORMAL); - } - self->close_pend = TRUE; - irttp_start_todo_timer(self, HZ/10); - - return 0; /* Will be back! */ - } - - tsap = hashbin_remove(irttp->tsaps, (long) self, NULL); - - IRDA_ASSERT(tsap == self, return -1;); - - /* Close corresponding LSAP */ - if (self->lsap) { - irlmp_close_lsap(self->lsap); - self->lsap = NULL; - } - - __irttp_close_tsap(self); - - return 0; -} -EXPORT_SYMBOL(irttp_close_tsap); - -/* - * Function irttp_udata_request (self, skb) - * - * Send unreliable data on this TSAP - * - */ -int irttp_udata_request(struct tsap_cb *self, struct sk_buff *skb) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - /* Take shortcut on zero byte packets */ - if (skb->len == 0) { - ret = 0; - goto err; - } - - /* Check that nothing bad happens */ - if (!self->connected) { - net_warn_ratelimited("%s(), Not connected\n", __func__); - ret = -ENOTCONN; - goto err; - } - - if (skb->len > self->max_seg_size) { - net_err_ratelimited("%s(), UData is too large for IrLAP!\n", - __func__); - ret = -EMSGSIZE; - goto err; - } - - irlmp_udata_request(self->lsap, skb); - self->stats.tx_packets++; - - return 0; - -err: - dev_kfree_skb(skb); - return ret; -} -EXPORT_SYMBOL(irttp_udata_request); - - -/* - * Function irttp_data_request (handle, skb) - * - * Queue frame for transmission. If SAR is enabled, fragement the frame - * and queue the fragments for transmission - */ -int irttp_data_request(struct tsap_cb *self, struct sk_buff *skb) -{ - __u8 *frame; - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - pr_debug("%s() : queue len = %d\n", __func__, - skb_queue_len(&self->tx_queue)); - - /* Take shortcut on zero byte packets */ - if (skb->len == 0) { - ret = 0; - goto err; - } - - /* Check that nothing bad happens */ - if (!self->connected) { - net_warn_ratelimited("%s: Not connected\n", __func__); - ret = -ENOTCONN; - goto err; - } - - /* - * Check if SAR is disabled, and the frame is larger than what fits - * inside an IrLAP frame - */ - if ((self->tx_max_sdu_size == 0) && (skb->len > self->max_seg_size)) { - net_err_ratelimited("%s: SAR disabled, and data is too large for IrLAP!\n", - __func__); - ret = -EMSGSIZE; - goto err; - } - - /* - * Check if SAR is enabled, and the frame is larger than the - * TxMaxSduSize - */ - if ((self->tx_max_sdu_size != 0) && - (self->tx_max_sdu_size != TTP_SAR_UNBOUND) && - (skb->len > self->tx_max_sdu_size)) { - net_err_ratelimited("%s: SAR enabled, but data is larger than TxMaxSduSize!\n", - __func__); - ret = -EMSGSIZE; - goto err; - } - /* - * Check if transmit queue is full - */ - if (skb_queue_len(&self->tx_queue) >= TTP_TX_MAX_QUEUE) { - /* - * Give it a chance to empty itself - */ - irttp_run_tx_queue(self); - - /* Drop packet. This error code should trigger the caller - * to resend the data in the client code - Jean II */ - ret = -ENOBUFS; - goto err; - } - - /* Queue frame, or queue frame segments */ - if ((self->tx_max_sdu_size == 0) || (skb->len < self->max_seg_size)) { - /* Queue frame */ - IRDA_ASSERT(skb_headroom(skb) >= TTP_HEADER, return -1;); - frame = skb_push(skb, TTP_HEADER); - frame[0] = 0x00; /* Clear more bit */ - - skb_queue_tail(&self->tx_queue, skb); - } else { - /* - * Fragment the frame, this function will also queue the - * fragments, we don't care about the fact the transmit - * queue may be overfilled by all the segments for a little - * while - */ - irttp_fragment_skb(self, skb); - } - - /* Check if we can accept more data from client */ - if ((!self->tx_sdu_busy) && - (skb_queue_len(&self->tx_queue) > TTP_TX_HIGH_THRESHOLD)) { - /* Tx queue filling up, so stop client. */ - if (self->notify.flow_indication) { - self->notify.flow_indication(self->notify.instance, - self, FLOW_STOP); - } - /* self->tx_sdu_busy is the state of the client. - * Update state after notifying client to avoid - * race condition with irttp_flow_indication(). - * If the queue empty itself after our test but before - * we set the flag, we will fix ourselves below in - * irttp_run_tx_queue(). - * Jean II */ - self->tx_sdu_busy = TRUE; - } - - /* Try to make some progress */ - irttp_run_tx_queue(self); - - return 0; - -err: - dev_kfree_skb(skb); - return ret; -} -EXPORT_SYMBOL(irttp_data_request); - -/* - * Function irttp_run_tx_queue (self) - * - * Transmit packets queued for transmission (if possible) - * - */ -static void irttp_run_tx_queue(struct tsap_cb *self) -{ - struct sk_buff *skb; - unsigned long flags; - int n; - - pr_debug("%s() : send_credit = %d, queue_len = %d\n", - __func__, - self->send_credit, skb_queue_len(&self->tx_queue)); - - /* Get exclusive access to the tx queue, otherwise don't touch it */ - if (irda_lock(&self->tx_queue_lock) == FALSE) - return; - - /* Try to send out frames as long as we have credits - * and as long as LAP is not full. If LAP is full, it will - * poll us through irttp_flow_indication() - Jean II */ - while ((self->send_credit > 0) && - (!irlmp_lap_tx_queue_full(self->lsap)) && - (skb = skb_dequeue(&self->tx_queue))) { - /* - * Since we can transmit and receive frames concurrently, - * the code below is a critical region and we must assure that - * nobody messes with the credits while we update them. - */ - spin_lock_irqsave(&self->lock, flags); - - n = self->avail_credit; - self->avail_credit = 0; - - /* Only room for 127 credits in frame */ - if (n > 127) { - self->avail_credit = n-127; - n = 127; - } - self->remote_credit += n; - self->send_credit--; - - spin_unlock_irqrestore(&self->lock, flags); - - /* - * More bit must be set by the data_request() or fragment() - * functions - */ - skb->data[0] |= (n & 0x7f); - - /* Detach from socket. - * The current skb has a reference to the socket that sent - * it (skb->sk). When we pass it to IrLMP, the skb will be - * stored in in IrLAP (self->wx_list). When we are within - * IrLAP, we lose the notion of socket, so we should not - * have a reference to a socket. So, we drop it here. - * - * Why does it matter ? - * When the skb is freed (kfree_skb), if it is associated - * with a socket, it release buffer space on the socket - * (through sock_wfree() and sock_def_write_space()). - * If the socket no longer exist, we may crash. Hard. - * When we close a socket, we make sure that associated packets - * in IrTTP are freed. However, we have no way to cancel - * the packet that we have passed to IrLAP. So, if a packet - * remains in IrLAP (retry on the link or else) after we - * close the socket, we are dead ! - * Jean II */ - if (skb->sk != NULL) { - /* IrSOCK application, IrOBEX, ... */ - skb_orphan(skb); - } - /* IrCOMM over IrTTP, IrLAN, ... */ - - /* Pass the skb to IrLMP - done */ - irlmp_data_request(self->lsap, skb); - self->stats.tx_packets++; - } - - /* Check if we can accept more frames from client. - * We don't want to wait until the todo timer to do that, and we - * can't use tasklets (grr...), so we are obliged to give control - * to client. That's ok, this test will be true not too often - * (max once per LAP window) and we are called from places - * where we can spend a bit of time doing stuff. - Jean II */ - if ((self->tx_sdu_busy) && - (skb_queue_len(&self->tx_queue) < TTP_TX_LOW_THRESHOLD) && - (!self->close_pend)) { - if (self->notify.flow_indication) - self->notify.flow_indication(self->notify.instance, - self, FLOW_START); - - /* self->tx_sdu_busy is the state of the client. - * We don't really have a race here, but it's always safer - * to update our state after the client - Jean II */ - self->tx_sdu_busy = FALSE; - } - - /* Reset lock */ - self->tx_queue_lock = 0; -} - -/* - * Function irttp_give_credit (self) - * - * Send a dataless flowdata TTP-PDU and give available credit to peer - * TSAP - */ -static inline void irttp_give_credit(struct tsap_cb *self) -{ - struct sk_buff *tx_skb = NULL; - unsigned long flags; - int n; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - pr_debug("%s() send=%d,avail=%d,remote=%d\n", - __func__, - self->send_credit, self->avail_credit, self->remote_credit); - - /* Give credit to peer */ - tx_skb = alloc_skb(TTP_MAX_HEADER, GFP_ATOMIC); - if (!tx_skb) - return; - - /* Reserve space for LMP, and LAP header */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - - /* - * Since we can transmit and receive frames concurrently, - * the code below is a critical region and we must assure that - * nobody messes with the credits while we update them. - */ - spin_lock_irqsave(&self->lock, flags); - - n = self->avail_credit; - self->avail_credit = 0; - - /* Only space for 127 credits in frame */ - if (n > 127) { - self->avail_credit = n - 127; - n = 127; - } - self->remote_credit += n; - - spin_unlock_irqrestore(&self->lock, flags); - - skb_put(tx_skb, 1); - tx_skb->data[0] = (__u8) (n & 0x7f); - - irlmp_data_request(self->lsap, tx_skb); - self->stats.tx_packets++; -} - -/* - * Function irttp_udata_indication (instance, sap, skb) - * - * Received some unit-data (unreliable) - * - */ -static int irttp_udata_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct tsap_cb *self; - int err; - - self = instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - IRDA_ASSERT(skb != NULL, return -1;); - - self->stats.rx_packets++; - - /* Just pass data to layer above */ - if (self->notify.udata_indication) { - err = self->notify.udata_indication(self->notify.instance, - self, skb); - /* Same comment as in irttp_do_data_indication() */ - if (!err) - return 0; - } - /* Either no handler, or handler returns an error */ - dev_kfree_skb(skb); - - return 0; -} - -/* - * Function irttp_data_indication (instance, sap, skb) - * - * Receive segment from IrLMP. - * - */ -static int irttp_data_indication(void *instance, void *sap, - struct sk_buff *skb) -{ - struct tsap_cb *self; - unsigned long flags; - int n; - - self = instance; - - n = skb->data[0] & 0x7f; /* Extract the credits */ - - self->stats.rx_packets++; - - /* Deal with inbound credit - * Since we can transmit and receive frames concurrently, - * the code below is a critical region and we must assure that - * nobody messes with the credits while we update them. - */ - spin_lock_irqsave(&self->lock, flags); - self->send_credit += n; - if (skb->len > 1) - self->remote_credit--; - spin_unlock_irqrestore(&self->lock, flags); - - /* - * Data or dataless packet? Dataless frames contains only the - * TTP_HEADER. - */ - if (skb->len > 1) { - /* - * We don't remove the TTP header, since we must preserve the - * more bit, so the defragment routing knows what to do - */ - skb_queue_tail(&self->rx_queue, skb); - } else { - /* Dataless flowdata TTP-PDU */ - dev_kfree_skb(skb); - } - - - /* Push data to the higher layer. - * We do it synchronously because running the todo timer for each - * receive packet would be too much overhead and latency. - * By passing control to the higher layer, we run the risk that - * it may take time or grab a lock. Most often, the higher layer - * will only put packet in a queue. - * Anyway, packets are only dripping through the IrDA, so we can - * have time before the next packet. - * Further, we are run from NET_BH, so the worse that can happen is - * us missing the optimal time to send back the PF bit in LAP. - * Jean II */ - irttp_run_rx_queue(self); - - /* We now give credits to peer in irttp_run_rx_queue(). - * We need to send credit *NOW*, otherwise we are going - * to miss the next Tx window. The todo timer may take - * a while before it's run... - Jean II */ - - /* - * If the peer device has given us some credits and we didn't have - * anyone from before, then we need to shedule the tx queue. - * We need to do that because our Tx have stopped (so we may not - * get any LAP flow indication) and the user may be stopped as - * well. - Jean II - */ - if (self->send_credit == n) { - /* Restart pushing stuff to LAP */ - irttp_run_tx_queue(self); - /* Note : we don't want to schedule the todo timer - * because it has horrible latency. No tasklets - * because the tasklet API is broken. - Jean II */ - } - - return 0; -} - -/* - * Function irttp_status_indication (self, reason) - * - * Status_indication, just pass to the higher layer... - * - */ -static void irttp_status_indication(void *instance, - LINK_STATUS link, LOCK_STATUS lock) -{ - struct tsap_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - /* Check if client has already closed the TSAP and gone away */ - if (self->close_pend) - return; - - /* - * Inform service user if he has requested it - */ - if (self->notify.status_indication != NULL) - self->notify.status_indication(self->notify.instance, - link, lock); - else - pr_debug("%s(), no handler\n", __func__); -} - -/* - * Function irttp_flow_indication (self, reason) - * - * Flow_indication : IrLAP tells us to send more data. - * - */ -static void irttp_flow_indication(void *instance, void *sap, LOCAL_FLOW flow) -{ - struct tsap_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - pr_debug("%s(instance=%p)\n", __func__, self); - - /* We are "polled" directly from LAP, and the LAP want to fill - * its Tx window. We want to do our best to send it data, so that - * we maximise the window. On the other hand, we want to limit the - * amount of work here so that LAP doesn't hang forever waiting - * for packets. - Jean II */ - - /* Try to send some packets. Currently, LAP calls us every time - * there is one free slot, so we will send only one packet. - * This allow the scheduler to do its round robin - Jean II */ - irttp_run_tx_queue(self); - - /* Note regarding the interraction with higher layer. - * irttp_run_tx_queue() may call the client when its queue - * start to empty, via notify.flow_indication(). Initially. - * I wanted this to happen in a tasklet, to avoid client - * grabbing the CPU, but we can't use tasklets safely. And timer - * is definitely too slow. - * This will happen only once per LAP window, and usually at - * the third packet (unless window is smaller). LAP is still - * doing mtt and sending first packet so it's sort of OK - * to do that. Jean II */ - - /* If we need to send disconnect. try to do it now */ - if (self->disconnect_pend) - irttp_start_todo_timer(self, 0); -} - -/* - * Function irttp_flow_request (self, command) - * - * This function could be used by the upper layers to tell IrTTP to stop - * delivering frames if the receive queues are starting to get full, or - * to tell IrTTP to start delivering frames again. - */ -void irttp_flow_request(struct tsap_cb *self, LOCAL_FLOW flow) -{ - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - switch (flow) { - case FLOW_STOP: - pr_debug("%s(), flow stop\n", __func__); - self->rx_sdu_busy = TRUE; - break; - case FLOW_START: - pr_debug("%s(), flow start\n", __func__); - self->rx_sdu_busy = FALSE; - - /* Client say he can accept more data, try to free our - * queues ASAP - Jean II */ - irttp_run_rx_queue(self); - - break; - default: - pr_debug("%s(), Unknown flow command!\n", __func__); - } -} -EXPORT_SYMBOL(irttp_flow_request); - -/* - * Function irttp_connect_request (self, dtsap_sel, daddr, qos) - * - * Try to connect to remote destination TSAP selector - * - */ -int irttp_connect_request(struct tsap_cb *self, __u8 dtsap_sel, - __u32 saddr, __u32 daddr, - struct qos_info *qos, __u32 max_sdu_size, - struct sk_buff *userdata) -{ - struct sk_buff *tx_skb; - __u8 *frame; - __u8 n; - - pr_debug("%s(), max_sdu_size=%d\n", __func__, max_sdu_size); - - IRDA_ASSERT(self != NULL, return -EBADR;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -EBADR;); - - if (self->connected) { - if (userdata) - dev_kfree_skb(userdata); - return -EISCONN; - } - - /* Any userdata supplied? */ - if (userdata == NULL) { - tx_skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER, - GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* Reserve space for MUX_CONTROL and LAP header */ - skb_reserve(tx_skb, TTP_MAX_HEADER + TTP_SAR_HEADER); - } else { - tx_skb = userdata; - /* - * Check that the client has reserved enough space for - * headers - */ - IRDA_ASSERT(skb_headroom(userdata) >= TTP_MAX_HEADER, - { dev_kfree_skb(userdata); return -1; }); - } - - /* Initialize connection parameters */ - self->connected = FALSE; - self->avail_credit = 0; - self->rx_max_sdu_size = max_sdu_size; - self->rx_sdu_size = 0; - self->rx_sdu_busy = FALSE; - self->dtsap_sel = dtsap_sel; - - n = self->initial_credit; - - self->remote_credit = 0; - self->send_credit = 0; - - /* - * Give away max 127 credits for now - */ - if (n > 127) { - self->avail_credit = n - 127; - n = 127; - } - - self->remote_credit = n; - - /* SAR enabled? */ - if (max_sdu_size > 0) { - IRDA_ASSERT(skb_headroom(tx_skb) >= (TTP_MAX_HEADER + TTP_SAR_HEADER), - { dev_kfree_skb(tx_skb); return -1; }); - - /* Insert SAR parameters */ - frame = skb_push(tx_skb, TTP_HEADER + TTP_SAR_HEADER); - - frame[0] = TTP_PARAMETERS | n; - frame[1] = 0x04; /* Length */ - frame[2] = 0x01; /* MaxSduSize */ - frame[3] = 0x02; /* Value length */ - - put_unaligned(cpu_to_be16((__u16) max_sdu_size), - (__be16 *)(frame+4)); - } else { - /* Insert plain TTP header */ - frame = skb_push(tx_skb, TTP_HEADER); - - /* Insert initial credit in frame */ - frame[0] = n & 0x7f; - } - - /* Connect with IrLMP. No QoS parameters for now */ - return irlmp_connect_request(self->lsap, dtsap_sel, saddr, daddr, qos, - tx_skb); -} -EXPORT_SYMBOL(irttp_connect_request); - -/* - * Function irttp_connect_confirm (handle, qos, skb) - * - * Service user confirms TSAP connection with peer. - * - */ -static void irttp_connect_confirm(void *instance, void *sap, - struct qos_info *qos, __u32 max_seg_size, - __u8 max_header_size, struct sk_buff *skb) -{ - struct tsap_cb *self; - int parameters; - int ret; - __u8 plen; - __u8 n; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - self->max_seg_size = max_seg_size - TTP_HEADER; - self->max_header_size = max_header_size + TTP_HEADER; - - /* - * Check if we have got some QoS parameters back! This should be the - * negotiated QoS for the link. - */ - if (qos) { - pr_debug("IrTTP, Negotiated BAUD_RATE: %02x\n", - qos->baud_rate.bits); - pr_debug("IrTTP, Negotiated BAUD_RATE: %d bps.\n", - qos->baud_rate.value); - } - - n = skb->data[0] & 0x7f; - - pr_debug("%s(), Initial send_credit=%d\n", __func__, n); - - self->send_credit = n; - self->tx_max_sdu_size = 0; - self->connected = TRUE; - - parameters = skb->data[0] & 0x80; - - IRDA_ASSERT(skb->len >= TTP_HEADER, return;); - skb_pull(skb, TTP_HEADER); - - if (parameters) { - plen = skb->data[0]; - - ret = irda_param_extract_all(self, skb->data+1, - IRDA_MIN(skb->len-1, plen), - ¶m_info); - - /* Any errors in the parameter list? */ - if (ret < 0) { - net_warn_ratelimited("%s: error extracting parameters\n", - __func__); - dev_kfree_skb(skb); - - /* Do not accept this connection attempt */ - return; - } - /* Remove parameters */ - skb_pull(skb, IRDA_MIN(skb->len, plen+1)); - } - - pr_debug("%s() send=%d,avail=%d,remote=%d\n", __func__, - self->send_credit, self->avail_credit, self->remote_credit); - - pr_debug("%s(), MaxSduSize=%d\n", __func__, - self->tx_max_sdu_size); - - if (self->notify.connect_confirm) { - self->notify.connect_confirm(self->notify.instance, self, qos, - self->tx_max_sdu_size, - self->max_header_size, skb); - } else - dev_kfree_skb(skb); -} - -/* - * Function irttp_connect_indication (handle, skb) - * - * Some other device is connecting to this TSAP - * - */ -static void irttp_connect_indication(void *instance, void *sap, - struct qos_info *qos, __u32 max_seg_size, __u8 max_header_size, - struct sk_buff *skb) -{ - struct tsap_cb *self; - struct lsap_cb *lsap; - int parameters; - int ret; - __u8 plen; - __u8 n; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - IRDA_ASSERT(skb != NULL, return;); - - lsap = sap; - - self->max_seg_size = max_seg_size - TTP_HEADER; - self->max_header_size = max_header_size+TTP_HEADER; - - pr_debug("%s(), TSAP sel=%02x\n", __func__, self->stsap_sel); - - /* Need to update dtsap_sel if its equal to LSAP_ANY */ - self->dtsap_sel = lsap->dlsap_sel; - - n = skb->data[0] & 0x7f; - - self->send_credit = n; - self->tx_max_sdu_size = 0; - - parameters = skb->data[0] & 0x80; - - IRDA_ASSERT(skb->len >= TTP_HEADER, return;); - skb_pull(skb, TTP_HEADER); - - if (parameters) { - plen = skb->data[0]; - - ret = irda_param_extract_all(self, skb->data+1, - IRDA_MIN(skb->len-1, plen), - ¶m_info); - - /* Any errors in the parameter list? */ - if (ret < 0) { - net_warn_ratelimited("%s: error extracting parameters\n", - __func__); - dev_kfree_skb(skb); - - /* Do not accept this connection attempt */ - return; - } - - /* Remove parameters */ - skb_pull(skb, IRDA_MIN(skb->len, plen+1)); - } - - if (self->notify.connect_indication) { - self->notify.connect_indication(self->notify.instance, self, - qos, self->tx_max_sdu_size, - self->max_header_size, skb); - } else - dev_kfree_skb(skb); -} - -/* - * Function irttp_connect_response (handle, userdata) - * - * Service user is accepting the connection, just pass it down to - * IrLMP! - * - */ -int irttp_connect_response(struct tsap_cb *self, __u32 max_sdu_size, - struct sk_buff *userdata) -{ - struct sk_buff *tx_skb; - __u8 *frame; - int ret; - __u8 n; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - - pr_debug("%s(), Source TSAP selector=%02x\n", __func__, - self->stsap_sel); - - /* Any userdata supplied? */ - if (userdata == NULL) { - tx_skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER, - GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* Reserve space for MUX_CONTROL and LAP header */ - skb_reserve(tx_skb, TTP_MAX_HEADER + TTP_SAR_HEADER); - } else { - tx_skb = userdata; - /* - * Check that the client has reserved enough space for - * headers - */ - IRDA_ASSERT(skb_headroom(userdata) >= TTP_MAX_HEADER, - { dev_kfree_skb(userdata); return -1; }); - } - - self->avail_credit = 0; - self->remote_credit = 0; - self->rx_max_sdu_size = max_sdu_size; - self->rx_sdu_size = 0; - self->rx_sdu_busy = FALSE; - - n = self->initial_credit; - - /* Frame has only space for max 127 credits (7 bits) */ - if (n > 127) { - self->avail_credit = n - 127; - n = 127; - } - - self->remote_credit = n; - self->connected = TRUE; - - /* SAR enabled? */ - if (max_sdu_size > 0) { - IRDA_ASSERT(skb_headroom(tx_skb) >= (TTP_MAX_HEADER + TTP_SAR_HEADER), - { dev_kfree_skb(tx_skb); return -1; }); - - /* Insert TTP header with SAR parameters */ - frame = skb_push(tx_skb, TTP_HEADER + TTP_SAR_HEADER); - - frame[0] = TTP_PARAMETERS | n; - frame[1] = 0x04; /* Length */ - - /* irda_param_insert(self, IRTTP_MAX_SDU_SIZE, frame+1, */ -/* TTP_SAR_HEADER, ¶m_info) */ - - frame[2] = 0x01; /* MaxSduSize */ - frame[3] = 0x02; /* Value length */ - - put_unaligned(cpu_to_be16((__u16) max_sdu_size), - (__be16 *)(frame+4)); - } else { - /* Insert TTP header */ - frame = skb_push(tx_skb, TTP_HEADER); - - frame[0] = n & 0x7f; - } - - ret = irlmp_connect_response(self->lsap, tx_skb); - - return ret; -} -EXPORT_SYMBOL(irttp_connect_response); - -/* - * Function irttp_dup (self, instance) - * - * Duplicate TSAP, can be used by servers to confirm a connection on a - * new TSAP so it can keep listening on the old one. - */ -struct tsap_cb *irttp_dup(struct tsap_cb *orig, void *instance) -{ - struct tsap_cb *new; - unsigned long flags; - - /* Protect our access to the old tsap instance */ - spin_lock_irqsave(&irttp->tsaps->hb_spinlock, flags); - - /* Find the old instance */ - if (!hashbin_find(irttp->tsaps, (long) orig, NULL)) { - pr_debug("%s(), unable to find TSAP\n", __func__); - spin_unlock_irqrestore(&irttp->tsaps->hb_spinlock, flags); - return NULL; - } - - /* Allocate a new instance */ - new = kmemdup(orig, sizeof(struct tsap_cb), GFP_ATOMIC); - if (!new) { - pr_debug("%s(), unable to kmalloc\n", __func__); - spin_unlock_irqrestore(&irttp->tsaps->hb_spinlock, flags); - return NULL; - } - spin_lock_init(&new->lock); - - /* We don't need the old instance any more */ - spin_unlock_irqrestore(&irttp->tsaps->hb_spinlock, flags); - - /* Try to dup the LSAP (may fail if we were too slow) */ - new->lsap = irlmp_dup(orig->lsap, new); - if (!new->lsap) { - pr_debug("%s(), dup failed!\n", __func__); - kfree(new); - return NULL; - } - - /* Not everything should be copied */ - new->notify.instance = instance; - - /* Initialize internal objects */ - irttp_init_tsap(new); - - /* This is locked */ - hashbin_insert(irttp->tsaps, (irda_queue_t *) new, (long) new, NULL); - - return new; -} -EXPORT_SYMBOL(irttp_dup); - -/* - * Function irttp_disconnect_request (self) - * - * Close this connection please! If priority is high, the queued data - * segments, if any, will be deallocated first - * - */ -int irttp_disconnect_request(struct tsap_cb *self, struct sk_buff *userdata, - int priority) -{ - int ret; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return -1;); - - /* Already disconnected? */ - if (!self->connected) { - pr_debug("%s(), already disconnected!\n", __func__); - if (userdata) - dev_kfree_skb(userdata); - return -1; - } - - /* Disconnect already pending ? - * We need to use an atomic operation to prevent reentry. This - * function may be called from various context, like user, timer - * for following a disconnect_indication() (i.e. net_bh). - * Jean II */ - if (test_and_set_bit(0, &self->disconnect_pend)) { - pr_debug("%s(), disconnect already pending\n", - __func__); - if (userdata) - dev_kfree_skb(userdata); - - /* Try to make some progress */ - irttp_run_tx_queue(self); - return -1; - } - - /* - * Check if there is still data segments in the transmit queue - */ - if (!skb_queue_empty(&self->tx_queue)) { - if (priority == P_HIGH) { - /* - * No need to send the queued data, if we are - * disconnecting right now since the data will - * not have any usable connection to be sent on - */ - pr_debug("%s(): High priority!!()\n", __func__); - irttp_flush_queues(self); - } else if (priority == P_NORMAL) { - /* - * Must delay disconnect until after all data segments - * have been sent and the tx_queue is empty - */ - /* We'll reuse this one later for the disconnect */ - self->disconnect_skb = userdata; /* May be NULL */ - - irttp_run_tx_queue(self); - - irttp_start_todo_timer(self, HZ/10); - return -1; - } - } - /* Note : we don't need to check if self->rx_queue is full and the - * state of self->rx_sdu_busy because the disconnect response will - * be sent at the LMP level (so even if the peer has its Tx queue - * full of data). - Jean II */ - - pr_debug("%s(), Disconnecting ...\n", __func__); - self->connected = FALSE; - - if (!userdata) { - struct sk_buff *tx_skb; - tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC); - if (!tx_skb) - return -ENOMEM; - - /* - * Reserve space for MUX and LAP header - */ - skb_reserve(tx_skb, LMP_MAX_HEADER); - - userdata = tx_skb; - } - ret = irlmp_disconnect_request(self->lsap, userdata); - - /* The disconnect is no longer pending */ - clear_bit(0, &self->disconnect_pend); /* FALSE */ - - return ret; -} -EXPORT_SYMBOL(irttp_disconnect_request); - -/* - * Function irttp_disconnect_indication (self, reason) - * - * Disconnect indication, TSAP disconnected by peer? - * - */ -static void irttp_disconnect_indication(void *instance, void *sap, - LM_REASON reason, struct sk_buff *skb) -{ - struct tsap_cb *self; - - self = instance; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == TTP_TSAP_MAGIC, return;); - - /* Prevent higher layer to send more data */ - self->connected = FALSE; - - /* Check if client has already tried to close the TSAP */ - if (self->close_pend) { - /* In this case, the higher layer is probably gone. Don't - * bother it and clean up the remains - Jean II */ - if (skb) - dev_kfree_skb(skb); - irttp_close_tsap(self); - return; - } - - /* If we are here, we assume that is the higher layer is still - * waiting for the disconnect notification and able to process it, - * even if he tried to disconnect. Otherwise, it would have already - * attempted to close the tsap and self->close_pend would be TRUE. - * Jean II */ - - /* No need to notify the client if has already tried to disconnect */ - if (self->notify.disconnect_indication) - self->notify.disconnect_indication(self->notify.instance, self, - reason, skb); - else - if (skb) - dev_kfree_skb(skb); -} - -/* - * Function irttp_do_data_indication (self, skb) - * - * Try to deliver reassembled skb to layer above, and requeue it if that - * for some reason should fail. We mark rx sdu as busy to apply back - * pressure is necessary. - */ -static void irttp_do_data_indication(struct tsap_cb *self, struct sk_buff *skb) -{ - int err; - - /* Check if client has already closed the TSAP and gone away */ - if (self->close_pend) { - dev_kfree_skb(skb); - return; - } - - err = self->notify.data_indication(self->notify.instance, self, skb); - - /* Usually the layer above will notify that it's input queue is - * starting to get filled by using the flow request, but this may - * be difficult, so it can instead just refuse to eat it and just - * give an error back - */ - if (err) { - pr_debug("%s() requeueing skb!\n", __func__); - - /* Make sure we take a break */ - self->rx_sdu_busy = TRUE; - - /* Need to push the header in again */ - skb_push(skb, TTP_HEADER); - skb->data[0] = 0x00; /* Make sure MORE bit is cleared */ - - /* Put skb back on queue */ - skb_queue_head(&self->rx_queue, skb); - } -} - -/* - * Function irttp_run_rx_queue (self) - * - * Check if we have any frames to be transmitted, or if we have any - * available credit to give away. - */ -static void irttp_run_rx_queue(struct tsap_cb *self) -{ - struct sk_buff *skb; - int more = 0; - - pr_debug("%s() send=%d,avail=%d,remote=%d\n", __func__, - self->send_credit, self->avail_credit, self->remote_credit); - - /* Get exclusive access to the rx queue, otherwise don't touch it */ - if (irda_lock(&self->rx_queue_lock) == FALSE) - return; - - /* - * Reassemble all frames in receive queue and deliver them - */ - while (!self->rx_sdu_busy && (skb = skb_dequeue(&self->rx_queue))) { - /* This bit will tell us if it's the last fragment or not */ - more = skb->data[0] & 0x80; - - /* Remove TTP header */ - skb_pull(skb, TTP_HEADER); - - /* Add the length of the remaining data */ - self->rx_sdu_size += skb->len; - - /* - * If SAR is disabled, or user has requested no reassembly - * of received fragments then we just deliver them - * immediately. This can be requested by clients that - * implements byte streams without any message boundaries - */ - if (self->rx_max_sdu_size == TTP_SAR_DISABLE) { - irttp_do_data_indication(self, skb); - self->rx_sdu_size = 0; - - continue; - } - - /* Check if this is a fragment, and not the last fragment */ - if (more) { - /* - * Queue the fragment if we still are within the - * limits of the maximum size of the rx_sdu - */ - if (self->rx_sdu_size <= self->rx_max_sdu_size) { - pr_debug("%s(), queueing frag\n", - __func__); - skb_queue_tail(&self->rx_fragments, skb); - } else { - /* Free the part of the SDU that is too big */ - dev_kfree_skb(skb); - } - continue; - } - /* - * This is the last fragment, so time to reassemble! - */ - if ((self->rx_sdu_size <= self->rx_max_sdu_size) || - (self->rx_max_sdu_size == TTP_SAR_UNBOUND)) { - /* - * A little optimizing. Only queue the fragment if - * there are other fragments. Since if this is the - * last and only fragment, there is no need to - * reassemble :-) - */ - if (!skb_queue_empty(&self->rx_fragments)) { - skb_queue_tail(&self->rx_fragments, - skb); - - skb = irttp_reassemble_skb(self); - } - - /* Now we can deliver the reassembled skb */ - irttp_do_data_indication(self, skb); - } else { - pr_debug("%s(), Truncated frame\n", __func__); - - /* Free the part of the SDU that is too big */ - dev_kfree_skb(skb); - - /* Deliver only the valid but truncated part of SDU */ - skb = irttp_reassemble_skb(self); - - irttp_do_data_indication(self, skb); - } - self->rx_sdu_size = 0; - } - - /* - * It's not trivial to keep track of how many credits are available - * by incrementing at each packet, because delivery may fail - * (irttp_do_data_indication() may requeue the frame) and because - * we need to take care of fragmentation. - * We want the other side to send up to initial_credit packets. - * We have some frames in our queues, and we have already allowed it - * to send remote_credit. - * No need to spinlock, write is atomic and self correcting... - * Jean II - */ - self->avail_credit = (self->initial_credit - - (self->remote_credit + - skb_queue_len(&self->rx_queue) + - skb_queue_len(&self->rx_fragments))); - - /* Do we have too much credits to send to peer ? */ - if ((self->remote_credit <= TTP_RX_MIN_CREDIT) && - (self->avail_credit > 0)) { - /* Send explicit credit frame */ - irttp_give_credit(self); - /* Note : do *NOT* check if tx_queue is non-empty, that - * will produce deadlocks. I repeat : send a credit frame - * even if we have something to send in our Tx queue. - * If we have credits, it means that our Tx queue is blocked. - * - * Let's suppose the peer can't keep up with our Tx. He will - * flow control us by not sending us any credits, and we - * will stop Tx and start accumulating credits here. - * Up to the point where the peer will stop its Tx queue, - * for lack of credits. - * Let's assume the peer application is single threaded. - * It will block on Tx and never consume any Rx buffer. - * Deadlock. Guaranteed. - Jean II - */ - } - - /* Reset lock */ - self->rx_queue_lock = 0; -} - -#ifdef CONFIG_PROC_FS -struct irttp_iter_state { - int id; -}; - -static void *irttp_seq_start(struct seq_file *seq, loff_t *pos) -{ - struct irttp_iter_state *iter = seq->private; - struct tsap_cb *self; - - /* Protect our access to the tsap list */ - spin_lock_irq(&irttp->tsaps->hb_spinlock); - iter->id = 0; - - for (self = (struct tsap_cb *) hashbin_get_first(irttp->tsaps); - self != NULL; - self = (struct tsap_cb *) hashbin_get_next(irttp->tsaps)) { - if (iter->id == *pos) - break; - ++iter->id; - } - - return self; -} - -static void *irttp_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - struct irttp_iter_state *iter = seq->private; - - ++*pos; - ++iter->id; - return (void *) hashbin_get_next(irttp->tsaps); -} - -static void irttp_seq_stop(struct seq_file *seq, void *v) -{ - spin_unlock_irq(&irttp->tsaps->hb_spinlock); -} - -static int irttp_seq_show(struct seq_file *seq, void *v) -{ - const struct irttp_iter_state *iter = seq->private; - const struct tsap_cb *self = v; - - seq_printf(seq, "TSAP %d, ", iter->id); - seq_printf(seq, "stsap_sel: %02x, ", - self->stsap_sel); - seq_printf(seq, "dtsap_sel: %02x\n", - self->dtsap_sel); - seq_printf(seq, " connected: %s, ", - self->connected ? "TRUE" : "FALSE"); - seq_printf(seq, "avail credit: %d, ", - self->avail_credit); - seq_printf(seq, "remote credit: %d, ", - self->remote_credit); - seq_printf(seq, "send credit: %d\n", - self->send_credit); - seq_printf(seq, " tx packets: %lu, ", - self->stats.tx_packets); - seq_printf(seq, "rx packets: %lu, ", - self->stats.rx_packets); - seq_printf(seq, "tx_queue len: %u ", - skb_queue_len(&self->tx_queue)); - seq_printf(seq, "rx_queue len: %u\n", - skb_queue_len(&self->rx_queue)); - seq_printf(seq, " tx_sdu_busy: %s, ", - self->tx_sdu_busy ? "TRUE" : "FALSE"); - seq_printf(seq, "rx_sdu_busy: %s\n", - self->rx_sdu_busy ? "TRUE" : "FALSE"); - seq_printf(seq, " max_seg_size: %u, ", - self->max_seg_size); - seq_printf(seq, "tx_max_sdu_size: %u, ", - self->tx_max_sdu_size); - seq_printf(seq, "rx_max_sdu_size: %u\n", - self->rx_max_sdu_size); - - seq_printf(seq, " Used by (%s)\n\n", - self->notify.name); - return 0; -} - -static const struct seq_operations irttp_seq_ops = { - .start = irttp_seq_start, - .next = irttp_seq_next, - .stop = irttp_seq_stop, - .show = irttp_seq_show, -}; - -static int irttp_seq_open(struct inode *inode, struct file *file) -{ - return seq_open_private(file, &irttp_seq_ops, - sizeof(struct irttp_iter_state)); -} - -const struct file_operations irttp_seq_fops = { - .owner = THIS_MODULE, - .open = irttp_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, -}; - -#endif /* PROC_FS */ diff --git a/net/irda/parameters.c b/net/irda/parameters.c deleted file mode 100644 index 16ce32ffe004..000000000000 --- a/net/irda/parameters.c +++ /dev/null @@ -1,584 +0,0 @@ -/********************************************************************* - * - * Filename: parameters.c - * Version: 1.0 - * Description: A more general way to handle (pi,pl,pv) parameters - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Jun 7 10:25:11 1999 - * Modified at: Sun Jan 30 14:08:39 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1999-2000 Dag Brattli, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/types.h> -#include <linux/module.h> - -#include <asm/unaligned.h> -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/parameters.h> - -static int irda_extract_integer(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); -static int irda_extract_string(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); -static int irda_extract_octseq(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); -static int irda_extract_no_value(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); - -static int irda_insert_integer(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); -static int irda_insert_no_value(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func); - -static int irda_param_unpack(__u8 *buf, char *fmt, ...); - -/* Parameter value call table. Must match PV_TYPE */ -static const PV_HANDLER pv_extract_table[] = { - irda_extract_integer, /* Handler for any length integers */ - irda_extract_integer, /* Handler for 8 bits integers */ - irda_extract_integer, /* Handler for 16 bits integers */ - irda_extract_string, /* Handler for strings */ - irda_extract_integer, /* Handler for 32 bits integers */ - irda_extract_octseq, /* Handler for octet sequences */ - irda_extract_no_value /* Handler for no value parameters */ -}; - -static const PV_HANDLER pv_insert_table[] = { - irda_insert_integer, /* Handler for any length integers */ - irda_insert_integer, /* Handler for 8 bits integers */ - irda_insert_integer, /* Handler for 16 bits integers */ - NULL, /* Handler for strings */ - irda_insert_integer, /* Handler for 32 bits integers */ - NULL, /* Handler for octet sequences */ - irda_insert_no_value /* Handler for no value parameters */ -}; - -/* - * Function irda_insert_no_value (self, buf, len, pi, type, func) - */ -static int irda_insert_no_value(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - irda_param_t p; - int ret; - - p.pi = pi; - p.pl = 0; - - /* Call handler for this parameter */ - ret = (*func)(self, &p, PV_GET); - - /* Extract values anyway, since handler may need them */ - irda_param_pack(buf, "bb", p.pi, p.pl); - - if (ret < 0) - return ret; - - return 2; /* Inserted pl+2 bytes */ -} - -/* - * Function irda_extract_no_value (self, buf, len, type, func) - * - * Extracts a parameter without a pv field (pl=0) - * - */ -static int irda_extract_no_value(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - irda_param_t p; - int ret; - - /* Extract values anyway, since handler may need them */ - irda_param_unpack(buf, "bb", &p.pi, &p.pl); - - /* Call handler for this parameter */ - ret = (*func)(self, &p, PV_PUT); - - if (ret < 0) - return ret; - - return 2; /* Extracted pl+2 bytes */ -} - -/* - * Function irda_insert_integer (self, buf, len, pi, type, func) - */ -static int irda_insert_integer(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - irda_param_t p; - int n = 0; - int err; - - p.pi = pi; /* In case handler needs to know */ - p.pl = type & PV_MASK; /* The integer type codes the length as well */ - p.pv.i = 0; /* Clear value */ - - /* Call handler for this parameter */ - err = (*func)(self, &p, PV_GET); - if (err < 0) - return err; - - /* - * If parameter length is still 0, then (1) this is an any length - * integer, and (2) the handler function does not care which length - * we choose to use, so we pick the one the gives the fewest bytes. - */ - if (p.pl == 0) { - if (p.pv.i < 0xff) { - pr_debug("%s(), using 1 byte\n", __func__); - p.pl = 1; - } else if (p.pv.i < 0xffff) { - pr_debug("%s(), using 2 bytes\n", __func__); - p.pl = 2; - } else { - pr_debug("%s(), using 4 bytes\n", __func__); - p.pl = 4; /* Default length */ - } - } - /* Check if buffer is long enough for insertion */ - if (len < (2+p.pl)) { - net_warn_ratelimited("%s: buffer too short for insertion!\n", - __func__); - return -1; - } - pr_debug("%s(), pi=%#x, pl=%d, pi=%d\n", __func__, - p.pi, p.pl, p.pv.i); - switch (p.pl) { - case 1: - n += irda_param_pack(buf, "bbb", p.pi, p.pl, (__u8) p.pv.i); - break; - case 2: - if (type & PV_BIG_ENDIAN) - p.pv.i = cpu_to_be16((__u16) p.pv.i); - else - p.pv.i = cpu_to_le16((__u16) p.pv.i); - n += irda_param_pack(buf, "bbs", p.pi, p.pl, (__u16) p.pv.i); - break; - case 4: - if (type & PV_BIG_ENDIAN) - cpu_to_be32s(&p.pv.i); - else - cpu_to_le32s(&p.pv.i); - n += irda_param_pack(buf, "bbi", p.pi, p.pl, p.pv.i); - - break; - default: - net_warn_ratelimited("%s: length %d not supported\n", - __func__, p.pl); - /* Skip parameter */ - return -1; - } - - return p.pl+2; /* Inserted pl+2 bytes */ -} - -/* - * Function irda_extract integer (self, buf, len, pi, type, func) - * - * Extract a possibly variable length integer from buffer, and call - * handler for processing of the parameter - */ -static int irda_extract_integer(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - irda_param_t p; - int n = 0; - int extract_len; /* Real length we extract */ - int err; - - p.pi = pi; /* In case handler needs to know */ - p.pl = buf[1]; /* Extract length of value */ - p.pv.i = 0; /* Clear value */ - extract_len = p.pl; /* Default : extract all */ - - /* Check if buffer is long enough for parsing */ - if (len < (2+p.pl)) { - net_warn_ratelimited("%s: buffer too short for parsing! Need %d bytes, but len is only %d\n", - __func__, p.pl, len); - return -1; - } - - /* - * Check that the integer length is what we expect it to be. If the - * handler want a 16 bits integer then a 32 bits is not good enough - * PV_INTEGER means that the handler is flexible. - */ - if (((type & PV_MASK) != PV_INTEGER) && ((type & PV_MASK) != p.pl)) { - net_err_ratelimited("%s: invalid parameter length! Expected %d bytes, but value had %d bytes!\n", - __func__, type & PV_MASK, p.pl); - - /* Most parameters are bit/byte fields or little endian, - * so it's ok to only extract a subset of it (the subset - * that the handler expect). This is necessary, as some - * broken implementations seems to add extra undefined bits. - * If the parameter is shorter than we expect or is big - * endian, we can't play those tricks. Jean II */ - if((p.pl < (type & PV_MASK)) || (type & PV_BIG_ENDIAN)) { - /* Skip parameter */ - return p.pl+2; - } else { - /* Extract subset of it, fallthrough */ - extract_len = type & PV_MASK; - } - } - - - switch (extract_len) { - case 1: - n += irda_param_unpack(buf+2, "b", &p.pv.i); - break; - case 2: - n += irda_param_unpack(buf+2, "s", &p.pv.i); - if (type & PV_BIG_ENDIAN) - p.pv.i = be16_to_cpu((__u16) p.pv.i); - else - p.pv.i = le16_to_cpu((__u16) p.pv.i); - break; - case 4: - n += irda_param_unpack(buf+2, "i", &p.pv.i); - if (type & PV_BIG_ENDIAN) - be32_to_cpus(&p.pv.i); - else - le32_to_cpus(&p.pv.i); - break; - default: - net_warn_ratelimited("%s: length %d not supported\n", - __func__, p.pl); - - /* Skip parameter */ - return p.pl+2; - } - - pr_debug("%s(), pi=%#x, pl=%d, pi=%d\n", __func__, - p.pi, p.pl, p.pv.i); - /* Call handler for this parameter */ - err = (*func)(self, &p, PV_PUT); - if (err < 0) - return err; - - return p.pl+2; /* Extracted pl+2 bytes */ -} - -/* - * Function irda_extract_string (self, buf, len, type, func) - */ -static int irda_extract_string(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - char str[33]; - irda_param_t p; - int err; - - p.pi = pi; /* In case handler needs to know */ - p.pl = buf[1]; /* Extract length of value */ - if (p.pl > 32) - p.pl = 32; - - pr_debug("%s(), pi=%#x, pl=%d\n", __func__, - p.pi, p.pl); - - /* Check if buffer is long enough for parsing */ - if (len < (2+p.pl)) { - net_warn_ratelimited("%s: buffer too short for parsing! Need %d bytes, but len is only %d\n", - __func__, p.pl, len); - return -1; - } - - /* Should be safe to copy string like this since we have already - * checked that the buffer is long enough */ - strncpy(str, buf+2, p.pl); - - pr_debug("%s(), str=0x%02x 0x%02x\n", - __func__, (__u8)str[0], (__u8)str[1]); - - /* Null terminate string */ - str[p.pl] = '\0'; - - p.pv.c = str; /* Handler will need to take a copy */ - - /* Call handler for this parameter */ - err = (*func)(self, &p, PV_PUT); - if (err < 0) - return err; - - return p.pl+2; /* Extracted pl+2 bytes */ -} - -/* - * Function irda_extract_octseq (self, buf, len, type, func) - */ -static int irda_extract_octseq(void *self, __u8 *buf, int len, __u8 pi, - PV_TYPE type, PI_HANDLER func) -{ - irda_param_t p; - - p.pi = pi; /* In case handler needs to know */ - p.pl = buf[1]; /* Extract length of value */ - - /* Check if buffer is long enough for parsing */ - if (len < (2+p.pl)) { - net_warn_ratelimited("%s: buffer too short for parsing! Need %d bytes, but len is only %d\n", - __func__, p.pl, len); - return -1; - } - - pr_debug("%s(), not impl\n", __func__); - - return p.pl+2; /* Extracted pl+2 bytes */ -} - -/* - * Function irda_param_pack (skb, fmt, ...) - * - * Format: - * 'i' = 32 bits integer - * 's' = string - * - */ -int irda_param_pack(__u8 *buf, char *fmt, ...) -{ - irda_pv_t arg; - va_list args; - char *p; - int n = 0; - - va_start(args, fmt); - - for (p = fmt; *p != '\0'; p++) { - switch (*p) { - case 'b': /* 8 bits unsigned byte */ - buf[n++] = (__u8)va_arg(args, int); - break; - case 's': /* 16 bits unsigned short */ - arg.i = (__u16)va_arg(args, int); - put_unaligned((__u16)arg.i, (__u16 *)(buf+n)); n+=2; - break; - case 'i': /* 32 bits unsigned integer */ - arg.i = va_arg(args, __u32); - put_unaligned(arg.i, (__u32 *)(buf+n)); n+=4; - break; -#if 0 - case 'c': /* \0 terminated string */ - arg.c = va_arg(args, char *); - strcpy(buf+n, arg.c); - n += strlen(arg.c) + 1; - break; -#endif - default: - va_end(args); - return -1; - } - } - va_end(args); - - return 0; -} -EXPORT_SYMBOL(irda_param_pack); - -/* - * Function irda_param_unpack (skb, fmt, ...) - */ -static int irda_param_unpack(__u8 *buf, char *fmt, ...) -{ - irda_pv_t arg; - va_list args; - char *p; - int n = 0; - - va_start(args, fmt); - - for (p = fmt; *p != '\0'; p++) { - switch (*p) { - case 'b': /* 8 bits byte */ - arg.ip = va_arg(args, __u32 *); - *arg.ip = buf[n++]; - break; - case 's': /* 16 bits short */ - arg.ip = va_arg(args, __u32 *); - *arg.ip = get_unaligned((__u16 *)(buf+n)); n+=2; - break; - case 'i': /* 32 bits unsigned integer */ - arg.ip = va_arg(args, __u32 *); - *arg.ip = get_unaligned((__u32 *)(buf+n)); n+=4; - break; -#if 0 - case 'c': /* \0 terminated string */ - arg.c = va_arg(args, char *); - strcpy(arg.c, buf+n); - n += strlen(arg.c) + 1; - break; -#endif - default: - va_end(args); - return -1; - } - - } - va_end(args); - - return 0; -} - -/* - * Function irda_param_insert (self, pi, buf, len, info) - * - * Insert the specified parameter (pi) into buffer. Returns number of - * bytes inserted - */ -int irda_param_insert(void *self, __u8 pi, __u8 *buf, int len, - pi_param_info_t *info) -{ - const pi_minor_info_t *pi_minor_info; - __u8 pi_minor; - __u8 pi_major; - int type; - int ret = -1; - int n = 0; - - IRDA_ASSERT(buf != NULL, return ret;); - IRDA_ASSERT(info != NULL, return ret;); - - pi_minor = pi & info->pi_mask; - pi_major = pi >> info->pi_major_offset; - - /* Check if the identifier value (pi) is valid */ - if ((pi_major > info->len-1) || - (pi_minor > info->tables[pi_major].len-1)) - { - pr_debug("%s(), no handler for parameter=0x%02x\n", - __func__, pi); - - /* Skip this parameter */ - return -1; - } - - /* Lookup the info on how to parse this parameter */ - pi_minor_info = &info->tables[pi_major].pi_minor_call_table[pi_minor]; - - /* Find expected data type for this parameter identifier (pi)*/ - type = pi_minor_info->type; - - /* Check if handler has been implemented */ - if (!pi_minor_info->func) { - net_info_ratelimited("%s: no handler for pi=%#x\n", - __func__, pi); - /* Skip this parameter */ - return -1; - } - - /* Insert parameter value */ - ret = (*pv_insert_table[type & PV_MASK])(self, buf+n, len, pi, type, - pi_minor_info->func); - return ret; -} -EXPORT_SYMBOL(irda_param_insert); - -/* - * Function irda_param_extract (self, buf, len, info) - * - * Parse all parameters. If len is correct, then everything should be - * safe. Returns the number of bytes that was parsed - * - */ -static int irda_param_extract(void *self, __u8 *buf, int len, - pi_param_info_t *info) -{ - const pi_minor_info_t *pi_minor_info; - __u8 pi_minor; - __u8 pi_major; - int type; - int ret = -1; - int n = 0; - - IRDA_ASSERT(buf != NULL, return ret;); - IRDA_ASSERT(info != NULL, return ret;); - - pi_minor = buf[n] & info->pi_mask; - pi_major = buf[n] >> info->pi_major_offset; - - /* Check if the identifier value (pi) is valid */ - if ((pi_major > info->len-1) || - (pi_minor > info->tables[pi_major].len-1)) - { - pr_debug("%s(), no handler for parameter=0x%02x\n", - __func__, buf[0]); - - /* Skip this parameter */ - return 2 + buf[n + 1]; /* Continue */ - } - - /* Lookup the info on how to parse this parameter */ - pi_minor_info = &info->tables[pi_major].pi_minor_call_table[pi_minor]; - - /* Find expected data type for this parameter identifier (pi)*/ - type = pi_minor_info->type; - - pr_debug("%s(), pi=[%d,%d], type=%d\n", __func__, - pi_major, pi_minor, type); - - /* Check if handler has been implemented */ - if (!pi_minor_info->func) { - net_info_ratelimited("%s: no handler for pi=%#x\n", - __func__, buf[n]); - /* Skip this parameter */ - return 2 + buf[n + 1]; /* Continue */ - } - - /* Parse parameter value */ - ret = (*pv_extract_table[type & PV_MASK])(self, buf+n, len, buf[n], - type, pi_minor_info->func); - return ret; -} - -/* - * Function irda_param_extract_all (self, buf, len, info) - * - * Parse all parameters. If len is correct, then everything should be - * safe. Returns the number of bytes that was parsed - * - */ -int irda_param_extract_all(void *self, __u8 *buf, int len, - pi_param_info_t *info) -{ - int ret = -1; - int n = 0; - - IRDA_ASSERT(buf != NULL, return ret;); - IRDA_ASSERT(info != NULL, return ret;); - - /* - * Parse all parameters. Each parameter must be at least two bytes - * long or else there is no point in trying to parse it - */ - while (len > 2) { - ret = irda_param_extract(self, buf+n, len, info); - if (ret < 0) - return ret; - - n += ret; - len -= ret; - } - return n; -} -EXPORT_SYMBOL(irda_param_extract_all); diff --git a/net/irda/qos.c b/net/irda/qos.c deleted file mode 100644 index 25ba8509ad3e..000000000000 --- a/net/irda/qos.c +++ /dev/null @@ -1,771 +0,0 @@ -/********************************************************************* - * - * Filename: qos.c - * Version: 1.0 - * Description: IrLAP QoS parameter negotiation - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Tue Sep 9 00:00:26 1997 - * Modified at: Sun Jan 30 14:29:16 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1998-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2001 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - * - ********************************************************************/ - -#include <linux/export.h> - -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/parameters.h> -#include <net/irda/qos.h> -#include <net/irda/irlap.h> -#include <net/irda/irlap_frame.h> - -/* - * Maximum values of the baud rate we negotiate with the other end. - * Most often, you don't have to change that, because Linux-IrDA will - * use the maximum offered by the link layer, which usually works fine. - * In some very rare cases, you may want to limit it to lower speeds... - */ -int sysctl_max_baud_rate = 16000000; -/* - * Maximum value of the lap disconnect timer we negotiate with the other end. - * Most often, the value below represent the best compromise, but some user - * may want to keep the LAP alive longer or shorter in case of link failure. - * Remember that the threshold time (early warning) is fixed to 3s... - */ -int sysctl_max_noreply_time = 12; -/* - * Minimum turn time to be applied before transmitting to the peer. - * Nonzero values (usec) are used as lower limit to the per-connection - * mtt value which was announced by the other end during negotiation. - * Might be helpful if the peer device provides too short mtt. - * Default is 10us which means using the unmodified value given by the - * peer except if it's 0 (0 is likely a bug in the other stack). - */ -unsigned int sysctl_min_tx_turn_time = 10; -/* - * Maximum data size to be used in transmission in payload of LAP frame. - * There is a bit of confusion in the IrDA spec : - * The LAP spec defines the payload of a LAP frame (I field) to be - * 2048 bytes max (IrLAP 1.1, chapt 6.6.5, p40). - * On the other hand, the PHY mention frames of 2048 bytes max (IrPHY - * 1.2, chapt 5.3.2.1, p41). But, this number includes the LAP header - * (2 bytes), and CRC (32 bits at 4 Mb/s). So, for the I field (LAP - * payload), that's only 2042 bytes. Oups ! - * My nsc-ircc hardware has troubles receiving 2048 bytes frames at 4 Mb/s, - * so adjust to 2042... I don't know if this bug applies only for 2048 - * bytes frames or all negotiated frame sizes, but you can use the sysctl - * to play with this value anyway. - * Jean II */ -unsigned int sysctl_max_tx_data_size = 2042; -/* - * Maximum transmit window, i.e. number of LAP frames between turn-around. - * This allow to override what the peer told us. Some peers are buggy and - * don't always support what they tell us. - * Jean II */ -unsigned int sysctl_max_tx_window = 7; - -static int irlap_param_baud_rate(void *instance, irda_param_t *param, int get); -static int irlap_param_link_disconnect(void *instance, irda_param_t *parm, - int get); -static int irlap_param_max_turn_time(void *instance, irda_param_t *param, - int get); -static int irlap_param_data_size(void *instance, irda_param_t *param, int get); -static int irlap_param_window_size(void *instance, irda_param_t *param, - int get); -static int irlap_param_additional_bofs(void *instance, irda_param_t *parm, - int get); -static int irlap_param_min_turn_time(void *instance, irda_param_t *param, - int get); - -#ifndef CONFIG_IRDA_DYNAMIC_WINDOW -static __u32 irlap_requested_line_capacity(struct qos_info *qos); -#endif - -static __u32 min_turn_times[] = { 10000, 5000, 1000, 500, 100, 50, 10, 0 }; /* us */ -static __u32 baud_rates[] = { 2400, 9600, 19200, 38400, 57600, 115200, 576000, - 1152000, 4000000, 16000000 }; /* bps */ -static __u32 data_sizes[] = { 64, 128, 256, 512, 1024, 2048 }; /* bytes */ -static __u32 add_bofs[] = { 48, 24, 12, 5, 3, 2, 1, 0 }; /* bytes */ -static __u32 max_turn_times[] = { 500, 250, 100, 50 }; /* ms */ -static __u32 link_disc_times[] = { 3, 8, 12, 16, 20, 25, 30, 40 }; /* secs */ - -static __u32 max_line_capacities[10][4] = { - /* 500 ms 250 ms 100 ms 50 ms (max turn time) */ - { 100, 0, 0, 0 }, /* 2400 bps */ - { 400, 0, 0, 0 }, /* 9600 bps */ - { 800, 0, 0, 0 }, /* 19200 bps */ - { 1600, 0, 0, 0 }, /* 38400 bps */ - { 2360, 0, 0, 0 }, /* 57600 bps */ - { 4800, 2400, 960, 480 }, /* 115200 bps */ - { 28800, 11520, 5760, 2880 }, /* 576000 bps */ - { 57600, 28800, 11520, 5760 }, /* 1152000 bps */ - { 200000, 100000, 40000, 20000 }, /* 4000000 bps */ - { 800000, 400000, 160000, 80000 }, /* 16000000 bps */ -}; - -static const pi_minor_info_t pi_minor_call_table_type_0[] = { - { NULL, 0 }, -/* 01 */{ irlap_param_baud_rate, PV_INTEGER | PV_LITTLE_ENDIAN }, - { NULL, 0 }, - { NULL, 0 }, - { NULL, 0 }, - { NULL, 0 }, - { NULL, 0 }, - { NULL, 0 }, -/* 08 */{ irlap_param_link_disconnect, PV_INT_8_BITS } -}; - -static const pi_minor_info_t pi_minor_call_table_type_1[] = { - { NULL, 0 }, - { NULL, 0 }, -/* 82 */{ irlap_param_max_turn_time, PV_INT_8_BITS }, -/* 83 */{ irlap_param_data_size, PV_INT_8_BITS }, -/* 84 */{ irlap_param_window_size, PV_INT_8_BITS }, -/* 85 */{ irlap_param_additional_bofs, PV_INT_8_BITS }, -/* 86 */{ irlap_param_min_turn_time, PV_INT_8_BITS }, -}; - -static const pi_major_info_t pi_major_call_table[] = { - { pi_minor_call_table_type_0, 9 }, - { pi_minor_call_table_type_1, 7 }, -}; - -static pi_param_info_t irlap_param_info = { pi_major_call_table, 2, 0x7f, 7 }; - -/* ---------------------- LOCAL SUBROUTINES ---------------------- */ -/* Note : we start with a bunch of local subroutines. - * As the compiler is "one pass", this is the only way to get them to - * inline properly... - * Jean II - */ -/* - * Function value_index (value, array, size) - * - * Returns the index to the value in the specified array - */ -static inline int value_index(__u32 value, __u32 *array, int size) -{ - int i; - - for (i=0; i < size; i++) - if (array[i] == value) - break; - return i; -} - -/* - * Function index_value (index, array) - * - * Returns value to index in array, easy! - * - */ -static inline __u32 index_value(int index, __u32 *array) -{ - return array[index]; -} - -/* - * Function msb_index (word) - * - * Returns index to most significant bit (MSB) in word - * - */ -static int msb_index (__u16 word) -{ - __u16 msb = 0x8000; - int index = 15; /* Current MSB */ - - /* Check for buggy peers. - * Note : there is a small probability that it could be us, but I - * would expect driver authors to catch that pretty early and be - * able to check precisely what's going on. If a end user sees this, - * it's very likely the peer. - Jean II */ - if (word == 0) { - net_warn_ratelimited("%s(), Detected buggy peer, adjust null PV to 0x1!\n", - __func__); - /* The only safe choice (we don't know the array size) */ - word = 0x1; - } - - while (msb) { - if (word & msb) - break; /* Found it! */ - msb >>=1; - index--; - } - return index; -} - -/* - * Function value_lower_bits (value, array) - * - * Returns a bit field marking all possibility lower than value. - */ -static inline int value_lower_bits(__u32 value, __u32 *array, int size, __u16 *field) -{ - int i; - __u16 mask = 0x1; - __u16 result = 0x0; - - for (i=0; i < size; i++) { - /* Add the current value to the bit field, shift mask */ - result |= mask; - mask <<= 1; - /* Finished ? */ - if (array[i] >= value) - break; - } - /* Send back a valid index */ - if(i >= size) - i = size - 1; /* Last item */ - *field = result; - return i; -} - -/* - * Function value_highest_bit (value, array) - * - * Returns a bit field marking the highest possibility lower than value. - */ -static inline int value_highest_bit(__u32 value, __u32 *array, int size, __u16 *field) -{ - int i; - __u16 mask = 0x1; - __u16 result = 0x0; - - for (i=0; i < size; i++) { - /* Finished ? */ - if (array[i] <= value) - break; - /* Shift mask */ - mask <<= 1; - } - /* Set the current value to the bit field */ - result |= mask; - /* Send back a valid index */ - if(i >= size) - i = size - 1; /* Last item */ - *field = result; - return i; -} - -/* -------------------------- MAIN CALLS -------------------------- */ - -/* - * Function irda_qos_compute_intersection (qos, new) - * - * Compute the intersection of the old QoS capabilities with new ones - * - */ -void irda_qos_compute_intersection(struct qos_info *qos, struct qos_info *new) -{ - IRDA_ASSERT(qos != NULL, return;); - IRDA_ASSERT(new != NULL, return;); - - /* Apply */ - qos->baud_rate.bits &= new->baud_rate.bits; - qos->window_size.bits &= new->window_size.bits; - qos->min_turn_time.bits &= new->min_turn_time.bits; - qos->max_turn_time.bits &= new->max_turn_time.bits; - qos->data_size.bits &= new->data_size.bits; - qos->link_disc_time.bits &= new->link_disc_time.bits; - qos->additional_bofs.bits &= new->additional_bofs.bits; - - irda_qos_bits_to_value(qos); -} - -/* - * Function irda_init_max_qos_capabilies (qos) - * - * The purpose of this function is for layers and drivers to be able to - * set the maximum QoS possible and then "and in" their own limitations - * - */ -void irda_init_max_qos_capabilies(struct qos_info *qos) -{ - int i; - /* - * These are the maximum supported values as specified on pages - * 39-43 in IrLAP - */ - - /* Use sysctl to set some configurable values... */ - /* Set configured max speed */ - i = value_lower_bits(sysctl_max_baud_rate, baud_rates, 10, - &qos->baud_rate.bits); - sysctl_max_baud_rate = index_value(i, baud_rates); - - /* Set configured max disc time */ - i = value_lower_bits(sysctl_max_noreply_time, link_disc_times, 8, - &qos->link_disc_time.bits); - sysctl_max_noreply_time = index_value(i, link_disc_times); - - /* LSB is first byte, MSB is second byte */ - qos->baud_rate.bits &= 0x03ff; - - qos->window_size.bits = 0x7f; - qos->min_turn_time.bits = 0xff; - qos->max_turn_time.bits = 0x0f; - qos->data_size.bits = 0x3f; - qos->link_disc_time.bits &= 0xff; - qos->additional_bofs.bits = 0xff; -} -EXPORT_SYMBOL(irda_init_max_qos_capabilies); - -/* - * Function irlap_adjust_qos_settings (qos) - * - * Adjust QoS settings in case some values are not possible to use because - * of other settings - */ -static void irlap_adjust_qos_settings(struct qos_info *qos) -{ - __u32 line_capacity; - int index; - - /* - * Make sure the mintt is sensible. - * Main culprit : Ericsson T39. - Jean II - */ - if (sysctl_min_tx_turn_time > qos->min_turn_time.value) { - int i; - - net_warn_ratelimited("%s(), Detected buggy peer, adjust mtt to %dus!\n", - __func__, sysctl_min_tx_turn_time); - - /* We don't really need bits, but easier this way */ - i = value_highest_bit(sysctl_min_tx_turn_time, min_turn_times, - 8, &qos->min_turn_time.bits); - sysctl_min_tx_turn_time = index_value(i, min_turn_times); - qos->min_turn_time.value = sysctl_min_tx_turn_time; - } - - /* - * Not allowed to use a max turn time less than 500 ms if the baudrate - * is less than 115200 - */ - if ((qos->baud_rate.value < 115200) && - (qos->max_turn_time.value < 500)) - { - pr_debug("%s(), adjusting max turn time from %d to 500 ms\n", - __func__, qos->max_turn_time.value); - qos->max_turn_time.value = 500; - } - - /* - * The data size must be adjusted according to the baud rate and max - * turn time - */ - index = value_index(qos->data_size.value, data_sizes, 6); - line_capacity = irlap_max_line_capacity(qos->baud_rate.value, - qos->max_turn_time.value); - -#ifdef CONFIG_IRDA_DYNAMIC_WINDOW - while ((qos->data_size.value > line_capacity) && (index > 0)) { - qos->data_size.value = data_sizes[index--]; - pr_debug("%s(), reducing data size to %d\n", - __func__, qos->data_size.value); - } -#else /* Use method described in section 6.6.11 of IrLAP */ - while (irlap_requested_line_capacity(qos) > line_capacity) { - IRDA_ASSERT(index != 0, return;); - - /* Must be able to send at least one frame */ - if (qos->window_size.value > 1) { - qos->window_size.value--; - pr_debug("%s(), reducing window size to %d\n", - __func__, qos->window_size.value); - } else if (index > 1) { - qos->data_size.value = data_sizes[index--]; - pr_debug("%s(), reducing data size to %d\n", - __func__, qos->data_size.value); - } else { - net_warn_ratelimited("%s(), nothing more we can do!\n", - __func__); - } - } -#endif /* CONFIG_IRDA_DYNAMIC_WINDOW */ - /* - * Fix tx data size according to user limits - Jean II - */ - if (qos->data_size.value > sysctl_max_tx_data_size) - /* Allow non discrete adjustement to avoid losing capacity */ - qos->data_size.value = sysctl_max_tx_data_size; - /* - * Override Tx window if user request it. - Jean II - */ - if (qos->window_size.value > sysctl_max_tx_window) - qos->window_size.value = sysctl_max_tx_window; -} - -/* - * Function irlap_negotiate (qos_device, qos_session, skb) - * - * Negotiate QoS values, not really that much negotiation :-) - * We just set the QoS capabilities for the peer station - * - */ -int irlap_qos_negotiate(struct irlap_cb *self, struct sk_buff *skb) -{ - int ret; - - ret = irda_param_extract_all(self, skb->data, skb->len, - &irlap_param_info); - - /* Convert the negotiated bits to values */ - irda_qos_bits_to_value(&self->qos_tx); - irda_qos_bits_to_value(&self->qos_rx); - - irlap_adjust_qos_settings(&self->qos_tx); - - pr_debug("Setting BAUD_RATE to %d bps.\n", - self->qos_tx.baud_rate.value); - pr_debug("Setting DATA_SIZE to %d bytes\n", - self->qos_tx.data_size.value); - pr_debug("Setting WINDOW_SIZE to %d\n", - self->qos_tx.window_size.value); - pr_debug("Setting XBOFS to %d\n", - self->qos_tx.additional_bofs.value); - pr_debug("Setting MAX_TURN_TIME to %d ms.\n", - self->qos_tx.max_turn_time.value); - pr_debug("Setting MIN_TURN_TIME to %d usecs.\n", - self->qos_tx.min_turn_time.value); - pr_debug("Setting LINK_DISC to %d secs.\n", - self->qos_tx.link_disc_time.value); - return ret; -} - -/* - * Function irlap_insert_negotiation_params (qos, fp) - * - * Insert QoS negotiaion pararameters into frame - * - */ -int irlap_insert_qos_negotiation_params(struct irlap_cb *self, - struct sk_buff *skb) -{ - int ret; - - /* Insert data rate */ - ret = irda_param_insert(self, PI_BAUD_RATE, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert max turnaround time */ - ret = irda_param_insert(self, PI_MAX_TURN_TIME, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert data size */ - ret = irda_param_insert(self, PI_DATA_SIZE, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert window size */ - ret = irda_param_insert(self, PI_WINDOW_SIZE, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert additional BOFs */ - ret = irda_param_insert(self, PI_ADD_BOFS, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert minimum turnaround time */ - ret = irda_param_insert(self, PI_MIN_TURN_TIME, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - /* Insert link disconnect/threshold time */ - ret = irda_param_insert(self, PI_LINK_DISC, skb_tail_pointer(skb), - skb_tailroom(skb), &irlap_param_info); - if (ret < 0) - return ret; - skb_put(skb, ret); - - return 0; -} - -/* - * Function irlap_param_baud_rate (instance, param, get) - * - * Negotiate data-rate - * - */ -static int irlap_param_baud_rate(void *instance, irda_param_t *param, int get) -{ - __u16 final; - - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) { - param->pv.i = self->qos_rx.baud_rate.bits; - pr_debug("%s(), baud rate = 0x%02x\n", - __func__, param->pv.i); - } else { - /* - * Stations must agree on baud rate, so calculate - * intersection - */ - pr_debug("Requested BAUD_RATE: 0x%04x\n", (__u16)param->pv.i); - final = (__u16) param->pv.i & self->qos_rx.baud_rate.bits; - - pr_debug("Final BAUD_RATE: 0x%04x\n", final); - self->qos_tx.baud_rate.bits = final; - self->qos_rx.baud_rate.bits = final; - } - - return 0; -} - -/* - * Function irlap_param_link_disconnect (instance, param, get) - * - * Negotiate link disconnect/threshold time. - * - */ -static int irlap_param_link_disconnect(void *instance, irda_param_t *param, - int get) -{ - __u16 final; - - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.link_disc_time.bits; - else { - /* - * Stations must agree on link disconnect/threshold - * time. - */ - pr_debug("LINK_DISC: %02x\n", (__u8)param->pv.i); - final = (__u8) param->pv.i & self->qos_rx.link_disc_time.bits; - - pr_debug("Final LINK_DISC: %02x\n", final); - self->qos_tx.link_disc_time.bits = final; - self->qos_rx.link_disc_time.bits = final; - } - return 0; -} - -/* - * Function irlap_param_max_turn_time (instance, param, get) - * - * Negotiate the maximum turnaround time. This is a type 1 parameter and - * will be negotiated independently for each station - * - */ -static int irlap_param_max_turn_time(void *instance, irda_param_t *param, - int get) -{ - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.max_turn_time.bits; - else - self->qos_tx.max_turn_time.bits = (__u8) param->pv.i; - - return 0; -} - -/* - * Function irlap_param_data_size (instance, param, get) - * - * Negotiate the data size. This is a type 1 parameter and - * will be negotiated independently for each station - * - */ -static int irlap_param_data_size(void *instance, irda_param_t *param, int get) -{ - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.data_size.bits; - else - self->qos_tx.data_size.bits = (__u8) param->pv.i; - - return 0; -} - -/* - * Function irlap_param_window_size (instance, param, get) - * - * Negotiate the window size. This is a type 1 parameter and - * will be negotiated independently for each station - * - */ -static int irlap_param_window_size(void *instance, irda_param_t *param, - int get) -{ - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.window_size.bits; - else - self->qos_tx.window_size.bits = (__u8) param->pv.i; - - return 0; -} - -/* - * Function irlap_param_additional_bofs (instance, param, get) - * - * Negotiate additional BOF characters. This is a type 1 parameter and - * will be negotiated independently for each station. - */ -static int irlap_param_additional_bofs(void *instance, irda_param_t *param, int get) -{ - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.additional_bofs.bits; - else - self->qos_tx.additional_bofs.bits = (__u8) param->pv.i; - - return 0; -} - -/* - * Function irlap_param_min_turn_time (instance, param, get) - * - * Negotiate the minimum turn around time. This is a type 1 parameter and - * will be negotiated independently for each station - */ -static int irlap_param_min_turn_time(void *instance, irda_param_t *param, - int get) -{ - struct irlap_cb *self = (struct irlap_cb *) instance; - - IRDA_ASSERT(self != NULL, return -1;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return -1;); - - if (get) - param->pv.i = self->qos_rx.min_turn_time.bits; - else - self->qos_tx.min_turn_time.bits = (__u8) param->pv.i; - - return 0; -} - -/* - * Function irlap_max_line_capacity (speed, max_turn_time, min_turn_time) - * - * Calculate the maximum line capacity - * - */ -__u32 irlap_max_line_capacity(__u32 speed, __u32 max_turn_time) -{ - __u32 line_capacity; - int i,j; - - pr_debug("%s(), speed=%d, max_turn_time=%d\n", - __func__, speed, max_turn_time); - - i = value_index(speed, baud_rates, 10); - j = value_index(max_turn_time, max_turn_times, 4); - - IRDA_ASSERT(((i >=0) && (i <10)), return 0;); - IRDA_ASSERT(((j >=0) && (j <4)), return 0;); - - line_capacity = max_line_capacities[i][j]; - - pr_debug("%s(), line capacity=%d bytes\n", - __func__, line_capacity); - - return line_capacity; -} - -#ifndef CONFIG_IRDA_DYNAMIC_WINDOW -static __u32 irlap_requested_line_capacity(struct qos_info *qos) -{ - __u32 line_capacity; - - line_capacity = qos->window_size.value * - (qos->data_size.value + 6 + qos->additional_bofs.value) + - irlap_min_turn_time_in_bytes(qos->baud_rate.value, - qos->min_turn_time.value); - - pr_debug("%s(), requested line capacity=%d\n", - __func__, line_capacity); - - return line_capacity; -} -#endif - -void irda_qos_bits_to_value(struct qos_info *qos) -{ - int index; - - IRDA_ASSERT(qos != NULL, return;); - - index = msb_index(qos->baud_rate.bits); - qos->baud_rate.value = baud_rates[index]; - - index = msb_index(qos->data_size.bits); - qos->data_size.value = data_sizes[index]; - - index = msb_index(qos->window_size.bits); - qos->window_size.value = index+1; - - index = msb_index(qos->min_turn_time.bits); - qos->min_turn_time.value = min_turn_times[index]; - - index = msb_index(qos->max_turn_time.bits); - qos->max_turn_time.value = max_turn_times[index]; - - index = msb_index(qos->link_disc_time.bits); - qos->link_disc_time.value = link_disc_times[index]; - - index = msb_index(qos->additional_bofs.bits); - qos->additional_bofs.value = add_bofs[index]; -} -EXPORT_SYMBOL(irda_qos_bits_to_value); diff --git a/net/irda/timer.c b/net/irda/timer.c deleted file mode 100644 index f2280f73b057..000000000000 --- a/net/irda/timer.c +++ /dev/null @@ -1,231 +0,0 @@ -/********************************************************************* - * - * Filename: timer.c - * Version: - * Description: - * Status: Experimental. - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Sat Aug 16 00:59:29 1997 - * Modified at: Wed Dec 8 12:50:34 1999 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * - * Copyright (c) 1997, 1999 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2002 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/delay.h> - -#include <net/irda/timer.h> -#include <net/irda/irda.h> -#include <net/irda/irda_device.h> -#include <net/irda/irlap.h> -#include <net/irda/irlmp.h> - -extern int sysctl_slot_timeout; - -static void irlap_slot_timer_expired(void* data); -static void irlap_query_timer_expired(void* data); -static void irlap_final_timer_expired(void* data); -static void irlap_wd_timer_expired(void* data); -static void irlap_backoff_timer_expired(void* data); -static void irlap_media_busy_expired(void* data); - -void irlap_start_slot_timer(struct irlap_cb *self, int timeout) -{ - irda_start_timer(&self->slot_timer, timeout, (void *) self, - irlap_slot_timer_expired); -} - -void irlap_start_query_timer(struct irlap_cb *self, int S, int s) -{ - int timeout; - - /* Calculate when the peer discovery should end. Normally, we - * get the end-of-discovery frame, so this is just in case - * we miss it. - * Basically, we multiply the number of remaining slots by our - * slot time, plus add some extra time to properly receive the last - * discovery packet (which is longer due to extra discovery info), - * to avoid messing with for incoming connections requests and - * to accommodate devices that perform discovery slower than us. - * Jean II */ - timeout = msecs_to_jiffies(sysctl_slot_timeout) * (S - s) - + XIDEXTRA_TIMEOUT + SMALLBUSY_TIMEOUT; - - /* Set or re-set the timer. We reset the timer for each received - * discovery query, which allow us to automatically adjust to - * the speed of the peer discovery (faster or slower). Jean II */ - irda_start_timer( &self->query_timer, timeout, (void *) self, - irlap_query_timer_expired); -} - -void irlap_start_final_timer(struct irlap_cb *self, int timeout) -{ - irda_start_timer(&self->final_timer, timeout, (void *) self, - irlap_final_timer_expired); -} - -void irlap_start_wd_timer(struct irlap_cb *self, int timeout) -{ - irda_start_timer(&self->wd_timer, timeout, (void *) self, - irlap_wd_timer_expired); -} - -void irlap_start_backoff_timer(struct irlap_cb *self, int timeout) -{ - irda_start_timer(&self->backoff_timer, timeout, (void *) self, - irlap_backoff_timer_expired); -} - -void irlap_start_mbusy_timer(struct irlap_cb *self, int timeout) -{ - irda_start_timer(&self->media_busy_timer, timeout, - (void *) self, irlap_media_busy_expired); -} - -void irlap_stop_mbusy_timer(struct irlap_cb *self) -{ - /* If timer is activated, kill it! */ - del_timer(&self->media_busy_timer); - - /* If we are in NDM, there is a bunch of events in LAP that - * that be pending due to the media_busy condition, such as - * CONNECT_REQUEST and SEND_UI_FRAME. If we don't generate - * an event, they will wait forever... - * Jean II */ - if (self->state == LAP_NDM) - irlap_do_event(self, MEDIA_BUSY_TIMER_EXPIRED, NULL, NULL); -} - -void irlmp_start_watchdog_timer(struct lsap_cb *self, int timeout) -{ - irda_start_timer(&self->watchdog_timer, timeout, (void *) self, - irlmp_watchdog_timer_expired); -} - -void irlmp_start_discovery_timer(struct irlmp_cb *self, int timeout) -{ - irda_start_timer(&self->discovery_timer, timeout, (void *) self, - irlmp_discovery_timer_expired); -} - -void irlmp_start_idle_timer(struct lap_cb *self, int timeout) -{ - irda_start_timer(&self->idle_timer, timeout, (void *) self, - irlmp_idle_timer_expired); -} - -void irlmp_stop_idle_timer(struct lap_cb *self) -{ - /* If timer is activated, kill it! */ - del_timer(&self->idle_timer); -} - -/* - * Function irlap_slot_timer_expired (data) - * - * IrLAP slot timer has expired - * - */ -static void irlap_slot_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, SLOT_TIMER_EXPIRED, NULL, NULL); -} - -/* - * Function irlap_query_timer_expired (data) - * - * IrLAP query timer has expired - * - */ -static void irlap_query_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, QUERY_TIMER_EXPIRED, NULL, NULL); -} - -/* - * Function irda_final_timer_expired (data) - * - * - * - */ -static void irlap_final_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, FINAL_TIMER_EXPIRED, NULL, NULL); -} - -/* - * Function irda_wd_timer_expired (data) - * - * - * - */ -static void irlap_wd_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, WD_TIMER_EXPIRED, NULL, NULL); -} - -/* - * Function irda_backoff_timer_expired (data) - * - * - * - */ -static void irlap_backoff_timer_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - IRDA_ASSERT(self->magic == LAP_MAGIC, return;); - - irlap_do_event(self, BACKOFF_TIMER_EXPIRED, NULL, NULL); -} - - -/* - * Function irtty_media_busy_expired (data) - * - * - */ -static void irlap_media_busy_expired(void *data) -{ - struct irlap_cb *self = (struct irlap_cb *) data; - - IRDA_ASSERT(self != NULL, return;); - - irda_device_set_media_busy(self->netdev, FALSE); - /* Note : the LAP event will be send in irlap_stop_mbusy_timer(), - * to catch other cases where the flag is cleared (for example - * after a discovery) - Jean II */ -} diff --git a/net/irda/wrapper.c b/net/irda/wrapper.c deleted file mode 100644 index 40a0f993bf13..000000000000 --- a/net/irda/wrapper.c +++ /dev/null @@ -1,492 +0,0 @@ -/********************************************************************* - * - * Filename: wrapper.c - * Version: 1.2 - * Description: IrDA SIR async wrapper layer - * Status: Stable - * Author: Dag Brattli <dagb@cs.uit.no> - * Created at: Mon Aug 4 20:40:53 1997 - * Modified at: Fri Jan 28 13:21:09 2000 - * Modified by: Dag Brattli <dagb@cs.uit.no> - * Modified at: Fri May 28 3:11 CST 1999 - * Modified by: Horst von Brand <vonbrand@sleipnir.valparaiso.cl> - * - * Copyright (c) 1998-2000 Dag Brattli <dagb@cs.uit.no>, - * All Rights Reserved. - * Copyright (c) 2000-2002 Jean Tourrilhes <jt@hpl.hp.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * Neither Dag Brattli nor University of Tromsø admit liability nor - * provide warranty for any of this software. This material is - * provided "AS-IS" and at no charge. - * - ********************************************************************/ - -#include <linux/skbuff.h> -#include <linux/string.h> -#include <linux/module.h> -#include <asm/byteorder.h> - -#include <net/irda/irda.h> -#include <net/irda/wrapper.h> -#include <net/irda/crc.h> -#include <net/irda/irlap.h> -#include <net/irda/irlap_frame.h> -#include <net/irda/irda_device.h> - -/************************** FRAME WRAPPING **************************/ -/* - * Unwrap and unstuff SIR frames - * - * Note : at FIR and MIR, HDLC framing is used and usually handled - * by the controller, so we come here only for SIR... Jean II - */ - -/* - * Function stuff_byte (byte, buf) - * - * Byte stuff one single byte and put the result in buffer pointed to by - * buf. The buffer must at all times be able to have two bytes inserted. - * - * This is in a tight loop, better inline it, so need to be prior to callers. - * (2000 bytes on P6 200MHz, non-inlined ~370us, inline ~170us) - Jean II - */ -static inline int stuff_byte(__u8 byte, __u8 *buf) -{ - switch (byte) { - case BOF: /* FALLTHROUGH */ - case EOF: /* FALLTHROUGH */ - case CE: - /* Insert transparently coded */ - buf[0] = CE; /* Send link escape */ - buf[1] = byte^IRDA_TRANS; /* Complement bit 5 */ - return 2; - /* break; */ - default: - /* Non-special value, no transparency required */ - buf[0] = byte; - return 1; - /* break; */ - } -} - -/* - * Function async_wrap (skb, *tx_buff, buffsize) - * - * Makes a new buffer with wrapping and stuffing, should check that - * we don't get tx buffer overflow. - */ -int async_wrap_skb(struct sk_buff *skb, __u8 *tx_buff, int buffsize) -{ - struct irda_skb_cb *cb = (struct irda_skb_cb *) skb->cb; - int xbofs; - int i; - int n; - union { - __u16 value; - __u8 bytes[2]; - } fcs; - - /* Initialize variables */ - fcs.value = INIT_FCS; - n = 0; - - /* - * Send XBOF's for required min. turn time and for the negotiated - * additional XBOFS - */ - - if (cb->magic != LAP_MAGIC) { - /* - * This will happen for all frames sent from user-space. - * Nothing to worry about, but we set the default number of - * BOF's - */ - pr_debug("%s(), wrong magic in skb!\n", __func__); - xbofs = 10; - } else - xbofs = cb->xbofs + cb->xbofs_delay; - - pr_debug("%s(), xbofs=%d\n", __func__, xbofs); - - /* Check that we never use more than 115 + 48 xbofs */ - if (xbofs > 163) { - pr_debug("%s(), too many xbofs (%d)\n", __func__, - xbofs); - xbofs = 163; - } - - memset(tx_buff + n, XBOF, xbofs); - n += xbofs; - - /* Start of packet character BOF */ - tx_buff[n++] = BOF; - - /* Insert frame and calc CRC */ - for (i=0; i < skb->len; i++) { - /* - * Check for the possibility of tx buffer overflow. We use - * bufsize-5 since the maximum number of bytes that can be - * transmitted after this point is 5. - */ - if(n >= (buffsize-5)) { - net_err_ratelimited("%s(), tx buffer overflow (n=%d)\n", - __func__, n); - return n; - } - - n += stuff_byte(skb->data[i], tx_buff+n); - fcs.value = irda_fcs(fcs.value, skb->data[i]); - } - - /* Insert CRC in little endian format (LSB first) */ - fcs.value = ~fcs.value; -#ifdef __LITTLE_ENDIAN - n += stuff_byte(fcs.bytes[0], tx_buff+n); - n += stuff_byte(fcs.bytes[1], tx_buff+n); -#else /* ifdef __BIG_ENDIAN */ - n += stuff_byte(fcs.bytes[1], tx_buff+n); - n += stuff_byte(fcs.bytes[0], tx_buff+n); -#endif - tx_buff[n++] = EOF; - - return n; -} -EXPORT_SYMBOL(async_wrap_skb); - -/************************* FRAME UNWRAPPING *************************/ -/* - * Unwrap and unstuff SIR frames - * - * Complete rewrite by Jean II : - * More inline, faster, more compact, more logical. Jean II - * (16 bytes on P6 200MHz, old 5 to 7 us, new 4 to 6 us) - * (24 bytes on P6 200MHz, old 9 to 10 us, new 7 to 8 us) - * (for reference, 115200 b/s is 1 byte every 69 us) - * And reduce wrapper.o by ~900B in the process ;-) - * - * Then, we have the addition of ZeroCopy, which is optional - * (i.e. the driver must initiate it) and improve final processing. - * (2005 B frame + EOF on P6 200MHz, without 30 to 50 us, with 10 to 25 us) - * - * Note : at FIR and MIR, HDLC framing is used and usually handled - * by the controller, so we come here only for SIR... Jean II - */ - -/* - * We can also choose where we want to do the CRC calculation. We can - * do it "inline", as we receive the bytes, or "postponed", when - * receiving the End-Of-Frame. - * (16 bytes on P6 200MHz, inlined 4 to 6 us, postponed 4 to 5 us) - * (24 bytes on P6 200MHz, inlined 7 to 8 us, postponed 5 to 7 us) - * With ZeroCopy : - * (2005 B frame on P6 200MHz, inlined 10 to 25 us, postponed 140 to 180 us) - * Without ZeroCopy : - * (2005 B frame on P6 200MHz, inlined 30 to 50 us, postponed 150 to 180 us) - * (Note : numbers taken with irq disabled) - * - * From those numbers, it's not clear which is the best strategy, because - * we end up running through a lot of data one way or another (i.e. cache - * misses). I personally prefer to avoid the huge latency spike of the - * "postponed" solution, because it come just at the time when we have - * lot's of protocol processing to do and it will hurt our ability to - * reach low link turnaround times... Jean II - */ -//#define POSTPONE_RX_CRC - -/* - * Function async_bump (buf, len, stats) - * - * Got a frame, make a copy of it, and pass it up the stack! We can try - * to inline it since it's only called from state_inside_frame - */ -static inline void -async_bump(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff) -{ - struct sk_buff *newskb; - struct sk_buff *dataskb; - int docopy; - - /* Check if we need to copy the data to a new skb or not. - * If the driver doesn't use ZeroCopy Rx, we have to do it. - * With ZeroCopy Rx, the rx_buff already point to a valid - * skb. But, if the frame is small, it is more efficient to - * copy it to save memory (copy will be fast anyway - that's - * called Rx-copy-break). Jean II */ - docopy = ((rx_buff->skb == NULL) || - (rx_buff->len < IRDA_RX_COPY_THRESHOLD)); - - /* Allocate a new skb */ - newskb = dev_alloc_skb(docopy ? rx_buff->len + 1 : rx_buff->truesize); - if (!newskb) { - stats->rx_dropped++; - /* We could deliver the current skb if doing ZeroCopy Rx, - * but this would stall the Rx path. Better drop the - * packet... Jean II */ - return; - } - - /* Align IP header to 20 bytes (i.e. increase skb->data) - * Note this is only useful with IrLAN, as PPP has a variable - * header size (2 or 1 bytes) - Jean II */ - skb_reserve(newskb, 1); - - if(docopy) { - /* Copy data without CRC (length already checked) */ - skb_copy_to_linear_data(newskb, rx_buff->data, - rx_buff->len - 2); - /* Deliver this skb */ - dataskb = newskb; - } else { - /* We are using ZeroCopy. Deliver old skb */ - dataskb = rx_buff->skb; - /* And hook the new skb to the rx_buff */ - rx_buff->skb = newskb; - rx_buff->head = newskb->data; /* NOT newskb->head */ - //printk(KERN_DEBUG "ZeroCopy : len = %d, dataskb = %p, newskb = %p\n", rx_buff->len, dataskb, newskb); - } - - /* Set proper length on skb (without CRC) */ - skb_put(dataskb, rx_buff->len - 2); - - /* Feed it to IrLAP layer */ - dataskb->dev = dev; - skb_reset_mac_header(dataskb); - dataskb->protocol = htons(ETH_P_IRDA); - - netif_rx(dataskb); - - stats->rx_packets++; - stats->rx_bytes += rx_buff->len; - - /* Clean up rx_buff (redundant with async_unwrap_bof() ???) */ - rx_buff->data = rx_buff->head; - rx_buff->len = 0; -} - -/* - * Function async_unwrap_bof(dev, byte) - * - * Handle Beginning Of Frame character received within a frame - * - */ -static inline void -async_unwrap_bof(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff, __u8 byte) -{ - switch(rx_buff->state) { - case LINK_ESCAPE: - case INSIDE_FRAME: - /* Not supposed to happen, the previous frame is not - * finished - Jean II */ - pr_debug("%s(), Discarding incomplete frame\n", - __func__); - stats->rx_errors++; - stats->rx_missed_errors++; - irda_device_set_media_busy(dev, TRUE); - break; - - case OUTSIDE_FRAME: - case BEGIN_FRAME: - default: - /* We may receive multiple BOF at the start of frame */ - break; - } - - /* Now receiving frame */ - rx_buff->state = BEGIN_FRAME; - rx_buff->in_frame = TRUE; - - /* Time to initialize receive buffer */ - rx_buff->data = rx_buff->head; - rx_buff->len = 0; - rx_buff->fcs = INIT_FCS; -} - -/* - * Function async_unwrap_eof(dev, byte) - * - * Handle End Of Frame character received within a frame - * - */ -static inline void -async_unwrap_eof(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff, __u8 byte) -{ -#ifdef POSTPONE_RX_CRC - int i; -#endif - - switch(rx_buff->state) { - case OUTSIDE_FRAME: - /* Probably missed the BOF */ - stats->rx_errors++; - stats->rx_missed_errors++; - irda_device_set_media_busy(dev, TRUE); - break; - - case BEGIN_FRAME: - case LINK_ESCAPE: - case INSIDE_FRAME: - default: - /* Note : in the case of BEGIN_FRAME and LINK_ESCAPE, - * the fcs will most likely not match and generate an - * error, as expected - Jean II */ - rx_buff->state = OUTSIDE_FRAME; - rx_buff->in_frame = FALSE; - -#ifdef POSTPONE_RX_CRC - /* If we haven't done the CRC as we receive bytes, we - * must do it now... Jean II */ - for(i = 0; i < rx_buff->len; i++) - rx_buff->fcs = irda_fcs(rx_buff->fcs, - rx_buff->data[i]); -#endif - - /* Test FCS and signal success if the frame is good */ - if (rx_buff->fcs == GOOD_FCS) { - /* Deliver frame */ - async_bump(dev, stats, rx_buff); - break; - } else { - /* Wrong CRC, discard frame! */ - irda_device_set_media_busy(dev, TRUE); - - pr_debug("%s(), crc error\n", __func__); - stats->rx_errors++; - stats->rx_crc_errors++; - } - break; - } -} - -/* - * Function async_unwrap_ce(dev, byte) - * - * Handle Character Escape character received within a frame - * - */ -static inline void -async_unwrap_ce(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff, __u8 byte) -{ - switch(rx_buff->state) { - case OUTSIDE_FRAME: - /* Activate carrier sense */ - irda_device_set_media_busy(dev, TRUE); - break; - - case LINK_ESCAPE: - net_warn_ratelimited("%s: state not defined\n", __func__); - break; - - case BEGIN_FRAME: - case INSIDE_FRAME: - default: - /* Stuffed byte coming */ - rx_buff->state = LINK_ESCAPE; - break; - } -} - -/* - * Function async_unwrap_other(dev, byte) - * - * Handle other characters received within a frame - * - */ -static inline void -async_unwrap_other(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff, __u8 byte) -{ - switch(rx_buff->state) { - /* This is on the critical path, case are ordered by - * probability (most frequent first) - Jean II */ - case INSIDE_FRAME: - /* Must be the next byte of the frame */ - if (rx_buff->len < rx_buff->truesize) { - rx_buff->data[rx_buff->len++] = byte; -#ifndef POSTPONE_RX_CRC - rx_buff->fcs = irda_fcs(rx_buff->fcs, byte); -#endif - } else { - pr_debug("%s(), Rx buffer overflow, aborting\n", - __func__); - rx_buff->state = OUTSIDE_FRAME; - } - break; - - case LINK_ESCAPE: - /* - * Stuffed char, complement bit 5 of byte - * following CE, IrLAP p.114 - */ - byte ^= IRDA_TRANS; - if (rx_buff->len < rx_buff->truesize) { - rx_buff->data[rx_buff->len++] = byte; -#ifndef POSTPONE_RX_CRC - rx_buff->fcs = irda_fcs(rx_buff->fcs, byte); -#endif - rx_buff->state = INSIDE_FRAME; - } else { - pr_debug("%s(), Rx buffer overflow, aborting\n", - __func__); - rx_buff->state = OUTSIDE_FRAME; - } - break; - - case OUTSIDE_FRAME: - /* Activate carrier sense */ - if(byte != XBOF) - irda_device_set_media_busy(dev, TRUE); - break; - - case BEGIN_FRAME: - default: - rx_buff->data[rx_buff->len++] = byte; -#ifndef POSTPONE_RX_CRC - rx_buff->fcs = irda_fcs(rx_buff->fcs, byte); -#endif - rx_buff->state = INSIDE_FRAME; - break; - } -} - -/* - * Function async_unwrap_char (dev, rx_buff, byte) - * - * Parse and de-stuff frame received from the IrDA-port - * - * This is the main entry point for SIR drivers. - */ -void async_unwrap_char(struct net_device *dev, - struct net_device_stats *stats, - iobuff_t *rx_buff, __u8 byte) -{ - switch(byte) { - case CE: - async_unwrap_ce(dev, stats, rx_buff, byte); - break; - case BOF: - async_unwrap_bof(dev, stats, rx_buff, byte); - break; - case EOF: - async_unwrap_eof(dev, stats, rx_buff, byte); - break; - default: - async_unwrap_other(dev, stats, rx_buff, byte); - break; - } -} -EXPORT_SYMBOL(async_unwrap_char); - diff --git a/net/kcm/kcmproc.c b/net/kcm/kcmproc.c index c343ac60bf50..c748e8a6a72c 100644 --- a/net/kcm/kcmproc.c +++ b/net/kcm/kcmproc.c @@ -155,8 +155,8 @@ static void kcm_format_psock(struct kcm_psock *psock, struct seq_file *seq, seq_printf(seq, " psock-%-5u %-10llu %-16llu %-10llu %-16llu %-8d %-8d %-8d %-8d ", psock->index, - psock->strp.stats.rx_msgs, - psock->strp.stats.rx_bytes, + psock->strp.stats.msgs, + psock->strp.stats.bytes, psock->stats.tx_msgs, psock->stats.tx_bytes, psock->sk->sk_receive_queue.qlen, @@ -170,22 +170,22 @@ static void kcm_format_psock(struct kcm_psock *psock, struct seq_file *seq, if (psock->tx_stopped) seq_puts(seq, "TxStop "); - if (psock->strp.rx_stopped) + if (psock->strp.stopped) seq_puts(seq, "RxStop "); if (psock->tx_kcm) seq_printf(seq, "Rsvd-%d ", psock->tx_kcm->index); - if (!psock->strp.rx_paused && !psock->ready_rx_msg) { + if (!psock->strp.paused && !psock->ready_rx_msg) { if (psock->sk->sk_receive_queue.qlen) { - if (psock->strp.rx_need_bytes) + if (psock->strp.need_bytes) seq_printf(seq, "RxWait=%u ", - psock->strp.rx_need_bytes); + psock->strp.need_bytes); else seq_printf(seq, "RxWait "); } } else { - if (psock->strp.rx_paused) + if (psock->strp.paused) seq_puts(seq, "RxPause "); if (psock->ready_rx_msg) @@ -371,20 +371,20 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u %-10u %-10u %-10u %-10u\n", "", - strp_stats.rx_msgs, - strp_stats.rx_bytes, + strp_stats.msgs, + strp_stats.bytes, psock_stats.tx_msgs, psock_stats.tx_bytes, psock_stats.reserved, psock_stats.unreserved, - strp_stats.rx_aborts, - strp_stats.rx_interrupted, - strp_stats.rx_unrecov_intr, - strp_stats.rx_mem_fail, - strp_stats.rx_need_more_hdr, - strp_stats.rx_bad_hdr_len, - strp_stats.rx_msg_too_big, - strp_stats.rx_msg_timeouts, + strp_stats.aborts, + strp_stats.interrupted, + strp_stats.unrecov_intr, + strp_stats.mem_fail, + strp_stats.need_more_hdr, + strp_stats.bad_hdr_len, + strp_stats.msg_too_big, + strp_stats.msg_timeouts, psock_stats.tx_aborts); return 0; diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index da49191f7ad0..af4e76ac88ff 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -96,12 +96,12 @@ static void kcm_update_rx_mux_stats(struct kcm_mux *mux, struct kcm_psock *psock) { STRP_STATS_ADD(mux->stats.rx_bytes, - psock->strp.stats.rx_bytes - + psock->strp.stats.bytes - psock->saved_rx_bytes); mux->stats.rx_msgs += - psock->strp.stats.rx_msgs - psock->saved_rx_msgs; - psock->saved_rx_msgs = psock->strp.stats.rx_msgs; - psock->saved_rx_bytes = psock->strp.stats.rx_bytes; + psock->strp.stats.msgs - psock->saved_rx_msgs; + psock->saved_rx_msgs = psock->strp.stats.msgs; + psock->saved_rx_bytes = psock->strp.stats.bytes; } static void kcm_update_tx_mux_stats(struct kcm_mux *mux, @@ -1118,7 +1118,7 @@ static int kcm_recvmsg(struct socket *sock, struct msghdr *msg, struct kcm_sock *kcm = kcm_sk(sk); int err = 0; long timeo; - struct strp_rx_msg *rxm; + struct strp_msg *stm; int copied = 0; struct sk_buff *skb; @@ -1132,26 +1132,26 @@ static int kcm_recvmsg(struct socket *sock, struct msghdr *msg, /* Okay, have a message on the receive queue */ - rxm = strp_rx_msg(skb); + stm = strp_msg(skb); - if (len > rxm->full_len) - len = rxm->full_len; + if (len > stm->full_len) + len = stm->full_len; - err = skb_copy_datagram_msg(skb, rxm->offset, msg, len); + err = skb_copy_datagram_msg(skb, stm->offset, msg, len); if (err < 0) goto out; copied = len; if (likely(!(flags & MSG_PEEK))) { KCM_STATS_ADD(kcm->stats.rx_bytes, copied); - if (copied < rxm->full_len) { + if (copied < stm->full_len) { if (sock->type == SOCK_DGRAM) { /* Truncated message */ msg->msg_flags |= MSG_TRUNC; goto msg_finished; } - rxm->offset += copied; - rxm->full_len -= copied; + stm->offset += copied; + stm->full_len -= copied; } else { msg_finished: /* Finished with message */ @@ -1175,7 +1175,7 @@ static ssize_t kcm_splice_read(struct socket *sock, loff_t *ppos, struct sock *sk = sock->sk; struct kcm_sock *kcm = kcm_sk(sk); long timeo; - struct strp_rx_msg *rxm; + struct strp_msg *stm; int err = 0; ssize_t copied; struct sk_buff *skb; @@ -1192,12 +1192,12 @@ static ssize_t kcm_splice_read(struct socket *sock, loff_t *ppos, /* Okay, have a message on the receive queue */ - rxm = strp_rx_msg(skb); + stm = strp_msg(skb); - if (len > rxm->full_len) - len = rxm->full_len; + if (len > stm->full_len) + len = stm->full_len; - copied = skb_splice_bits(skb, sk, rxm->offset, pipe, len, flags); + copied = skb_splice_bits(skb, sk, stm->offset, pipe, len, flags); if (copied < 0) { err = copied; goto err_out; @@ -1205,8 +1205,8 @@ static ssize_t kcm_splice_read(struct socket *sock, loff_t *ppos, KCM_STATS_ADD(kcm->stats.rx_bytes, copied); - rxm->offset += copied; - rxm->full_len -= copied; + stm->offset += copied; + stm->full_len -= copied; /* We have no way to return MSG_EOR. If all the bytes have been * read we still leave the message in the receive socket buffer. @@ -1376,13 +1376,21 @@ static int kcm_attach(struct socket *sock, struct socket *csock, struct kcm_psock *psock = NULL, *tpsock; struct list_head *head; int index = 0; - struct strp_callbacks cb; + static const struct strp_callbacks cb = { + .rcv_msg = kcm_rcv_strparser, + .parse_msg = kcm_parse_func_strparser, + .read_sock_done = kcm_read_sock_done, + }; int err; csk = csock->sk; if (!csk) return -EINVAL; + /* We must prevent loops or risk deadlock ! */ + if (csk->sk_family == PF_KCM) + return -EOPNOTSUPP; + psock = kmem_cache_zalloc(kcm_psockp, GFP_KERNEL); if (!psock) return -ENOMEM; @@ -1391,11 +1399,6 @@ static int kcm_attach(struct socket *sock, struct socket *csock, psock->sk = csk; psock->bpf_prog = prog; - cb.rcv_msg = kcm_rcv_strparser; - cb.abort_parser = NULL; - cb.parse_msg = kcm_parse_func_strparser; - cb.read_sock_done = kcm_read_sock_done; - err = strp_init(&psock->strp, csk, &cb); if (err) { kmem_cache_free(kcm_psockp, psock); diff --git a/net/key/af_key.c b/net/key/af_key.c index ca9d3ae665e7..a00d607e7224 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -228,7 +228,7 @@ static int pfkey_broadcast_one(struct sk_buff *skb, struct sk_buff **skb2, #define BROADCAST_ONE 1 #define BROADCAST_REGISTERED 2 #define BROADCAST_PROMISC_ONLY 4 -static int pfkey_broadcast(struct sk_buff *skb, +static int pfkey_broadcast(struct sk_buff *skb, gfp_t allocation, int broadcast_flags, struct sock *one_sk, struct net *net) { @@ -278,7 +278,7 @@ static int pfkey_broadcast(struct sk_buff *skb, rcu_read_unlock(); if (one_sk != NULL) - err = pfkey_broadcast_one(skb, &skb2, GFP_KERNEL, one_sk); + err = pfkey_broadcast_one(skb, &skb2, allocation, one_sk); kfree_skb(skb2); kfree_skb(skb); @@ -311,7 +311,7 @@ static int pfkey_do_dump(struct pfkey_sock *pfk) hdr = (struct sadb_msg *) pfk->dump.skb->data; hdr->sadb_msg_seq = 0; hdr->sadb_msg_errno = rc; - pfkey_broadcast(pfk->dump.skb, BROADCAST_ONE, + pfkey_broadcast(pfk->dump.skb, GFP_ATOMIC, BROADCAST_ONE, &pfk->sk, sock_net(&pfk->sk)); pfk->dump.skb = NULL; } @@ -355,7 +355,7 @@ static int pfkey_error(const struct sadb_msg *orig, int err, struct sock *sk) hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - pfkey_broadcast(skb, BROADCAST_ONE, sk, sock_net(sk)); + pfkey_broadcast(skb, GFP_KERNEL, BROADCAST_ONE, sk, sock_net(sk)); return 0; } @@ -1389,7 +1389,7 @@ static int pfkey_getspi(struct sock *sk, struct sk_buff *skb, const struct sadb_ xfrm_state_put(x); - pfkey_broadcast(resp_skb, BROADCAST_ONE, sk, net); + pfkey_broadcast(resp_skb, GFP_KERNEL, BROADCAST_ONE, sk, net); return 0; } @@ -1476,7 +1476,7 @@ static int key_notify_sa(struct xfrm_state *x, const struct km_event *c) hdr->sadb_msg_seq = c->seq; hdr->sadb_msg_pid = c->portid; - pfkey_broadcast(skb, BROADCAST_ALL, NULL, xs_net(x)); + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, xs_net(x)); return 0; } @@ -1589,7 +1589,7 @@ static int pfkey_get(struct sock *sk, struct sk_buff *skb, const struct sadb_msg out_hdr->sadb_msg_reserved = 0; out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, BROADCAST_ONE, sk, sock_net(sk)); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk, sock_net(sk)); return 0; } @@ -1694,8 +1694,8 @@ static int pfkey_register(struct sock *sk, struct sk_buff *skb, const struct sad return -ENOBUFS; } - pfkey_broadcast(supp_skb, BROADCAST_REGISTERED, sk, sock_net(sk)); - + pfkey_broadcast(supp_skb, GFP_KERNEL, BROADCAST_REGISTERED, sk, + sock_net(sk)); return 0; } @@ -1712,7 +1712,8 @@ static int unicast_flush_resp(struct sock *sk, const struct sadb_msg *ihdr) hdr->sadb_msg_errno = (uint8_t) 0; hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - return pfkey_broadcast(skb, BROADCAST_ONE, sk, sock_net(sk)); + return pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ONE, sk, + sock_net(sk)); } static int key_notify_sa_flush(const struct km_event *c) @@ -1733,7 +1734,7 @@ static int key_notify_sa_flush(const struct km_event *c) hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); hdr->sadb_msg_reserved = 0; - pfkey_broadcast(skb, BROADCAST_ALL, NULL, c->net); + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, c->net); return 0; } @@ -1790,7 +1791,7 @@ static int dump_sa(struct xfrm_state *x, int count, void *ptr) out_hdr->sadb_msg_pid = pfk->dump.msg_portid; if (pfk->dump.skb) - pfkey_broadcast(pfk->dump.skb, BROADCAST_ONE, + pfkey_broadcast(pfk->dump.skb, GFP_ATOMIC, BROADCAST_ONE, &pfk->sk, sock_net(&pfk->sk)); pfk->dump.skb = out_skb; @@ -1878,7 +1879,7 @@ static int pfkey_promisc(struct sock *sk, struct sk_buff *skb, const struct sadb new_hdr->sadb_msg_errno = 0; } - pfkey_broadcast(skb, BROADCAST_ALL, NULL, sock_net(sk)); + pfkey_broadcast(skb, GFP_KERNEL, BROADCAST_ALL, NULL, sock_net(sk)); return 0; } @@ -2206,7 +2207,7 @@ static int key_notify_policy(struct xfrm_policy *xp, int dir, const struct km_ev out_hdr->sadb_msg_errno = 0; out_hdr->sadb_msg_seq = c->seq; out_hdr->sadb_msg_pid = c->portid; - pfkey_broadcast(out_skb, BROADCAST_ALL, NULL, xp_net(xp)); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, NULL, xp_net(xp)); return 0; } @@ -2398,8 +2399,6 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, const struct sa out: xfrm_pol_put(xp); - if (err == 0) - xfrm_garbage_collect(net); return err; } @@ -2426,7 +2425,7 @@ static int key_pol_get_resp(struct sock *sk, struct xfrm_policy *xp, const struc out_hdr->sadb_msg_errno = 0; out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, BROADCAST_ONE, sk, xp_net(xp)); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk, xp_net(xp)); err = 0; out: @@ -2650,8 +2649,6 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, const struct sadb_ out: xfrm_pol_put(xp); - if (delete && err == 0) - xfrm_garbage_collect(net); return err; } @@ -2682,7 +2679,7 @@ static int dump_sp(struct xfrm_policy *xp, int dir, int count, void *ptr) out_hdr->sadb_msg_pid = pfk->dump.msg_portid; if (pfk->dump.skb) - pfkey_broadcast(pfk->dump.skb, BROADCAST_ONE, + pfkey_broadcast(pfk->dump.skb, GFP_ATOMIC, BROADCAST_ONE, &pfk->sk, sock_net(&pfk->sk)); pfk->dump.skb = out_skb; @@ -2739,7 +2736,7 @@ static int key_notify_policy_flush(const struct km_event *c) hdr->sadb_msg_satype = SADB_SATYPE_UNSPEC; hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); hdr->sadb_msg_reserved = 0; - pfkey_broadcast(skb_out, BROADCAST_ALL, NULL, c->net); + pfkey_broadcast(skb_out, GFP_ATOMIC, BROADCAST_ALL, NULL, c->net); return 0; } @@ -2751,8 +2748,6 @@ static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, const struct sad int err, err2; err = xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, true); - if (!err) - xfrm_garbage_collect(net); err2 = unicast_flush_resp(sk, hdr); if (err || err2) { if (err == -ESRCH) /* empty table - old silent behavior */ @@ -2803,7 +2798,7 @@ static int pfkey_process(struct sock *sk, struct sk_buff *skb, const struct sadb void *ext_hdrs[SADB_EXT_MAX]; int err; - pfkey_broadcast(skb_clone(skb, GFP_KERNEL), + pfkey_broadcast(skb_clone(skb, GFP_KERNEL), GFP_KERNEL, BROADCAST_PROMISC_ONLY, NULL, sock_net(sk)); memset(ext_hdrs, 0, sizeof(ext_hdrs)); @@ -3024,7 +3019,8 @@ static int key_notify_sa_expire(struct xfrm_state *x, const struct km_event *c) out_hdr->sadb_msg_seq = 0; out_hdr->sadb_msg_pid = 0; - pfkey_broadcast(out_skb, BROADCAST_REGISTERED, NULL, xs_net(x)); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_REGISTERED, NULL, + xs_net(x)); return 0; } @@ -3212,7 +3208,8 @@ static int pfkey_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_ctx->ctx_len); } - return pfkey_broadcast(skb, BROADCAST_REGISTERED, NULL, xs_net(x)); + return pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_REGISTERED, NULL, + xs_net(x)); } static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt, @@ -3408,7 +3405,8 @@ static int pfkey_send_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, n_port->sadb_x_nat_t_port_port = sport; n_port->sadb_x_nat_t_port_reserved = 0; - return pfkey_broadcast(skb, BROADCAST_REGISTERED, NULL, xs_net(x)); + return pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_REGISTERED, NULL, + xs_net(x)); } #ifdef CONFIG_NET_KEY_MIGRATE @@ -3599,7 +3597,7 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, } /* broadcast migrate message to sockets */ - pfkey_broadcast(skb, BROADCAST_ALL, NULL, &init_net); + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, &init_net); return 0; diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index b0c2d4ae781d..ee485df73ccd 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -113,7 +113,6 @@ struct l2tp_net { spinlock_t l2tp_session_hlist_lock; }; -static void l2tp_tunnel_free(struct l2tp_tunnel *tunnel); static inline struct l2tp_tunnel *l2tp_tunnel(struct sock *sk) { @@ -127,39 +126,6 @@ static inline struct l2tp_net *l2tp_pernet(const struct net *net) return net_generic(net, l2tp_net_id); } -/* Tunnel reference counts. Incremented per session that is added to - * the tunnel. - */ -static inline void l2tp_tunnel_inc_refcount_1(struct l2tp_tunnel *tunnel) -{ - refcount_inc(&tunnel->ref_count); -} - -static inline void l2tp_tunnel_dec_refcount_1(struct l2tp_tunnel *tunnel) -{ - if (refcount_dec_and_test(&tunnel->ref_count)) - l2tp_tunnel_free(tunnel); -} -#ifdef L2TP_REFCNT_DEBUG -#define l2tp_tunnel_inc_refcount(_t) \ -do { \ - pr_debug("l2tp_tunnel_inc_refcount: %s:%d %s: cnt=%d\n", \ - __func__, __LINE__, (_t)->name, \ - refcount_read(&_t->ref_count)); \ - l2tp_tunnel_inc_refcount_1(_t); \ -} while (0) -#define l2tp_tunnel_dec_refcount(_t) \ -do { \ - pr_debug("l2tp_tunnel_dec_refcount: %s:%d %s: cnt=%d\n", \ - __func__, __LINE__, (_t)->name, \ - refcount_read(&_t->ref_count)); \ - l2tp_tunnel_dec_refcount_1(_t); \ -} while (0) -#else -#define l2tp_tunnel_inc_refcount(t) l2tp_tunnel_inc_refcount_1(t) -#define l2tp_tunnel_dec_refcount(t) l2tp_tunnel_dec_refcount_1(t) -#endif - /* Session hash global list for L2TPv3. * The session_id SHOULD be random according to RFC3931, but several * L2TP implementations use incrementing session_ids. So we do a real @@ -229,6 +195,27 @@ l2tp_session_id_hash(struct l2tp_tunnel *tunnel, u32 session_id) return &tunnel->session_hlist[hash_32(session_id, L2TP_HASH_BITS)]; } +/* Lookup a tunnel. A new reference is held on the returned tunnel. */ +struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id) +{ + const struct l2tp_net *pn = l2tp_pernet(net); + struct l2tp_tunnel *tunnel; + + rcu_read_lock_bh(); + list_for_each_entry_rcu(tunnel, &pn->l2tp_tunnel_list, list) { + if (tunnel->tunnel_id == tunnel_id) { + l2tp_tunnel_inc_refcount(tunnel); + rcu_read_unlock_bh(); + + return tunnel; + } + } + rcu_read_unlock_bh(); + + return NULL; +} +EXPORT_SYMBOL_GPL(l2tp_tunnel_get); + /* Lookup a session. A new reference is held on the returned session. * Optionally calls session->ref() too if do_ref is true. */ @@ -342,13 +329,21 @@ static int l2tp_session_add_to_tunnel(struct l2tp_tunnel *tunnel, struct hlist_head *g_head; struct hlist_head *head; struct l2tp_net *pn; + int err; head = l2tp_session_id_hash(tunnel, session->session_id); write_lock_bh(&tunnel->hlist_lock); + if (!tunnel->acpt_newsess) { + err = -ENODEV; + goto err_tlock; + } + hlist_for_each_entry(session_walk, head, hlist) - if (session_walk->session_id == session->session_id) - goto exist; + if (session_walk->session_id == session->session_id) { + err = -EEXIST; + goto err_tlock; + } if (tunnel->version == L2TP_HDR_VER_3) { pn = l2tp_pernet(tunnel->l2tp_net); @@ -356,12 +351,21 @@ static int l2tp_session_add_to_tunnel(struct l2tp_tunnel *tunnel, session->session_id); spin_lock_bh(&pn->l2tp_session_hlist_lock); + hlist_for_each_entry(session_walk, g_head, global_hlist) - if (session_walk->session_id == session->session_id) - goto exist_glob; + if (session_walk->session_id == session->session_id) { + err = -EEXIST; + goto err_tlock_pnlock; + } + l2tp_tunnel_inc_refcount(tunnel); + sock_hold(tunnel->sock); hlist_add_head_rcu(&session->global_hlist, g_head); + spin_unlock_bh(&pn->l2tp_session_hlist_lock); + } else { + l2tp_tunnel_inc_refcount(tunnel); + sock_hold(tunnel->sock); } hlist_add_head(&session->hlist, head); @@ -369,12 +373,12 @@ static int l2tp_session_add_to_tunnel(struct l2tp_tunnel *tunnel, return 0; -exist_glob: +err_tlock_pnlock: spin_unlock_bh(&pn->l2tp_session_hlist_lock); -exist: +err_tlock: write_unlock_bh(&tunnel->hlist_lock); - return -EEXIST; + return err; } /* Lookup a tunnel by id @@ -1264,7 +1268,6 @@ static void l2tp_tunnel_destruct(struct sock *sk) /* Remove hooks into tunnel socket */ sk->sk_destruct = tunnel->old_sk_destruct; sk->sk_user_data = NULL; - tunnel->sock = NULL; /* Remove the tunnel struct from the tunnel list */ pn = l2tp_pernet(tunnel->l2tp_net); @@ -1274,6 +1277,8 @@ static void l2tp_tunnel_destruct(struct sock *sk) atomic_dec(&l2tp_tunnel_count); l2tp_tunnel_closeall(tunnel); + + tunnel->sock = NULL; l2tp_tunnel_dec_refcount(tunnel); /* Call the original destructor */ @@ -1298,6 +1303,7 @@ void l2tp_tunnel_closeall(struct l2tp_tunnel *tunnel) tunnel->name); write_lock_bh(&tunnel->hlist_lock); + tunnel->acpt_newsess = false; for (hash = 0; hash < L2TP_HASH_SIZE; hash++) { again: hlist_for_each_safe(walk, tmp, &tunnel->session_hlist[hash]) { @@ -1348,17 +1354,6 @@ static void l2tp_udp_encap_destroy(struct sock *sk) } } -/* Really kill the tunnel. - * Come here only when all sessions have been cleared from the tunnel. - */ -static void l2tp_tunnel_free(struct l2tp_tunnel *tunnel) -{ - BUG_ON(refcount_read(&tunnel->ref_count) != 0); - BUG_ON(tunnel->sock != NULL); - l2tp_info(tunnel, L2TP_MSG_CONTROL, "%s: free...\n", tunnel->name); - kfree_rcu(tunnel, rcu); -} - /* Workqueue tunnel deletion function */ static void l2tp_tunnel_del_work(struct work_struct *work) { @@ -1605,6 +1600,7 @@ int l2tp_tunnel_create(struct net *net, int fd, int version, u32 tunnel_id, u32 tunnel->magic = L2TP_TUNNEL_MAGIC; sprintf(&tunnel->name[0], "tunl %u", tunnel_id); rwlock_init(&tunnel->hlist_lock); + tunnel->acpt_newsess = true; /* The net we belong to */ tunnel->l2tp_net = net; @@ -1844,6 +1840,8 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn l2tp_session_set_header_len(session, tunnel->version); + refcount_set(&session->ref_count, 1); + err = l2tp_session_add_to_tunnel(tunnel, session); if (err) { kfree(session); @@ -1851,15 +1849,6 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn return ERR_PTR(err); } - /* Bump the reference count. The session context is deleted - * only when this drops to zero. - */ - refcount_set(&session->ref_count, 1); - l2tp_tunnel_inc_refcount(tunnel); - - /* Ensure tunnel socket isn't deleted */ - sock_hold(tunnel->sock); - /* Ignore management session in session count value */ if (session->session_id != 0) atomic_inc(&l2tp_session_count); diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h index cdb6e3327f74..a305e0c5925a 100644 --- a/net/l2tp/l2tp_core.h +++ b/net/l2tp/l2tp_core.h @@ -162,6 +162,10 @@ struct l2tp_tunnel { int magic; /* Should be L2TP_TUNNEL_MAGIC */ struct rcu_head rcu; rwlock_t hlist_lock; /* protect session_hlist */ + bool acpt_newsess; /* Indicates whether this + * tunnel accepts new sessions. + * Protected by hlist_lock. + */ struct hlist_head session_hlist[L2TP_HASH_SIZE]; /* hashed list of sessions, * hashed by id */ @@ -197,7 +201,9 @@ struct l2tp_tunnel { }; struct l2tp_nl_cmd_ops { - int (*session_create)(struct net *net, u32 tunnel_id, u32 session_id, u32 peer_session_id, struct l2tp_session_cfg *cfg); + int (*session_create)(struct net *net, struct l2tp_tunnel *tunnel, + u32 session_id, u32 peer_session_id, + struct l2tp_session_cfg *cfg); int (*session_delete)(struct l2tp_session *session); }; @@ -231,6 +237,8 @@ out: return tunnel; } +struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id); + struct l2tp_session *l2tp_session_get(const struct net *net, struct l2tp_tunnel *tunnel, u32 session_id, bool do_ref); @@ -269,6 +277,17 @@ int l2tp_nl_register_ops(enum l2tp_pwtype pw_type, void l2tp_nl_unregister_ops(enum l2tp_pwtype pw_type); int l2tp_ioctl(struct sock *sk, int cmd, unsigned long arg); +static inline void l2tp_tunnel_inc_refcount(struct l2tp_tunnel *tunnel) +{ + refcount_inc(&tunnel->ref_count); +} + +static inline void l2tp_tunnel_dec_refcount(struct l2tp_tunnel *tunnel) +{ + if (refcount_dec_and_test(&tunnel->ref_count)) + kfree_rcu(tunnel, rcu); +} + /* Session reference counts. Incremented when code obtains a reference * to a session. */ diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index 4de2ec94b08c..87da9ef61860 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -262,24 +262,19 @@ static void l2tp_eth_adjust_mtu(struct l2tp_tunnel *tunnel, dev->needed_headroom += session->hdr_len; } -static int l2tp_eth_create(struct net *net, u32 tunnel_id, u32 session_id, u32 peer_session_id, struct l2tp_session_cfg *cfg) +static int l2tp_eth_create(struct net *net, struct l2tp_tunnel *tunnel, + u32 session_id, u32 peer_session_id, + struct l2tp_session_cfg *cfg) { unsigned char name_assign_type; struct net_device *dev; char name[IFNAMSIZ]; - struct l2tp_tunnel *tunnel; struct l2tp_session *session; struct l2tp_eth *priv; struct l2tp_eth_sess *spriv; int rc; struct l2tp_eth_net *pn; - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (!tunnel) { - rc = -ENODEV; - goto out; - } - if (cfg->ifname) { strlcpy(name, cfg->ifname, IFNAMSIZ); name_assign_type = NET_NAME_USER; diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index 12cfcd0ca807..7135f4645d3a 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -65,10 +65,12 @@ static struct l2tp_session *l2tp_nl_session_get(struct genl_info *info, (info->attrs[L2TP_ATTR_CONN_ID])) { tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); session_id = nla_get_u32(info->attrs[L2TP_ATTR_SESSION_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel) + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (tunnel) { session = l2tp_session_get(net, tunnel, session_id, do_ref); + l2tp_tunnel_dec_refcount(tunnel); + } } return session; @@ -271,8 +273,8 @@ static int l2tp_nl_cmd_tunnel_delete(struct sk_buff *skb, struct genl_info *info } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { ret = -ENODEV; goto out; } @@ -282,6 +284,8 @@ static int l2tp_nl_cmd_tunnel_delete(struct sk_buff *skb, struct genl_info *info (void) l2tp_tunnel_delete(tunnel); + l2tp_tunnel_dec_refcount(tunnel); + out: return ret; } @@ -299,8 +303,8 @@ static int l2tp_nl_cmd_tunnel_modify(struct sk_buff *skb, struct genl_info *info } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { ret = -ENODEV; goto out; } @@ -311,6 +315,8 @@ static int l2tp_nl_cmd_tunnel_modify(struct sk_buff *skb, struct genl_info *info ret = l2tp_tunnel_notify(&l2tp_nl_family, info, tunnel, L2TP_CMD_TUNNEL_MODIFY); + l2tp_tunnel_dec_refcount(tunnel); + out: return ret; } @@ -438,34 +444,37 @@ static int l2tp_nl_cmd_tunnel_get(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[L2TP_ATTR_CONN_ID]) { ret = -EINVAL; - goto out; + goto err; } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { - ret = -ENODEV; - goto out; - } - msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) { ret = -ENOMEM; - goto out; + goto err; + } + + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { + ret = -ENODEV; + goto err_nlmsg; } ret = l2tp_nl_tunnel_send(msg, info->snd_portid, info->snd_seq, NLM_F_ACK, tunnel, L2TP_CMD_TUNNEL_GET); if (ret < 0) - goto err_out; + goto err_nlmsg_tunnel; + + l2tp_tunnel_dec_refcount(tunnel); return genlmsg_unicast(net, msg, info->snd_portid); -err_out: +err_nlmsg_tunnel: + l2tp_tunnel_dec_refcount(tunnel); +err_nlmsg: nlmsg_free(msg); - -out: +err: return ret; } @@ -509,8 +518,9 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf ret = -EINVAL; goto out; } + tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); + tunnel = l2tp_tunnel_get(net, tunnel_id); if (!tunnel) { ret = -ENODEV; goto out; @@ -518,24 +528,24 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf if (!info->attrs[L2TP_ATTR_SESSION_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } session_id = nla_get_u32(info->attrs[L2TP_ATTR_SESSION_ID]); if (!info->attrs[L2TP_ATTR_PEER_SESSION_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } peer_session_id = nla_get_u32(info->attrs[L2TP_ATTR_PEER_SESSION_ID]); if (!info->attrs[L2TP_ATTR_PW_TYPE]) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.pw_type = nla_get_u16(info->attrs[L2TP_ATTR_PW_TYPE]); if (cfg.pw_type >= __L2TP_PWTYPE_MAX) { ret = -EINVAL; - goto out; + goto out_tunnel; } if (tunnel->version > 2) { @@ -557,7 +567,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf u16 len = nla_len(info->attrs[L2TP_ATTR_COOKIE]); if (len > 8) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.cookie_len = len; memcpy(&cfg.cookie[0], nla_data(info->attrs[L2TP_ATTR_COOKIE]), len); @@ -566,7 +576,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf u16 len = nla_len(info->attrs[L2TP_ATTR_PEER_COOKIE]); if (len > 8) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.peer_cookie_len = len; memcpy(&cfg.peer_cookie[0], nla_data(info->attrs[L2TP_ATTR_PEER_COOKIE]), len); @@ -609,7 +619,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf if ((l2tp_nl_cmd_ops[cfg.pw_type] == NULL) || (l2tp_nl_cmd_ops[cfg.pw_type]->session_create == NULL)) { ret = -EPROTONOSUPPORT; - goto out; + goto out_tunnel; } /* Check that pseudowire-specific params are present */ @@ -619,7 +629,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf case L2TP_PWTYPE_ETH_VLAN: if (!info->attrs[L2TP_ATTR_VLAN_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } break; case L2TP_PWTYPE_ETH: @@ -633,10 +643,10 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf break; } - ret = -EPROTONOSUPPORT; - if (l2tp_nl_cmd_ops[cfg.pw_type]->session_create) - ret = (*l2tp_nl_cmd_ops[cfg.pw_type]->session_create)(net, tunnel_id, - session_id, peer_session_id, &cfg); + ret = l2tp_nl_cmd_ops[cfg.pw_type]->session_create(net, tunnel, + session_id, + peer_session_id, + &cfg); if (ret >= 0) { session = l2tp_session_get(net, tunnel, session_id, false); @@ -647,6 +657,8 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf } } +out_tunnel: + l2tp_tunnel_dec_refcount(tunnel); out: return ret; } diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index f0edb7209079..50e3ee9a9d61 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -788,25 +788,20 @@ end: #ifdef CONFIG_L2TP_V3 -/* Called when creating sessions via the netlink interface. - */ -static int pppol2tp_session_create(struct net *net, u32 tunnel_id, u32 session_id, u32 peer_session_id, struct l2tp_session_cfg *cfg) +/* Called when creating sessions via the netlink interface. */ +static int pppol2tp_session_create(struct net *net, struct l2tp_tunnel *tunnel, + u32 session_id, u32 peer_session_id, + struct l2tp_session_cfg *cfg) { int error; - struct l2tp_tunnel *tunnel; struct l2tp_session *session; struct pppol2tp_session *ps; - tunnel = l2tp_tunnel_find(net, tunnel_id); - - /* Error if we can't find the tunnel */ - error = -ENOENT; - if (tunnel == NULL) - goto out; - /* Error if tunnel socket is not prepped */ - if (tunnel->sock == NULL) + if (!tunnel->sock) { + error = -ENOENT; goto out; + } /* Default MTU values. */ if (cfg->mtu == 0) diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 8708cbe8af5b..2b36eff5d97e 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -7,7 +7,7 @@ * Copyright 2006-2007 Jiri Benc <jbenc@suse.cz> * Copyright 2007, Michael Wu <flamingice@sourmilk.net> * Copyright 2007-2010, Intel Corporation - * Copyright(c) 2015 Intel Deutschland GmbH + * Copyright(c) 2015-2017 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -466,3 +466,23 @@ void ieee80211_manage_rx_ba_offl(struct ieee80211_vif *vif, rcu_read_unlock(); } EXPORT_SYMBOL(ieee80211_manage_rx_ba_offl); + +void ieee80211_rx_ba_timer_expired(struct ieee80211_vif *vif, + const u8 *addr, unsigned int tid) +{ + struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); + struct ieee80211_local *local = sdata->local; + struct sta_info *sta; + + rcu_read_lock(); + sta = sta_info_get_bss(sdata, addr); + if (!sta) + goto unlock; + + set_bit(tid, sta->ampdu_mlme.tid_rx_timer_expired); + ieee80211_queue_work(&local->hw, &sta->ampdu_mlme.work); + + unlock: + rcu_read_unlock(); +} +EXPORT_SYMBOL(ieee80211_rx_ba_timer_expired); diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index ea4f481839dd..c5b9ce41d66f 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -2479,12 +2479,12 @@ static int __init mpls_init(void) rtnl_af_register(&mpls_af_ops); - rtnl_register(PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, NULL); - rtnl_register(PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, NULL); + rtnl_register(PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, 0); + rtnl_register(PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, 0); rtnl_register(PF_MPLS, RTM_GETROUTE, mpls_getroute, mpls_dump_routes, - NULL); + 0); rtnl_register(PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf, - mpls_netconf_dump_devconf, NULL); + mpls_netconf_dump_devconf, 0); err = 0; out: return err; diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index 1308a56f2591..af3d636534ef 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -180,6 +180,7 @@ struct ncsi_channel { #define NCSI_CHANNEL_INACTIVE 1 #define NCSI_CHANNEL_ACTIVE 2 #define NCSI_CHANNEL_INVISIBLE 3 + bool reconfigure_needed; spinlock_t lock; /* Protect filters etc */ struct ncsi_package *package; struct ncsi_channel_version version; @@ -235,6 +236,9 @@ enum { ncsi_dev_state_probe_dp, ncsi_dev_state_config_sp = 0x0301, ncsi_dev_state_config_cis, + ncsi_dev_state_config_clear_vids, + ncsi_dev_state_config_svf, + ncsi_dev_state_config_ev, ncsi_dev_state_config_sma, ncsi_dev_state_config_ebf, #if IS_ENABLED(CONFIG_IPV6) @@ -253,6 +257,12 @@ enum { ncsi_dev_state_suspend_done }; +struct vlan_vid { + struct list_head list; + __be16 proto; + u16 vid; +}; + struct ncsi_dev_priv { struct ncsi_dev ndev; /* Associated NCSI device */ unsigned int flags; /* NCSI device flags */ @@ -276,6 +286,7 @@ struct ncsi_dev_priv { struct work_struct work; /* For channel management */ struct packet_type ptype; /* NCSI packet Rx handler */ struct list_head node; /* Form NCSI device list */ + struct list_head vlan_vids; /* List of active VLAN IDs */ }; struct ncsi_cmd_arg { diff --git a/net/ncsi/ncsi-cmd.c b/net/ncsi/ncsi-cmd.c index 5e03ed190e18..7567ca63aae2 100644 --- a/net/ncsi/ncsi-cmd.c +++ b/net/ncsi/ncsi-cmd.c @@ -139,9 +139,9 @@ static int ncsi_cmd_handler_svf(struct sk_buff *skb, struct ncsi_cmd_svf_pkt *cmd; cmd = skb_put_zero(skb, sizeof(*cmd)); - cmd->vlan = htons(nca->words[0]); - cmd->index = nca->bytes[2]; - cmd->enable = nca->bytes[3]; + cmd->vlan = htons(nca->words[1]); + cmd->index = nca->bytes[6]; + cmd->enable = nca->bytes[7]; ncsi_cmd_build_header(&cmd->cmd.common, nca); return 0; @@ -153,7 +153,7 @@ static int ncsi_cmd_handler_ev(struct sk_buff *skb, struct ncsi_cmd_ev_pkt *cmd; cmd = skb_put_zero(skb, sizeof(*cmd)); - cmd->mode = nca->bytes[0]; + cmd->mode = nca->bytes[3]; ncsi_cmd_build_header(&cmd->cmd.common, nca); return 0; @@ -228,7 +228,7 @@ static struct ncsi_cmd_handler { { NCSI_PKT_CMD_AE, 8, ncsi_cmd_handler_ae }, { NCSI_PKT_CMD_SL, 8, ncsi_cmd_handler_sl }, { NCSI_PKT_CMD_GLS, 0, ncsi_cmd_handler_default }, - { NCSI_PKT_CMD_SVF, 4, ncsi_cmd_handler_svf }, + { NCSI_PKT_CMD_SVF, 8, ncsi_cmd_handler_svf }, { NCSI_PKT_CMD_EV, 4, ncsi_cmd_handler_ev }, { NCSI_PKT_CMD_DV, 0, ncsi_cmd_handler_default }, { NCSI_PKT_CMD_SMA, 8, ncsi_cmd_handler_sma }, diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c index a3bd5fa8ad09..3fd3c39e6278 100644 --- a/net/ncsi/ncsi-manage.c +++ b/net/ncsi/ncsi-manage.c @@ -38,6 +38,25 @@ static inline int ncsi_filter_size(int table) return sizes[table]; } +u32 *ncsi_get_filter(struct ncsi_channel *nc, int table, int index) +{ + struct ncsi_channel_filter *ncf; + int size; + + ncf = nc->filters[table]; + if (!ncf) + return NULL; + + size = ncsi_filter_size(table); + if (size < 0) + return NULL; + + return ncf->data + size * index; +} + +/* Find the first active filter in a filter table that matches the given + * data parameter. If data is NULL, this returns the first active filter. + */ int ncsi_find_filter(struct ncsi_channel *nc, int table, void *data) { struct ncsi_channel_filter *ncf; @@ -58,7 +77,7 @@ int ncsi_find_filter(struct ncsi_channel *nc, int table, void *data) index = -1; while ((index = find_next_bit(bitmap, ncf->total, index + 1)) < ncf->total) { - if (!memcmp(ncf->data + size * index, data, size)) { + if (!data || !memcmp(ncf->data + size * index, data, size)) { spin_unlock_irqrestore(&nc->lock, flags); return index; } @@ -639,6 +658,95 @@ error: nd->state = ncsi_dev_state_functional; } +/* Check the VLAN filter bitmap for a set filter, and construct a + * "Set VLAN Filter - Disable" packet if found. + */ +static int clear_one_vid(struct ncsi_dev_priv *ndp, struct ncsi_channel *nc, + struct ncsi_cmd_arg *nca) +{ + int index; + u32 *data; + u16 vid; + + index = ncsi_find_filter(nc, NCSI_FILTER_VLAN, NULL); + if (index < 0) { + /* Filter table empty */ + return -1; + } + + data = ncsi_get_filter(nc, NCSI_FILTER_VLAN, index); + if (!data) { + netdev_err(ndp->ndev.dev, + "ncsi: failed to retrieve filter %d\n", index); + /* Set the VLAN id to 0 - this will still disable the entry in + * the filter table, but we won't know what it was. + */ + vid = 0; + } else { + vid = *(u16 *)data; + } + + netdev_printk(KERN_DEBUG, ndp->ndev.dev, + "ncsi: removed vlan tag %u at index %d\n", + vid, index + 1); + ncsi_remove_filter(nc, NCSI_FILTER_VLAN, index); + + nca->type = NCSI_PKT_CMD_SVF; + nca->words[1] = vid; + /* HW filter index starts at 1 */ + nca->bytes[6] = index + 1; + nca->bytes[7] = 0x00; + return 0; +} + +/* Find an outstanding VLAN tag and constuct a "Set VLAN Filter - Enable" + * packet. + */ +static int set_one_vid(struct ncsi_dev_priv *ndp, struct ncsi_channel *nc, + struct ncsi_cmd_arg *nca) +{ + struct vlan_vid *vlan = NULL; + int index = 0; + + list_for_each_entry_rcu(vlan, &ndp->vlan_vids, list) { + index = ncsi_find_filter(nc, NCSI_FILTER_VLAN, &vlan->vid); + if (index < 0) { + /* New tag to add */ + netdev_printk(KERN_DEBUG, ndp->ndev.dev, + "ncsi: new vlan id to set: %u\n", + vlan->vid); + break; + } + netdev_printk(KERN_DEBUG, ndp->ndev.dev, + "vid %u already at filter pos %d\n", + vlan->vid, index); + } + + if (!vlan || index >= 0) { + netdev_printk(KERN_DEBUG, ndp->ndev.dev, + "no vlan ids left to set\n"); + return -1; + } + + index = ncsi_add_filter(nc, NCSI_FILTER_VLAN, &vlan->vid); + if (index < 0) { + netdev_err(ndp->ndev.dev, + "Failed to add new VLAN tag, error %d\n", index); + return -1; + } + + netdev_printk(KERN_DEBUG, ndp->ndev.dev, + "ncsi: set vid %u in packet, index %u\n", + vlan->vid, index + 1); + nca->type = NCSI_PKT_CMD_SVF; + nca->words[1] = vlan->vid; + /* HW filter index starts at 1 */ + nca->bytes[6] = index + 1; + nca->bytes[7] = 0x01; + + return 0; +} + static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) { struct ncsi_dev *nd = &ndp->ndev; @@ -683,8 +791,11 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) if (ret) goto error; - nd->state = ncsi_dev_state_config_sma; + nd->state = ncsi_dev_state_config_clear_vids; break; + case ncsi_dev_state_config_clear_vids: + case ncsi_dev_state_config_svf: + case ncsi_dev_state_config_ev: case ncsi_dev_state_config_sma: case ncsi_dev_state_config_ebf: #if IS_ENABLED(CONFIG_IPV6) @@ -699,11 +810,40 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) nca.package = np->id; nca.channel = nc->id; + /* Clear any active filters on the channel before setting */ + if (nd->state == ncsi_dev_state_config_clear_vids) { + ret = clear_one_vid(ndp, nc, &nca); + if (ret) { + nd->state = ncsi_dev_state_config_svf; + schedule_work(&ndp->work); + break; + } + /* Repeat */ + nd->state = ncsi_dev_state_config_clear_vids; + /* Add known VLAN tags to the filter */ + } else if (nd->state == ncsi_dev_state_config_svf) { + ret = set_one_vid(ndp, nc, &nca); + if (ret) { + nd->state = ncsi_dev_state_config_ev; + schedule_work(&ndp->work); + break; + } + /* Repeat */ + nd->state = ncsi_dev_state_config_svf; + /* Enable/Disable the VLAN filter */ + } else if (nd->state == ncsi_dev_state_config_ev) { + if (list_empty(&ndp->vlan_vids)) { + nca.type = NCSI_PKT_CMD_DV; + } else { + nca.type = NCSI_PKT_CMD_EV; + nca.bytes[3] = NCSI_CAP_VLAN_NO; + } + nd->state = ncsi_dev_state_config_sma; + } else if (nd->state == ncsi_dev_state_config_sma) { /* Use first entry in unicast filter table. Note that * the MAC filter table starts from entry 1 instead of * 0. */ - if (nd->state == ncsi_dev_state_config_sma) { nca.type = NCSI_PKT_CMD_SMA; for (index = 0; index < 6; index++) nca.bytes[index] = dev->dev_addr[index]; @@ -751,6 +891,25 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) break; case ncsi_dev_state_config_done: spin_lock_irqsave(&nc->lock, flags); + if (nc->reconfigure_needed) { + /* This channel's configuration has been updated + * part-way during the config state - start the + * channel configuration over + */ + nc->reconfigure_needed = false; + nc->state = NCSI_CHANNEL_INACTIVE; + spin_unlock_irqrestore(&nc->lock, flags); + + spin_lock_irqsave(&ndp->lock, flags); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + netdev_printk(KERN_DEBUG, dev, + "Dirty NCSI channel state reset\n"); + ncsi_process_next_channel(ndp); + break; + } + if (nc->modes[NCSI_MODE_LINK].data[2] & 0x1) { hot_nc = nc; nc->state = NCSI_CHANNEL_ACTIVE; @@ -1191,6 +1350,150 @@ static struct notifier_block ncsi_inet6addr_notifier = { }; #endif /* CONFIG_IPV6 */ +static int ncsi_kick_channels(struct ncsi_dev_priv *ndp) +{ + struct ncsi_dev *nd = &ndp->ndev; + struct ncsi_channel *nc; + struct ncsi_package *np; + unsigned long flags; + unsigned int n = 0; + + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + spin_lock_irqsave(&nc->lock, flags); + + /* Channels may be busy, mark dirty instead of + * kicking if; + * a) not ACTIVE (configured) + * b) in the channel_queue (to be configured) + * c) it's ndev is in the config state + */ + if (nc->state != NCSI_CHANNEL_ACTIVE) { + if ((ndp->ndev.state & 0xff00) == + ncsi_dev_state_config || + !list_empty(&nc->link)) { + netdev_printk(KERN_DEBUG, nd->dev, + "ncsi: channel %p marked dirty\n", + nc); + nc->reconfigure_needed = true; + } + spin_unlock_irqrestore(&nc->lock, flags); + continue; + } + + spin_unlock_irqrestore(&nc->lock, flags); + + ncsi_stop_channel_monitor(nc); + spin_lock_irqsave(&nc->lock, flags); + nc->state = NCSI_CHANNEL_INACTIVE; + spin_unlock_irqrestore(&nc->lock, flags); + + spin_lock_irqsave(&ndp->lock, flags); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + netdev_printk(KERN_DEBUG, nd->dev, + "ncsi: kicked channel %p\n", nc); + n++; + } + } + + return n; +} + +int ncsi_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid) +{ + struct ncsi_channel_filter *ncf; + struct ncsi_dev_priv *ndp; + unsigned int n_vids = 0; + struct vlan_vid *vlan; + struct ncsi_dev *nd; + bool found = false; + + if (vid == 0) + return 0; + + nd = ncsi_find_dev(dev); + if (!nd) { + netdev_warn(dev, "ncsi: No net_device?\n"); + return 0; + } + + ndp = TO_NCSI_DEV_PRIV(nd); + ncf = ndp->hot_channel->filters[NCSI_FILTER_VLAN]; + + /* Add the VLAN id to our internal list */ + list_for_each_entry_rcu(vlan, &ndp->vlan_vids, list) { + n_vids++; + if (vlan->vid == vid) { + netdev_printk(KERN_DEBUG, dev, + "vid %u already registered\n", vid); + return 0; + } + } + + if (n_vids >= ncf->total) { + netdev_info(dev, + "NCSI Channel supports up to %u VLAN tags but %u are already set\n", + ncf->total, n_vids); + return -EINVAL; + } + + vlan = kzalloc(sizeof(*vlan), GFP_KERNEL); + if (!vlan) + return -ENOMEM; + + vlan->proto = proto; + vlan->vid = vid; + list_add_rcu(&vlan->list, &ndp->vlan_vids); + + netdev_printk(KERN_DEBUG, dev, "Added new vid %u\n", vid); + + found = ncsi_kick_channels(ndp) != 0; + + return found ? ncsi_process_next_channel(ndp) : 0; +} +EXPORT_SYMBOL_GPL(ncsi_vlan_rx_add_vid); + +int ncsi_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid) +{ + struct vlan_vid *vlan, *tmp; + struct ncsi_dev_priv *ndp; + struct ncsi_dev *nd; + bool found = false; + + if (vid == 0) + return 0; + + nd = ncsi_find_dev(dev); + if (!nd) { + netdev_warn(dev, "ncsi: no net_device?\n"); + return 0; + } + + ndp = TO_NCSI_DEV_PRIV(nd); + + /* Remove the VLAN id from our internal list */ + list_for_each_entry_safe(vlan, tmp, &ndp->vlan_vids, list) + if (vlan->vid == vid) { + netdev_printk(KERN_DEBUG, dev, + "vid %u found, removing\n", vid); + list_del_rcu(&vlan->list); + found = true; + kfree(vlan); + } + + if (!found) { + netdev_err(dev, "ncsi: vid %u wasn't registered!\n", vid); + return -EINVAL; + } + + found = ncsi_kick_channels(ndp) != 0; + + return found ? ncsi_process_next_channel(ndp) : 0; +} +EXPORT_SYMBOL_GPL(ncsi_vlan_rx_kill_vid); + struct ncsi_dev *ncsi_register_dev(struct net_device *dev, void (*handler)(struct ncsi_dev *ndev)) { @@ -1215,6 +1518,7 @@ struct ncsi_dev *ncsi_register_dev(struct net_device *dev, nd->handler = handler; ndp->pending_req_num = 0; INIT_LIST_HEAD(&ndp->channel_queue); + INIT_LIST_HEAD(&ndp->vlan_vids); INIT_WORK(&ndp->work, ncsi_dev_work); /* Initialize private NCSI device */ diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h index 3ea49ed0a935..91b4b66438df 100644 --- a/net/ncsi/ncsi-pkt.h +++ b/net/ncsi/ncsi-pkt.h @@ -104,7 +104,7 @@ struct ncsi_cmd_svf_pkt { unsigned char index; /* VLAN table index */ unsigned char enable; /* Enable or disable */ __be32 checksum; /* Checksum */ - unsigned char pad[14]; + unsigned char pad[18]; }; /* Enable VLAN */ diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c index 087db775b3dc..265b9a892d41 100644 --- a/net/ncsi/ncsi-rsp.c +++ b/net/ncsi/ncsi-rsp.c @@ -354,7 +354,8 @@ static int ncsi_rsp_handler_svf(struct ncsi_request *nr) /* Add or remove the VLAN filter */ if (!(cmd->enable & 0x1)) { - ret = ncsi_remove_filter(nc, NCSI_FILTER_VLAN, cmd->index); + /* HW indexes from 1 */ + ret = ncsi_remove_filter(nc, NCSI_FILTER_VLAN, cmd->index - 1); } else { vlan = ntohs(cmd->vlan); ret = ncsi_add_filter(nc, NCSI_FILTER_VLAN, &vlan); @@ -693,7 +694,14 @@ static int ncsi_rsp_handler_gc(struct ncsi_request *nr) ncf->index = i; ncf->total = cnt; - ncf->bitmap = 0x0ul; + if (i == NCSI_FILTER_VLAN) { + /* Set VLAN filters active so they are cleared in + * first configuration state + */ + ncf->bitmap = U64_MAX; + } else { + ncf->bitmap = 0x0ul; + } nc->filters[i] = ncf; } diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 9b28864cc36a..e4a13cc8a2e7 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -636,6 +636,15 @@ config NFT_FWD_NETDEV help This option enables packet forwarding for the "netdev" family. +config NFT_FIB_NETDEV + depends on NFT_FIB_IPV4 + depends on NFT_FIB_IPV6 + tristate "Netfilter nf_tables netdev fib lookups support" + help + This option allows using the FIB expression from the netdev table. + The lookup will be delegated to the IPv4 or IPv6 FIB depending + on the protocol of the packet. + endif # NF_TABLES_NETDEV endif # NF_TABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 913380919301..d3891c93edd6 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -100,6 +100,7 @@ obj-$(CONFIG_NFT_REDIR) += nft_redir.o obj-$(CONFIG_NFT_HASH) += nft_hash.o obj-$(CONFIG_NFT_FIB) += nft_fib.o obj-$(CONFIG_NFT_FIB_INET) += nft_fib_inet.o +obj-$(CONFIG_NFT_FIB_NETDEV) += nft_fib_netdev.o # nf_tables netdev obj-$(CONFIG_NFT_DUP_NETDEV) += nft_dup_netdev.o diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 552d606e57ca..04fe25abc5f6 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -21,7 +21,7 @@ #include <linux/inetdevice.h> #include <linux/proc_fs.h> #include <linux/mutex.h> -#include <linux/slab.h> +#include <linux/mm.h> #include <linux/rcupdate.h> #include <net/net_namespace.h> #include <net/sock.h> @@ -62,10 +62,182 @@ EXPORT_SYMBOL(nf_hooks_needed); #endif static DEFINE_MUTEX(nf_hook_mutex); + +/* max hooks per family/hooknum */ +#define MAX_HOOK_COUNT 1024 + #define nf_entry_dereference(e) \ rcu_dereference_protected(e, lockdep_is_held(&nf_hook_mutex)) -static struct nf_hook_entry __rcu **nf_hook_entry_head(struct net *net, const struct nf_hook_ops *reg) +static struct nf_hook_entries *allocate_hook_entries_size(u16 num) +{ + struct nf_hook_entries *e; + size_t alloc = sizeof(*e) + + sizeof(struct nf_hook_entry) * num + + sizeof(struct nf_hook_ops *) * num; + + if (num == 0) + return NULL; + + e = kvzalloc(alloc, GFP_KERNEL); + if (e) + e->num_hook_entries = num; + return e; +} + +static unsigned int accept_all(void *priv, + struct sk_buff *skb, + const struct nf_hook_state *state) +{ + return NF_ACCEPT; /* ACCEPT makes nf_hook_slow call next hook */ +} + +static const struct nf_hook_ops dummy_ops = { + .hook = accept_all, + .priority = INT_MIN, +}; + +static struct nf_hook_entries * +nf_hook_entries_grow(const struct nf_hook_entries *old, + const struct nf_hook_ops *reg) +{ + unsigned int i, alloc_entries, nhooks, old_entries; + struct nf_hook_ops **orig_ops = NULL; + struct nf_hook_ops **new_ops; + struct nf_hook_entries *new; + bool inserted = false; + + alloc_entries = 1; + old_entries = old ? old->num_hook_entries : 0; + + if (old) { + orig_ops = nf_hook_entries_get_hook_ops(old); + + for (i = 0; i < old_entries; i++) { + if (orig_ops[i] != &dummy_ops) + alloc_entries++; + } + } + + if (alloc_entries > MAX_HOOK_COUNT) + return ERR_PTR(-E2BIG); + + new = allocate_hook_entries_size(alloc_entries); + if (!new) + return ERR_PTR(-ENOMEM); + + new_ops = nf_hook_entries_get_hook_ops(new); + + i = 0; + nhooks = 0; + while (i < old_entries) { + if (orig_ops[i] == &dummy_ops) { + ++i; + continue; + } + if (inserted || reg->priority > orig_ops[i]->priority) { + new_ops[nhooks] = (void *)orig_ops[i]; + new->hooks[nhooks] = old->hooks[i]; + i++; + } else { + new_ops[nhooks] = (void *)reg; + new->hooks[nhooks].hook = reg->hook; + new->hooks[nhooks].priv = reg->priv; + inserted = true; + } + nhooks++; + } + + if (!inserted) { + new_ops[nhooks] = (void *)reg; + new->hooks[nhooks].hook = reg->hook; + new->hooks[nhooks].priv = reg->priv; + } + + return new; +} + +static void hooks_validate(const struct nf_hook_entries *hooks) +{ +#ifdef CONFIG_DEBUG_KERNEL + struct nf_hook_ops **orig_ops; + int prio = INT_MIN; + size_t i = 0; + + orig_ops = nf_hook_entries_get_hook_ops(hooks); + + for (i = 0; i < hooks->num_hook_entries; i++) { + if (orig_ops[i] == &dummy_ops) + continue; + + WARN_ON(orig_ops[i]->priority < prio); + + if (orig_ops[i]->priority > prio) + prio = orig_ops[i]->priority; + } +#endif +} + +/* + * __nf_hook_entries_try_shrink - try to shrink hook array + * + * @pp -- location of hook blob + * + * Hook unregistration must always succeed, so to-be-removed hooks + * are replaced by a dummy one that will just move to next hook. + * + * This counts the current dummy hooks, attempts to allocate new blob, + * copies the live hooks, then replaces and discards old one. + * + * return values: + * + * Returns address to free, or NULL. + */ +static void *__nf_hook_entries_try_shrink(struct nf_hook_entries __rcu **pp) +{ + struct nf_hook_entries *old, *new = NULL; + unsigned int i, j, skip = 0, hook_entries; + struct nf_hook_ops **orig_ops; + struct nf_hook_ops **new_ops; + + old = nf_entry_dereference(*pp); + if (WARN_ON_ONCE(!old)) + return NULL; + + orig_ops = nf_hook_entries_get_hook_ops(old); + for (i = 0; i < old->num_hook_entries; i++) { + if (orig_ops[i] == &dummy_ops) + skip++; + } + + /* if skip == hook_entries all hooks have been removed */ + hook_entries = old->num_hook_entries; + if (skip == hook_entries) + goto out_assign; + + if (WARN_ON(skip == 0)) + return NULL; + + hook_entries -= skip; + new = allocate_hook_entries_size(hook_entries); + if (!new) + return NULL; + + new_ops = nf_hook_entries_get_hook_ops(new); + for (i = 0, j = 0; i < old->num_hook_entries; i++) { + if (orig_ops[i] == &dummy_ops) + continue; + new->hooks[j] = old->hooks[i]; + new_ops[j] = (void *)orig_ops[i]; + j++; + } + hooks_validate(new); +out_assign: + rcu_assign_pointer(*pp, new); + return old; +} + +static struct nf_hook_entries __rcu **nf_hook_entry_head(struct net *net, const struct nf_hook_ops *reg) { if (reg->pf != NFPROTO_NETDEV) return net->nf.hooks[reg->pf]+reg->hooknum; @@ -76,13 +248,14 @@ static struct nf_hook_entry __rcu **nf_hook_entry_head(struct net *net, const st return ®->dev->nf_hooks_ingress; } #endif + WARN_ON_ONCE(1); return NULL; } int nf_register_net_hook(struct net *net, const struct nf_hook_ops *reg) { - struct nf_hook_entry __rcu **pp; - struct nf_hook_entry *entry, *p; + struct nf_hook_entries *p, *new_hooks; + struct nf_hook_entries __rcu **pp; if (reg->pf == NFPROTO_NETDEV) { #ifndef CONFIG_NETFILTER_INGRESS @@ -98,23 +271,19 @@ int nf_register_net_hook(struct net *net, const struct nf_hook_ops *reg) if (!pp) return -EINVAL; - entry = kmalloc(sizeof(*entry), GFP_KERNEL); - if (!entry) - return -ENOMEM; - - nf_hook_entry_init(entry, reg); - mutex_lock(&nf_hook_mutex); - /* Find the spot in the list */ - for (; (p = nf_entry_dereference(*pp)) != NULL; pp = &p->next) { - if (reg->priority < nf_hook_entry_priority(p)) - break; - } - rcu_assign_pointer(entry->next, p); - rcu_assign_pointer(*pp, entry); + p = nf_entry_dereference(*pp); + new_hooks = nf_hook_entries_grow(p, reg); + + if (!IS_ERR(new_hooks)) + rcu_assign_pointer(*pp, new_hooks); mutex_unlock(&nf_hook_mutex); + if (IS_ERR(new_hooks)) + return PTR_ERR(new_hooks); + + hooks_validate(new_hooks); #ifdef CONFIG_NETFILTER_INGRESS if (reg->pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) net_inc_ingress_queue(); @@ -122,48 +291,74 @@ int nf_register_net_hook(struct net *net, const struct nf_hook_ops *reg) #ifdef HAVE_JUMP_LABEL static_key_slow_inc(&nf_hooks_needed[reg->pf][reg->hooknum]); #endif + synchronize_net(); + BUG_ON(p == new_hooks); + kvfree(p); return 0; } EXPORT_SYMBOL(nf_register_net_hook); -static struct nf_hook_entry * -__nf_unregister_net_hook(struct net *net, const struct nf_hook_ops *reg) +/* + * __nf_unregister_net_hook - remove a hook from blob + * + * @oldp: current address of hook blob + * @unreg: hook to unregister + * + * This cannot fail, hook unregistration must always succeed. + * Therefore replace the to-be-removed hook with a dummy hook. + */ +static void __nf_unregister_net_hook(struct nf_hook_entries *old, + const struct nf_hook_ops *unreg) { - struct nf_hook_entry __rcu **pp; - struct nf_hook_entry *p; - - pp = nf_hook_entry_head(net, reg); - if (WARN_ON_ONCE(!pp)) - return NULL; + struct nf_hook_ops **orig_ops; + bool found = false; + unsigned int i; - mutex_lock(&nf_hook_mutex); - for (; (p = nf_entry_dereference(*pp)) != NULL; pp = &p->next) { - if (nf_hook_entry_ops(p) == reg) { - rcu_assign_pointer(*pp, p->next); - break; - } - } - mutex_unlock(&nf_hook_mutex); - if (!p) { - WARN(1, "nf_unregister_net_hook: hook not found!\n"); - return NULL; + orig_ops = nf_hook_entries_get_hook_ops(old); + for (i = 0; i < old->num_hook_entries; i++) { + if (orig_ops[i] != unreg) + continue; + WRITE_ONCE(old->hooks[i].hook, accept_all); + WRITE_ONCE(orig_ops[i], &dummy_ops); + found = true; + break; } + + if (found) { #ifdef CONFIG_NETFILTER_INGRESS - if (reg->pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) - net_dec_ingress_queue(); + if (unreg->pf == NFPROTO_NETDEV && unreg->hooknum == NF_NETDEV_INGRESS) + net_dec_ingress_queue(); #endif #ifdef HAVE_JUMP_LABEL - static_key_slow_dec(&nf_hooks_needed[reg->pf][reg->hooknum]); + static_key_slow_dec(&nf_hooks_needed[unreg->pf][unreg->hooknum]); #endif - - return p; + } else { + WARN_ONCE(1, "hook not found, pf %d num %d", unreg->pf, unreg->hooknum); + } } void nf_unregister_net_hook(struct net *net, const struct nf_hook_ops *reg) { - struct nf_hook_entry *p = __nf_unregister_net_hook(net, reg); + struct nf_hook_entries __rcu **pp; + struct nf_hook_entries *p; unsigned int nfq; + pp = nf_hook_entry_head(net, reg); + if (!pp) + return; + + mutex_lock(&nf_hook_mutex); + + p = nf_entry_dereference(*pp); + if (WARN_ON_ONCE(!p)) { + mutex_unlock(&nf_hook_mutex); + return; + } + + __nf_unregister_net_hook(p, reg); + + p = __nf_hook_entries_try_shrink(pp); + mutex_unlock(&nf_hook_mutex); if (!p) return; @@ -173,7 +368,7 @@ void nf_unregister_net_hook(struct net *net, const struct nf_hook_ops *reg) nfq = nf_queue_nf_hook_drop(net); if (nfq) synchronize_net(); - kfree(p); + kvfree(p); } EXPORT_SYMBOL(nf_unregister_net_hook); @@ -200,154 +395,78 @@ EXPORT_SYMBOL(nf_register_net_hooks); void nf_unregister_net_hooks(struct net *net, const struct nf_hook_ops *reg, unsigned int hookcount) { - struct nf_hook_entry *to_free[16]; - unsigned int i, n, nfq; - - do { - n = min_t(unsigned int, hookcount, ARRAY_SIZE(to_free)); - - for (i = 0; i < n; i++) - to_free[i] = __nf_unregister_net_hook(net, ®[i]); - - synchronize_net(); - - /* need 2nd synchronize_net() if nfqueue is used, skb - * can get reinjected right before nf_queue_hook_drop() - */ - nfq = nf_queue_nf_hook_drop(net); - if (nfq) - synchronize_net(); - - for (i = 0; i < n; i++) - kfree(to_free[i]); - - reg += n; - hookcount -= n; - } while (hookcount > 0); -} -EXPORT_SYMBOL(nf_unregister_net_hooks); - -static LIST_HEAD(nf_hook_list); - -static int _nf_register_hook(struct nf_hook_ops *reg) -{ - struct net *net, *last; - int ret; - - for_each_net(net) { - ret = nf_register_net_hook(net, reg); - if (ret && ret != -ENOENT) - goto rollback; - } - list_add_tail(®->list, &nf_hook_list); + struct nf_hook_entries *to_free[16], *p; + struct nf_hook_entries __rcu **pp; + unsigned int i, j, n; - return 0; -rollback: - last = net; - for_each_net(net) { - if (net == last) - break; - nf_unregister_net_hook(net, reg); + mutex_lock(&nf_hook_mutex); + for (i = 0; i < hookcount; i++) { + pp = nf_hook_entry_head(net, ®[i]); + if (!pp) + continue; + + p = nf_entry_dereference(*pp); + if (WARN_ON_ONCE(!p)) + continue; + __nf_unregister_net_hook(p, ®[i]); } - return ret; -} - -int nf_register_hook(struct nf_hook_ops *reg) -{ - int ret; - - rtnl_lock(); - ret = _nf_register_hook(reg); - rtnl_unlock(); - - return ret; -} -EXPORT_SYMBOL(nf_register_hook); + mutex_unlock(&nf_hook_mutex); -static void _nf_unregister_hook(struct nf_hook_ops *reg) -{ - struct net *net; + do { + n = min_t(unsigned int, hookcount, ARRAY_SIZE(to_free)); - list_del(®->list); - for_each_net(net) - nf_unregister_net_hook(net, reg); -} + mutex_lock(&nf_hook_mutex); -void nf_unregister_hook(struct nf_hook_ops *reg) -{ - rtnl_lock(); - _nf_unregister_hook(reg); - rtnl_unlock(); -} -EXPORT_SYMBOL(nf_unregister_hook); + for (i = 0, j = 0; i < hookcount && j < n; i++) { + pp = nf_hook_entry_head(net, ®[i]); + if (!pp) + continue; -int nf_register_hooks(struct nf_hook_ops *reg, unsigned int n) -{ - unsigned int i; - int err = 0; + p = nf_entry_dereference(*pp); + if (!p) + continue; - for (i = 0; i < n; i++) { - err = nf_register_hook(®[i]); - if (err) - goto err; - } - return err; + to_free[j] = __nf_hook_entries_try_shrink(pp); + if (to_free[j]) + ++j; + } -err: - if (i > 0) - nf_unregister_hooks(reg, i); - return err; -} -EXPORT_SYMBOL(nf_register_hooks); + mutex_unlock(&nf_hook_mutex); -/* Caller MUST take rtnl_lock() */ -int _nf_register_hooks(struct nf_hook_ops *reg, unsigned int n) -{ - unsigned int i; - int err = 0; + if (j) { + unsigned int nfq; - for (i = 0; i < n; i++) { - err = _nf_register_hook(®[i]); - if (err) - goto err; - } - return err; + synchronize_net(); -err: - if (i > 0) - _nf_unregister_hooks(reg, i); - return err; -} -EXPORT_SYMBOL(_nf_register_hooks); + /* need 2nd synchronize_net() if nfqueue is used, skb + * can get reinjected right before nf_queue_hook_drop() + */ + nfq = nf_queue_nf_hook_drop(net); + if (nfq) + synchronize_net(); -void nf_unregister_hooks(struct nf_hook_ops *reg, unsigned int n) -{ - while (n-- > 0) - nf_unregister_hook(®[n]); -} -EXPORT_SYMBOL(nf_unregister_hooks); + for (i = 0; i < j; i++) + kvfree(to_free[i]); + } -/* Caller MUST take rtnl_lock */ -void _nf_unregister_hooks(struct nf_hook_ops *reg, unsigned int n) -{ - while (n-- > 0) - _nf_unregister_hook(®[n]); + reg += n; + hookcount -= n; + } while (hookcount > 0); } -EXPORT_SYMBOL(_nf_unregister_hooks); +EXPORT_SYMBOL(nf_unregister_net_hooks); /* Returns 1 if okfn() needs to be executed by the caller, * -EPERM for NF_DROP, 0 otherwise. Caller must hold rcu_read_lock. */ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state, - struct nf_hook_entry *entry) + const struct nf_hook_entries *e, unsigned int s) { unsigned int verdict; int ret; - do { - verdict = nf_hook_entry_hookfn(entry, skb, state); + for (; s < e->num_hook_entries; s++) { + verdict = nf_hook_entry_hookfn(&e->hooks[s], skb, state); switch (verdict & NF_VERDICT_MASK) { case NF_ACCEPT: - entry = rcu_dereference(entry->next); break; case NF_DROP: kfree_skb(skb); @@ -356,8 +475,8 @@ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state, ret = -EPERM; return ret; case NF_QUEUE: - ret = nf_queue(skb, state, &entry, verdict); - if (ret == 1 && entry) + ret = nf_queue(skb, state, e, s, verdict); + if (ret == 1) continue; return ret; default: @@ -366,7 +485,7 @@ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state, */ return 0; } - } while (entry); + } return 1; } @@ -450,40 +569,9 @@ void (*nf_nat_decode_session_hook)(struct sk_buff *, struct flowi *); EXPORT_SYMBOL(nf_nat_decode_session_hook); #endif -static int nf_register_hook_list(struct net *net) -{ - struct nf_hook_ops *elem; - int ret; - - rtnl_lock(); - list_for_each_entry(elem, &nf_hook_list, list) { - ret = nf_register_net_hook(net, elem); - if (ret && ret != -ENOENT) - goto out_undo; - } - rtnl_unlock(); - return 0; - -out_undo: - list_for_each_entry_continue_reverse(elem, &nf_hook_list, list) - nf_unregister_net_hook(net, elem); - rtnl_unlock(); - return ret; -} - -static void nf_unregister_hook_list(struct net *net) -{ - struct nf_hook_ops *elem; - - rtnl_lock(); - list_for_each_entry(elem, &nf_hook_list, list) - nf_unregister_net_hook(net, elem); - rtnl_unlock(); -} - static int __net_init netfilter_net_init(struct net *net) { - int i, h, ret; + int i, h; for (i = 0; i < ARRAY_SIZE(net->nf.hooks); i++) { for (h = 0; h < NF_MAX_HOOKS; h++) @@ -500,16 +588,12 @@ static int __net_init netfilter_net_init(struct net *net) return -ENOMEM; } #endif - ret = nf_register_hook_list(net); - if (ret) - remove_proc_entry("netfilter", net->proc_net); - return ret; + return 0; } static void __net_exit netfilter_net_exit(struct net *net) { - nf_unregister_hook_list(net); remove_proc_entry("netfilter", net->proc_net); } diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index e31956b58aba..5cb7cac9177d 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -125,14 +125,12 @@ ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb) s->cnt.inbytes += skb->len; u64_stats_update_end(&s->syncp); - rcu_read_lock(); svc = rcu_dereference(dest->svc); s = this_cpu_ptr(svc->stats.cpustats); u64_stats_update_begin(&s->syncp); s->cnt.inpkts++; s->cnt.inbytes += skb->len; u64_stats_update_end(&s->syncp); - rcu_read_unlock(); s = this_cpu_ptr(ipvs->tot_stats.cpustats); u64_stats_update_begin(&s->syncp); @@ -159,14 +157,12 @@ ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) s->cnt.outbytes += skb->len; u64_stats_update_end(&s->syncp); - rcu_read_lock(); svc = rcu_dereference(dest->svc); s = this_cpu_ptr(svc->stats.cpustats); u64_stats_update_begin(&s->syncp); s->cnt.outpkts++; s->cnt.outbytes += skb->len; u64_stats_update_end(&s->syncp); - rcu_read_unlock(); s = this_cpu_ptr(ipvs->tot_stats.cpustats); u64_stats_update_begin(&s->syncp); @@ -1222,7 +1218,6 @@ static struct ip_vs_conn *__ip_vs_rs_conn_out(unsigned int hooknum, if (!pptr) return NULL; - rcu_read_lock(); dest = ip_vs_find_real_service(ipvs, af, iph->protocol, &iph->saddr, pptr[0]); if (dest) { @@ -1237,7 +1232,6 @@ static struct ip_vs_conn *__ip_vs_rs_conn_out(unsigned int hooknum, pptr[0], pptr[1]); } } - rcu_read_unlock(); return cp; } @@ -1689,11 +1683,9 @@ ip_vs_in_icmp(struct netns_ipvs *ipvs, struct sk_buff *skb, int *related, if (dest) { struct ip_vs_dest_dst *dest_dst; - rcu_read_lock(); dest_dst = rcu_dereference(dest->dest_dst); if (dest_dst) mtu = dst_mtu(dest_dst->dst_cache); - rcu_read_unlock(); } if (mtu > 68 + sizeof(struct iphdr)) mtu -= sizeof(struct iphdr); @@ -2109,7 +2101,7 @@ ip_vs_forward_icmp_v6(void *priv, struct sk_buff *skb, #endif -static struct nf_hook_ops ip_vs_ops[] __read_mostly = { +static const struct nf_hook_ops ip_vs_ops[] = { /* After packet filtering, change source only for VS/NAT */ { .hook = ip_vs_reply4, diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 1fa3c2307b6e..4f940d7eb2f7 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -550,18 +550,15 @@ bool ip_vs_has_real_service(struct netns_ipvs *ipvs, int af, __u16 protocol, /* Check for "full" addressed entries */ hash = ip_vs_rs_hashkey(af, daddr, dport); - rcu_read_lock(); hlist_for_each_entry_rcu(dest, &ipvs->rs_table[hash], d_list) { if (dest->port == dport && dest->af == af && ip_vs_addr_equal(af, &dest->addr, daddr) && (dest->protocol == protocol || dest->vfwmark)) { /* HIT */ - rcu_read_unlock(); return true; } } - rcu_read_unlock(); return false; } diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index fb780be76d15..3e17d32b629d 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -269,13 +269,11 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, * hopefully it will succeed on the retransmitted * packet. */ - rcu_read_lock(); mangled = nf_nat_mangle_tcp_packet(skb, ct, ctinfo, iph->ihl * 4, start - data, end - start, buf, buf_len); - rcu_read_unlock(); if (mangled) { ip_vs_nfct_expect_related(skb, ct, n_cp, IPPROTO_TCP, 0, 0); diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 3ffad4adaddf..e1efa446b305 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -38,7 +38,6 @@ sctp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, return 0; } - rcu_read_lock(); if (likely(!ip_vs_iph_inverse(iph))) svc = ip_vs_service_find(ipvs, af, skb->mark, iph->protocol, &iph->daddr, ports[1]); @@ -53,7 +52,6 @@ sctp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, * It seems that we are very loaded. * We have to drop this packet :( */ - rcu_read_unlock(); *verdict = NF_DROP; return 0; } @@ -67,11 +65,9 @@ sctp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, *verdict = ip_vs_leave(svc, skb, pd, iph); else *verdict = NF_DROP; - rcu_read_unlock(); return 0; } } - rcu_read_unlock(); /* NF_ACCEPT */ return 1; } @@ -526,12 +522,10 @@ static int sctp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = sctp_app_hashkey(cp->vport); - rcu_read_lock(); list_for_each_entry_rcu(inc, &ipvs->sctp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - rcu_read_unlock(); IP_VS_DBG_BUF(9, "%s: Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -544,11 +538,10 @@ static int sctp_app_conn_bind(struct ip_vs_conn *cp) cp->app = inc; if (inc->init_conn) result = inc->init_conn(inc, cp); - goto out; + break; } } - rcu_read_unlock(); -out: + return result; } diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 12dc8d5bc37d..121a321b91be 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -63,7 +63,6 @@ tcp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, } /* No !th->ack check to allow scheduling on SYN+ACK for Active FTP */ - rcu_read_lock(); if (likely(!ip_vs_iph_inverse(iph))) svc = ip_vs_service_find(ipvs, af, skb->mark, iph->protocol, @@ -80,7 +79,6 @@ tcp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, * It seems that we are very loaded. * We have to drop this packet :( */ - rcu_read_unlock(); *verdict = NF_DROP; return 0; } @@ -95,11 +93,9 @@ tcp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, *verdict = ip_vs_leave(svc, skb, pd, iph); else *verdict = NF_DROP; - rcu_read_unlock(); return 0; } } - rcu_read_unlock(); /* NF_ACCEPT */ return 1; } @@ -661,12 +657,10 @@ tcp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = tcp_app_hashkey(cp->vport); - rcu_read_lock(); list_for_each_entry_rcu(inc, &ipvs->tcp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - rcu_read_unlock(); IP_VS_DBG_BUF(9, "%s(): Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -680,12 +674,10 @@ tcp_app_conn_bind(struct ip_vs_conn *cp) cp->app = inc; if (inc->init_conn) result = inc->init_conn(inc, cp); - goto out; + break; } } - rcu_read_unlock(); - out: return result; } diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index e494e9a88c7f..30e11cd6aa8a 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -53,7 +53,6 @@ udp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, return 0; } - rcu_read_lock(); if (likely(!ip_vs_iph_inverse(iph))) svc = ip_vs_service_find(ipvs, af, skb->mark, iph->protocol, &iph->daddr, ports[1]); @@ -69,7 +68,6 @@ udp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, * It seems that we are very loaded. * We have to drop this packet :( */ - rcu_read_unlock(); *verdict = NF_DROP; return 0; } @@ -84,11 +82,9 @@ udp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, *verdict = ip_vs_leave(svc, skb, pd, iph); else *verdict = NF_DROP; - rcu_read_unlock(); return 0; } } - rcu_read_unlock(); /* NF_ACCEPT */ return 1; } @@ -410,12 +406,10 @@ static int udp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = udp_app_hashkey(cp->vport); - rcu_read_lock(); list_for_each_entry_rcu(inc, &ipvs->udp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - rcu_read_unlock(); IP_VS_DBG_BUF(9, "%s(): Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -429,12 +423,10 @@ static int udp_app_conn_bind(struct ip_vs_conn *cp) cp->app = inc; if (inc->init_conn) result = inc->init_conn(inc, cp); - goto out; + break; } } - rcu_read_unlock(); - out: return result; } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 2eab1e0400f4..90d396814798 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -678,7 +678,6 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); if (__ip_vs_get_out_rt(cp->ipvs, cp->af, skb, NULL, iph->daddr, IP_VS_RT_MODE_NON_LOCAL, NULL, ipvsh) < 0) goto tx_error; @@ -689,14 +688,12 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -710,7 +707,6 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); if (__ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, NULL, &iph->daddr, NULL, ipvsh, 0, IP_VS_RT_MODE_NON_LOCAL) < 0) @@ -720,14 +716,12 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -746,7 +740,6 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); /* check if it is a connection of no-client-port */ if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) { __be16 _pt, *p; @@ -815,14 +808,12 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); - rcu_read_unlock(); LeaveFunction(10); return rc; tx_error: kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -837,7 +828,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); /* check if it is a connection of no-client-port */ if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT && !ipvsh->fragoffs)) { __be16 _pt, *p; @@ -906,7 +896,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); - rcu_read_unlock(); LeaveFunction(10); return rc; @@ -914,7 +903,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, tx_error: LeaveFunction(10); kfree_skb(skb); - rcu_read_unlock(); return NF_STOLEN; } #endif @@ -1035,7 +1023,6 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); local = __ip_vs_get_out_rt(ipvs, cp->af, skb, cp->dest, cp->daddr.ip, IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | @@ -1043,10 +1030,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_RT_MODE_TUNNEL, &saddr, ipvsh); if (local < 0) goto tx_error; - if (local) { - rcu_read_unlock(); + if (local) return ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 1); - } rt = skb_rtable(skb); tdev = rt->dst.dev; @@ -1095,7 +1080,6 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_local_out(net, skb->sk, skb); else if (ret == NF_DROP) kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); @@ -1104,7 +1088,6 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, tx_error: if (!IS_ERR(skb)) kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -1127,7 +1110,6 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); local = __ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, cp->dest, &cp->daddr.in6, &saddr, ipvsh, 1, @@ -1136,10 +1118,8 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_RT_MODE_TUNNEL); if (local < 0) goto tx_error; - if (local) { - rcu_read_unlock(); + if (local) return ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 1); - } rt = (struct rt6_info *) skb_dst(skb); tdev = rt->dst.dev; @@ -1185,7 +1165,6 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ip6_local_out(cp->ipvs->net, skb->sk, skb); else if (ret == NF_DROP) kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); @@ -1194,7 +1173,6 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, tx_error: if (!IS_ERR(skb)) kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -1213,17 +1191,14 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); local = __ip_vs_get_out_rt(cp->ipvs, cp->af, skb, cp->dest, cp->daddr.ip, IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | IP_VS_RT_MODE_KNOWN_NH, NULL, ipvsh); if (local < 0) goto tx_error; - if (local) { - rcu_read_unlock(); + if (local) return ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 1); - } ip_send_check(ip_hdr(skb)); @@ -1231,14 +1206,12 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -1252,7 +1225,6 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - rcu_read_lock(); local = __ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, cp->dest, &cp->daddr.in6, NULL, ipvsh, 0, @@ -1261,23 +1233,19 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_RT_MODE_KNOWN_NH); if (local < 0) goto tx_error; - if (local) { - rcu_read_unlock(); + if (local) return ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 1); - } /* Another hack: avoid icmp_send in ip_fragment */ skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - rcu_read_unlock(); LeaveFunction(10); return NF_STOLEN; } @@ -1322,7 +1290,6 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, rt_mode = (hooknum != NF_INET_FORWARD) ? IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | IP_VS_RT_MODE_RDR : IP_VS_RT_MODE_NON_LOCAL; - rcu_read_lock(); local = __ip_vs_get_out_rt(cp->ipvs, cp->af, skb, cp->dest, cp->daddr.ip, rt_mode, NULL, iph); if (local < 0) @@ -1368,12 +1335,10 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); - rcu_read_unlock(); goto out; tx_error: kfree_skb(skb); - rcu_read_unlock(); rc = NF_STOLEN; out: LeaveFunction(10); @@ -1414,7 +1379,6 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, rt_mode = (hooknum != NF_INET_FORWARD) ? IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | IP_VS_RT_MODE_RDR : IP_VS_RT_MODE_NON_LOCAL; - rcu_read_lock(); local = __ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, cp->dest, &cp->daddr.in6, NULL, ipvsh, 0, rt_mode); if (local < 0) @@ -1460,12 +1424,10 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); - rcu_read_unlock(); goto out; tx_error: kfree_skb(skb); - rcu_read_unlock(); rc = NF_STOLEN; out: LeaveFunction(10); diff --git a/net/netfilter/nf_conntrack_broadcast.c b/net/netfilter/nf_conntrack_broadcast.c index 4e99cca61612..ecc3ab784633 100644 --- a/net/netfilter/nf_conntrack_broadcast.c +++ b/net/netfilter/nf_conntrack_broadcast.c @@ -40,7 +40,6 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) goto out; - rcu_read_lock(); in_dev = __in_dev_get_rcu(rt->dst.dev); if (in_dev != NULL) { for_primary_ifa(in_dev) { @@ -50,7 +49,6 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, } } endfor_ifa(in_dev); } - rcu_read_unlock(); if (mask == 0) goto out; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 9979f46c81dc..01130392b7c0 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -56,6 +56,8 @@ #include <net/netfilter/nf_nat_helper.h> #include <net/netns/hash.h> +#include "nf_internals.h" + #define NF_CONNTRACK_VERSION "0.5.0" int (*nfnetlink_parse_nat_setup_hook)(struct nf_conn *ct, @@ -96,19 +98,26 @@ static struct conntrack_gc_work conntrack_gc_work; void nf_conntrack_lock(spinlock_t *lock) __acquires(lock) { + /* 1) Acquire the lock */ spin_lock(lock); - while (unlikely(nf_conntrack_locks_all)) { - spin_unlock(lock); - /* - * Order the 'nf_conntrack_locks_all' load vs. the - * spin_unlock_wait() loads below, to ensure - * that 'nf_conntrack_locks_all_lock' is indeed held: - */ - smp_rmb(); /* spin_lock(&nf_conntrack_locks_all_lock) */ - spin_unlock_wait(&nf_conntrack_locks_all_lock); - spin_lock(lock); - } + /* 2) read nf_conntrack_locks_all, with ACQUIRE semantics + * It pairs with the smp_store_release() in nf_conntrack_all_unlock() + */ + if (likely(smp_load_acquire(&nf_conntrack_locks_all) == false)) + return; + + /* fast path failed, unlock */ + spin_unlock(lock); + + /* Slow path 1) get global lock */ + spin_lock(&nf_conntrack_locks_all_lock); + + /* Slow path 2) get the lock we want */ + spin_lock(lock); + + /* Slow path 3) release the global lock */ + spin_unlock(&nf_conntrack_locks_all_lock); } EXPORT_SYMBOL_GPL(nf_conntrack_lock); @@ -149,28 +158,27 @@ static void nf_conntrack_all_lock(void) int i; spin_lock(&nf_conntrack_locks_all_lock); - nf_conntrack_locks_all = true; - /* - * Order the above store of 'nf_conntrack_locks_all' against - * the spin_unlock_wait() loads below, such that if - * nf_conntrack_lock() observes 'nf_conntrack_locks_all' - * we must observe nf_conntrack_locks[] held: - */ - smp_mb(); /* spin_lock(&nf_conntrack_locks_all_lock) */ + nf_conntrack_locks_all = true; for (i = 0; i < CONNTRACK_LOCKS; i++) { - spin_unlock_wait(&nf_conntrack_locks[i]); + spin_lock(&nf_conntrack_locks[i]); + + /* This spin_unlock provides the "release" to ensure that + * nf_conntrack_locks_all==true is visible to everyone that + * acquired spin_lock(&nf_conntrack_locks[]). + */ + spin_unlock(&nf_conntrack_locks[i]); } } static void nf_conntrack_all_unlock(void) { - /* - * All prior stores must be complete before we clear + /* All prior stores must be complete before we clear * 'nf_conntrack_locks_all'. Otherwise nf_conntrack_lock() * might observe the false value but not the entire - * critical section: + * critical section. + * It pairs with the smp_load_acquire() in nf_conntrack_lock() */ smp_store_release(&nf_conntrack_locks_all, false); spin_unlock(&nf_conntrack_locks_all_lock); @@ -248,8 +256,8 @@ bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff, u_int16_t l3num, struct net *net, struct nf_conntrack_tuple *tuple) { - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; unsigned int protoff; u_int8_t protonum; int ret; @@ -398,22 +406,19 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; pr_debug("destroy_conntrack(%p)\n", ct); - NF_CT_ASSERT(atomic_read(&nfct->use) == 0); + WARN_ON(atomic_read(&nfct->use) != 0); if (unlikely(nf_ct_is_template(ct))) { nf_ct_tmpl_free(ct); return; } - rcu_read_lock(); l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto->destroy) l4proto->destroy(ct); - rcu_read_unlock(); - local_bh_disable(); /* Expectations will have been removed in clean_from_lists, * except TFTP can create an expectation on the first packet, @@ -695,7 +700,7 @@ static int nf_ct_resolve_clash(struct net *net, struct sk_buff *skb, { /* This is the conntrack entry already in hashes that won race. */ struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto->allow_clash && @@ -757,12 +762,11 @@ __nf_conntrack_confirm(struct sk_buff *skb) * connections for unconfirmed conns. But packet copies and * REJECT will give spurious warnings here. */ - /* NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 1); */ /* No external references means no one else could have * confirmed us. */ - NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + WARN_ON(nf_ct_is_confirmed(ct)); pr_debug("Confirming conntrack %p\n", ct); /* We have to check the DYING flag after unlink to prevent * a race against nf_ct_get_next_corpse() possibly called from @@ -1084,7 +1088,7 @@ static void gc_worker(struct work_struct *work) static void conntrack_gc_work_init(struct conntrack_gc_work *gc_work) { - INIT_DELAYED_WORK(&gc_work->dwork, gc_worker); + INIT_DEFERRABLE_WORK(&gc_work->dwork, gc_worker); gc_work->next_gc_run = HZ; gc_work->exiting = false; } @@ -1161,7 +1165,7 @@ void nf_conntrack_free(struct nf_conn *ct) /* A freed object has refcnt == 0, that's * the golden rule for SLAB_TYPESAFE_BY_RCU */ - NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 0); + WARN_ON(atomic_read(&ct->ct_general.use) != 0); nf_ct_ext_destroy(ct); nf_ct_ext_free(ct); @@ -1177,8 +1181,8 @@ EXPORT_SYMBOL_GPL(nf_conntrack_free); static noinline struct nf_conntrack_tuple_hash * init_conntrack(struct net *net, struct nf_conn *tmpl, const struct nf_conntrack_tuple *tuple, - struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_l4proto *l4proto, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, unsigned int dataoff, u32 hash) { @@ -1289,8 +1293,8 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl, unsigned int dataoff, u_int16_t l3num, u_int8_t protonum, - struct nf_conntrack_l3proto *l3proto, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_l4proto *l4proto) { const struct nf_conntrack_zone *zone; struct nf_conntrack_tuple tuple; @@ -1345,10 +1349,10 @@ unsigned int nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, struct sk_buff *skb) { + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; struct nf_conn *ct, *tmpl; enum ip_conntrack_info ctinfo; - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; unsigned int *timeouts; unsigned int dataoff; u_int8_t protonum; @@ -1415,7 +1419,7 @@ repeat: /* Decide what timeout policy we want to apply to this flow. */ timeouts = nf_ct_timeout_lookup(net, ct, l4proto); - ret = l4proto->packet(ct, skb, dataoff, ctinfo, pf, hooknum, timeouts); + ret = l4proto->packet(ct, skb, dataoff, ctinfo, pf, timeouts); if (ret <= 0) { /* Invalid: inverse of the return code tells * the netfilter core what to do */ @@ -1469,7 +1473,7 @@ void nf_conntrack_alter_reply(struct nf_conn *ct, struct nf_conn_help *help = nfct_help(ct); /* Should be unconfirmed, so not in hash table yet */ - NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + WARN_ON(nf_ct_is_confirmed(ct)); pr_debug("Altering reply tuple of %p to ", ct); nf_ct_dump_tuple(newreply); @@ -1491,7 +1495,7 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, unsigned long extra_jiffies, int do_acct) { - NF_CT_ASSERT(skb); + WARN_ON(!skb); /* Only update if this is not a fixed timeout */ if (test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status)) @@ -1689,6 +1693,18 @@ __nf_ct_unconfirmed_destroy(struct net *net) } } +void nf_ct_unconfirmed_destroy(struct net *net) +{ + might_sleep(); + + if (atomic_read(&net->ct.count) > 0) { + __nf_ct_unconfirmed_destroy(net); + nf_queue_nf_hook_drop(net); + synchronize_net(); + } +} +EXPORT_SYMBOL_GPL(nf_ct_unconfirmed_destroy); + void nf_ct_iterate_cleanup_net(struct net *net, int (*iter)(struct nf_conn *i, void *data), void *data, u32 portid, int report) @@ -1700,14 +1716,10 @@ void nf_ct_iterate_cleanup_net(struct net *net, if (atomic_read(&net->ct.count) == 0) return; - __nf_ct_unconfirmed_destroy(net); - d.iter = iter; d.data = data; d.net = net; - synchronize_net(); - nf_ct_iterate_cleanup(iter_net_only, &d, portid, report); } EXPORT_SYMBOL_GPL(nf_ct_iterate_cleanup_net); @@ -1733,6 +1745,7 @@ nf_ct_iterate_destroy(int (*iter)(struct nf_conn *i, void *data), void *data) if (atomic_read(&net->ct.count) == 0) continue; __nf_ct_unconfirmed_destroy(net); + nf_queue_nf_hook_drop(net); } rtnl_unlock(); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index e03d16ed550d..64778f9a8548 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -51,8 +51,8 @@ void nf_ct_unlink_expect_report(struct nf_conntrack_expect *exp, struct nf_conn_help *master_help = nfct_help(exp->master); struct net *net = nf_ct_exp_net(exp); - NF_CT_ASSERT(master_help); - NF_CT_ASSERT(!timer_pending(&exp->timeout)); + WARN_ON(!master_help); + WARN_ON(timer_pending(&exp->timeout)); hlist_del_rcu(&exp->hnode); net->ct.expect_count--; @@ -368,12 +368,6 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) /* two references : one for hash insert, one for the timer */ refcount_add(2, &exp->use); - hlist_add_head_rcu(&exp->lnode, &master_help->expectations); - master_help->expecting[exp->class]++; - - hlist_add_head_rcu(&exp->hnode, &nf_ct_expect_hash[h]); - net->ct.expect_count++; - setup_timer(&exp->timeout, nf_ct_expectation_timed_out, (unsigned long)exp); helper = rcu_dereference_protected(master_help->helper, @@ -384,6 +378,12 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) } add_timer(&exp->timeout); + hlist_add_head_rcu(&exp->lnode, &master_help->expectations); + master_help->expecting[exp->class]++; + + hlist_add_head_rcu(&exp->hnode, &nf_ct_expect_hash[h]); + net->ct.expect_count++; + NF_CT_STAT_INC(net, expect_create); } @@ -422,7 +422,7 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect) h = nf_ct_expect_dst_hash(net, &expect->tuple); hlist_for_each_entry_safe(i, next, &nf_ct_expect_hash[h], hnode) { if (expect_matches(i, expect)) { - if (nf_ct_remove_expect(expect)) + if (nf_ct_remove_expect(i)) break; } else if (expect_clash(i, expect)) { ret = -EBUSY; @@ -474,6 +474,60 @@ out: } EXPORT_SYMBOL_GPL(nf_ct_expect_related_report); +void nf_ct_expect_iterate_destroy(bool (*iter)(struct nf_conntrack_expect *e, void *data), + void *data) +{ + struct nf_conntrack_expect *exp; + const struct hlist_node *next; + unsigned int i; + + spin_lock_bh(&nf_conntrack_expect_lock); + + for (i = 0; i < nf_ct_expect_hsize; i++) { + hlist_for_each_entry_safe(exp, next, + &nf_ct_expect_hash[i], + hnode) { + if (iter(exp, data) && del_timer(&exp->timeout)) { + nf_ct_unlink_expect(exp); + nf_ct_expect_put(exp); + } + } + } + + spin_unlock_bh(&nf_conntrack_expect_lock); +} +EXPORT_SYMBOL_GPL(nf_ct_expect_iterate_destroy); + +void nf_ct_expect_iterate_net(struct net *net, + bool (*iter)(struct nf_conntrack_expect *e, void *data), + void *data, + u32 portid, int report) +{ + struct nf_conntrack_expect *exp; + const struct hlist_node *next; + unsigned int i; + + spin_lock_bh(&nf_conntrack_expect_lock); + + for (i = 0; i < nf_ct_expect_hsize; i++) { + hlist_for_each_entry_safe(exp, next, + &nf_ct_expect_hash[i], + hnode) { + + if (!net_eq(nf_ct_exp_net(exp), net)) + continue; + + if (iter(exp, data) && del_timer(&exp->timeout)) { + nf_ct_unlink_expect_report(exp, portid, report); + nf_ct_expect_put(exp); + } + } + } + + spin_unlock_bh(&nf_conntrack_expect_lock); +} +EXPORT_SYMBOL_GPL(nf_ct_expect_iterate_net); + #ifdef CONFIG_NF_CONNTRACK_PROCFS struct ct_expect_iter_state { struct seq_net_private p; diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 6c605e88ebae..9fe0ddc333fb 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -47,7 +47,7 @@ void *nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp) struct nf_ct_ext_type *t; /* Conntrack must not be confirmed to avoid races on reallocation. */ - NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + WARN_ON(nf_ct_is_confirmed(ct)); old = ct->ext; diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 9129bb3b5153..551a1eddf0fa 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -437,12 +437,22 @@ out: } EXPORT_SYMBOL_GPL(nf_conntrack_helper_register); -void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +static bool expect_iter_me(struct nf_conntrack_expect *exp, void *data) { - struct nf_conntrack_expect *exp; - const struct hlist_node *next; - unsigned int i; + struct nf_conn_help *help = nfct_help(exp->master); + const struct nf_conntrack_helper *me = data; + const struct nf_conntrack_helper *this; + + if (exp->helper == me) + return true; + this = rcu_dereference_protected(help->helper, + lockdep_is_held(&nf_conntrack_expect_lock)); + return this == me; +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ mutex_lock(&nf_ct_helper_mutex); hlist_del_rcu(&me->hnode); nf_ct_helper_count--; @@ -453,21 +463,7 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) */ synchronize_rcu(); - /* Get rid of expectations */ - spin_lock_bh(&nf_conntrack_expect_lock); - for (i = 0; i < nf_ct_expect_hsize; i++) { - hlist_for_each_entry_safe(exp, next, - &nf_ct_expect_hash[i], hnode) { - struct nf_conn_help *help = nfct_help(exp->master); - if ((rcu_dereference_protected( - help->helper, - lockdep_is_held(&nf_conntrack_expect_lock) - ) == me || exp->helper == me)) - nf_ct_remove_expect(exp); - } - } - spin_unlock_bh(&nf_conntrack_expect_lock); - + nf_ct_expect_iterate_destroy(expect_iter_me, NULL); nf_ct_iterate_destroy(unhelp, me); } EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister); diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c index cf9ace70bece..397e6911214f 100644 --- a/net/netfilter/nf_conntrack_l3proto_generic.c +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -49,11 +49,6 @@ static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -static void generic_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ -} - static int generic_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, unsigned int *dataoff, u_int8_t *protonum) { @@ -64,10 +59,8 @@ static int generic_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_l3proto nf_conntrack_l3proto_generic __read_mostly = { .l3proto = PF_UNSPEC, - .name = "unknown", .pkt_to_tuple = generic_pkt_to_tuple, .invert_tuple = generic_invert_tuple, - .print_tuple = generic_print_tuple, .get_l4proto = generic_get_l4proto, }; EXPORT_SYMBOL_GPL(nf_conntrack_l3proto_generic); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 7999e70c3bfb..de4053d84364 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -61,8 +61,8 @@ MODULE_LICENSE("GPL"); static char __initdata version[] = "0.93"; static int ctnetlink_dump_tuples_proto(struct sk_buff *skb, - const struct nf_conntrack_tuple *tuple, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_l4proto *l4proto) { int ret = 0; struct nlattr *nest_parms; @@ -86,7 +86,7 @@ nla_put_failure: static int ctnetlink_dump_tuples_ip(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple, - struct nf_conntrack_l3proto *l3proto) + const struct nf_conntrack_l3proto *l3proto) { int ret = 0; struct nlattr *nest_parms; @@ -109,9 +109,9 @@ nla_put_failure: static int ctnetlink_dump_tuples(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple) { + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; int ret; - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; rcu_read_lock(); l3proto = __nf_ct_l3proto_find(tuple->src.l3num); @@ -163,7 +163,7 @@ nla_put_failure: static int ctnetlink_dump_protoinfo(struct sk_buff *skb, struct nf_conn *ct) { - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; struct nlattr *nest_proto; int ret; @@ -535,17 +535,16 @@ nla_put_failure: static inline size_t ctnetlink_proto_size(const struct nf_conn *ct) { - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; - size_t len = 0; + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; + size_t len; - rcu_read_lock(); l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct)); - len += l3proto->nla_size; + len = l3proto->nla_size; + len *= 3u; /* ORIG, REPLY, MASTER */ l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); len += l4proto->nla_size; - rcu_read_unlock(); return len; } @@ -664,7 +663,6 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; - rcu_read_lock(); zone = nf_ct_zone(ct); nest_parms = nla_nest_start(skb, CTA_TUPLE_ORIG | NLA_F_NESTED); @@ -736,8 +734,6 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) && ctnetlink_dump_mark(skb, ct) < 0) goto nla_put_failure; #endif - rcu_read_unlock(); - nlmsg_end(skb, nlh); err = nfnetlink_send(skb, net, item->portid, group, item->report, GFP_ATOMIC); @@ -747,7 +743,6 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) return 0; nla_put_failure: - rcu_read_unlock(); nlmsg_cancel(skb, nlh); nlmsg_failure: kfree_skb(skb); @@ -941,8 +936,8 @@ static const struct nla_policy proto_nla_policy[CTA_PROTO_MAX+1] = { static int ctnetlink_parse_tuple_proto(struct nlattr *attr, struct nf_conntrack_tuple *tuple) { + const struct nf_conntrack_l4proto *l4proto; struct nlattr *tb[CTA_PROTO_MAX+1]; - struct nf_conntrack_l4proto *l4proto; int ret = 0; ret = nla_parse_nested(tb, CTA_PROTO_MAX, attr, proto_nla_policy, @@ -1585,8 +1580,8 @@ static int ctnetlink_change_protoinfo(struct nf_conn *ct, const struct nlattr * const cda[]) { const struct nlattr *attr = cda[CTA_PROTOINFO]; + const struct nf_conntrack_l4proto *l4proto; struct nlattr *tb[CTA_PROTOINFO_MAX+1]; - struct nf_conntrack_l4proto *l4proto; int err = 0; err = nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, protoinfo_policy, @@ -2213,7 +2208,6 @@ static int __ctnetlink_glue_build(struct sk_buff *skb, struct nf_conn *ct) const struct nf_conntrack_zone *zone; struct nlattr *nest_parms; - rcu_read_lock(); zone = nf_ct_zone(ct); nest_parms = nla_nest_start(skb, CTA_TUPLE_ORIG | NLA_F_NESTED); @@ -2272,11 +2266,9 @@ static int __ctnetlink_glue_build(struct sk_buff *skb, struct nf_conn *ct) #endif if (ctnetlink_dump_labels(skb, ct) < 0) goto nla_put_failure; - rcu_read_unlock(); return 0; nla_put_failure: - rcu_read_unlock(); return -ENOSPC; } @@ -2483,11 +2475,11 @@ static int ctnetlink_exp_dump_mask(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple, const struct nf_conntrack_tuple_mask *mask) { - int ret; - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; struct nf_conntrack_tuple m; struct nlattr *nest_parms; + int ret; memset(&m, 0xFF, sizeof(m)); memcpy(&m.src.u3, &mask->src.u3, sizeof(m.src.u3)); @@ -2661,17 +2653,14 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item) nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; - rcu_read_lock(); if (ctnetlink_exp_dump_expect(skb, exp) < 0) goto nla_put_failure; - rcu_read_unlock(); nlmsg_end(skb, nlh); nfnetlink_send(skb, net, item->portid, group, item->report, GFP_ATOMIC); return 0; nla_put_failure: - rcu_read_unlock(); nlmsg_cancel(skb, nlh); nlmsg_failure: kfree_skb(skb); @@ -2910,6 +2899,21 @@ out: return err == -EAGAIN ? -ENOBUFS : err; } +static bool expect_iter_name(struct nf_conntrack_expect *exp, void *data) +{ + const struct nf_conn_help *m_help; + const char *name = data; + + m_help = nfct_help(exp->master); + + return strcmp(m_help->helper->name, name) == 0; +} + +static bool expect_iter_all(struct nf_conntrack_expect *exp, void *data) +{ + return true; +} + static int ctnetlink_del_expect(struct net *net, struct sock *ctnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[], @@ -2918,10 +2922,8 @@ static int ctnetlink_del_expect(struct net *net, struct sock *ctnl, struct nf_conntrack_expect *exp; struct nf_conntrack_tuple tuple; struct nfgenmsg *nfmsg = nlmsg_data(nlh); - struct hlist_node *next; u_int8_t u3 = nfmsg->nfgen_family; struct nf_conntrack_zone zone; - unsigned int i; int err; if (cda[CTA_EXPECT_TUPLE]) { @@ -2961,49 +2963,15 @@ static int ctnetlink_del_expect(struct net *net, struct sock *ctnl, nf_ct_expect_put(exp); } else if (cda[CTA_EXPECT_HELP_NAME]) { char *name = nla_data(cda[CTA_EXPECT_HELP_NAME]); - struct nf_conn_help *m_help; - - /* delete all expectations for this helper */ - spin_lock_bh(&nf_conntrack_expect_lock); - for (i = 0; i < nf_ct_expect_hsize; i++) { - hlist_for_each_entry_safe(exp, next, - &nf_ct_expect_hash[i], - hnode) { - - if (!net_eq(nf_ct_exp_net(exp), net)) - continue; - m_help = nfct_help(exp->master); - if (!strcmp(m_help->helper->name, name) && - del_timer(&exp->timeout)) { - nf_ct_unlink_expect_report(exp, - NETLINK_CB(skb).portid, - nlmsg_report(nlh)); - nf_ct_expect_put(exp); - } - } - } - spin_unlock_bh(&nf_conntrack_expect_lock); + nf_ct_expect_iterate_net(net, expect_iter_name, name, + NETLINK_CB(skb).portid, + nlmsg_report(nlh)); } else { /* This basically means we have to flush everything*/ - spin_lock_bh(&nf_conntrack_expect_lock); - for (i = 0; i < nf_ct_expect_hsize; i++) { - hlist_for_each_entry_safe(exp, next, - &nf_ct_expect_hash[i], - hnode) { - - if (!net_eq(nf_ct_exp_net(exp), net)) - continue; - - if (del_timer(&exp->timeout)) { - nf_ct_unlink_expect_report(exp, - NETLINK_CB(skb).portid, - nlmsg_report(nlh)); - nf_ct_expect_put(exp); - } - } - } - spin_unlock_bh(&nf_conntrack_expect_lock); + nf_ct_expect_iterate_net(net, expect_iter_all, NULL, + NETLINK_CB(skb).portid, + nlmsg_report(nlh)); } return 0; diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 6959e93063d4..11562f2a08bb 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -113,7 +113,6 @@ static void pptp_expectfn(struct nf_conn *ct, /* Can you see how rusty this code is, compared with the pre-2.6.11 * one? That's what happened to my shiny newnat of 2002 ;( -HW */ - rcu_read_lock(); nf_nat_pptp_expectfn = rcu_dereference(nf_nat_pptp_hook_expectfn); if (nf_nat_pptp_expectfn && ct->master->status & IPS_NAT_MASK) nf_nat_pptp_expectfn(ct, exp); @@ -136,7 +135,6 @@ static void pptp_expectfn(struct nf_conn *ct, pr_debug("not found\n"); } } - rcu_read_unlock(); } static int destroy_sibling_or_exp(struct net *net, struct nf_conn *ct, diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 1dcad229c3cc..b3e489c859ec 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -65,7 +65,7 @@ nf_ct_unregister_sysctl(struct ctl_table_header **header, } #endif -struct nf_conntrack_l4proto * +const struct nf_conntrack_l4proto * __nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto) { if (unlikely(l3proto >= NFPROTO_NUMPROTO || nf_ct_protos[l3proto] == NULL)) @@ -77,7 +77,7 @@ EXPORT_SYMBOL_GPL(__nf_ct_l4proto_find); /* this is guaranteed to always return a valid protocol helper, since * it falls back to generic_protocol */ -struct nf_conntrack_l3proto * +const struct nf_conntrack_l3proto * nf_ct_l3proto_find_get(u_int16_t l3proto) { struct nf_conntrack_l3proto *p; @@ -95,8 +95,8 @@ EXPORT_SYMBOL_GPL(nf_ct_l3proto_find_get); int nf_ct_l3proto_try_module_get(unsigned short l3proto) { + const struct nf_conntrack_l3proto *p; int ret; - struct nf_conntrack_l3proto *p; retry: p = nf_ct_l3proto_find_get(l3proto); if (p == &nf_conntrack_l3proto_generic) { @@ -173,10 +173,10 @@ void nf_ct_netns_put(struct net *net, u8 nfproto) } EXPORT_SYMBOL_GPL(nf_ct_netns_put); -struct nf_conntrack_l4proto * +const struct nf_conntrack_l4proto * nf_ct_l4proto_find_get(u_int16_t l3num, u_int8_t l4num) { - struct nf_conntrack_l4proto *p; + const struct nf_conntrack_l4proto *p; rcu_read_lock(); p = __nf_ct_l4proto_find(l3num, l4num); @@ -188,7 +188,7 @@ nf_ct_l4proto_find_get(u_int16_t l3num, u_int8_t l4num) } EXPORT_SYMBOL_GPL(nf_ct_l4proto_find_get); -void nf_ct_l4proto_put(struct nf_conntrack_l4proto *p) +void nf_ct_l4proto_put(const struct nf_conntrack_l4proto *p) { module_put(p->me); } @@ -196,28 +196,28 @@ EXPORT_SYMBOL_GPL(nf_ct_l4proto_put); static int kill_l3proto(struct nf_conn *i, void *data) { - return nf_ct_l3num(i) == ((struct nf_conntrack_l3proto *)data)->l3proto; + return nf_ct_l3num(i) == ((const struct nf_conntrack_l3proto *)data)->l3proto; } static int kill_l4proto(struct nf_conn *i, void *data) { - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; l4proto = data; return nf_ct_protonum(i) == l4proto->l4proto && nf_ct_l3num(i) == l4proto->l3proto; } -int nf_ct_l3proto_register(struct nf_conntrack_l3proto *proto) +int nf_ct_l3proto_register(const struct nf_conntrack_l3proto *proto) { int ret = 0; struct nf_conntrack_l3proto *old; if (proto->l3proto >= NFPROTO_NUMPROTO) return -EBUSY; - - if (proto->tuple_to_nlattr && !proto->nlattr_tuple_size) +#if IS_ENABLED(CONFIG_NF_CT_NETLINK) + if (proto->tuple_to_nlattr && proto->nla_size == 0) return -EINVAL; - +#endif mutex_lock(&nf_ct_proto_mutex); old = rcu_dereference_protected(nf_ct_l3protos[proto->l3proto], lockdep_is_held(&nf_ct_proto_mutex)); @@ -226,9 +226,6 @@ int nf_ct_l3proto_register(struct nf_conntrack_l3proto *proto) goto out_unlock; } - if (proto->nlattr_tuple_size) - proto->nla_size = 3 * proto->nlattr_tuple_size(); - rcu_assign_pointer(nf_ct_l3protos[proto->l3proto], proto); out_unlock: @@ -238,21 +235,7 @@ out_unlock: } EXPORT_SYMBOL_GPL(nf_ct_l3proto_register); -#ifdef CONFIG_SYSCTL -extern unsigned int nf_conntrack_default_on; - -int nf_ct_l3proto_pernet_register(struct net *net, - struct nf_conntrack_l3proto *proto) -{ - if (nf_conntrack_default_on == 0) - return 0; - - return proto->net_ns_get ? proto->net_ns_get(net) : 0; -} -EXPORT_SYMBOL_GPL(nf_ct_l3proto_pernet_register); -#endif - -void nf_ct_l3proto_unregister(struct nf_conntrack_l3proto *proto) +void nf_ct_l3proto_unregister(const struct nf_conntrack_l3proto *proto) { BUG_ON(proto->l3proto >= NFPROTO_NUMPROTO); @@ -266,27 +249,12 @@ void nf_ct_l3proto_unregister(struct nf_conntrack_l3proto *proto) synchronize_rcu(); /* Remove all contrack entries for this protocol */ - nf_ct_iterate_destroy(kill_l3proto, proto); + nf_ct_iterate_destroy(kill_l3proto, (void*)proto); } EXPORT_SYMBOL_GPL(nf_ct_l3proto_unregister); -void nf_ct_l3proto_pernet_unregister(struct net *net, - struct nf_conntrack_l3proto *proto) -{ - /* - * nf_conntrack_default_on *might* have registered hooks. - * ->net_ns_put must cope with more puts() than get(), i.e. - * if nf_conntrack_default_on was 0 at time of - * nf_ct_l3proto_pernet_register invocation this net_ns_put() - * should be a noop. - */ - if (proto->net_ns_put) - proto->net_ns_put(net); -} -EXPORT_SYMBOL_GPL(nf_ct_l3proto_pernet_unregister); - static struct nf_proto_net *nf_ct_l4proto_net(struct net *net, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto) { if (l4proto->get_net_proto) { /* statically built-in protocols use static per-net */ @@ -301,7 +269,7 @@ static struct nf_proto_net *nf_ct_l4proto_net(struct net *net, static int nf_ct_l4proto_register_sysctl(struct net *net, struct nf_proto_net *pn, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto) { int err = 0; @@ -324,8 +292,8 @@ int nf_ct_l4proto_register_sysctl(struct net *net, static void nf_ct_l4proto_unregister_sysctl(struct net *net, - struct nf_proto_net *pn, - struct nf_conntrack_l4proto *l4proto) + struct nf_proto_net *pn, + const struct nf_conntrack_l4proto *l4proto) { #ifdef CONFIG_SYSCTL if (pn->ctl_table_header != NULL) @@ -395,7 +363,7 @@ out_unlock: EXPORT_SYMBOL_GPL(nf_ct_l4proto_register_one); int nf_ct_l4proto_pernet_register_one(struct net *net, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto) { int ret = 0; struct nf_proto_net *pn = NULL; @@ -420,7 +388,7 @@ out: } EXPORT_SYMBOL_GPL(nf_ct_l4proto_pernet_register_one); -static void __nf_ct_l4proto_unregister_one(struct nf_conntrack_l4proto *l4proto) +static void __nf_ct_l4proto_unregister_one(const struct nf_conntrack_l4proto *l4proto) { BUG_ON(l4proto->l3proto >= ARRAY_SIZE(nf_ct_protos)); @@ -433,7 +401,7 @@ static void __nf_ct_l4proto_unregister_one(struct nf_conntrack_l4proto *l4proto) &nf_conntrack_l4proto_generic); } -void nf_ct_l4proto_unregister_one(struct nf_conntrack_l4proto *l4proto) +void nf_ct_l4proto_unregister_one(const struct nf_conntrack_l4proto *l4proto) { mutex_lock(&nf_ct_proto_mutex); __nf_ct_l4proto_unregister_one(l4proto); @@ -444,7 +412,7 @@ void nf_ct_l4proto_unregister_one(struct nf_conntrack_l4proto *l4proto) EXPORT_SYMBOL_GPL(nf_ct_l4proto_unregister_one); void nf_ct_l4proto_pernet_unregister_one(struct net *net, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto) { struct nf_proto_net *pn = nf_ct_l4proto_net(net, l4proto); @@ -469,8 +437,8 @@ int nf_ct_l4proto_register(struct nf_conntrack_l4proto *l4proto[], } if (i != num_proto) { ver = l4proto[i]->l3proto == PF_INET6 ? 6 : 4; - pr_err("nf_conntrack_ipv%d: can't register %s%d proto.\n", - ver, l4proto[i]->name, ver); + pr_err("nf_conntrack_ipv%d: can't register l4 %d proto.\n", + ver, l4proto[i]->l4proto); nf_ct_l4proto_unregister(l4proto, i); } return ret; @@ -478,7 +446,7 @@ int nf_ct_l4proto_register(struct nf_conntrack_l4proto *l4proto[], EXPORT_SYMBOL_GPL(nf_ct_l4proto_register); int nf_ct_l4proto_pernet_register(struct net *net, - struct nf_conntrack_l4proto *l4proto[], + struct nf_conntrack_l4proto *const l4proto[], unsigned int num_proto) { int ret = -EINVAL; @@ -490,8 +458,8 @@ int nf_ct_l4proto_pernet_register(struct net *net, break; } if (i != num_proto) { - pr_err("nf_conntrack_%s%d: pernet registration failed\n", - l4proto[i]->name, + pr_err("nf_conntrack_proto_%d %d: pernet registration failed\n", + l4proto[i]->l4proto, l4proto[i]->l3proto == PF_INET6 ? 6 : 4); nf_ct_l4proto_pernet_unregister(net, l4proto, i); } @@ -514,8 +482,8 @@ void nf_ct_l4proto_unregister(struct nf_conntrack_l4proto *l4proto[], EXPORT_SYMBOL_GPL(nf_ct_l4proto_unregister); void nf_ct_l4proto_pernet_unregister(struct net *net, - struct nf_conntrack_l4proto *l4proto[], - unsigned int num_proto) + struct nf_conntrack_l4proto *const l4proto[], + unsigned int num_proto) { while (num_proto-- != 0) nf_ct_l4proto_pernet_unregister_one(net, l4proto[num_proto]); diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index 4707d997558a..0f5a4d79f6b8 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -469,7 +469,7 @@ static unsigned int *dccp_get_timeouts(struct net *net) static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int hooknum, + u_int8_t pf, unsigned int *timeouts) { struct net *net = nf_ct_net(ct); @@ -623,18 +623,12 @@ static bool dccp_can_early_drop(const struct nf_conn *ct) return false; } -static void dccp_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "sport=%hu dport=%hu ", - ntohs(tuple->src.u.dccp.port), - ntohs(tuple->dst.u.dccp.port)); -} - +#ifdef CONFIG_NF_CONNTRACK_PROCFS static void dccp_print_conntrack(struct seq_file *s, struct nf_conn *ct) { seq_printf(s, "%s ", dccp_state_names[ct->proto.dccp.state]); } +#endif #if IS_ENABLED(CONFIG_NF_CT_NETLINK) static int dccp_to_nlattr(struct sk_buff *skb, struct nlattr *nla, @@ -880,7 +874,6 @@ static struct nf_proto_net *dccp_get_net_proto(struct net *net) struct nf_conntrack_l4proto nf_conntrack_l4proto_dccp4 __read_mostly = { .l3proto = AF_INET, .l4proto = IPPROTO_DCCP, - .name = "dccp", .pkt_to_tuple = dccp_pkt_to_tuple, .invert_tuple = dccp_invert_tuple, .new = dccp_new, @@ -888,8 +881,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_dccp4 __read_mostly = { .get_timeouts = dccp_get_timeouts, .error = dccp_error, .can_early_drop = dccp_can_early_drop, - .print_tuple = dccp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = dccp_print_conntrack, +#endif #if IS_ENABLED(CONFIG_NF_CT_NETLINK) .to_nlattr = dccp_to_nlattr, .nlattr_size = dccp_nlattr_size, @@ -916,7 +910,6 @@ EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_dccp4); struct nf_conntrack_l4proto nf_conntrack_l4proto_dccp6 __read_mostly = { .l3proto = AF_INET6, .l4proto = IPPROTO_DCCP, - .name = "dccp", .pkt_to_tuple = dccp_pkt_to_tuple, .invert_tuple = dccp_invert_tuple, .new = dccp_new, @@ -924,8 +917,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_dccp6 __read_mostly = { .get_timeouts = dccp_get_timeouts, .error = dccp_error, .can_early_drop = dccp_can_early_drop, - .print_tuple = dccp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = dccp_print_conntrack, +#endif #if IS_ENABLED(CONFIG_NF_CT_NETLINK) .to_nlattr = dccp_to_nlattr, .nlattr_size = dccp_nlattr_size, diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index d5868bad33a7..9cd40700842e 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -17,22 +17,10 @@ static unsigned int nf_ct_generic_timeout __read_mostly = 600*HZ; static bool nf_generic_should_process(u8 proto) { switch (proto) { -#ifdef CONFIG_NF_CT_PROTO_SCTP_MODULE - case IPPROTO_SCTP: - return false; -#endif -#ifdef CONFIG_NF_CT_PROTO_DCCP_MODULE - case IPPROTO_DCCP: - return false; -#endif #ifdef CONFIG_NF_CT_PROTO_GRE_MODULE case IPPROTO_GRE: return false; #endif -#ifdef CONFIG_NF_CT_PROTO_UDPLITE_MODULE - case IPPROTO_UDPLITE: - return false; -#endif default: return true; } @@ -62,12 +50,6 @@ static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void generic_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ -} - static unsigned int *generic_get_timeouts(struct net *net) { return &(generic_pernet(net)->timeout); @@ -79,7 +61,6 @@ static int generic_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeout) { nf_ct_refresh_acct(ct, ctinfo, skb, *timeout); @@ -187,10 +168,8 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_generic __read_mostly = { .l3proto = PF_UNSPEC, .l4proto = 255, - .name = "unknown", .pkt_to_tuple = generic_pkt_to_tuple, .invert_tuple = generic_invert_tuple, - .print_tuple = generic_print_tuple, .packet = generic_packet, .get_timeouts = generic_get_timeouts, .new = generic_new, diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 87bb40a3feb5..09a90484c27d 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -224,15 +224,7 @@ static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, return true; } -/* print gre part of tuple */ -static void gre_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "srckey=0x%x dstkey=0x%x ", - ntohs(tuple->src.u.gre.key), - ntohs(tuple->dst.u.gre.key)); -} - +#ifdef CONFIG_NF_CONNTRACK_PROCFS /* print private data for conntrack */ static void gre_print_conntrack(struct seq_file *s, struct nf_conn *ct) { @@ -240,6 +232,7 @@ static void gre_print_conntrack(struct seq_file *s, struct nf_conn *ct) (ct->proto.gre.timeout / HZ), (ct->proto.gre.stream_timeout / HZ)); } +#endif static unsigned int *gre_get_timeouts(struct net *net) { @@ -252,7 +245,6 @@ static int gre_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeouts) { /* If we've seen traffic both ways, this is a GRE connection. @@ -364,11 +356,11 @@ static int gre_init_net(struct net *net, u_int16_t proto) static struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 __read_mostly = { .l3proto = AF_INET, .l4proto = IPPROTO_GRE, - .name = "gre", .pkt_to_tuple = gre_pkt_to_tuple, .invert_tuple = gre_invert_tuple, - .print_tuple = gre_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = gre_print_conntrack, +#endif .get_timeouts = gre_get_timeouts, .packet = gre_packet, .new = gre_new, diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 6eef29d2eec4..6303a88af12b 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -174,20 +174,13 @@ static bool sctp_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void sctp_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "sport=%hu dport=%hu ", - ntohs(tuple->src.u.sctp.port), - ntohs(tuple->dst.u.sctp.port)); -} - +#ifdef CONFIG_NF_CONNTRACK_PROCFS /* Print out the private part of the conntrack. */ static void sctp_print_conntrack(struct seq_file *s, struct nf_conn *ct) { seq_printf(s, "%s ", sctp_conntrack_names[ct->proto.sctp.state]); } +#endif #define for_each_sctp_chunk(skb, sch, _sch, offset, dataoff, count) \ for ((offset) = (dataoff) + sizeof(struct sctphdr), (count) = 0; \ @@ -314,7 +307,6 @@ static int sctp_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeouts) { enum sctp_conntrack new_state, old_state; @@ -791,11 +783,11 @@ static struct nf_proto_net *sctp_get_net_proto(struct net *net) struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 __read_mostly = { .l3proto = PF_INET, .l4proto = IPPROTO_SCTP, - .name = "sctp", .pkt_to_tuple = sctp_pkt_to_tuple, .invert_tuple = sctp_invert_tuple, - .print_tuple = sctp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = sctp_print_conntrack, +#endif .packet = sctp_packet, .get_timeouts = sctp_get_timeouts, .new = sctp_new, @@ -828,11 +820,11 @@ EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_sctp4); struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp6 __read_mostly = { .l3proto = PF_INET6, .l4proto = IPPROTO_SCTP, - .name = "sctp", .pkt_to_tuple = sctp_pkt_to_tuple, .invert_tuple = sctp_invert_tuple, - .print_tuple = sctp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = sctp_print_conntrack, +#endif .packet = sctp_packet, .get_timeouts = sctp_get_timeouts, .new = sctp_new, diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 9758a7dfd83e..cba1c6ffe51a 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -301,20 +301,13 @@ static bool tcp_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void tcp_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "sport=%hu dport=%hu ", - ntohs(tuple->src.u.tcp.port), - ntohs(tuple->dst.u.tcp.port)); -} - +#ifdef CONFIG_NF_CONNTRACK_PROCFS /* Print out the private part of the conntrack. */ static void tcp_print_conntrack(struct seq_file *s, struct nf_conn *ct) { seq_printf(s, "%s ", tcp_conntrack_names[ct->proto.tcp.state]); } +#endif static unsigned int get_conntrack_index(const struct tcphdr *tcph) { @@ -810,7 +803,6 @@ static int tcp_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeouts) { struct net *net = nf_ct_net(ct); @@ -1556,11 +1548,11 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4 __read_mostly = { .l3proto = PF_INET, .l4proto = IPPROTO_TCP, - .name = "tcp", .pkt_to_tuple = tcp_pkt_to_tuple, .invert_tuple = tcp_invert_tuple, - .print_tuple = tcp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = tcp_print_conntrack, +#endif .packet = tcp_packet, .get_timeouts = tcp_get_timeouts, .new = tcp_new, @@ -1594,11 +1586,11 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6 __read_mostly = { .l3proto = PF_INET6, .l4proto = IPPROTO_TCP, - .name = "tcp", .pkt_to_tuple = tcp_pkt_to_tuple, .invert_tuple = tcp_invert_tuple, - .print_tuple = tcp_print_tuple, +#ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = tcp_print_conntrack, +#endif .packet = tcp_packet, .get_timeouts = tcp_get_timeouts, .new = tcp_new, diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index f6ebce6178ca..8af734cd1a94 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -63,15 +63,6 @@ static bool udp_invert_tuple(struct nf_conntrack_tuple *tuple, return true; } -/* Print out the per-protocol part of the tuple. */ -static void udp_print_tuple(struct seq_file *s, - const struct nf_conntrack_tuple *tuple) -{ - seq_printf(s, "sport=%hu dport=%hu ", - ntohs(tuple->src.u.udp.port), - ntohs(tuple->dst.u.udp.port)); -} - static unsigned int *udp_get_timeouts(struct net *net) { return udp_pernet(net)->timeouts; @@ -83,7 +74,6 @@ static int udp_packet(struct nf_conn *ct, unsigned int dataoff, enum ip_conntrack_info ctinfo, u_int8_t pf, - unsigned int hooknum, unsigned int *timeouts) { /* If we've seen traffic both ways, this is some kind of UDP @@ -313,11 +303,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4 __read_mostly = { .l3proto = PF_INET, .l4proto = IPPROTO_UDP, - .name = "udp", .allow_clash = true, .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, - .print_tuple = udp_print_tuple, .packet = udp_packet, .get_timeouts = udp_get_timeouts, .new = udp_new, @@ -347,11 +335,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_udplite4 __read_mostly = { .l3proto = PF_INET, .l4proto = IPPROTO_UDPLITE, - .name = "udplite", .allow_clash = true, .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, - .print_tuple = udp_print_tuple, .packet = udp_packet, .get_timeouts = udp_get_timeouts, .new = udp_new, @@ -381,11 +367,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6 __read_mostly = { .l3proto = PF_INET6, .l4proto = IPPROTO_UDP, - .name = "udp", .allow_clash = true, .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, - .print_tuple = udp_print_tuple, .packet = udp_packet, .get_timeouts = udp_get_timeouts, .new = udp_new, @@ -415,11 +399,9 @@ struct nf_conntrack_l4proto nf_conntrack_l4proto_udplite6 __read_mostly = { .l3proto = PF_INET6, .l4proto = IPPROTO_UDPLITE, - .name = "udplite", .allow_clash = true, .pkt_to_tuple = udp_pkt_to_tuple, .invert_tuple = udp_invert_tuple, - .print_tuple = udp_print_tuple, .packet = udp_packet, .get_timeouts = udp_get_timeouts, .new = udp_new, diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index d38af4274335..4dbb5bad4363 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -884,7 +884,6 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int protoff, tuple.dst.u3 = *daddr; tuple.dst.u.udp.port = port; - rcu_read_lock(); do { exp = __nf_ct_expect_find(net, nf_ct_zone(ct), &tuple); @@ -918,10 +917,8 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int protoff, goto err1; } - if (skip_expect) { - rcu_read_unlock(); + if (skip_expect) return NF_ACCEPT; - } rtp_exp = nf_ct_expect_alloc(ct); if (rtp_exp == NULL) @@ -952,7 +949,6 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int protoff, err2: nf_ct_expect_put(rtp_exp); err1: - rcu_read_unlock(); return ret; } diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index ccb5cb9043e0..5a101caa3e12 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -41,8 +41,62 @@ print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, const struct nf_conntrack_l3proto *l3proto, const struct nf_conntrack_l4proto *l4proto) { - l3proto->print_tuple(s, tuple); - l4proto->print_tuple(s, tuple); + switch (l3proto->l3proto) { + case NFPROTO_IPV4: + seq_printf(s, "src=%pI4 dst=%pI4 ", + &tuple->src.u3.ip, &tuple->dst.u3.ip); + break; + case NFPROTO_IPV6: + seq_printf(s, "src=%pI6 dst=%pI6 ", + tuple->src.u3.ip6, tuple->dst.u3.ip6); + break; + default: + break; + } + + switch (l4proto->l4proto) { + case IPPROTO_ICMP: + seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); + break; + case IPPROTO_TCP: + seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.tcp.port), + ntohs(tuple->dst.u.tcp.port)); + break; + case IPPROTO_UDPLITE: /* fallthrough */ + case IPPROTO_UDP: + seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.udp.port), + ntohs(tuple->dst.u.udp.port)); + + break; + case IPPROTO_DCCP: + seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.dccp.port), + ntohs(tuple->dst.u.dccp.port)); + break; + case IPPROTO_SCTP: + seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.sctp.port), + ntohs(tuple->dst.u.sctp.port)); + break; + case IPPROTO_ICMPV6: + seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); + break; + case IPPROTO_GRE: + seq_printf(s, "srckey=0x%x dstkey=0x%x ", + ntohs(tuple->src.u.gre.key), + ntohs(tuple->dst.u.gre.key)); + break; + default: + break; + } } EXPORT_SYMBOL_GPL(print_tuple); @@ -198,6 +252,31 @@ ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) } #endif +static const char* l3proto_name(u16 proto) +{ + switch (proto) { + case AF_INET: return "ipv4"; + case AF_INET6: return "ipv6"; + } + + return "unknown"; +} + +static const char* l4proto_name(u16 proto) +{ + switch (proto) { + case IPPROTO_ICMP: return "icmp"; + case IPPROTO_TCP: return "tcp"; + case IPPROTO_UDP: return "udp"; + case IPPROTO_DCCP: return "dccp"; + case IPPROTO_GRE: return "gre"; + case IPPROTO_SCTP: return "sctp"; + case IPPROTO_UDPLITE: return "udplite"; + } + + return "unknown"; +} + /* return 0 on success, 1 in case of error */ static int ct_seq_show(struct seq_file *s, void *v) { @@ -208,7 +287,7 @@ static int ct_seq_show(struct seq_file *s, void *v) struct net *net = seq_file_net(s); int ret = 0; - NF_CT_ASSERT(ct); + WARN_ON(!ct); if (unlikely(!atomic_inc_not_zero(&ct->ct_general.use))) return 0; @@ -225,14 +304,14 @@ static int ct_seq_show(struct seq_file *s, void *v) goto release; l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct)); - NF_CT_ASSERT(l3proto); + WARN_ON(!l3proto); l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); - NF_CT_ASSERT(l4proto); + WARN_ON(!l4proto); ret = -ENOSPC; seq_printf(s, "%-8s %u %-8s %u %ld ", - l3proto->name, nf_ct_l3num(ct), - l4proto->name, nf_ct_protonum(ct), + l3proto_name(l3proto->l3proto), nf_ct_l3num(ct), + l4proto_name(l4proto->l4proto), nf_ct_protonum(ct), nf_ct_expires(ct) / HZ); if (l4proto->print_conntrack) @@ -452,9 +531,6 @@ static int log_invalid_proto_max __read_mostly = 255; /* size the user *wants to set */ static unsigned int nf_conntrack_htable_size_user __read_mostly; -extern unsigned int nf_conntrack_default_on; -unsigned int nf_conntrack_default_on __read_mostly = 1; - static int nf_conntrack_hash_sysctl(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) @@ -520,13 +596,6 @@ static struct ctl_table nf_ct_sysctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { - .procname = "nf_conntrack_default_on", - .data = &nf_conntrack_default_on, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { } }; diff --git a/net/netfilter/nf_internals.h b/net/netfilter/nf_internals.h index bfa742da83af..49f87ec093a3 100644 --- a/net/netfilter/nf_internals.h +++ b/net/netfilter/nf_internals.h @@ -5,17 +5,11 @@ #include <linux/skbuff.h> #include <linux/netdevice.h> -#ifdef CONFIG_NETFILTER_DEBUG -#define NFDEBUG(format, args...) printk(KERN_DEBUG format , ## args) -#else -#define NFDEBUG(format, args...) -#endif - /* nf_queue.c */ int nf_queue(struct sk_buff *skb, struct nf_hook_state *state, - struct nf_hook_entry **entryp, unsigned int verdict); + const struct nf_hook_entries *entries, unsigned int index, + unsigned int verdict); unsigned int nf_queue_nf_hook_drop(struct net *net); -int __init netfilter_queue_init(void); /* nf_log.c */ int __init netfilter_log_init(void); diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 832c5a08d9a5..40573aa6c133 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -222,20 +222,21 @@ find_appropriate_src(struct net *net, .tuple = tuple, .zone = zone }; - struct rhlist_head *hl; + struct rhlist_head *hl, *h; hl = rhltable_lookup(&nf_nat_bysource_table, &key, nf_nat_bysource_params); - if (!hl) - return 0; - ct = container_of(hl, typeof(*ct), nat_bysource); + rhl_for_each_entry_rcu(ct, h, hl, nat_bysource) { + nf_ct_invert_tuplepr(result, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + result->dst = tuple->dst; - nf_ct_invert_tuplepr(result, - &ct->tuplehash[IP_CT_DIR_REPLY].tuple); - result->dst = tuple->dst; + if (in_range(l3proto, l4proto, result, range)) + return 1; + } - return in_range(l3proto, l4proto, result, range); + return 0; } /* For [FUTURE] fragmentation handling, we want the least-used @@ -413,8 +414,8 @@ nf_nat_setup_info(struct nf_conn *ct, if (nf_ct_is_confirmed(ct)) return NF_ACCEPT; - NF_CT_ASSERT(maniptype == NF_NAT_MANIP_SRC || - maniptype == NF_NAT_MANIP_DST); + WARN_ON(maniptype != NF_NAT_MANIP_SRC && + maniptype != NF_NAT_MANIP_DST); BUG_ON(nf_nat_initialized(ct, maniptype)); /* What we've got will look like inverse of reply. Normally @@ -440,7 +441,7 @@ nf_nat_setup_info(struct nf_conn *ct, else ct->status |= IPS_DST_NAT; - if (nfct_help(ct)) + if (nfct_help(ct) && !nfct_seqadj(ct)) if (!nfct_seqadj_ext_add(ct)) return NF_DROP; } diff --git a/net/netfilter/nf_nat_redirect.c b/net/netfilter/nf_nat_redirect.c index 86067560a318..25b06b959118 100644 --- a/net/netfilter/nf_nat_redirect.c +++ b/net/netfilter/nf_nat_redirect.c @@ -38,11 +38,11 @@ nf_nat_redirect_ipv4(struct sk_buff *skb, __be32 newdst; struct nf_nat_range newrange; - NF_CT_ASSERT(hooknum == NF_INET_PRE_ROUTING || - hooknum == NF_INET_LOCAL_OUT); + WARN_ON(hooknum != NF_INET_PRE_ROUTING && + hooknum != NF_INET_LOCAL_OUT); ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)); + WARN_ON(!(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED))); /* Local packets: make them go to loopback */ if (hooknum == NF_INET_LOCAL_OUT) { diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 043850c9d154..f7e21953b1de 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -109,9 +109,11 @@ unsigned int nf_queue_nf_hook_drop(struct net *net) return count; } +EXPORT_SYMBOL_GPL(nf_queue_nf_hook_drop); static int __nf_queue(struct sk_buff *skb, const struct nf_hook_state *state, - struct nf_hook_entry *hook_entry, unsigned int queuenum) + const struct nf_hook_entries *entries, + unsigned int index, unsigned int queuenum) { int status = -ENOENT; struct nf_queue_entry *entry = NULL; @@ -139,7 +141,7 @@ static int __nf_queue(struct sk_buff *skb, const struct nf_hook_state *state, *entry = (struct nf_queue_entry) { .skb = skb, .state = *state, - .hook = hook_entry, + .hook_index = index, .size = sizeof(*entry) + afinfo->route_key_size, }; @@ -162,18 +164,16 @@ err: /* Packets leaving via this function must come back through nf_reinject(). */ int nf_queue(struct sk_buff *skb, struct nf_hook_state *state, - struct nf_hook_entry **entryp, unsigned int verdict) + const struct nf_hook_entries *entries, unsigned int index, + unsigned int verdict) { - struct nf_hook_entry *entry = *entryp; int ret; - ret = __nf_queue(skb, state, entry, verdict >> NF_VERDICT_QBITS); + ret = __nf_queue(skb, state, entries, index, verdict >> NF_VERDICT_QBITS); if (ret < 0) { if (ret == -ESRCH && - (verdict & NF_VERDICT_FLAG_QUEUE_BYPASS)) { - *entryp = rcu_dereference(entry->next); + (verdict & NF_VERDICT_FLAG_QUEUE_BYPASS)) return 1; - } kfree_skb(skb); } @@ -182,33 +182,56 @@ int nf_queue(struct sk_buff *skb, struct nf_hook_state *state, static unsigned int nf_iterate(struct sk_buff *skb, struct nf_hook_state *state, - struct nf_hook_entry **entryp) + const struct nf_hook_entries *hooks, + unsigned int *index) { - unsigned int verdict; + const struct nf_hook_entry *hook; + unsigned int verdict, i = *index; - do { + while (i < hooks->num_hook_entries) { + hook = &hooks->hooks[i]; repeat: - verdict = nf_hook_entry_hookfn((*entryp), skb, state); + verdict = nf_hook_entry_hookfn(hook, skb, state); if (verdict != NF_ACCEPT) { if (verdict != NF_REPEAT) return verdict; goto repeat; } - *entryp = rcu_dereference((*entryp)->next); - } while (*entryp); + i++; + } + *index = i; return NF_ACCEPT; } +/* Caller must hold rcu read-side lock */ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) { - struct nf_hook_entry *hook_entry = entry->hook; + const struct nf_hook_entry *hook_entry; + const struct nf_hook_entries *hooks; struct sk_buff *skb = entry->skb; const struct nf_afinfo *afinfo; + const struct net *net; + unsigned int i; int err; + u8 pf; + + net = entry->state.net; + pf = entry->state.pf; + + hooks = rcu_dereference(net->nf.hooks[pf][entry->state.hook]); nf_queue_entry_release_refs(entry); + i = entry->hook_index; + if (WARN_ON_ONCE(i >= hooks->num_hook_entries)) { + kfree_skb(skb); + kfree(entry); + return; + } + + hook_entry = &hooks->hooks[i]; + /* Continue traversal iff userspace said ok... */ if (verdict == NF_REPEAT) verdict = nf_hook_entry_hookfn(hook_entry, skb, &entry->state); @@ -220,27 +243,22 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) } if (verdict == NF_ACCEPT) { - hook_entry = rcu_dereference(hook_entry->next); - if (hook_entry) next_hook: - verdict = nf_iterate(skb, &entry->state, &hook_entry); + ++i; + verdict = nf_iterate(skb, &entry->state, hooks, &i); } switch (verdict & NF_VERDICT_MASK) { case NF_ACCEPT: case NF_STOP: -okfn: local_bh_disable(); entry->state.okfn(entry->state.net, entry->state.sk, skb); local_bh_enable(); break; case NF_QUEUE: - err = nf_queue(skb, &entry->state, &hook_entry, verdict); - if (err == 1) { - if (hook_entry) - goto next_hook; - goto okfn; - } + err = nf_queue(skb, &entry->state, hooks, i, verdict); + if (err == 1) + goto next_hook; break; case NF_STOLEN: break; diff --git a/net/netfilter/nf_sockopt.c b/net/netfilter/nf_sockopt.c index c68c1e58b362..d2a9e6b5d01f 100644 --- a/net/netfilter/nf_sockopt.c +++ b/net/netfilter/nf_sockopt.c @@ -33,7 +33,7 @@ int nf_register_sockopt(struct nf_sockopt_ops *reg) reg->set_optmin, reg->set_optmax) || overlap(ops->get_optmin, ops->get_optmax, reg->get_optmin, reg->get_optmax))) { - NFDEBUG("nf_sock overlap: %u-%u/%u-%u v %u-%u/%u-%u\n", + pr_debug("nf_sock overlap: %u-%u/%u-%u v %u-%u/%u-%u\n", ops->set_optmin, ops->set_optmax, ops->get_optmin, ops->get_optmax, reg->set_optmin, reg->set_optmax, diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 7843efa33c59..929927171426 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -726,7 +726,10 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, if (table == NULL) goto err2; - nla_strlcpy(table->name, name, NFT_TABLE_MAXNAMELEN); + table->name = nla_strdup(name, GFP_KERNEL); + if (table->name == NULL) + goto err3; + INIT_LIST_HEAD(&table->chains); INIT_LIST_HEAD(&table->sets); INIT_LIST_HEAD(&table->objects); @@ -735,10 +738,12 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, nft_ctx_init(&ctx, net, skb, nlh, afi, table, NULL, nla); err = nft_trans_table_add(&ctx, NFT_MSG_NEWTABLE); if (err < 0) - goto err3; + goto err4; list_add_tail_rcu(&table->list, &afi->tables); return 0; +err4: + kfree(table->name); err3: kfree(table); err2: @@ -855,6 +860,10 @@ static int nf_tables_deltable(struct net *net, struct sock *nlsk, if (IS_ERR(table)) return PTR_ERR(table); + if (nlh->nlmsg_flags & NLM_F_NONREC && + table->use > 0) + return -EBUSY; + ctx.afi = afi; ctx.table = table; @@ -865,6 +874,7 @@ static void nf_tables_table_destroy(struct nft_ctx *ctx) { BUG_ON(ctx->table->use > 0); + kfree(ctx->table->name); kfree(ctx->table); module_put(ctx->afi->owner); } @@ -1240,10 +1250,14 @@ static void nf_tables_chain_destroy(struct nft_chain *chain) module_put(basechain->type->owner); free_percpu(basechain->stats); + if (basechain->stats) + static_branch_dec(&nft_counters_enabled); if (basechain->ops[0].dev != NULL) dev_put(basechain->ops[0].dev); + kfree(chain->name); kfree(basechain); } else { + kfree(chain->name); kfree(chain); } } @@ -1325,155 +1339,18 @@ static void nft_chain_release_hook(struct nft_chain_hook *hook) dev_put(hook->dev); } -static int nf_tables_newchain(struct net *net, struct sock *nlsk, - struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[], - struct netlink_ext_ack *extack) +static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, + u8 policy, bool create) { - const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nlattr * uninitialized_var(name); - struct nft_af_info *afi; - struct nft_table *table; + const struct nlattr * const *nla = ctx->nla; + struct nft_table *table = ctx->table; + struct nft_af_info *afi = ctx->afi; + struct nft_base_chain *basechain; + struct nft_stats __percpu *stats; + struct net *net = ctx->net; struct nft_chain *chain; - struct nft_base_chain *basechain = NULL; - u8 genmask = nft_genmask_next(net); - int family = nfmsg->nfgen_family; - u8 policy = NF_ACCEPT; - u64 handle = 0; unsigned int i; - struct nft_stats __percpu *stats; int err; - bool create; - struct nft_ctx ctx; - - create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; - - afi = nf_tables_afinfo_lookup(net, family, true); - if (IS_ERR(afi)) - return PTR_ERR(afi); - - table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE], genmask); - if (IS_ERR(table)) - return PTR_ERR(table); - - chain = NULL; - name = nla[NFTA_CHAIN_NAME]; - - if (nla[NFTA_CHAIN_HANDLE]) { - handle = be64_to_cpu(nla_get_be64(nla[NFTA_CHAIN_HANDLE])); - chain = nf_tables_chain_lookup_byhandle(table, handle, genmask); - if (IS_ERR(chain)) - return PTR_ERR(chain); - } else { - chain = nf_tables_chain_lookup(table, name, genmask); - if (IS_ERR(chain)) { - if (PTR_ERR(chain) != -ENOENT) - return PTR_ERR(chain); - chain = NULL; - } - } - - if (nla[NFTA_CHAIN_POLICY]) { - if (chain != NULL && - !nft_is_base_chain(chain)) - return -EOPNOTSUPP; - - if (chain == NULL && - nla[NFTA_CHAIN_HOOK] == NULL) - return -EOPNOTSUPP; - - policy = ntohl(nla_get_be32(nla[NFTA_CHAIN_POLICY])); - switch (policy) { - case NF_DROP: - case NF_ACCEPT: - break; - default: - return -EINVAL; - } - } - - if (chain != NULL) { - struct nft_stats *stats = NULL; - struct nft_trans *trans; - - if (nlh->nlmsg_flags & NLM_F_EXCL) - return -EEXIST; - if (nlh->nlmsg_flags & NLM_F_REPLACE) - return -EOPNOTSUPP; - - if (nla[NFTA_CHAIN_HOOK]) { - struct nft_base_chain *basechain; - struct nft_chain_hook hook; - struct nf_hook_ops *ops; - - if (!nft_is_base_chain(chain)) - return -EBUSY; - - err = nft_chain_parse_hook(net, nla, afi, &hook, - create); - if (err < 0) - return err; - - basechain = nft_base_chain(chain); - if (basechain->type != hook.type) { - nft_chain_release_hook(&hook); - return -EBUSY; - } - - for (i = 0; i < afi->nops; i++) { - ops = &basechain->ops[i]; - if (ops->hooknum != hook.num || - ops->priority != hook.priority || - ops->dev != hook.dev) { - nft_chain_release_hook(&hook); - return -EBUSY; - } - } - nft_chain_release_hook(&hook); - } - - if (nla[NFTA_CHAIN_HANDLE] && name) { - struct nft_chain *chain2; - - chain2 = nf_tables_chain_lookup(table, - nla[NFTA_CHAIN_NAME], - genmask); - if (IS_ERR(chain2)) - return PTR_ERR(chain2); - } - - if (nla[NFTA_CHAIN_COUNTERS]) { - if (!nft_is_base_chain(chain)) - return -EOPNOTSUPP; - - stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); - if (IS_ERR(stats)) - return PTR_ERR(stats); - } - - nft_ctx_init(&ctx, net, skb, nlh, afi, table, chain, nla); - trans = nft_trans_alloc(&ctx, NFT_MSG_NEWCHAIN, - sizeof(struct nft_trans_chain)); - if (trans == NULL) { - free_percpu(stats); - return -ENOMEM; - } - - nft_trans_chain_stats(trans) = stats; - nft_trans_chain_update(trans) = true; - - if (nla[NFTA_CHAIN_POLICY]) - nft_trans_chain_policy(trans) = policy; - else - nft_trans_chain_policy(trans) = -1; - - if (nla[NFTA_CHAIN_HANDLE] && name) { - nla_strlcpy(nft_trans_chain_name(trans), name, - NFT_CHAIN_MAXNAMELEN); - } - list_add_tail(&trans->list, &net->nft.commit_list); - return 0; - } if (table->use == UINT_MAX) return -EOVERFLOW; @@ -1504,14 +1381,7 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, return PTR_ERR(stats); } basechain->stats = stats; - } else { - stats = netdev_alloc_pcpu_stats(struct nft_stats); - if (stats == NULL) { - nft_chain_release_hook(&hook); - kfree(basechain); - return -ENOMEM; - } - rcu_assign_pointer(basechain->stats, stats); + static_branch_inc(&nft_counters_enabled); } hookfn = hook.type->hooks[hook.num]; @@ -1539,31 +1409,204 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, if (chain == NULL) return -ENOMEM; } - INIT_LIST_HEAD(&chain->rules); chain->handle = nf_tables_alloc_handle(table); chain->table = table; - nla_strlcpy(chain->name, name, NFT_CHAIN_MAXNAMELEN); + chain->name = nla_strdup(nla[NFTA_CHAIN_NAME], GFP_KERNEL); + if (!chain->name) { + err = -ENOMEM; + goto err1; + } err = nf_tables_register_hooks(net, table, chain, afi->nops); if (err < 0) goto err1; - nft_ctx_init(&ctx, net, skb, nlh, afi, table, chain, nla); - err = nft_trans_chain_add(&ctx, NFT_MSG_NEWCHAIN); + ctx->chain = chain; + err = nft_trans_chain_add(ctx, NFT_MSG_NEWCHAIN); if (err < 0) goto err2; table->use++; list_add_tail_rcu(&chain->list, &table->chains); + return 0; err2: nf_tables_unregister_hooks(net, table, chain, afi->nops); err1: nf_tables_chain_destroy(chain); + return err; } +static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, + bool create) +{ + const struct nlattr * const *nla = ctx->nla; + struct nft_table *table = ctx->table; + struct nft_chain *chain = ctx->chain; + struct nft_af_info *afi = ctx->afi; + struct nft_base_chain *basechain; + struct nft_stats *stats = NULL; + struct nft_chain_hook hook; + const struct nlattr *name; + struct nf_hook_ops *ops; + struct nft_trans *trans; + int err, i; + + if (nla[NFTA_CHAIN_HOOK]) { + if (!nft_is_base_chain(chain)) + return -EBUSY; + + err = nft_chain_parse_hook(ctx->net, nla, ctx->afi, &hook, + create); + if (err < 0) + return err; + + basechain = nft_base_chain(chain); + if (basechain->type != hook.type) { + nft_chain_release_hook(&hook); + return -EBUSY; + } + + for (i = 0; i < afi->nops; i++) { + ops = &basechain->ops[i]; + if (ops->hooknum != hook.num || + ops->priority != hook.priority || + ops->dev != hook.dev) { + nft_chain_release_hook(&hook); + return -EBUSY; + } + } + nft_chain_release_hook(&hook); + } + + if (nla[NFTA_CHAIN_HANDLE] && + nla[NFTA_CHAIN_NAME]) { + struct nft_chain *chain2; + + chain2 = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME], + genmask); + if (IS_ERR(chain2)) + return PTR_ERR(chain2); + } + + if (nla[NFTA_CHAIN_COUNTERS]) { + if (!nft_is_base_chain(chain)) + return -EOPNOTSUPP; + + stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); + if (IS_ERR(stats)) + return PTR_ERR(stats); + } + + trans = nft_trans_alloc(ctx, NFT_MSG_NEWCHAIN, + sizeof(struct nft_trans_chain)); + if (trans == NULL) { + free_percpu(stats); + return -ENOMEM; + } + + nft_trans_chain_stats(trans) = stats; + nft_trans_chain_update(trans) = true; + + if (nla[NFTA_CHAIN_POLICY]) + nft_trans_chain_policy(trans) = policy; + else + nft_trans_chain_policy(trans) = -1; + + name = nla[NFTA_CHAIN_NAME]; + if (nla[NFTA_CHAIN_HANDLE] && name) { + nft_trans_chain_name(trans) = + nla_strdup(name, GFP_KERNEL); + if (!nft_trans_chain_name(trans)) { + kfree(trans); + free_percpu(stats); + return -ENOMEM; + } + } + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + + return 0; +} + +static int nf_tables_newchain(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nla[], + struct netlink_ext_ack *extack) +{ + const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + const struct nlattr * uninitialized_var(name); + u8 genmask = nft_genmask_next(net); + int family = nfmsg->nfgen_family; + struct nft_af_info *afi; + struct nft_table *table; + struct nft_chain *chain; + u8 policy = NF_ACCEPT; + struct nft_ctx ctx; + u64 handle = 0; + bool create; + + create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; + + afi = nf_tables_afinfo_lookup(net, family, true); + if (IS_ERR(afi)) + return PTR_ERR(afi); + + table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE], genmask); + if (IS_ERR(table)) + return PTR_ERR(table); + + chain = NULL; + name = nla[NFTA_CHAIN_NAME]; + + if (nla[NFTA_CHAIN_HANDLE]) { + handle = be64_to_cpu(nla_get_be64(nla[NFTA_CHAIN_HANDLE])); + chain = nf_tables_chain_lookup_byhandle(table, handle, genmask); + if (IS_ERR(chain)) + return PTR_ERR(chain); + } else { + chain = nf_tables_chain_lookup(table, name, genmask); + if (IS_ERR(chain)) { + if (PTR_ERR(chain) != -ENOENT) + return PTR_ERR(chain); + chain = NULL; + } + } + + if (nla[NFTA_CHAIN_POLICY]) { + if (chain != NULL && + !nft_is_base_chain(chain)) + return -EOPNOTSUPP; + + if (chain == NULL && + nla[NFTA_CHAIN_HOOK] == NULL) + return -EOPNOTSUPP; + + policy = ntohl(nla_get_be32(nla[NFTA_CHAIN_POLICY])); + switch (policy) { + case NF_DROP: + case NF_ACCEPT: + break; + default: + return -EINVAL; + } + } + + nft_ctx_init(&ctx, net, skb, nlh, afi, table, chain, nla); + + if (chain != NULL) { + if (nlh->nlmsg_flags & NLM_F_EXCL) + return -EEXIST; + if (nlh->nlmsg_flags & NLM_F_REPLACE) + return -EOPNOTSUPP; + + return nf_tables_updchain(&ctx, genmask, policy, create); + } + + return nf_tables_addchain(&ctx, family, genmask, policy, create); +} + static int nf_tables_delchain(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[], @@ -1574,8 +1617,11 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; + struct nft_rule *rule; int family = nfmsg->nfgen_family; struct nft_ctx ctx; + u32 use; + int err; afi = nf_tables_afinfo_lookup(net, family, false); if (IS_ERR(afi)) @@ -1588,11 +1634,30 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); - if (chain->use > 0) + + if (nlh->nlmsg_flags & NLM_F_NONREC && + chain->use > 0) return -EBUSY; nft_ctx_init(&ctx, net, skb, nlh, afi, table, chain, nla); + use = chain->use; + list_for_each_entry(rule, &chain->rules, list) { + if (!nft_is_active_next(net, rule)) + continue; + use--; + + err = nft_delrule(&ctx, rule); + if (err < 0) + return err; + } + + /* There are rules and elements that are still holding references to us, + * we cannot do a recursive removal in this case. + */ + if (use > 0) + return -EBUSY; + return nft_delchain(&ctx); } @@ -1977,8 +2042,8 @@ err: } struct nft_rule_dump_ctx { - char table[NFT_TABLE_MAXNAMELEN]; - char chain[NFT_CHAIN_MAXNAMELEN]; + char *table; + char *chain; }; static int nf_tables_dump_rules(struct sk_buff *skb, @@ -2002,7 +2067,7 @@ static int nf_tables_dump_rules(struct sk_buff *skb, continue; list_for_each_entry_rcu(table, &afi->tables, list) { - if (ctx && ctx->table[0] && + if (ctx && ctx->table && strcmp(ctx->table, table->name) != 0) continue; @@ -2042,7 +2107,13 @@ done: static int nf_tables_dump_rules_done(struct netlink_callback *cb) { - kfree(cb->data); + struct nft_rule_dump_ctx *ctx = cb->data; + + if (ctx) { + kfree(ctx->table); + kfree(ctx->chain); + kfree(ctx); + } return 0; } @@ -2074,12 +2145,23 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, if (!ctx) return -ENOMEM; - if (nla[NFTA_RULE_TABLE]) - nla_strlcpy(ctx->table, nla[NFTA_RULE_TABLE], - sizeof(ctx->table)); - if (nla[NFTA_RULE_CHAIN]) - nla_strlcpy(ctx->chain, nla[NFTA_RULE_CHAIN], - sizeof(ctx->chain)); + if (nla[NFTA_RULE_TABLE]) { + ctx->table = nla_strdup(nla[NFTA_RULE_TABLE], + GFP_KERNEL); + if (!ctx->table) { + kfree(ctx); + return -ENOMEM; + } + } + if (nla[NFTA_RULE_CHAIN]) { + ctx->chain = nla_strdup(nla[NFTA_RULE_CHAIN], + GFP_KERNEL); + if (!ctx->chain) { + kfree(ctx->table); + kfree(ctx); + return -ENOMEM; + } + } c.data = ctx; } @@ -2621,7 +2703,7 @@ static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, unsigned long *inuse; unsigned int n = 0, min = 0; - p = strnchr(name, NFT_SET_MAXNAMELEN, '%'); + p = strchr(name, '%'); if (p != NULL) { if (p[1] != 'd' || strchr(p + 2, '%')) return -EINVAL; @@ -2652,7 +2734,10 @@ cont: free_page((unsigned long)inuse); } - snprintf(set->name, sizeof(set->name), name, min + n); + set->name = kasprintf(GFP_KERNEL, name, min + n); + if (!set->name) + return -ENOMEM; + list_for_each_entry(i, &ctx->table->sets, list) { if (!nft_is_active_next(ctx->net, i)) continue; @@ -2929,7 +3014,7 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, struct nft_table *table; struct nft_set *set; struct nft_ctx ctx; - char name[NFT_SET_MAXNAMELEN]; + char *name; unsigned int size; bool create; u64 timeout; @@ -3075,8 +3160,14 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, goto err1; } - nla_strlcpy(name, nla[NFTA_SET_NAME], sizeof(set->name)); + name = nla_strdup(nla[NFTA_SET_NAME], GFP_KERNEL); + if (!name) { + err = -ENOMEM; + goto err2; + } + err = nf_tables_set_alloc_name(&ctx, set, name); + kfree(name); if (err < 0) goto err2; @@ -3126,6 +3217,7 @@ static void nft_set_destroy(struct nft_set *set) { set->ops->destroy(set); module_put(set->ops->type->owner); + kfree(set->name); kvfree(set); } @@ -3159,7 +3251,9 @@ static int nf_tables_delset(struct net *net, struct sock *nlsk, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME], genmask); if (IS_ERR(set)) return PTR_ERR(set); - if (!list_empty(&set->bindings)) + + if (!list_empty(&set->bindings) || + (nlh->nlmsg_flags & NLM_F_NONREC && atomic_read(&set->nelems) > 0)) return -EBUSY; return nft_delset(&ctx, set); @@ -4209,7 +4303,7 @@ struct nft_object *nf_tables_obj_lookup(const struct nft_table *table, list_for_each_entry(obj, &table->objects, list) { if (!nla_strcmp(nla, obj->name) && - objtype == obj->type->type && + objtype == obj->ops->type->type && nft_active_genmask(obj, genmask)) return obj; } @@ -4231,6 +4325,7 @@ static struct nft_object *nft_obj_init(const struct nft_ctx *ctx, const struct nlattr *attr) { struct nlattr *tb[type->maxattr + 1]; + const struct nft_object_ops *ops; struct nft_object *obj; int err; @@ -4243,16 +4338,27 @@ static struct nft_object *nft_obj_init(const struct nft_ctx *ctx, memset(tb, 0, sizeof(tb[0]) * (type->maxattr + 1)); } + if (type->select_ops) { + ops = type->select_ops(ctx, (const struct nlattr * const *)tb); + if (IS_ERR(ops)) { + err = PTR_ERR(ops); + goto err1; + } + } else { + ops = type->ops; + } + err = -ENOMEM; - obj = kzalloc(sizeof(struct nft_object) + type->size, GFP_KERNEL); + obj = kzalloc(sizeof(*obj) + ops->size, GFP_KERNEL); if (obj == NULL) goto err1; - err = type->init(ctx, (const struct nlattr * const *)tb, obj); + err = ops->init(ctx, (const struct nlattr * const *)tb, obj); if (err < 0) goto err2; - obj->type = type; + obj->ops = ops; + return obj; err2: kfree(obj); @@ -4268,7 +4374,7 @@ static int nft_object_dump(struct sk_buff *skb, unsigned int attr, nest = nla_nest_start(skb, attr); if (!nest) goto nla_put_failure; - if (obj->type->dump(skb, obj, reset) < 0) + if (obj->ops->dump(skb, obj, reset) < 0) goto nla_put_failure; nla_nest_end(skb, nest); return 0; @@ -4363,18 +4469,24 @@ static int nf_tables_newobj(struct net *net, struct sock *nlsk, goto err1; } obj->table = table; - nla_strlcpy(obj->name, nla[NFTA_OBJ_NAME], NFT_OBJ_MAXNAMELEN); + obj->name = nla_strdup(nla[NFTA_OBJ_NAME], GFP_KERNEL); + if (!obj->name) { + err = -ENOMEM; + goto err2; + } err = nft_trans_obj_add(&ctx, NFT_MSG_NEWOBJ, obj); if (err < 0) - goto err2; + goto err3; list_add_tail_rcu(&obj->list, &table->objects); table->use++; return 0; +err3: + kfree(obj->name); err2: - if (obj->type->destroy) - obj->type->destroy(obj); + if (obj->ops->destroy) + obj->ops->destroy(obj); kfree(obj); err1: module_put(type->owner); @@ -4401,7 +4513,7 @@ static int nf_tables_fill_obj_info(struct sk_buff *skb, struct net *net, if (nla_put_string(skb, NFTA_OBJ_TABLE, table->name) || nla_put_string(skb, NFTA_OBJ_NAME, obj->name) || - nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->type->type)) || + nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->ops->type->type)) || nla_put_be32(skb, NFTA_OBJ_USE, htonl(obj->use)) || nft_object_dump(skb, NFTA_OBJ_DATA, obj, reset)) goto nla_put_failure; @@ -4415,7 +4527,7 @@ nla_put_failure: } struct nft_obj_filter { - char table[NFT_OBJ_MAXNAMELEN]; + char *table; u32 type; }; @@ -4455,7 +4567,7 @@ static int nf_tables_dump_obj(struct sk_buff *skb, struct netlink_callback *cb) goto cont; if (filter && filter->type != NFT_OBJECT_UNSPEC && - obj->type->type != filter->type) + obj->ops->type->type != filter->type) goto cont; if (nf_tables_fill_obj_info(skb, net, NETLINK_CB(cb->skb).portid, @@ -4480,7 +4592,10 @@ done: static int nf_tables_dump_obj_done(struct netlink_callback *cb) { - kfree(cb->data); + struct nft_obj_filter *filter = cb->data; + + kfree(filter->table); + kfree(filter); return 0; } @@ -4494,9 +4609,13 @@ nft_obj_filter_alloc(const struct nlattr * const nla[]) if (!filter) return ERR_PTR(-ENOMEM); - if (nla[NFTA_OBJ_TABLE]) - nla_strlcpy(filter->table, nla[NFTA_OBJ_TABLE], - NFT_TABLE_MAXNAMELEN); + if (nla[NFTA_OBJ_TABLE]) { + filter->table = nla_strdup(nla[NFTA_OBJ_TABLE], GFP_KERNEL); + if (!filter->table) { + kfree(filter); + return ERR_PTR(-ENOMEM); + } + } if (nla[NFTA_OBJ_TYPE]) filter->type = ntohl(nla_get_be32(nla[NFTA_OBJ_TYPE])); @@ -4576,10 +4695,11 @@ err: static void nft_obj_destroy(struct nft_object *obj) { - if (obj->type->destroy) - obj->type->destroy(obj); + if (obj->ops->destroy) + obj->ops->destroy(obj); - module_put(obj->type->owner); + module_put(obj->ops->type->owner); + kfree(obj->name); kfree(obj); } @@ -4662,6 +4782,7 @@ static int nf_tables_fill_gen_info(struct sk_buff *skb, struct net *net, { struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; + char buf[TASK_COMM_LEN]; int event = nfnl_msg_type(NFNL_SUBSYS_NFTABLES, NFT_MSG_NEWGEN); nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct nfgenmsg), 0); @@ -4673,7 +4794,9 @@ static int nf_tables_fill_gen_info(struct sk_buff *skb, struct net *net, nfmsg->version = NFNETLINK_V0; nfmsg->res_id = htons(net->nft.base_seq & 0xffff); - if (nla_put_be32(skb, NFTA_GEN_ID, htonl(net->nft.base_seq))) + if (nla_put_be32(skb, NFTA_GEN_ID, htonl(net->nft.base_seq)) || + nla_put_be32(skb, NFTA_GEN_PROC_PID, htonl(task_pid_nr(current))) || + nla_put_string(skb, NFTA_GEN_PROC_NAME, get_task_comm(buf, current))) goto nla_put_failure; nlmsg_end(skb, nlh); @@ -4842,7 +4965,7 @@ static void nft_chain_commit_update(struct nft_trans *trans) { struct nft_base_chain *basechain; - if (nft_trans_chain_name(trans)[0]) + if (nft_trans_chain_name(trans)) strcpy(trans->ctx.chain->name, nft_trans_chain_name(trans)); if (!nft_is_base_chain(trans->ctx.chain)) diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 65dbeadcb118..dfd0bf3810d2 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -29,7 +29,7 @@ static const char *const comments[__NFT_TRACETYPE_MAX] = { [NFT_TRACETYPE_RULE] = "rule", }; -static struct nf_loginfo trace_loginfo = { +static const struct nf_loginfo trace_loginfo = { .type = NF_LOG_TYPE_LOG, .u = { .log = { @@ -114,6 +114,22 @@ static bool nft_payload_fast_eval(const struct nft_expr *expr, return true; } +DEFINE_STATIC_KEY_FALSE(nft_counters_enabled); + +static noinline void nft_update_chain_stats(const struct nft_chain *chain, + const struct nft_pktinfo *pkt) +{ + struct nft_stats *stats; + + local_bh_disable(); + stats = this_cpu_ptr(rcu_dereference(nft_base_chain(chain)->stats)); + u64_stats_update_begin(&stats->syncp); + stats->pkts++; + stats->bytes += pkt->skb->len; + u64_stats_update_end(&stats->syncp); + local_bh_enable(); +} + struct nft_jumpstack { const struct nft_chain *chain; const struct nft_rule *rule; @@ -130,7 +146,6 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv) struct nft_regs regs; unsigned int stackptr = 0; struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE]; - struct nft_stats *stats; int rulenum; unsigned int gencursor = nft_genmask_cur(net); struct nft_traceinfo info; @@ -220,13 +235,8 @@ next_rule: nft_trace_packet(&info, basechain, NULL, -1, NFT_TRACETYPE_POLICY); - rcu_read_lock_bh(); - stats = this_cpu_ptr(rcu_dereference(nft_base_chain(basechain)->stats)); - u64_stats_update_begin(&stats->syncp); - stats->pkts++; - stats->bytes += pkt->skb->len; - u64_stats_update_end(&stats->syncp); - rcu_read_unlock_bh(); + if (static_branch_unlikely(&nft_counters_enabled)) + nft_update_chain_stats(basechain, pkt); return nft_base_chain(basechain)->policy; } diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index e1b15e7a5793..e1dc527a493b 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -162,6 +162,27 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb, NFTA_TRACE_PAD); } +static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info) +{ + switch (info->type) { + case NFT_TRACETYPE_RETURN: + case NFT_TRACETYPE_RULE: + break; + default: + return false; + } + + switch (info->verdict->code) { + case NFT_JUMP: + case NFT_GOTO: + break; + default: + return false; + } + + return true; +} + void nft_trace_notify(struct nft_traceinfo *info) { const struct nft_pktinfo *pkt = info->pkt; @@ -175,13 +196,12 @@ void nft_trace_notify(struct nft_traceinfo *info) return; size = nlmsg_total_size(sizeof(struct nfgenmsg)) + - nla_total_size(NFT_TABLE_MAXNAMELEN) + - nla_total_size(NFT_CHAIN_MAXNAMELEN) + + nla_total_size(strlen(info->chain->table->name)) + + nla_total_size(strlen(info->chain->name)) + nla_total_size_64bit(sizeof(__be64)) + /* rule handle */ nla_total_size(sizeof(__be32)) + /* trace type */ nla_total_size(0) + /* VERDICT, nested */ nla_total_size(sizeof(u32)) + /* verdict code */ - nla_total_size(NFT_CHAIN_MAXNAMELEN) + /* jump target */ nla_total_size(sizeof(u32)) + /* id */ nla_total_size(NFT_TRACETYPE_LL_HSIZE) + nla_total_size(NFT_TRACETYPE_NETWORK_HSIZE) + @@ -194,6 +214,9 @@ void nft_trace_notify(struct nft_traceinfo *info) nla_total_size(sizeof(u32)) + /* nfproto */ nla_total_size(sizeof(u32)); /* policy */ + if (nft_trace_have_verdict_chain(info)) + size += nla_total_size(strlen(info->verdict->chain->name)); /* jump target */ + skb = nlmsg_new(size, GFP_ATOMIC); if (!skb) return; @@ -217,14 +240,11 @@ void nft_trace_notify(struct nft_traceinfo *info) if (trace_fill_id(skb, pkt->skb)) goto nla_put_failure; - if (info->chain) { - if (nla_put_string(skb, NFTA_TRACE_CHAIN, - info->chain->name)) - goto nla_put_failure; - if (nla_put_string(skb, NFTA_TRACE_TABLE, - info->chain->table->name)) - goto nla_put_failure; - } + if (nla_put_string(skb, NFTA_TRACE_CHAIN, info->chain->name)) + goto nla_put_failure; + + if (nla_put_string(skb, NFTA_TRACE_TABLE, info->chain->table->name)) + goto nla_put_failure; if (nf_trace_fill_rule_info(skb, info)) goto nla_put_failure; diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 92b05e188fd1..733d3e4a30d8 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -472,8 +472,7 @@ static void nfnetlink_rcv_skb_batch(struct sk_buff *skb, struct nlmsghdr *nlh) if (msglen > skb->len) msglen = skb->len; - if (nlh->nlmsg_len < NLMSG_HDRLEN || - skb->len < NLMSG_HDRLEN + sizeof(struct nfgenmsg)) + if (skb->len < NLMSG_HDRLEN + sizeof(struct nfgenmsg)) return; err = nla_parse(cda, NFNL_BATCH_MAX, attr, attrlen, nfnl_batch_policy, @@ -500,7 +499,8 @@ static void nfnetlink_rcv(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); - if (nlh->nlmsg_len < NLMSG_HDRLEN || + if (skb->len < NLMSG_HDRLEN || + nlh->nlmsg_len < NLMSG_HDRLEN || skb->len < nlh->nlmsg_len) return; diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 400e9ae97153..32b1c0b44e79 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -47,7 +47,8 @@ static const struct nla_policy cttimeout_nla_policy[CTA_TIMEOUT_MAX+1] = { }; static int -ctnl_timeout_parse_policy(void *timeouts, struct nf_conntrack_l4proto *l4proto, +ctnl_timeout_parse_policy(void *timeouts, + const struct nf_conntrack_l4proto *l4proto, struct net *net, const struct nlattr *attr) { int ret = 0; @@ -74,7 +75,7 @@ static int cttimeout_new_timeout(struct net *net, struct sock *ctnl, { __u16 l3num; __u8 l4num; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; struct ctnl_timeout *timeout, *matching = NULL; char *name; int ret; @@ -158,7 +159,7 @@ ctnl_timeout_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; unsigned int flags = portid ? NLM_F_MULTI : 0; - struct nf_conntrack_l4proto *l4proto = timeout->l4proto; + const struct nf_conntrack_l4proto *l4proto = timeout->l4proto; event = nfnl_msg_type(NFNL_SUBSYS_CTNETLINK_TIMEOUT, event); nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nfmsg), flags); @@ -363,10 +364,10 @@ static int cttimeout_default_set(struct net *net, struct sock *ctnl, const struct nlattr * const cda[], struct netlink_ext_ack *extack) { + const struct nf_conntrack_l4proto *l4proto; + unsigned int *timeouts; __u16 l3num; __u8 l4num; - struct nf_conntrack_l4proto *l4proto; - unsigned int *timeouts; int ret; if (!cda[CTA_TIMEOUT_L3PROTO] || @@ -401,7 +402,7 @@ err: static int cttimeout_default_fill_info(struct net *net, struct sk_buff *skb, u32 portid, u32 seq, u32 type, int event, - struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto) { struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; @@ -453,11 +454,11 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl, const struct nlattr * const cda[], struct netlink_ext_ack *extack) { - __u16 l3num; - __u8 l4num; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l4proto *l4proto; struct sk_buff *skb2; int ret, err; + __u16 l3num; + __u8 l4num; if (!cda[CTA_TIMEOUT_L3PROTO] || !cda[CTA_TIMEOUT_L4PROTO]) return -EINVAL; @@ -505,7 +506,6 @@ ctnl_timeout_find_get(struct net *net, const char *name) { struct ctnl_timeout *timeout, *matching = NULL; - rcu_read_lock(); list_for_each_entry_rcu(timeout, &net->nfct_timeout_list, head) { if (strncmp(timeout->name, name, CTNL_TIMEOUT_NAME_MAX) != 0) continue; @@ -521,7 +521,6 @@ ctnl_timeout_find_get(struct net *net, const char *name) break; } err: - rcu_read_unlock(); return matching; } @@ -572,6 +571,7 @@ static void __net_exit cttimeout_net_exit(struct net *net) { struct ctnl_timeout *cur, *tmp; + nf_ct_unconfirmed_destroy(net); ctnl_untimeout(net, NULL); list_for_each_entry_safe(cur, tmp, &net->nfct_timeout_list, head) { diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index c684ba95dbb4..cad6498f10b0 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -606,7 +606,7 @@ nla_put_failure: return -1; } -static struct nf_loginfo default_loginfo = { +static const struct nf_loginfo default_loginfo = { .type = NF_LOG_TYPE_ULOG, .u = { .ulog = { diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 16fa04086880..c9796629858f 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -41,6 +41,10 @@ #include "../bridge/br_private.h" #endif +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +#include <net/netfilter/nf_conntrack.h> +#endif + #define NFQNL_QMAX_DEFAULT 1024 /* We're using struct nlattr which has 16bit nla_len. Note that nla_len @@ -612,6 +616,18 @@ nlmsg_failure: return NULL; } +static bool nf_ct_drop_unconfirmed(const struct nf_queue_entry *entry) +{ +#if IS_ENABLED(CONFIG_NF_CONNTRACK) + static const unsigned long flags = IPS_CONFIRMED | IPS_DYING; + const struct nf_conn *ct = (void *)skb_nfct(entry->skb); + + if (ct && ((ct->status & flags) == IPS_DYING)) + return true; +#endif + return false; +} + static int __nfqnl_enqueue_packet(struct net *net, struct nfqnl_instance *queue, struct nf_queue_entry *entry) @@ -628,6 +644,9 @@ __nfqnl_enqueue_packet(struct net *net, struct nfqnl_instance *queue, } spin_lock_bh(&queue->lock); + if (nf_ct_drop_unconfirmed(entry)) + goto err_out_free_nskb; + if (queue->queue_total >= queue->queue_maxlen) { if (queue->flags & NFQA_CFG_F_FAIL_OPEN) { failopen = 1; @@ -928,7 +947,6 @@ static unsigned int nfqnl_nf_hook_drop(struct net *net) unsigned int instances = 0; int i; - rcu_read_lock(); for (i = 0; i < INSTANCE_BUCKETS; i++) { struct nfqnl_instance *inst; struct hlist_head *head = &q->instance_table[i]; @@ -938,7 +956,6 @@ static unsigned int nfqnl_nf_hook_drop(struct net *net) instances++; } } - rcu_read_unlock(); return instances; } diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index f5a7cb68694e..b89f4f65b2a0 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -305,7 +305,7 @@ static int nft_target_validate(const struct nft_ctx *ctx, const struct nf_hook_ops *ops = &basechain->ops[0]; hook_mask = 1 << ops->hooknum; - if (!(hook_mask & target->hooks)) + if (target->hooks && !(hook_mask & target->hooks)) return -EINVAL; ret = nft_compat_chain_validate_dependency(target->table, @@ -484,7 +484,7 @@ static int nft_match_validate(const struct nft_ctx *ctx, const struct nf_hook_ops *ops = &basechain->ops[0]; hook_mask = 1 << ops->hooknum; - if (!(hook_mask & match->hooks)) + if (match->hooks && !(hook_mask & match->hooks)) return -EINVAL; ret = nft_compat_chain_validate_dependency(match->table, diff --git a/net/netfilter/nft_counter.c b/net/netfilter/nft_counter.c index 67a710ebde09..eefe3b409925 100644 --- a/net/netfilter/nft_counter.c +++ b/net/netfilter/nft_counter.c @@ -175,15 +175,21 @@ static const struct nla_policy nft_counter_policy[NFTA_COUNTER_MAX + 1] = { [NFTA_COUNTER_BYTES] = { .type = NLA_U64 }, }; -static struct nft_object_type nft_counter_obj __read_mostly = { - .type = NFT_OBJECT_COUNTER, +static struct nft_object_type nft_counter_obj_type; +static const struct nft_object_ops nft_counter_obj_ops = { + .type = &nft_counter_obj_type, .size = sizeof(struct nft_counter_percpu_priv), - .maxattr = NFTA_COUNTER_MAX, - .policy = nft_counter_policy, .eval = nft_counter_obj_eval, .init = nft_counter_obj_init, .destroy = nft_counter_obj_destroy, .dump = nft_counter_obj_dump, +}; + +static struct nft_object_type nft_counter_obj_type __read_mostly = { + .type = NFT_OBJECT_COUNTER, + .ops = &nft_counter_obj_ops, + .maxattr = NFTA_COUNTER_MAX, + .policy = nft_counter_policy, .owner = THIS_MODULE, }; @@ -271,7 +277,7 @@ static int __init nft_counter_module_init(void) for_each_possible_cpu(cpu) seqcount_init(per_cpu_ptr(&nft_counter_seq, cpu)); - err = nft_register_obj(&nft_counter_obj); + err = nft_register_obj(&nft_counter_obj_type); if (err < 0) return err; @@ -281,14 +287,14 @@ static int __init nft_counter_module_init(void) return 0; err1: - nft_unregister_obj(&nft_counter_obj); + nft_unregister_obj(&nft_counter_obj_type); return err; } static void __exit nft_counter_module_exit(void) { nft_unregister_expr(&nft_counter_type); - nft_unregister_obj(&nft_counter_obj); + nft_unregister_obj(&nft_counter_obj_type); } module_init(nft_counter_module_init); diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 1678e9e75e8e..bd0975d7dd6f 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -904,15 +904,21 @@ static const struct nla_policy nft_ct_helper_policy[NFTA_CT_HELPER_MAX + 1] = { [NFTA_CT_HELPER_L4PROTO] = { .type = NLA_U8 }, }; -static struct nft_object_type nft_ct_helper_obj __read_mostly = { - .type = NFT_OBJECT_CT_HELPER, +static struct nft_object_type nft_ct_helper_obj_type; +static const struct nft_object_ops nft_ct_helper_obj_ops = { + .type = &nft_ct_helper_obj_type, .size = sizeof(struct nft_ct_helper_obj), - .maxattr = NFTA_CT_HELPER_MAX, - .policy = nft_ct_helper_policy, .eval = nft_ct_helper_obj_eval, .init = nft_ct_helper_obj_init, .destroy = nft_ct_helper_obj_destroy, .dump = nft_ct_helper_obj_dump, +}; + +static struct nft_object_type nft_ct_helper_obj_type __read_mostly = { + .type = NFT_OBJECT_CT_HELPER, + .ops = &nft_ct_helper_obj_ops, + .maxattr = NFTA_CT_HELPER_MAX, + .policy = nft_ct_helper_policy, .owner = THIS_MODULE, }; @@ -930,7 +936,7 @@ static int __init nft_ct_module_init(void) if (err < 0) goto err1; - err = nft_register_obj(&nft_ct_helper_obj); + err = nft_register_obj(&nft_ct_helper_obj_type); if (err < 0) goto err2; @@ -945,7 +951,7 @@ err1: static void __exit nft_ct_module_exit(void) { - nft_unregister_obj(&nft_ct_helper_obj); + nft_unregister_obj(&nft_ct_helper_obj_type); nft_unregister_expr(&nft_notrack_type); nft_unregister_expr(&nft_ct_type); } diff --git a/net/netfilter/nft_exthdr.c b/net/netfilter/nft_exthdr.c index 1ec49fe5845f..a0a93d987a3b 100644 --- a/net/netfilter/nft_exthdr.c +++ b/net/netfilter/nft_exthdr.c @@ -8,6 +8,7 @@ * Development of this code funded by Astaro AG (http://www.astaro.com/) */ +#include <asm/unaligned.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> @@ -23,6 +24,7 @@ struct nft_exthdr { u8 len; u8 op; enum nft_registers dreg:8; + enum nft_registers sreg:8; u8 flags; }; @@ -61,6 +63,26 @@ err: regs->verdict.code = NFT_BREAK; } +static void * +nft_tcp_header_pointer(const struct nft_pktinfo *pkt, + unsigned int len, void *buffer, unsigned int *tcphdr_len) +{ + struct tcphdr *tcph; + + if (!pkt->tprot_set || pkt->tprot != IPPROTO_TCP) + return NULL; + + tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff, sizeof(*tcph), buffer); + if (!tcph) + return NULL; + + *tcphdr_len = __tcp_hdrlen(tcph); + if (*tcphdr_len < sizeof(*tcph) || *tcphdr_len > len) + return NULL; + + return skb_header_pointer(pkt->skb, pkt->xt.thoff, *tcphdr_len, buffer); +} + static void nft_exthdr_tcp_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) @@ -72,18 +94,7 @@ static void nft_exthdr_tcp_eval(const struct nft_expr *expr, struct tcphdr *tcph; u8 *opt; - if (!pkt->tprot_set || pkt->tprot != IPPROTO_TCP) - goto err; - - tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff, sizeof(*tcph), buff); - if (!tcph) - goto err; - - tcphdr_len = __tcp_hdrlen(tcph); - if (tcphdr_len < sizeof(*tcph)) - goto err; - - tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff, tcphdr_len, buff); + tcph = nft_tcp_header_pointer(pkt, sizeof(buff), buff, &tcphdr_len); if (!tcph) goto err; @@ -115,6 +126,88 @@ err: regs->verdict.code = NFT_BREAK; } +static void nft_exthdr_tcp_set_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + u8 buff[sizeof(struct tcphdr) + MAX_TCP_OPTION_SPACE]; + struct nft_exthdr *priv = nft_expr_priv(expr); + unsigned int i, optl, tcphdr_len, offset; + struct tcphdr *tcph; + u8 *opt; + u32 src; + + tcph = nft_tcp_header_pointer(pkt, sizeof(buff), buff, &tcphdr_len); + if (!tcph) + return; + + opt = (u8 *)tcph; + for (i = sizeof(*tcph); i < tcphdr_len - 1; i += optl) { + union { + u8 octet; + __be16 v16; + __be32 v32; + } old, new; + + optl = optlen(opt, i); + + if (priv->type != opt[i]) + continue; + + if (i + optl > tcphdr_len || priv->len + priv->offset > optl) + return; + + if (!skb_make_writable(pkt->skb, pkt->xt.thoff + i + priv->len)) + return; + + tcph = nft_tcp_header_pointer(pkt, sizeof(buff), buff, + &tcphdr_len); + if (!tcph) + return; + + src = regs->data[priv->sreg]; + offset = i + priv->offset; + + switch (priv->len) { + case 2: + old.v16 = get_unaligned((u16 *)(opt + offset)); + new.v16 = src; + + switch (priv->type) { + case TCPOPT_MSS: + /* increase can cause connection to stall */ + if (ntohs(old.v16) <= ntohs(new.v16)) + return; + break; + } + + if (old.v16 == new.v16) + return; + + put_unaligned(new.v16, (u16*)(opt + offset)); + inet_proto_csum_replace2(&tcph->check, pkt->skb, + old.v16, new.v16, false); + break; + case 4: + new.v32 = src; + old.v32 = get_unaligned((u32 *)(opt + offset)); + + if (old.v32 == new.v32) + return; + + put_unaligned(new.v32, (u32*)(opt + offset)); + inet_proto_csum_replace4(&tcph->check, pkt->skb, + old.v32, new.v32, false); + break; + default: + WARN_ON_ONCE(1); + break; + } + + return; + } +} + static const struct nla_policy nft_exthdr_policy[NFTA_EXTHDR_MAX + 1] = { [NFTA_EXTHDR_DREG] = { .type = NLA_U32 }, [NFTA_EXTHDR_TYPE] = { .type = NLA_U8 }, @@ -171,12 +264,57 @@ static int nft_exthdr_init(const struct nft_ctx *ctx, NFT_DATA_VALUE, priv->len); } -static int nft_exthdr_dump(struct sk_buff *skb, const struct nft_expr *expr) +static int nft_exthdr_tcp_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { - const struct nft_exthdr *priv = nft_expr_priv(expr); + struct nft_exthdr *priv = nft_expr_priv(expr); + u32 offset, len, flags = 0, op = NFT_EXTHDR_OP_IPV6; + int err; - if (nft_dump_register(skb, NFTA_EXTHDR_DREG, priv->dreg)) - goto nla_put_failure; + if (!tb[NFTA_EXTHDR_SREG] || + !tb[NFTA_EXTHDR_TYPE] || + !tb[NFTA_EXTHDR_OFFSET] || + !tb[NFTA_EXTHDR_LEN]) + return -EINVAL; + + if (tb[NFTA_EXTHDR_DREG] || tb[NFTA_EXTHDR_FLAGS]) + return -EINVAL; + + err = nft_parse_u32_check(tb[NFTA_EXTHDR_OFFSET], U8_MAX, &offset); + if (err < 0) + return err; + + err = nft_parse_u32_check(tb[NFTA_EXTHDR_LEN], U8_MAX, &len); + if (err < 0) + return err; + + if (offset < 2) + return -EOPNOTSUPP; + + switch (len) { + case 2: break; + case 4: break; + default: + return -EOPNOTSUPP; + } + + err = nft_parse_u32_check(tb[NFTA_EXTHDR_OP], U8_MAX, &op); + if (err < 0) + return err; + + priv->type = nla_get_u8(tb[NFTA_EXTHDR_TYPE]); + priv->offset = offset; + priv->len = len; + priv->sreg = nft_parse_register(tb[NFTA_EXTHDR_SREG]); + priv->flags = flags; + priv->op = op; + + return nft_validate_register_load(priv->sreg, priv->len); +} + +static int nft_exthdr_dump_common(struct sk_buff *skb, const struct nft_exthdr *priv) +{ if (nla_put_u8(skb, NFTA_EXTHDR_TYPE, priv->type)) goto nla_put_failure; if (nla_put_be32(skb, NFTA_EXTHDR_OFFSET, htonl(priv->offset))) @@ -193,6 +331,26 @@ nla_put_failure: return -1; } +static int nft_exthdr_dump(struct sk_buff *skb, const struct nft_expr *expr) +{ + const struct nft_exthdr *priv = nft_expr_priv(expr); + + if (nft_dump_register(skb, NFTA_EXTHDR_DREG, priv->dreg)) + return -1; + + return nft_exthdr_dump_common(skb, priv); +} + +static int nft_exthdr_dump_set(struct sk_buff *skb, const struct nft_expr *expr) +{ + const struct nft_exthdr *priv = nft_expr_priv(expr); + + if (nft_dump_register(skb, NFTA_EXTHDR_SREG, priv->sreg)) + return -1; + + return nft_exthdr_dump_common(skb, priv); +} + static struct nft_expr_type nft_exthdr_type; static const struct nft_expr_ops nft_exthdr_ipv6_ops = { .type = &nft_exthdr_type, @@ -210,6 +368,14 @@ static const struct nft_expr_ops nft_exthdr_tcp_ops = { .dump = nft_exthdr_dump, }; +static const struct nft_expr_ops nft_exthdr_tcp_set_ops = { + .type = &nft_exthdr_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_exthdr)), + .eval = nft_exthdr_tcp_set_eval, + .init = nft_exthdr_tcp_set_init, + .dump = nft_exthdr_dump_set, +}; + static const struct nft_expr_ops * nft_exthdr_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[]) @@ -219,12 +385,21 @@ nft_exthdr_select_ops(const struct nft_ctx *ctx, if (!tb[NFTA_EXTHDR_OP]) return &nft_exthdr_ipv6_ops; - op = ntohl(nla_get_u32(tb[NFTA_EXTHDR_OP])); + if (tb[NFTA_EXTHDR_SREG] && tb[NFTA_EXTHDR_DREG]) + return ERR_PTR(-EOPNOTSUPP); + + op = ntohl(nla_get_be32(tb[NFTA_EXTHDR_OP])); switch (op) { case NFT_EXTHDR_OP_TCPOPT: - return &nft_exthdr_tcp_ops; + if (tb[NFTA_EXTHDR_SREG]) + return &nft_exthdr_tcp_set_ops; + if (tb[NFTA_EXTHDR_DREG]) + return &nft_exthdr_tcp_ops; + break; case NFT_EXTHDR_OP_IPV6: - return &nft_exthdr_ipv6_ops; + if (tb[NFTA_EXTHDR_DREG]) + return &nft_exthdr_ipv6_ops; + break; } return ERR_PTR(-EOPNOTSUPP); diff --git a/net/netfilter/nft_fib_netdev.c b/net/netfilter/nft_fib_netdev.c new file mode 100644 index 000000000000..3997ee36cfbd --- /dev/null +++ b/net/netfilter/nft_fib_netdev.c @@ -0,0 +1,87 @@ +/* + * Copyright (c) 2017 Pablo M. Bermudo Garay <pablombg@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This code is based on net/netfilter/nft_fib_inet.c, written by + * Florian Westphal <fw@strlen.de>. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/netlink.h> +#include <linux/netfilter.h> +#include <linux/netfilter/nf_tables.h> +#include <net/netfilter/nf_tables_core.h> +#include <net/netfilter/nf_tables.h> + +#include <net/netfilter/nft_fib.h> + +static void nft_fib_netdev_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + const struct nft_fib *priv = nft_expr_priv(expr); + + switch (ntohs(pkt->skb->protocol)) { + case ETH_P_IP: + switch (priv->result) { + case NFT_FIB_RESULT_OIF: + case NFT_FIB_RESULT_OIFNAME: + return nft_fib4_eval(expr, regs, pkt); + case NFT_FIB_RESULT_ADDRTYPE: + return nft_fib4_eval_type(expr, regs, pkt); + } + break; + case ETH_P_IPV6: + switch (priv->result) { + case NFT_FIB_RESULT_OIF: + case NFT_FIB_RESULT_OIFNAME: + return nft_fib6_eval(expr, regs, pkt); + case NFT_FIB_RESULT_ADDRTYPE: + return nft_fib6_eval_type(expr, regs, pkt); + } + break; + } + + regs->verdict.code = NFT_BREAK; +} + +static struct nft_expr_type nft_fib_netdev_type; +static const struct nft_expr_ops nft_fib_netdev_ops = { + .type = &nft_fib_netdev_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_fib)), + .eval = nft_fib_netdev_eval, + .init = nft_fib_init, + .dump = nft_fib_dump, + .validate = nft_fib_validate, +}; + +static struct nft_expr_type nft_fib_netdev_type __read_mostly = { + .family = NFPROTO_NETDEV, + .name = "fib", + .ops = &nft_fib_netdev_ops, + .policy = nft_fib_policy, + .maxattr = NFTA_FIB_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_fib_netdev_module_init(void) +{ + return nft_register_expr(&nft_fib_netdev_type); +} + +static void __exit nft_fib_netdev_module_exit(void) +{ + nft_unregister_expr(&nft_fib_netdev_type); +} + +module_init(nft_fib_netdev_module_init); +module_exit(nft_fib_netdev_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Pablo M. Bermudo Garay <pablombg@gmail.com>"); +MODULE_ALIAS_NFT_AF_EXPR(5, "fib"); diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c index 18dd57a52651..a9fc298ef4c3 100644 --- a/net/netfilter/nft_limit.c +++ b/net/netfilter/nft_limit.c @@ -65,19 +65,23 @@ static int nft_limit_init(struct nft_limit *limit, limit->nsecs = unit * NSEC_PER_SEC; if (limit->rate == 0 || limit->nsecs < unit) return -EOVERFLOW; - limit->tokens = limit->tokens_max = limit->nsecs; - - if (tb[NFTA_LIMIT_BURST]) { - u64 rate; + if (tb[NFTA_LIMIT_BURST]) limit->burst = ntohl(nla_get_be32(tb[NFTA_LIMIT_BURST])); + else + limit->burst = 0; - rate = limit->rate + limit->burst; - if (rate < limit->rate) - return -EOVERFLOW; + if (limit->rate + limit->burst < limit->rate) + return -EOVERFLOW; + + /* The token bucket size limits the number of tokens can be + * accumulated. tokens_max specifies the bucket size. + * tokens_max = unit * (rate + burst) / rate. + */ + limit->tokens = div_u64(limit->nsecs * (limit->rate + limit->burst), + limit->rate); + limit->tokens_max = limit->tokens; - limit->rate = rate; - } if (tb[NFTA_LIMIT_FLAGS]) { u32 flags = ntohl(nla_get_be32(tb[NFTA_LIMIT_FLAGS])); @@ -95,9 +99,8 @@ static int nft_limit_dump(struct sk_buff *skb, const struct nft_limit *limit, { u32 flags = limit->invert ? NFT_LIMIT_F_INV : 0; u64 secs = div_u64(limit->nsecs, NSEC_PER_SEC); - u64 rate = limit->rate - limit->burst; - if (nla_put_be64(skb, NFTA_LIMIT_RATE, cpu_to_be64(rate), + if (nla_put_be64(skb, NFTA_LIMIT_RATE, cpu_to_be64(limit->rate), NFTA_LIMIT_PAD) || nla_put_be64(skb, NFTA_LIMIT_UNIT, cpu_to_be64(secs), NFTA_LIMIT_PAD) || @@ -165,9 +168,9 @@ static const struct nft_expr_ops nft_limit_pkts_ops = { .dump = nft_limit_pkts_dump, }; -static void nft_limit_pkt_bytes_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) +static void nft_limit_bytes_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) { struct nft_limit *priv = nft_expr_priv(expr); u64 cost = div64_u64(priv->nsecs * pkt->skb->len, priv->rate); @@ -176,29 +179,29 @@ static void nft_limit_pkt_bytes_eval(const struct nft_expr *expr, regs->verdict.code = NFT_BREAK; } -static int nft_limit_pkt_bytes_init(const struct nft_ctx *ctx, - const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_limit_bytes_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_limit *priv = nft_expr_priv(expr); return nft_limit_init(priv, tb); } -static int nft_limit_pkt_bytes_dump(struct sk_buff *skb, - const struct nft_expr *expr) +static int nft_limit_bytes_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_limit *priv = nft_expr_priv(expr); return nft_limit_dump(skb, priv, NFT_LIMIT_PKT_BYTES); } -static const struct nft_expr_ops nft_limit_pkt_bytes_ops = { +static const struct nft_expr_ops nft_limit_bytes_ops = { .type = &nft_limit_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_limit)), - .eval = nft_limit_pkt_bytes_eval, - .init = nft_limit_pkt_bytes_init, - .dump = nft_limit_pkt_bytes_dump, + .eval = nft_limit_bytes_eval, + .init = nft_limit_bytes_init, + .dump = nft_limit_bytes_dump, }; static const struct nft_expr_ops * @@ -212,7 +215,7 @@ nft_limit_select_ops(const struct nft_ctx *ctx, case NFT_LIMIT_PKTS: return &nft_limit_pkts_ops; case NFT_LIMIT_PKT_BYTES: - return &nft_limit_pkt_bytes_ops; + return &nft_limit_bytes_ops; } return ERR_PTR(-EOPNOTSUPP); } @@ -226,14 +229,133 @@ static struct nft_expr_type nft_limit_type __read_mostly = { .owner = THIS_MODULE, }; +static void nft_limit_obj_pkts_eval(struct nft_object *obj, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + struct nft_limit_pkts *priv = nft_obj_data(obj); + + if (nft_limit_eval(&priv->limit, priv->cost)) + regs->verdict.code = NFT_BREAK; +} + +static int nft_limit_obj_pkts_init(const struct nft_ctx *ctx, + const struct nlattr * const tb[], + struct nft_object *obj) +{ + struct nft_limit_pkts *priv = nft_obj_data(obj); + int err; + + err = nft_limit_init(&priv->limit, tb); + if (err < 0) + return err; + + priv->cost = div64_u64(priv->limit.nsecs, priv->limit.rate); + return 0; +} + +static int nft_limit_obj_pkts_dump(struct sk_buff *skb, + struct nft_object *obj, + bool reset) +{ + const struct nft_limit_pkts *priv = nft_obj_data(obj); + + return nft_limit_dump(skb, &priv->limit, NFT_LIMIT_PKTS); +} + +static struct nft_object_type nft_limit_obj_type; +static const struct nft_object_ops nft_limit_obj_pkts_ops = { + .type = &nft_limit_obj_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_limit_pkts)), + .init = nft_limit_obj_pkts_init, + .eval = nft_limit_obj_pkts_eval, + .dump = nft_limit_obj_pkts_dump, +}; + +static void nft_limit_obj_bytes_eval(struct nft_object *obj, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + struct nft_limit *priv = nft_obj_data(obj); + u64 cost = div64_u64(priv->nsecs * pkt->skb->len, priv->rate); + + if (nft_limit_eval(priv, cost)) + regs->verdict.code = NFT_BREAK; +} + +static int nft_limit_obj_bytes_init(const struct nft_ctx *ctx, + const struct nlattr * const tb[], + struct nft_object *obj) +{ + struct nft_limit *priv = nft_obj_data(obj); + + return nft_limit_init(priv, tb); +} + +static int nft_limit_obj_bytes_dump(struct sk_buff *skb, + struct nft_object *obj, + bool reset) +{ + const struct nft_limit *priv = nft_obj_data(obj); + + return nft_limit_dump(skb, priv, NFT_LIMIT_PKT_BYTES); +} + +static struct nft_object_type nft_limit_obj_type; +static const struct nft_object_ops nft_limit_obj_bytes_ops = { + .type = &nft_limit_obj_type, + .size = sizeof(struct nft_limit), + .init = nft_limit_obj_bytes_init, + .eval = nft_limit_obj_bytes_eval, + .dump = nft_limit_obj_bytes_dump, +}; + +static const struct nft_object_ops * +nft_limit_obj_select_ops(const struct nft_ctx *ctx, + const struct nlattr * const tb[]) +{ + if (!tb[NFTA_LIMIT_TYPE]) + return &nft_limit_obj_pkts_ops; + + switch (ntohl(nla_get_be32(tb[NFTA_LIMIT_TYPE]))) { + case NFT_LIMIT_PKTS: + return &nft_limit_obj_pkts_ops; + case NFT_LIMIT_PKT_BYTES: + return &nft_limit_obj_bytes_ops; + } + return ERR_PTR(-EOPNOTSUPP); +} + +static struct nft_object_type nft_limit_obj_type __read_mostly = { + .select_ops = nft_limit_obj_select_ops, + .type = NFT_OBJECT_LIMIT, + .maxattr = NFTA_LIMIT_MAX, + .policy = nft_limit_policy, + .owner = THIS_MODULE, +}; + static int __init nft_limit_module_init(void) { - return nft_register_expr(&nft_limit_type); + int err; + + err = nft_register_obj(&nft_limit_obj_type); + if (err < 0) + return err; + + err = nft_register_expr(&nft_limit_type); + if (err < 0) + goto err1; + + return 0; +err1: + nft_unregister_obj(&nft_limit_obj_type); + return err; } static void __exit nft_limit_module_exit(void) { nft_unregister_expr(&nft_limit_type); + nft_unregister_obj(&nft_limit_obj_type); } module_init(nft_limit_module_init); @@ -242,3 +364,4 @@ module_exit(nft_limit_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); MODULE_ALIAS_NFT_EXPR("limit"); +MODULE_ALIAS_NFT_OBJ(NFT_OBJECT_LIMIT); diff --git a/net/netfilter/nft_objref.c b/net/netfilter/nft_objref.c index 1dd428fbaaa3..7bcdc48f3d73 100644 --- a/net/netfilter/nft_objref.c +++ b/net/netfilter/nft_objref.c @@ -22,7 +22,7 @@ static void nft_objref_eval(const struct nft_expr *expr, { struct nft_object *obj = nft_objref_priv(expr); - obj->type->eval(obj, regs, pkt); + obj->ops->eval(obj, regs, pkt); } static int nft_objref_init(const struct nft_ctx *ctx, @@ -54,7 +54,8 @@ static int nft_objref_dump(struct sk_buff *skb, const struct nft_expr *expr) const struct nft_object *obj = nft_objref_priv(expr); if (nla_put_string(skb, NFTA_OBJREF_IMM_NAME, obj->name) || - nla_put_be32(skb, NFTA_OBJREF_IMM_TYPE, htonl(obj->type->type))) + nla_put_be32(skb, NFTA_OBJREF_IMM_TYPE, + htonl(obj->ops->type->type))) goto nla_put_failure; return 0; @@ -104,7 +105,7 @@ static void nft_objref_map_eval(const struct nft_expr *expr, return; } obj = *nft_set_ext_obj(ext); - obj->type->eval(obj, regs, pkt); + obj->ops->eval(obj, regs, pkt); } static int nft_objref_map_init(const struct nft_ctx *ctx, diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index 7d699bbd45b0..e110b0ebbf58 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -184,7 +184,7 @@ static bool nft_payload_udp_checksum(struct sk_buff *skb, unsigned int thoff) if (!uh) return false; - return uh->check; + return (__force bool)uh->check; } static int nft_payload_l4csum_offset(const struct nft_pktinfo *pkt, diff --git a/net/netfilter/nft_quota.c b/net/netfilter/nft_quota.c index 25e33159be57..0ed124a93fcf 100644 --- a/net/netfilter/nft_quota.c +++ b/net/netfilter/nft_quota.c @@ -151,14 +151,20 @@ static int nft_quota_obj_dump(struct sk_buff *skb, struct nft_object *obj, return nft_quota_do_dump(skb, priv, reset); } -static struct nft_object_type nft_quota_obj __read_mostly = { - .type = NFT_OBJECT_QUOTA, +static struct nft_object_type nft_quota_obj_type; +static const struct nft_object_ops nft_quota_obj_ops = { + .type = &nft_quota_obj_type, .size = sizeof(struct nft_quota), - .maxattr = NFTA_QUOTA_MAX, - .policy = nft_quota_policy, .init = nft_quota_obj_init, .eval = nft_quota_obj_eval, .dump = nft_quota_obj_dump, +}; + +static struct nft_object_type nft_quota_obj_type __read_mostly = { + .type = NFT_OBJECT_QUOTA, + .ops = &nft_quota_obj_ops, + .maxattr = NFTA_QUOTA_MAX, + .policy = nft_quota_policy, .owner = THIS_MODULE, }; @@ -209,7 +215,7 @@ static int __init nft_quota_module_init(void) { int err; - err = nft_register_obj(&nft_quota_obj); + err = nft_register_obj(&nft_quota_obj_type); if (err < 0) return err; @@ -219,14 +225,14 @@ static int __init nft_quota_module_init(void) return 0; err1: - nft_unregister_obj(&nft_quota_obj); + nft_unregister_obj(&nft_quota_obj_type); return err; } static void __exit nft_quota_module_exit(void) { nft_unregister_expr(&nft_quota_type); - nft_unregister_obj(&nft_quota_obj); + nft_unregister_obj(&nft_quota_obj_type); } module_init(nft_quota_module_init); diff --git a/net/netfilter/nft_rt.c b/net/netfilter/nft_rt.c index c7383d8f88d0..a6b7d05aeacf 100644 --- a/net/netfilter/nft_rt.c +++ b/net/netfilter/nft_rt.c @@ -23,6 +23,43 @@ struct nft_rt { enum nft_registers dreg:8; }; +static u16 get_tcpmss(const struct nft_pktinfo *pkt, const struct dst_entry *skbdst) +{ + u32 minlen = sizeof(struct ipv6hdr), mtu = dst_mtu(skbdst); + const struct sk_buff *skb = pkt->skb; + const struct nf_afinfo *ai; + struct flowi fl; + + memset(&fl, 0, sizeof(fl)); + + switch (nft_pf(pkt)) { + case NFPROTO_IPV4: + fl.u.ip4.daddr = ip_hdr(skb)->saddr; + minlen = sizeof(struct iphdr) + sizeof(struct tcphdr); + break; + case NFPROTO_IPV6: + fl.u.ip6.daddr = ipv6_hdr(skb)->saddr; + minlen = sizeof(struct ipv6hdr) + sizeof(struct tcphdr); + break; + } + + ai = nf_get_afinfo(nft_pf(pkt)); + if (ai) { + struct dst_entry *dst = NULL; + + ai->route(nft_net(pkt), &dst, &fl, false); + if (dst) { + mtu = min(mtu, dst_mtu(dst)); + dst_release(dst); + } + } + + if (mtu <= minlen || mtu > 0xffff) + return TCP_MSS_DEFAULT; + + return mtu - minlen; +} + static void nft_rt_get_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) @@ -46,8 +83,8 @@ static void nft_rt_get_eval(const struct nft_expr *expr, if (nft_pf(pkt) != NFPROTO_IPV4) goto err; - *dest = rt_nexthop((const struct rtable *)dst, - ip_hdr(skb)->daddr); + *dest = (__force u32)rt_nexthop((const struct rtable *)dst, + ip_hdr(skb)->daddr); break; case NFT_RT_NEXTHOP6: if (nft_pf(pkt) != NFPROTO_IPV6) @@ -57,6 +94,9 @@ static void nft_rt_get_eval(const struct nft_expr *expr, &ipv6_hdr(skb)->daddr), sizeof(struct in6_addr)); break; + case NFT_RT_TCPMSS: + nft_reg_store16(dest, get_tcpmss(pkt, dst)); + break; default: WARN_ON(1); goto err; @@ -67,7 +107,7 @@ err: regs->verdict.code = NFT_BREAK; } -const struct nla_policy nft_rt_policy[NFTA_RT_MAX + 1] = { +static const struct nla_policy nft_rt_policy[NFTA_RT_MAX + 1] = { [NFTA_RT_DREG] = { .type = NLA_U32 }, [NFTA_RT_KEY] = { .type = NLA_U32 }, }; @@ -94,6 +134,9 @@ static int nft_rt_get_init(const struct nft_ctx *ctx, case NFT_RT_NEXTHOP6: len = sizeof(struct in6_addr); break; + case NFT_RT_TCPMSS: + len = sizeof(u16); + break; default: return -EOPNOTSUPP; } @@ -118,6 +161,29 @@ nla_put_failure: return -1; } +static int nft_rt_validate(const struct nft_ctx *ctx, const struct nft_expr *expr, + const struct nft_data **data) +{ + const struct nft_rt *priv = nft_expr_priv(expr); + unsigned int hooks; + + switch (priv->key) { + case NFT_RT_NEXTHOP4: + case NFT_RT_NEXTHOP6: + case NFT_RT_CLASSID: + return 0; + case NFT_RT_TCPMSS: + hooks = (1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING); + break; + default: + return -EINVAL; + } + + return nft_chain_validate_hooks(ctx->chain, hooks); +} + static struct nft_expr_type nft_rt_type; static const struct nft_expr_ops nft_rt_get_ops = { .type = &nft_rt_type, @@ -125,6 +191,7 @@ static const struct nft_expr_ops nft_rt_get_ops = { .eval = nft_rt_get_eval, .init = nft_rt_get_init, .dump = nft_rt_get_dump, + .validate = nft_rt_validate, }; static struct nft_expr_type nft_rt_type __read_mostly = { diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index bce5382f1d49..d83a4ec5900d 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -19,8 +19,9 @@ #include <net/netfilter/nf_tables.h> struct nft_rbtree { - rwlock_t lock; struct rb_root root; + rwlock_t lock; + seqcount_t count; }; struct nft_rbtree_elem { @@ -40,8 +41,9 @@ static bool nft_rbtree_equal(const struct nft_set *set, const void *this, return memcmp(this, nft_set_ext_key(&interval->ext), set->klen) == 0; } -static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, - const u32 *key, const struct nft_set_ext **ext) +static bool __nft_rbtree_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext, + unsigned int seq) { struct nft_rbtree *priv = nft_set_priv(set); const struct nft_rbtree_elem *rbe, *interval = NULL; @@ -50,15 +52,17 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, const void *this; int d; - read_lock_bh(&priv->lock); - parent = priv->root.rb_node; + parent = rcu_dereference_raw(priv->root.rb_node); while (parent != NULL) { + if (read_seqcount_retry(&priv->count, seq)) + return false; + rbe = rb_entry(parent, struct nft_rbtree_elem, node); this = nft_set_ext_key(&rbe->ext); d = memcmp(this, key, set->klen); if (d < 0) { - parent = parent->rb_left; + parent = rcu_dereference_raw(parent->rb_left); if (interval && nft_rbtree_equal(set, this, interval) && nft_rbtree_interval_end(this) && @@ -66,15 +70,14 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, continue; interval = rbe; } else if (d > 0) - parent = parent->rb_right; + parent = rcu_dereference_raw(parent->rb_right); else { if (!nft_set_elem_active(&rbe->ext, genmask)) { - parent = parent->rb_left; + parent = rcu_dereference_raw(parent->rb_left); continue; } if (nft_rbtree_interval_end(rbe)) goto out; - read_unlock_bh(&priv->lock); *ext = &rbe->ext; return true; @@ -84,15 +87,32 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, if (set->flags & NFT_SET_INTERVAL && interval != NULL && nft_set_elem_active(&interval->ext, genmask) && !nft_rbtree_interval_end(interval)) { - read_unlock_bh(&priv->lock); *ext = &interval->ext; return true; } out: - read_unlock_bh(&priv->lock); return false; } +static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext) +{ + struct nft_rbtree *priv = nft_set_priv(set); + unsigned int seq = read_seqcount_begin(&priv->count); + bool ret; + + ret = __nft_rbtree_lookup(net, set, key, ext, seq); + if (ret || !read_seqcount_retry(&priv->count, seq)) + return ret; + + read_lock_bh(&priv->lock); + seq = read_seqcount_begin(&priv->count); + ret = __nft_rbtree_lookup(net, set, key, ext, seq); + read_unlock_bh(&priv->lock); + + return ret; +} + static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, struct nft_rbtree_elem *new, struct nft_set_ext **ext) @@ -130,7 +150,7 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, } } } - rb_link_node(&new->node, parent, p); + rb_link_node_rcu(&new->node, parent, p); rb_insert_color(&new->node, &priv->root); return 0; } @@ -144,7 +164,9 @@ static int nft_rbtree_insert(const struct net *net, const struct nft_set *set, int err; write_lock_bh(&priv->lock); + write_seqcount_begin(&priv->count); err = __nft_rbtree_insert(net, set, rbe, ext); + write_seqcount_end(&priv->count); write_unlock_bh(&priv->lock); return err; @@ -158,7 +180,9 @@ static void nft_rbtree_remove(const struct net *net, struct nft_rbtree_elem *rbe = elem->priv; write_lock_bh(&priv->lock); + write_seqcount_begin(&priv->count); rb_erase(&rbe->node, &priv->root); + write_seqcount_end(&priv->count); write_unlock_bh(&priv->lock); } @@ -264,6 +288,7 @@ static int nft_rbtree_init(const struct nft_set *set, struct nft_rbtree *priv = nft_set_priv(set); rwlock_init(&priv->lock); + seqcount_init(&priv->count); priv->root = RB_ROOT; return 0; } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index e1648238a9c9..c83a3b5e1c6c 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1192,16 +1192,10 @@ xt_replace_table(struct xt_table *table, #ifdef CONFIG_AUDIT if (audit_enabled) { - struct audit_buffer *ab; - - ab = audit_log_start(current->audit_context, GFP_KERNEL, - AUDIT_NETFILTER_CFG); - if (ab) { - audit_log_format(ab, "table=%s family=%u entries=%u", - table->name, table->af, - private->number); - audit_log_end(ab); - } + audit_log(current->audit_context, GFP_KERNEL, + AUDIT_NETFILTER_CFG, + "table=%s family=%u entries=%u", + table->name, table->af, private->number); } #endif diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 623ef37de886..5a152e2acfd5 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -121,9 +121,9 @@ xt_ct_set_timeout(struct nf_conn *ct, const struct xt_tgchk_param *par, { #ifdef CONFIG_NF_CONNTRACK_TIMEOUT typeof(nf_ct_timeout_find_get_hook) timeout_find_get; + const struct nf_conntrack_l4proto *l4proto; struct ctnl_timeout *timeout; struct nf_conn_timeout *timeout_ext; - struct nf_conntrack_l4proto *l4proto; int ret = 0; u8 proto; diff --git a/net/netfilter/xt_NETMAP.c b/net/netfilter/xt_NETMAP.c index e45a01255e70..58aa9dd3c5b7 100644 --- a/net/netfilter/xt_NETMAP.c +++ b/net/netfilter/xt_NETMAP.c @@ -77,10 +77,10 @@ netmap_tg4(struct sk_buff *skb, const struct xt_action_param *par) const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo; struct nf_nat_range newrange; - NF_CT_ASSERT(xt_hooknum(par) == NF_INET_PRE_ROUTING || - xt_hooknum(par) == NF_INET_POST_ROUTING || - xt_hooknum(par) == NF_INET_LOCAL_OUT || - xt_hooknum(par) == NF_INET_LOCAL_IN); + WARN_ON(xt_hooknum(par) != NF_INET_PRE_ROUTING && + xt_hooknum(par) != NF_INET_POST_ROUTING && + xt_hooknum(par) != NF_INET_LOCAL_OUT && + xt_hooknum(par) != NF_INET_LOCAL_IN); ct = nf_ct_get(skb, &ctinfo); netmask = ~(mr->range[0].min_ip ^ mr->range[0].max_ip); diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index c64aca611ac5..9dae4d665965 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -62,11 +62,9 @@ static u_int32_t tcpmss_reverse_mtu(struct net *net, memset(fl6, 0, sizeof(*fl6)); fl6->daddr = ipv6_hdr(skb)->saddr; } - rcu_read_lock(); ai = nf_get_afinfo(family); if (ai != NULL) ai->route(net, (struct dst_entry **)&rt, &fl, false); - rcu_read_unlock(); if (rt != NULL) { mtu = dst_mtu(&rt->dst); diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c index d767e35fff6b..17d7705e3bd4 100644 --- a/net/netfilter/xt_TPROXY.c +++ b/net/netfilter/xt_TPROXY.c @@ -70,13 +70,11 @@ tproxy_laddr4(struct sk_buff *skb, __be32 user_laddr, __be32 daddr) return user_laddr; laddr = 0; - rcu_read_lock(); indev = __in_dev_get_rcu(skb->dev); for_primary_ifa(indev) { laddr = ifa->ifa_local; break; } endfor_ifa(indev); - rcu_read_unlock(); return laddr ? laddr : daddr; } @@ -125,7 +123,7 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp, __tcp_hdrlen(tcph), saddr, sport, daddr, dport, - in->ifindex); + in->ifindex, 0); if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; @@ -195,7 +193,7 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp, thoff + __tcp_hdrlen(tcph), saddr, sport, daddr, ntohs(dport), - in->ifindex); + in->ifindex, 0); if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; @@ -208,7 +206,7 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp, case NFT_LOOKUP_ESTABLISHED: sk = __inet6_lookup_established(net, &tcp_hashinfo, saddr, sport, daddr, ntohs(dport), - in->ifindex); + in->ifindex, 0); break; default: BUG(); @@ -391,7 +389,6 @@ tproxy_laddr6(struct sk_buff *skb, const struct in6_addr *user_laddr, return user_laddr; laddr = NULL; - rcu_read_lock(); indev = __in6_dev_get(skb->dev); if (indev) { read_lock_bh(&indev->lock); @@ -404,7 +401,6 @@ tproxy_laddr6(struct sk_buff *skb, const struct in6_addr *user_laddr, } read_unlock_bh(&indev->lock); } - rcu_read_unlock(); return laddr ? laddr : daddr; } diff --git a/net/netfilter/xt_addrtype.c b/net/netfilter/xt_addrtype.c index e329dabde35f..3b2be2ae6987 100644 --- a/net/netfilter/xt_addrtype.c +++ b/net/netfilter/xt_addrtype.c @@ -47,8 +47,6 @@ static u32 match_lookup_rt6(struct net *net, const struct net_device *dev, if (dev) flow.flowi6_oif = dev->ifindex; - rcu_read_lock(); - afinfo = nf_get_afinfo(NFPROTO_IPV6); if (afinfo != NULL) { const struct nf_ipv6_ops *v6ops; @@ -63,7 +61,6 @@ static u32 match_lookup_rt6(struct net *net, const struct net_device *dev, } else { route_err = 1; } - rcu_read_unlock(); if (route_err) return XT_ADDRTYPE_UNREACHABLE; diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index b8fd4ab762ed..ffa8eec980e9 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -58,8 +58,7 @@ struct xt_connlimit_rb { static spinlock_t xt_connlimit_locks[CONNLIMIT_LOCK_SLOTS] __cacheline_aligned_in_smp; struct xt_connlimit_data { - struct rb_root climit_root4[CONNLIMIT_SLOTS]; - struct rb_root climit_root6[CONNLIMIT_SLOTS]; + struct rb_root climit_root[CONNLIMIT_SLOTS]; }; static u_int32_t connlimit_rnd __read_mostly; @@ -144,7 +143,6 @@ static unsigned int check_hlist(struct net *net, unsigned int length = 0; *addit = true; - rcu_read_lock(); /* check the saved connections */ hlist_for_each_entry_safe(conn, n, head, node) { @@ -179,8 +177,6 @@ static unsigned int check_hlist(struct net *net, length++; } - rcu_read_unlock(); - return length; } @@ -297,13 +293,11 @@ static int count_them(struct net *net, int count; u32 hash; - if (family == NFPROTO_IPV6) { + if (family == NFPROTO_IPV6) hash = connlimit_iphash6(addr, mask); - root = &data->climit_root6[hash]; - } else { + else hash = connlimit_iphash(addr->ip & mask->ip); - root = &data->climit_root4[hash]; - } + root = &data->climit_root[hash]; spin_lock_bh(&xt_connlimit_locks[hash % CONNLIMIT_LOCK_SLOTS]); @@ -382,10 +376,8 @@ static int connlimit_mt_check(const struct xt_mtchk_param *par) return -ENOMEM; } - for (i = 0; i < ARRAY_SIZE(info->data->climit_root4); ++i) - info->data->climit_root4[i] = RB_ROOT; - for (i = 0; i < ARRAY_SIZE(info->data->climit_root6); ++i) - info->data->climit_root6[i] = RB_ROOT; + for (i = 0; i < ARRAY_SIZE(info->data->climit_root); ++i) + info->data->climit_root[i] = RB_ROOT; return 0; } @@ -416,10 +408,8 @@ static void connlimit_mt_destroy(const struct xt_mtdtor_param *par) nf_ct_netns_put(par->net, par->family); - for (i = 0; i < ARRAY_SIZE(info->data->climit_root4); ++i) - destroy_tree(&info->data->climit_root4[i]); - for (i = 0; i < ARRAY_SIZE(info->data->climit_root6); ++i) - destroy_tree(&info->data->climit_root6[i]); + for (i = 0; i < ARRAY_SIZE(info->data->climit_root); ++i) + destroy_tree(&info->data->climit_root[i]); kfree(info->data); } diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index 762e1874f28b..10d48234f5f4 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -56,6 +56,7 @@ static inline struct hashlimit_net *hashlimit_pernet(struct net *net) } /* need to declare this at the top */ +static const struct file_operations dl_file_ops_v2; static const struct file_operations dl_file_ops_v1; static const struct file_operations dl_file_ops; @@ -87,8 +88,19 @@ struct dsthash_ent { unsigned long expires; /* precalculated expiry time */ struct { unsigned long prev; /* last modification */ - u_int64_t credit; - u_int64_t credit_cap, cost; + union { + struct { + u_int64_t credit; + u_int64_t credit_cap; + u_int64_t cost; + }; + struct { + u_int32_t interval, prev_window; + u_int64_t current_rate; + u_int64_t rate; + int64_t burst; + }; + }; } rateinfo; struct rcu_head rcu; }; @@ -99,7 +111,7 @@ struct xt_hashlimit_htable { u_int8_t family; bool rnd_initialized; - struct hashlimit_cfg2 cfg; /* config */ + struct hashlimit_cfg3 cfg; /* config */ /* used internally */ spinlock_t lock; /* lock for list_head */ @@ -116,10 +128,10 @@ struct xt_hashlimit_htable { }; static int -cfg_copy(struct hashlimit_cfg2 *to, void *from, int revision) +cfg_copy(struct hashlimit_cfg3 *to, const void *from, int revision) { if (revision == 1) { - struct hashlimit_cfg1 *cfg = from; + struct hashlimit_cfg1 *cfg = (struct hashlimit_cfg1 *)from; to->mode = cfg->mode; to->avg = cfg->avg; @@ -131,7 +143,19 @@ cfg_copy(struct hashlimit_cfg2 *to, void *from, int revision) to->srcmask = cfg->srcmask; to->dstmask = cfg->dstmask; } else if (revision == 2) { - memcpy(to, from, sizeof(struct hashlimit_cfg2)); + struct hashlimit_cfg2 *cfg = (struct hashlimit_cfg2 *)from; + + to->mode = cfg->mode; + to->avg = cfg->avg; + to->burst = cfg->burst; + to->size = cfg->size; + to->max = cfg->max; + to->gc_interval = cfg->gc_interval; + to->expire = cfg->expire; + to->srcmask = cfg->srcmask; + to->dstmask = cfg->dstmask; + } else if (revision == 3) { + memcpy(to, from, sizeof(struct hashlimit_cfg3)); } else { return -EINVAL; } @@ -240,13 +264,14 @@ dsthash_free(struct xt_hashlimit_htable *ht, struct dsthash_ent *ent) } static void htable_gc(struct work_struct *work); -static int htable_create(struct net *net, struct hashlimit_cfg2 *cfg, +static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg, const char *name, u_int8_t family, struct xt_hashlimit_htable **out_hinfo, int revision) { struct hashlimit_net *hashlimit_net = hashlimit_pernet(net); struct xt_hashlimit_htable *hinfo; + const struct file_operations *fops; unsigned int size, i; int ret; @@ -268,7 +293,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg2 *cfg, *out_hinfo = hinfo; /* copy match config into hashtable config */ - ret = cfg_copy(&hinfo->cfg, (void *)cfg, 2); + ret = cfg_copy(&hinfo->cfg, (void *)cfg, 3); if (ret) return ret; @@ -293,11 +318,21 @@ static int htable_create(struct net *net, struct hashlimit_cfg2 *cfg, } spin_lock_init(&hinfo->lock); + switch (revision) { + case 1: + fops = &dl_file_ops_v1; + break; + case 2: + fops = &dl_file_ops_v2; + break; + default: + fops = &dl_file_ops; + } + hinfo->pde = proc_create_data(name, 0, (family == NFPROTO_IPV4) ? hashlimit_net->ipt_hashlimit : hashlimit_net->ip6t_hashlimit, - (revision == 1) ? &dl_file_ops_v1 : &dl_file_ops, - hinfo); + fops, hinfo); if (hinfo->pde == NULL) { kfree(hinfo->name); vfree(hinfo); @@ -482,6 +517,25 @@ static u32 user2credits_byte(u32 user) return (u32) (us >> 32); } +static u64 user2rate(u64 user) +{ + if (user != 0) { + return div64_u64(XT_HASHLIMIT_SCALE_v2, user); + } else { + pr_warn("invalid rate from userspace: %llu\n", user); + return 0; + } +} + +static u64 user2rate_bytes(u64 user) +{ + u64 r; + + r = user ? 0xFFFFFFFFULL / user : 0xFFFFFFFFULL; + r = (r - 1) << 4; + return r; +} + static void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now, u32 mode, int revision) { @@ -491,6 +545,21 @@ static void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now, if (delta == 0) return; + if (revision >= 3 && mode & XT_HASHLIMIT_RATE_MATCH) { + u64 interval = dh->rateinfo.interval * HZ; + + if (delta < interval) + return; + + dh->rateinfo.prev = now; + dh->rateinfo.prev_window = + ((dh->rateinfo.current_rate * interval) > + (delta * dh->rateinfo.rate)); + dh->rateinfo.current_rate = 0; + + return; + } + dh->rateinfo.prev = now; if (mode & XT_HASHLIMIT_BYTES) { @@ -515,7 +584,23 @@ static void rateinfo_init(struct dsthash_ent *dh, struct xt_hashlimit_htable *hinfo, int revision) { dh->rateinfo.prev = jiffies; - if (hinfo->cfg.mode & XT_HASHLIMIT_BYTES) { + if (revision >= 3 && hinfo->cfg.mode & XT_HASHLIMIT_RATE_MATCH) { + dh->rateinfo.prev_window = 0; + dh->rateinfo.current_rate = 0; + if (hinfo->cfg.mode & XT_HASHLIMIT_BYTES) { + dh->rateinfo.rate = user2rate_bytes(hinfo->cfg.avg); + if (hinfo->cfg.burst) + dh->rateinfo.burst = + hinfo->cfg.burst * dh->rateinfo.rate; + else + dh->rateinfo.burst = dh->rateinfo.rate; + } else { + dh->rateinfo.rate = user2rate(hinfo->cfg.avg); + dh->rateinfo.burst = + hinfo->cfg.burst + dh->rateinfo.rate; + } + dh->rateinfo.interval = hinfo->cfg.interval; + } else if (hinfo->cfg.mode & XT_HASHLIMIT_BYTES) { dh->rateinfo.credit = CREDITS_PER_JIFFY_BYTES * HZ; dh->rateinfo.cost = user2credits_byte(hinfo->cfg.avg); dh->rateinfo.credit_cap = hinfo->cfg.burst; @@ -648,7 +733,7 @@ static u32 hashlimit_byte_cost(unsigned int len, struct dsthash_ent *dh) static bool hashlimit_mt_common(const struct sk_buff *skb, struct xt_action_param *par, struct xt_hashlimit_htable *hinfo, - const struct hashlimit_cfg2 *cfg, int revision) + const struct hashlimit_cfg3 *cfg, int revision) { unsigned long now = jiffies; struct dsthash_ent *dh; @@ -659,12 +744,12 @@ hashlimit_mt_common(const struct sk_buff *skb, struct xt_action_param *par, if (hashlimit_init_dst(hinfo, &dst, skb, par->thoff) < 0) goto hotdrop; - rcu_read_lock_bh(); + local_bh_disable(); dh = dsthash_find(hinfo, &dst); if (dh == NULL) { dh = dsthash_alloc_init(hinfo, &dst, &race); if (dh == NULL) { - rcu_read_unlock_bh(); + local_bh_enable(); goto hotdrop; } else if (race) { /* Already got an entry, update expiration timeout */ @@ -680,6 +765,20 @@ hashlimit_mt_common(const struct sk_buff *skb, struct xt_action_param *par, rateinfo_recalc(dh, now, hinfo->cfg.mode, revision); } + if (cfg->mode & XT_HASHLIMIT_RATE_MATCH) { + cost = (cfg->mode & XT_HASHLIMIT_BYTES) ? skb->len : 1; + dh->rateinfo.current_rate += cost; + + if (!dh->rateinfo.prev_window && + (dh->rateinfo.current_rate <= dh->rateinfo.burst)) { + spin_unlock(&dh->lock); + rcu_read_unlock_bh(); + return !(cfg->mode & XT_HASHLIMIT_INVERT); + } else { + goto overlimit; + } + } + if (cfg->mode & XT_HASHLIMIT_BYTES) cost = hashlimit_byte_cost(skb->len, dh); else @@ -689,12 +788,13 @@ hashlimit_mt_common(const struct sk_buff *skb, struct xt_action_param *par, /* below the limit */ dh->rateinfo.credit -= cost; spin_unlock(&dh->lock); - rcu_read_unlock_bh(); + local_bh_enable(); return !(cfg->mode & XT_HASHLIMIT_INVERT); } +overlimit: spin_unlock(&dh->lock); - rcu_read_unlock_bh(); + local_bh_enable(); /* default match is underlimit - so over the limit, we need to invert */ return cfg->mode & XT_HASHLIMIT_INVERT; @@ -708,7 +808,7 @@ hashlimit_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_hashlimit_mtinfo1 *info = par->matchinfo; struct xt_hashlimit_htable *hinfo = info->hinfo; - struct hashlimit_cfg2 cfg = {}; + struct hashlimit_cfg3 cfg = {}; int ret; ret = cfg_copy(&cfg, (void *)&info->cfg, 1); @@ -720,17 +820,33 @@ hashlimit_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) } static bool -hashlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) +hashlimit_mt_v2(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_hashlimit_mtinfo2 *info = par->matchinfo; struct xt_hashlimit_htable *hinfo = info->hinfo; + struct hashlimit_cfg3 cfg = {}; + int ret; + + ret = cfg_copy(&cfg, (void *)&info->cfg, 2); + + if (ret) + return ret; + + return hashlimit_mt_common(skb, par, hinfo, &cfg, 2); +} + +static bool +hashlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_hashlimit_mtinfo3 *info = par->matchinfo; + struct xt_hashlimit_htable *hinfo = info->hinfo; - return hashlimit_mt_common(skb, par, hinfo, &info->cfg, 2); + return hashlimit_mt_common(skb, par, hinfo, &info->cfg, 3); } static int hashlimit_mt_check_common(const struct xt_mtchk_param *par, struct xt_hashlimit_htable **hinfo, - struct hashlimit_cfg2 *cfg, + struct hashlimit_cfg3 *cfg, const char *name, int revision) { struct net *net = par->net; @@ -753,7 +869,17 @@ static int hashlimit_mt_check_common(const struct xt_mtchk_param *par, } /* Check for overflow. */ - if (cfg->mode & XT_HASHLIMIT_BYTES) { + if (revision >= 3 && cfg->mode & XT_HASHLIMIT_RATE_MATCH) { + if (cfg->avg == 0) { + pr_info("hashlimit invalid rate\n"); + return -ERANGE; + } + + if (cfg->interval == 0) { + pr_info("hashlimit invalid interval\n"); + return -EINVAL; + } + } else if (cfg->mode & XT_HASHLIMIT_BYTES) { if (user2credits_byte(cfg->avg) == 0) { pr_info("overflow, rate too high: %llu\n", cfg->avg); return -EINVAL; @@ -784,7 +910,7 @@ static int hashlimit_mt_check_common(const struct xt_mtchk_param *par, static int hashlimit_mt_check_v1(const struct xt_mtchk_param *par) { struct xt_hashlimit_mtinfo1 *info = par->matchinfo; - struct hashlimit_cfg2 cfg = {}; + struct hashlimit_cfg3 cfg = {}; int ret; if (info->name[sizeof(info->name) - 1] != '\0') @@ -799,15 +925,40 @@ static int hashlimit_mt_check_v1(const struct xt_mtchk_param *par) &cfg, info->name, 1); } -static int hashlimit_mt_check(const struct xt_mtchk_param *par) +static int hashlimit_mt_check_v2(const struct xt_mtchk_param *par) { struct xt_hashlimit_mtinfo2 *info = par->matchinfo; + struct hashlimit_cfg3 cfg = {}; + int ret; + + if (info->name[sizeof(info->name) - 1] != '\0') + return -EINVAL; + + ret = cfg_copy(&cfg, (void *)&info->cfg, 2); + + if (ret) + return ret; + + return hashlimit_mt_check_common(par, &info->hinfo, + &cfg, info->name, 2); +} + +static int hashlimit_mt_check(const struct xt_mtchk_param *par) +{ + struct xt_hashlimit_mtinfo3 *info = par->matchinfo; if (info->name[sizeof(info->name) - 1] != '\0') return -EINVAL; return hashlimit_mt_check_common(par, &info->hinfo, &info->cfg, - info->name, 2); + info->name, 3); +} + +static void hashlimit_mt_destroy_v2(const struct xt_mtdtor_param *par) +{ + const struct xt_hashlimit_mtinfo2 *info = par->matchinfo; + + htable_put(info->hinfo); } static void hashlimit_mt_destroy_v1(const struct xt_mtdtor_param *par) @@ -819,7 +970,7 @@ static void hashlimit_mt_destroy_v1(const struct xt_mtdtor_param *par) static void hashlimit_mt_destroy(const struct xt_mtdtor_param *par) { - const struct xt_hashlimit_mtinfo2 *info = par->matchinfo; + const struct xt_hashlimit_mtinfo3 *info = par->matchinfo; htable_put(info->hinfo); } @@ -840,9 +991,20 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { .name = "hashlimit", .revision = 2, .family = NFPROTO_IPV4, - .match = hashlimit_mt, + .match = hashlimit_mt_v2, .matchsize = sizeof(struct xt_hashlimit_mtinfo2), .usersize = offsetof(struct xt_hashlimit_mtinfo2, hinfo), + .checkentry = hashlimit_mt_check_v2, + .destroy = hashlimit_mt_destroy_v2, + .me = THIS_MODULE, + }, + { + .name = "hashlimit", + .revision = 3, + .family = NFPROTO_IPV4, + .match = hashlimit_mt, + .matchsize = sizeof(struct xt_hashlimit_mtinfo3), + .usersize = offsetof(struct xt_hashlimit_mtinfo3, hinfo), .checkentry = hashlimit_mt_check, .destroy = hashlimit_mt_destroy, .me = THIS_MODULE, @@ -863,9 +1025,20 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { .name = "hashlimit", .revision = 2, .family = NFPROTO_IPV6, - .match = hashlimit_mt, + .match = hashlimit_mt_v2, .matchsize = sizeof(struct xt_hashlimit_mtinfo2), .usersize = offsetof(struct xt_hashlimit_mtinfo2, hinfo), + .checkentry = hashlimit_mt_check_v2, + .destroy = hashlimit_mt_destroy_v2, + .me = THIS_MODULE, + }, + { + .name = "hashlimit", + .revision = 3, + .family = NFPROTO_IPV6, + .match = hashlimit_mt, + .matchsize = sizeof(struct xt_hashlimit_mtinfo3), + .usersize = offsetof(struct xt_hashlimit_mtinfo3, hinfo), .checkentry = hashlimit_mt_check, .destroy = hashlimit_mt_destroy, .me = THIS_MODULE, @@ -947,6 +1120,21 @@ static void dl_seq_print(struct dsthash_ent *ent, u_int8_t family, } } +static int dl_seq_real_show_v2(struct dsthash_ent *ent, u_int8_t family, + struct seq_file *s) +{ + const struct xt_hashlimit_htable *ht = s->private; + + spin_lock(&ent->lock); + /* recalculate to show accurate numbers */ + rateinfo_recalc(ent, jiffies, ht->cfg.mode, 2); + + dl_seq_print(ent, family, s); + + spin_unlock(&ent->lock); + return seq_has_overflowed(s); +} + static int dl_seq_real_show_v1(struct dsthash_ent *ent, u_int8_t family, struct seq_file *s) { @@ -969,7 +1157,7 @@ static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family, spin_lock(&ent->lock); /* recalculate to show accurate numbers */ - rateinfo_recalc(ent, jiffies, ht->cfg.mode, 2); + rateinfo_recalc(ent, jiffies, ht->cfg.mode, 3); dl_seq_print(ent, family, s); @@ -977,6 +1165,20 @@ static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family, return seq_has_overflowed(s); } +static int dl_seq_show_v2(struct seq_file *s, void *v) +{ + struct xt_hashlimit_htable *htable = s->private; + unsigned int *bucket = (unsigned int *)v; + struct dsthash_ent *ent; + + if (!hlist_empty(&htable->hash[*bucket])) { + hlist_for_each_entry(ent, &htable->hash[*bucket], node) + if (dl_seq_real_show_v2(ent, htable->family, s)) + return -1; + } + return 0; +} + static int dl_seq_show_v1(struct seq_file *s, void *v) { struct xt_hashlimit_htable *htable = s->private; @@ -1012,6 +1214,13 @@ static const struct seq_operations dl_seq_ops_v1 = { .show = dl_seq_show_v1 }; +static const struct seq_operations dl_seq_ops_v2 = { + .start = dl_seq_start, + .next = dl_seq_next, + .stop = dl_seq_stop, + .show = dl_seq_show_v2 +}; + static const struct seq_operations dl_seq_ops = { .start = dl_seq_start, .next = dl_seq_next, @@ -1019,6 +1228,18 @@ static const struct seq_operations dl_seq_ops = { .show = dl_seq_show }; +static int dl_proc_open_v2(struct inode *inode, struct file *file) +{ + int ret = seq_open(file, &dl_seq_ops_v2); + + if (!ret) { + struct seq_file *sf = file->private_data; + + sf->private = PDE_DATA(inode); + } + return ret; +} + static int dl_proc_open_v1(struct inode *inode, struct file *file) { int ret = seq_open(file, &dl_seq_ops_v1); @@ -1042,6 +1263,14 @@ static int dl_proc_open(struct inode *inode, struct file *file) return ret; } +static const struct file_operations dl_file_ops_v2 = { + .owner = THIS_MODULE, + .open = dl_proc_open_v2, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + static const struct file_operations dl_file_ops_v1 = { .owner = THIS_MODULE, .open = dl_proc_open_v1, diff --git a/net/netfilter/xt_nat.c b/net/netfilter/xt_nat.c index 8107b3eb865f..0fd14d1eb09d 100644 --- a/net/netfilter/xt_nat.c +++ b/net/netfilter/xt_nat.c @@ -58,9 +58,9 @@ xt_snat_target_v0(struct sk_buff *skb, const struct xt_action_param *par) struct nf_conn *ct; ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct != NULL && - (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || - ctinfo == IP_CT_RELATED_REPLY)); + WARN_ON(!(ct != NULL && + (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || + ctinfo == IP_CT_RELATED_REPLY))); xt_nat_convert_range(&range, &mr->range[0]); return nf_nat_setup_info(ct, &range, NF_NAT_MANIP_SRC); @@ -75,8 +75,8 @@ xt_dnat_target_v0(struct sk_buff *skb, const struct xt_action_param *par) struct nf_conn *ct; ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct != NULL && - (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)); + WARN_ON(!(ct != NULL && + (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED))); xt_nat_convert_range(&range, &mr->range[0]); return nf_nat_setup_info(ct, &range, NF_NAT_MANIP_DST); @@ -90,9 +90,9 @@ xt_snat_target_v1(struct sk_buff *skb, const struct xt_action_param *par) struct nf_conn *ct; ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct != NULL && - (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || - ctinfo == IP_CT_RELATED_REPLY)); + WARN_ON(!(ct != NULL && + (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || + ctinfo == IP_CT_RELATED_REPLY))); return nf_nat_setup_info(ct, range, NF_NAT_MANIP_SRC); } @@ -105,8 +105,8 @@ xt_dnat_target_v1(struct sk_buff *skb, const struct xt_action_param *par) struct nf_conn *ct; ct = nf_ct_get(skb, &ctinfo); - NF_CT_ASSERT(ct != NULL && - (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)); + WARN_ON(!(ct != NULL && + (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED))); return nf_nat_setup_info(ct, range, NF_NAT_MANIP_DST); } diff --git a/net/netfilter/xt_osf.c b/net/netfilter/xt_osf.c index 71cfa9551d08..36e14b1f061d 100644 --- a/net/netfilter/xt_osf.c +++ b/net/netfilter/xt_osf.c @@ -226,7 +226,6 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p) sizeof(struct tcphdr), optsize, opts); } - rcu_read_lock(); list_for_each_entry_rcu(kf, &xt_osf_fingers[df], finger_entry) { int foptsize, optnum; @@ -340,7 +339,6 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p) info->loglevel == XT_OSF_LOGLEVEL_FIRST) break; } - rcu_read_unlock(); if (!fcount && (info->flags & XT_OSF_LOG)) nf_log_packet(net, xt_family(p), xt_hooknum(p), skb, xt_in(p), diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 3f6c4fa78bdb..245fa350a7a8 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -106,7 +106,7 @@ static DEFINE_SPINLOCK(recent_lock); static DEFINE_MUTEX(recent_mutex); #ifdef CONFIG_PROC_FS -static const struct file_operations recent_old_fops, recent_mt_fops; +static const struct file_operations recent_mt_fops; #endif static u_int32_t hash_rnd __read_mostly; diff --git a/net/nsh/Kconfig b/net/nsh/Kconfig new file mode 100644 index 000000000000..bafc3dd60c2c --- /dev/null +++ b/net/nsh/Kconfig @@ -0,0 +1,9 @@ +menuconfig NET_NSH + tristate "Network Service Header (NSH) protocol" + default n + ---help--- + Network Service Header is an implementation of Service Function + Chaining (RFC 7665). The current implementation in Linux supports + only MD type 1 and only with the openvswitch module. + + If unsure, say N. diff --git a/net/nsh/Makefile b/net/nsh/Makefile new file mode 100644 index 000000000000..c93c787385ca --- /dev/null +++ b/net/nsh/Makefile @@ -0,0 +1 @@ +obj-$(CONFIG_NET_NSH) += nsh.o diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c new file mode 100644 index 000000000000..58fb827439a8 --- /dev/null +++ b/net/nsh/nsh.c @@ -0,0 +1,91 @@ +/* + * Network Service Header + * + * Copyright (c) 2017 Red Hat, Inc. -- Jiri Benc <jbenc@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/module.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> +#include <net/nsh.h> +#include <net/tun_proto.h> + +static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, + netdev_features_t features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + unsigned int nsh_len, mac_len; + __be16 proto; + int nhoff; + + skb_reset_network_header(skb); + + nhoff = skb->network_header - skb->mac_header; + mac_len = skb->mac_len; + + if (unlikely(!pskb_may_pull(skb, NSH_BASE_HDR_LEN))) + goto out; + nsh_len = nsh_hdr_len(nsh_hdr(skb)); + if (unlikely(!pskb_may_pull(skb, nsh_len))) + goto out; + + proto = tun_p_to_eth_p(nsh_hdr(skb)->np); + if (!proto) + goto out; + + __skb_pull(skb, nsh_len); + + skb_reset_mac_header(skb); + skb_reset_mac_len(skb); + skb->protocol = proto; + + features &= NETIF_F_SG; + segs = skb_mac_gso_segment(skb, features); + if (IS_ERR_OR_NULL(segs)) { + skb_gso_error_unwind(skb, htons(ETH_P_NSH), nsh_len, + skb->network_header - nhoff, + mac_len); + goto out; + } + + for (skb = segs; skb; skb = skb->next) { + skb->protocol = htons(ETH_P_NSH); + __skb_push(skb, nsh_len); + skb_set_mac_header(skb, -nhoff); + skb->network_header = skb->mac_header + mac_len; + skb->mac_len = mac_len; + } + +out: + return segs; +} + +static struct packet_offload nsh_packet_offload __read_mostly = { + .type = htons(ETH_P_NSH), + .priority = 15, + .callbacks = { + .gso_segment = nsh_gso_segment, + }, +}; + +static int __init nsh_init_module(void) +{ + dev_add_offload(&nsh_packet_offload); + return 0; +} + +static void __exit nsh_cleanup_module(void) +{ + dev_remove_offload(&nsh_packet_offload); +} + +module_init(nsh_init_module); +module_exit(nsh_cleanup_module); + +MODULE_AUTHOR("Jiri Benc <jbenc@redhat.com>"); +MODULE_DESCRIPTION("NSH protocol"); +MODULE_LICENSE("GPL v2"); diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index e4610676299b..a54a556fcdb5 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -1337,6 +1337,7 @@ int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb, goto out; } + OVS_CB(skb)->acts_origlen = acts->orig_len; err = do_execute_actions(dp, skb, key, acts->actions, acts->actions_len); diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 08679ebb3068..d558e882ca0c 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -579,8 +579,8 @@ static struct nf_conn * ovs_ct_find_existing(struct net *net, const struct nf_conntrack_zone *zone, u8 l3num, struct sk_buff *skb, bool natted) { - struct nf_conntrack_l3proto *l3proto; - struct nf_conntrack_l4proto *l4proto; + const struct nf_conntrack_l3proto *l3proto; + const struct nf_conntrack_l4proto *l4proto; struct nf_conntrack_tuple tuple; struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; @@ -629,6 +629,34 @@ ovs_ct_find_existing(struct net *net, const struct nf_conntrack_zone *zone, return ct; } +static +struct nf_conn *ovs_ct_executed(struct net *net, + const struct sw_flow_key *key, + const struct ovs_conntrack_info *info, + struct sk_buff *skb, + bool *ct_executed) +{ + struct nf_conn *ct = NULL; + + /* If no ct, check if we have evidence that an existing conntrack entry + * might be found for this skb. This happens when we lose a skb->_nfct + * due to an upcall, or if the direction is being forced. If the + * connection was not confirmed, it is not cached and needs to be run + * through conntrack again. + */ + *ct_executed = (key->ct_state & OVS_CS_F_TRACKED) && + !(key->ct_state & OVS_CS_F_INVALID) && + (key->ct_zone == info->zone.id); + + if (*ct_executed || (!key->ct_state && info->force)) { + ct = ovs_ct_find_existing(net, &info->zone, info->family, skb, + !!(key->ct_state & + OVS_CS_F_NAT_MASK)); + } + + return ct; +} + /* Determine whether skb->_nfct is equal to the result of conntrack lookup. */ static bool skb_nfct_cached(struct net *net, const struct sw_flow_key *key, @@ -637,24 +665,17 @@ static bool skb_nfct_cached(struct net *net, { enum ip_conntrack_info ctinfo; struct nf_conn *ct; + bool ct_executed = true; ct = nf_ct_get(skb, &ctinfo); - /* If no ct, check if we have evidence that an existing conntrack entry - * might be found for this skb. This happens when we lose a skb->_nfct - * due to an upcall. If the connection was not confirmed, it is not - * cached and needs to be run through conntrack again. - */ - if (!ct && key->ct_state & OVS_CS_F_TRACKED && - !(key->ct_state & OVS_CS_F_INVALID) && - key->ct_zone == info->zone.id) { - ct = ovs_ct_find_existing(net, &info->zone, info->family, skb, - !!(key->ct_state - & OVS_CS_F_NAT_MASK)); - if (ct) - nf_ct_get(skb, &ctinfo); - } if (!ct) + ct = ovs_ct_executed(net, key, info, skb, &ct_executed); + + if (ct) + nf_ct_get(skb, &ctinfo); + else return false; + if (!net_eq(net, read_pnet(&ct->ct_net))) return false; if (!nf_ct_zone_equal_any(info->ct, nf_ct_zone(ct))) @@ -679,7 +700,7 @@ static bool skb_nfct_cached(struct net *net, return false; } - return true; + return ct_executed; } #ifdef CONFIG_NF_NAT_NEEDED @@ -1159,15 +1180,13 @@ static int parse_nat(const struct nlattr *attr, int type = nla_type(a); if (type > OVS_NAT_ATTR_MAX) { - OVS_NLERR(log, - "Unknown NAT attribute (type=%d, max=%d).\n", + OVS_NLERR(log, "Unknown NAT attribute (type=%d, max=%d)", type, OVS_NAT_ATTR_MAX); return -EINVAL; } if (nla_len(a) != ovs_nat_attr_lens[type][ip_vers]) { - OVS_NLERR(log, - "NAT attribute type %d has unexpected length (%d != %d).\n", + OVS_NLERR(log, "NAT attribute type %d has unexpected length (%d != %d)", type, nla_len(a), ovs_nat_attr_lens[type][ip_vers]); return -EINVAL; @@ -1177,9 +1196,7 @@ static int parse_nat(const struct nlattr *attr, case OVS_NAT_ATTR_SRC: case OVS_NAT_ATTR_DST: if (info->nat) { - OVS_NLERR(log, - "Only one type of NAT may be specified.\n" - ); + OVS_NLERR(log, "Only one type of NAT may be specified"); return -ERANGE; } info->nat |= OVS_CT_NAT; @@ -1224,13 +1241,13 @@ static int parse_nat(const struct nlattr *attr, break; default: - OVS_NLERR(log, "Unknown nat attribute (%d).\n", type); + OVS_NLERR(log, "Unknown nat attribute (%d)", type); return -EINVAL; } } if (rem > 0) { - OVS_NLERR(log, "NAT attribute has %d unknown bytes.\n", rem); + OVS_NLERR(log, "NAT attribute has %d unknown bytes", rem); return -EINVAL; } if (!info->nat) { @@ -1289,8 +1306,8 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, nla_for_each_nested(a, attr, rem) { int type = nla_type(a); - int maxlen = ovs_ct_attr_lens[type].maxlen; - int minlen = ovs_ct_attr_lens[type].minlen; + int maxlen; + int minlen; if (type > OVS_CT_ATTR_MAX) { OVS_NLERR(log, @@ -1298,6 +1315,9 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, type, OVS_CT_ATTR_MAX); return -EINVAL; } + + maxlen = ovs_ct_attr_lens[type].maxlen; + minlen = ovs_ct_attr_lens[type].minlen; if (nla_len(a) < minlen || nla_len(a) > maxlen) { OVS_NLERR(log, "Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)", diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 45fe8c8a884d..76cf273a56c7 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -335,8 +335,6 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, const struct dp_upcall_info *upcall_info, uint32_t cutlen) { - unsigned short gso_type = skb_shinfo(skb)->gso_type; - struct sw_flow_key later_key; struct sk_buff *segs, *nskb; int err; @@ -347,21 +345,9 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, if (segs == NULL) return -EINVAL; - if (gso_type & SKB_GSO_UDP) { - /* The initial flow key extracted by ovs_flow_key_extract() - * in this case is for a first fragment, so we need to - * properly mark later fragments. - */ - later_key = *key; - later_key.ip.frag = OVS_FRAG_TYPE_LATER; - } - /* Queue all of the segments. */ skb = segs; do { - if (gso_type & SKB_GSO_UDP && skb != segs) - key = &later_key; - err = queue_userspace_packet(dp, skb, key, upcall_info, cutlen); if (err) break; @@ -381,7 +367,7 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, } static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info, - unsigned int hdrlen) + unsigned int hdrlen, int actions_attrlen) { size_t size = NLMSG_ALIGN(sizeof(struct ovs_header)) + nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */ @@ -398,7 +384,7 @@ static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info, /* OVS_PACKET_ATTR_ACTIONS */ if (upcall_info->actions_len) - size += nla_total_size(upcall_info->actions_len); + size += nla_total_size(actions_attrlen); /* OVS_PACKET_ATTR_MRU */ if (upcall_info->mru) @@ -465,7 +451,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, else hlen = skb->len; - len = upcall_msg_size(upcall_info, hlen - cutlen); + len = upcall_msg_size(upcall_info, hlen - cutlen, + OVS_CB(skb)->acts_origlen); user_skb = genlmsg_new(len, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 5d8dcd88815f..480600649d0b 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -99,11 +99,13 @@ struct datapath { * when a packet is received by OVS. * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. + * @acts_origlen: The netlink size of the flow actions applied to this skb. * @cutlen: The number of bytes from the packet end to be removed. */ struct ovs_skb_cb { struct vport *input_vport; u16 mru; + u16 acts_origlen; u32 cutlen; }; #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb) diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 3f76cb765e5b..8c94cef25a72 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -72,8 +72,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, const struct sk_buff *skb) { struct flow_stats *stats; - int node = numa_node_id(); - int cpu = smp_processor_id(); + unsigned int cpu = smp_processor_id(); int len = skb->len + (skb_vlan_tag_present(skb) ? VLAN_HLEN : 0); stats = rcu_dereference(flow->stats[cpu]); @@ -108,7 +107,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, __GFP_THISNODE | __GFP_NOWARN | __GFP_NOMEMALLOC, - node); + numa_node_id()); if (likely(new_stats)) { new_stats->used = jiffies; new_stats->packet_count = 1; @@ -118,6 +117,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, rcu_assign_pointer(flow->stats[cpu], new_stats); + cpumask_set_cpu(cpu, &flow->cpu_used_mask); goto unlock; } } @@ -145,7 +145,7 @@ void ovs_flow_stats_get(const struct sw_flow *flow, memset(ovs_stats, 0, sizeof(*ovs_stats)); /* We open code this to make sure cpu 0 is always considered */ - for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, cpu_possible_mask)) { + for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, &flow->cpu_used_mask)) { struct flow_stats *stats = rcu_dereference_ovsl(flow->stats[cpu]); if (stats) { @@ -169,7 +169,7 @@ void ovs_flow_stats_clear(struct sw_flow *flow) int cpu; /* We open code this to make sure cpu 0 is always considered */ - for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, cpu_possible_mask)) { + for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, &flow->cpu_used_mask)) { struct flow_stats *stats = ovsl_dereference(flow->stats[cpu]); if (stats) { @@ -584,8 +584,7 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) key->ip.frag = OVS_FRAG_TYPE_LATER; return 0; } - if (nh->frag_off & htons(IP_MF) || - skb_shinfo(skb)->gso_type & SKB_GSO_UDP) + if (nh->frag_off & htons(IP_MF)) key->ip.frag = OVS_FRAG_TYPE_FIRST; else key->ip.frag = OVS_FRAG_TYPE_NONE; @@ -701,9 +700,6 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) if (key->ip.frag == OVS_FRAG_TYPE_LATER) return 0; - if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP) - key->ip.frag = OVS_FRAG_TYPE_FIRST; - /* Transport layer. */ if (key->ip.proto == NEXTHDR_TCP) { if (tcphdr_ok(skb)) { diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index a9bc1c875965..1875bba4f865 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -31,6 +31,7 @@ #include <linux/jiffies.h> #include <linux/time.h> #include <linux/flex_array.h> +#include <linux/cpumask.h> #include <net/inet_ecn.h> #include <net/ip_tunnels.h> #include <net/dst_metadata.h> @@ -219,6 +220,7 @@ struct sw_flow { */ struct sw_flow_key key; struct sw_flow_id id; + struct cpumask cpu_used_mask; struct sw_flow_mask *mask; struct sw_flow_actions __rcu *sf_acts; struct flow_stats __rcu *stats[]; /* One for each CPU. First one diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index f07d10ac35d8..e8eb427ce6d1 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -1255,7 +1255,7 @@ static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match, } if (!is_mask && ipv6_key->ipv6_label & htonl(0xFFF00000)) { - OVS_NLERR(log, "IPv6 flow label %x is out of range (max=%x).\n", + OVS_NLERR(log, "IPv6 flow label %x is out of range (max=%x)", ntohl(ipv6_key->ipv6_label), (1 << 20) - 1); return -EINVAL; } diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index ea7a8073fa02..80ea2a71852e 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -98,6 +98,8 @@ struct sw_flow *ovs_flow_alloc(void) RCU_INIT_POINTER(flow->stats[0], stats); + cpumask_set_cpu(0, &flow->cpu_used_mask); + return flow; err: kmem_cache_free(flow_cache, flow); @@ -141,7 +143,7 @@ static void flow_free(struct sw_flow *flow) if (flow->sf_acts) ovs_nla_free_flow_actions((struct sw_flow_actions __force *)flow->sf_acts); /* We open code this to make sure cpu 0 is always considered */ - for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, cpu_possible_mask)) + for (cpu = 0; cpu < nr_cpu_ids; cpu = cpumask_next(cpu, &flow->cpu_used_mask)) if (flow->stats[cpu]) kmem_cache_free(flow_stats_cache, (struct flow_stats __force *)flow->stats[cpu]); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index e3beb28203eb..c26172995511 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -177,8 +177,6 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u, #define BLK_PLUS_PRIV(sz_of_priv) \ (BLK_HDR_LEN + ALIGN((sz_of_priv), V3_ALIGNMENT)) -#define PGV_FROM_VMALLOC 1 - #define BLOCK_STATUS(x) ((x)->hdr.bh1.block_status) #define BLOCK_NUM_PKTS(x) ((x)->hdr.bh1.num_pkts) #define BLOCK_O2FP(x) ((x)->hdr.bh1.offset_to_first_pkt) @@ -214,6 +212,7 @@ static void prb_clear_rxhash(struct tpacket_kbdq_core *, static void prb_fill_vlan_info(struct tpacket_kbdq_core *, struct tpacket3_hdr *); static void packet_flush_mclist(struct sock *sk); +static void packet_pick_tx_queue(struct net_device *dev, struct sk_buff *skb); struct packet_skb_cb { union { @@ -260,6 +259,7 @@ static int packet_direct_xmit(struct sk_buff *skb) if (skb != orig_skb) goto drop; + packet_pick_tx_queue(dev, skb); txq = skb_get_tx_queue(dev, skb); local_bh_disable(); @@ -2189,6 +2189,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct timespec ts; __u32 ts_status; bool is_drop_n_account = false; + bool do_vnet = false; /* struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT. * We may add members to them until current aligned size without forcing @@ -2239,8 +2240,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, netoff = TPACKET_ALIGN(po->tp_hdrlen + (maclen < 16 ? 16 : maclen)) + po->tp_reserve; - if (po->has_vnet_hdr) + if (po->has_vnet_hdr) { netoff += sizeof(struct virtio_net_hdr); + do_vnet = true; + } macoff = netoff - maclen; } if (po->tp_version <= TPACKET_V2) { @@ -2257,8 +2260,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, skb_set_owner_r(copy_skb, sk); } snaplen = po->rx_ring.frame_size - macoff; - if ((int)snaplen < 0) + if ((int)snaplen < 0) { snaplen = 0; + do_vnet = false; + } } } else if (unlikely(macoff + snaplen > GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len)) { @@ -2271,6 +2276,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, if (unlikely((int)snaplen < 0)) { snaplen = 0; macoff = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len; + do_vnet = false; } } spin_lock(&sk->sk_receive_queue.lock); @@ -2296,7 +2302,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, } spin_unlock(&sk->sk_receive_queue.lock); - if (po->has_vnet_hdr) { + if (do_vnet) { if (virtio_net_hdr_from_skb(skb, h.raw + macoff - sizeof(struct virtio_net_hdr), vio_le(), true)) { @@ -2747,8 +2753,6 @@ tpacket_error: goto tpacket_error; } - packet_pick_tx_queue(dev, skb); - skb->destructor = tpacket_destruct_skb; __packet_set_status(po, ph, TP_STATUS_SENDING); packet_inc_pending(&po->tx_ring); @@ -2931,8 +2935,6 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) skb->priority = sk->sk_priority; skb->mark = sockc.mark; - packet_pick_tx_queue(dev, skb); - if (po->has_vnet_hdr) { err = virtio_net_hdr_to_skb(skb, &vnet_hdr, vio_le()); if (err) @@ -3702,14 +3704,19 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv if (optlen != sizeof(val)) return -EINVAL; - if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) - return -EBUSY; if (copy_from_user(&val, optval, sizeof(val))) return -EFAULT; if (val > INT_MAX) return -EINVAL; - po->tp_reserve = val; - return 0; + lock_sock(sk); + if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) { + ret = -EBUSY; + } else { + po->tp_reserve = val; + ret = 0; + } + release_sock(sk); + return ret; } case PACKET_LOSS: { @@ -4331,7 +4338,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u, register_prot_hook(sk); } spin_unlock(&po->bind_lock); - if (closing && (po->tp_version > TPACKET_V2)) { + if (pg_vec && (po->tp_version > TPACKET_V2)) { /* Because we don't support block-based V3 on tx-ring */ if (!tx_ring) prb_shutdown_retire_blk_timer(po, rb_queue); diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c index 45b3af3080d8..da754fc926e7 100644 --- a/net/phonet/pn_netlink.c +++ b/net/phonet/pn_netlink.c @@ -300,15 +300,15 @@ out: int __init phonet_netlink_register(void) { int err = __rtnl_register(PF_PHONET, RTM_NEWADDR, addr_doit, - NULL, NULL); + NULL, 0); if (err) return err; /* Further __rtnl_register() cannot fail */ - __rtnl_register(PF_PHONET, RTM_DELADDR, addr_doit, NULL, NULL); - __rtnl_register(PF_PHONET, RTM_GETADDR, NULL, getaddr_dumpit, NULL); - __rtnl_register(PF_PHONET, RTM_NEWROUTE, route_doit, NULL, NULL); - __rtnl_register(PF_PHONET, RTM_DELROUTE, route_doit, NULL, NULL); - __rtnl_register(PF_PHONET, RTM_GETROUTE, NULL, route_dumpit, NULL); + __rtnl_register(PF_PHONET, RTM_DELADDR, addr_doit, NULL, 0); + __rtnl_register(PF_PHONET, RTM_GETADDR, NULL, getaddr_dumpit, 0); + __rtnl_register(PF_PHONET, RTM_NEWROUTE, route_doit, NULL, 0); + __rtnl_register(PF_PHONET, RTM_DELROUTE, route_doit, NULL, 0); + __rtnl_register(PF_PHONET, RTM_GETROUTE, NULL, route_dumpit, 0); return 0; } diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c index 5586609afa27..c2f5c13550c0 100644 --- a/net/qrtr/qrtr.c +++ b/net/qrtr/qrtr.c @@ -1081,7 +1081,7 @@ static int __init qrtr_proto_init(void) return rc; } - rtnl_register(PF_QIPCRTR, RTM_NEWADDR, qrtr_addr_doit, NULL, NULL); + rtnl_register(PF_QIPCRTR, RTM_NEWADDR, qrtr_addr_doit, NULL, 0); return 0; } diff --git a/net/rds/bind.c b/net/rds/bind.c index 3a915bedb76c..75d43dc8e96b 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -40,7 +40,7 @@ static struct rhashtable bind_hash_table; -static struct rhashtable_params ht_parms = { +static const struct rhashtable_params ht_parms = { .nelem_hint = 768, .key_len = sizeof(u64), .key_offset = offsetof(struct rds_sock, rs_bound_key), diff --git a/net/rds/connection.c b/net/rds/connection.c index 50a3789ac23e..7ee2d5d68b78 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -151,6 +151,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, struct rds_transport *loop_trans; unsigned long flags; int ret, i; + int npaths = (trans->t_mp_capable ? RDS_MPATH_WORKERS : 1); rcu_read_lock(); conn = rds_conn_lookup(net, head, laddr, faddr, trans); @@ -172,6 +173,12 @@ static struct rds_connection *__rds_conn_create(struct net *net, conn = ERR_PTR(-ENOMEM); goto out; } + conn->c_path = kcalloc(npaths, sizeof(struct rds_conn_path), gfp); + if (!conn->c_path) { + kmem_cache_free(rds_conn_slab, conn); + conn = ERR_PTR(-ENOMEM); + goto out; + } INIT_HLIST_NODE(&conn->c_hash_node); conn->c_laddr = laddr; @@ -181,6 +188,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, ret = rds_cong_get_maps(conn); if (ret) { + kfree(conn->c_path); kmem_cache_free(rds_conn_slab, conn); conn = ERR_PTR(ret); goto out; @@ -207,13 +215,14 @@ static struct rds_connection *__rds_conn_create(struct net *net, conn->c_trans = trans; init_waitqueue_head(&conn->c_hs_waitq); - for (i = 0; i < RDS_MPATH_WORKERS; i++) { + for (i = 0; i < npaths; i++) { __rds_conn_path_init(conn, &conn->c_path[i], is_outgoing); conn->c_path[i].cp_index = i; } ret = trans->conn_alloc(conn, gfp); if (ret) { + kfree(conn->c_path); kmem_cache_free(rds_conn_slab, conn); conn = ERR_PTR(ret); goto out; @@ -236,6 +245,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, /* Creating passive conn */ if (parent->c_passive) { trans->conn_free(conn->c_path[0].cp_transport_data); + kfree(conn->c_path); kmem_cache_free(rds_conn_slab, conn); conn = parent->c_passive; } else { @@ -252,7 +262,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, struct rds_conn_path *cp; int i; - for (i = 0; i < RDS_MPATH_WORKERS; i++) { + for (i = 0; i < npaths; i++) { cp = &conn->c_path[i]; /* The ->conn_alloc invocation may have * allocated resource for all paths, so all @@ -261,6 +271,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, if (cp->cp_transport_data) trans->conn_free(cp->cp_transport_data); } + kfree(conn->c_path); kmem_cache_free(rds_conn_slab, conn); conn = found; } else { @@ -374,13 +385,13 @@ static void rds_conn_path_destroy(struct rds_conn_path *cp) if (!cp->cp_transport_data) return; - rds_conn_path_drop(cp); - flush_work(&cp->cp_down_w); - /* make sure lingering queued work won't try to ref the conn */ cancel_delayed_work_sync(&cp->cp_send_w); cancel_delayed_work_sync(&cp->cp_recv_w); + rds_conn_path_drop(cp, true); + flush_work(&cp->cp_down_w); + /* tear down queued messages */ list_for_each_entry_safe(rm, rtmp, &cp->cp_send_queue, @@ -407,6 +418,7 @@ void rds_conn_destroy(struct rds_connection *conn) unsigned long flags; int i; struct rds_conn_path *cp; + int npaths = (conn->c_trans->t_mp_capable ? RDS_MPATH_WORKERS : 1); rdsdebug("freeing conn %p for %pI4 -> " "%pI4\n", conn, &conn->c_laddr, @@ -420,7 +432,7 @@ void rds_conn_destroy(struct rds_connection *conn) synchronize_rcu(); /* shut the connection down */ - for (i = 0; i < RDS_MPATH_WORKERS; i++) { + for (i = 0; i < npaths; i++) { cp = &conn->c_path[i]; rds_conn_path_destroy(cp); BUG_ON(!list_empty(&cp->cp_retrans)); @@ -434,6 +446,7 @@ void rds_conn_destroy(struct rds_connection *conn) rds_cong_remove_conn(conn); put_net(conn->c_net); + kfree(conn->c_path); kmem_cache_free(rds_conn_slab, conn); spin_lock_irqsave(&rds_conn_lock, flags); @@ -464,8 +477,12 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, i++, head++) { hlist_for_each_entry_rcu(conn, head, c_hash_node) { struct rds_conn_path *cp; + int npaths; + + npaths = (conn->c_trans->t_mp_capable ? + RDS_MPATH_WORKERS : 1); - for (j = 0; j < RDS_MPATH_WORKERS; j++) { + for (j = 0; j < npaths; j++) { cp = &conn->c_path[j]; if (want_send) list = &cp->cp_send_queue; @@ -486,8 +503,6 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, } spin_unlock_irqrestore(&cp->cp_lock, flags); - if (!conn->c_trans->t_mp_capable) - break; } } } @@ -571,15 +586,16 @@ static void rds_walk_conn_path_info(struct socket *sock, unsigned int len, i++, head++) { hlist_for_each_entry_rcu(conn, head, c_hash_node) { struct rds_conn_path *cp; + int npaths; - for (j = 0; j < RDS_MPATH_WORKERS; j++) { + npaths = (conn->c_trans->t_mp_capable ? + RDS_MPATH_WORKERS : 1); + for (j = 0; j < npaths; j++) { cp = &conn->c_path[j]; /* XXX no cp_lock usage.. */ if (!visitor(cp, buffer)) continue; - if (!conn->c_trans->t_mp_capable) - break; } /* We copy as much as we can fit in the buffer, @@ -664,9 +680,13 @@ void rds_conn_exit(void) /* * Force a disconnect */ -void rds_conn_path_drop(struct rds_conn_path *cp) +void rds_conn_path_drop(struct rds_conn_path *cp, bool destroy) { atomic_set(&cp->cp_state, RDS_CONN_ERROR); + + if (!destroy && cp->cp_conn->c_destroy_in_prog) + return; + queue_work(rds_wq, &cp->cp_down_w); } EXPORT_SYMBOL_GPL(rds_conn_path_drop); @@ -674,7 +694,7 @@ EXPORT_SYMBOL_GPL(rds_conn_path_drop); void rds_conn_drop(struct rds_connection *conn) { WARN_ON(conn->c_trans->t_mp_capable); - rds_conn_path_drop(&conn->c_path[0]); + rds_conn_path_drop(&conn->c_path[0], false); } EXPORT_SYMBOL_GPL(rds_conn_drop); @@ -706,5 +726,5 @@ __rds_conn_path_error(struct rds_conn_path *cp, const char *fmt, ...) vprintk(fmt, ap); va_end(ap); - rds_conn_path_drop(cp); + rds_conn_path_drop(cp, false); } diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index e10624aa6959..9722bf839d9d 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -1015,8 +1015,10 @@ void rds_ib_recv_cqe_handler(struct rds_ib_connection *ic, if (rds_ib_ring_empty(&ic->i_recv_ring)) rds_ib_stats_inc(s_ib_rx_ring_empty); - if (rds_ib_ring_low(&ic->i_recv_ring)) + if (rds_ib_ring_low(&ic->i_recv_ring)) { rds_ib_recv_refill(conn, 0, GFP_NOWAIT); + rds_ib_stats_inc(s_ib_rx_refill_from_cq); + } } int rds_ib_recv_path(struct rds_conn_path *cp) @@ -1029,6 +1031,7 @@ int rds_ib_recv_path(struct rds_conn_path *cp) if (rds_conn_up(conn)) { rds_ib_attempt_ack(ic); rds_ib_recv_refill(conn, 0, GFP_KERNEL); + rds_ib_stats_inc(s_ib_rx_refill_from_thread); } return ret; diff --git a/net/rds/rds.h b/net/rds/rds.h index 516bcc89b46f..2e0315b159cb 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -154,7 +154,7 @@ struct rds_connection { struct list_head c_map_item; unsigned long c_map_queued; - struct rds_conn_path c_path[RDS_MPATH_WORKERS]; + struct rds_conn_path *c_path; wait_queue_head_t c_hs_waitq; /* handshake waitq */ u32 c_my_gen_num; @@ -700,7 +700,7 @@ struct rds_connection *rds_conn_create_outgoing(struct net *net, void rds_conn_shutdown(struct rds_conn_path *cpath); void rds_conn_destroy(struct rds_connection *conn); void rds_conn_drop(struct rds_connection *conn); -void rds_conn_path_drop(struct rds_conn_path *cpath); +void rds_conn_path_drop(struct rds_conn_path *cpath, bool destroy); void rds_conn_connect_if_down(struct rds_connection *conn); void rds_conn_path_connect_if_down(struct rds_conn_path *cp); void rds_for_each_conn_info(struct socket *sock, unsigned int len, diff --git a/net/rds/send.c b/net/rds/send.c index e81aa176f4e2..058a40743041 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -170,8 +170,8 @@ restart: * The acquire_in_xmit() check above ensures that only one * caller can increment c_send_gen at any time. */ - cp->cp_send_gen++; - send_gen = cp->cp_send_gen; + send_gen = READ_ONCE(cp->cp_send_gen) + 1; + WRITE_ONCE(cp->cp_send_gen, send_gen); /* * rds_conn_shutdown() sets the conn state and then tests RDS_IN_XMIT, @@ -273,7 +273,7 @@ restart: len = ntohl(rm->m_inc.i_hdr.h_len); if (cp->cp_unacked_packets == 0 || cp->cp_unacked_bytes < len) { - __set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); + set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); cp->cp_unacked_packets = rds_sysctl_max_unacked_packets; @@ -431,7 +431,7 @@ over_batch: smp_mb(); if ((test_bit(0, &conn->c_map_queued) || !list_empty(&cp->cp_send_queue)) && - send_gen == cp->cp_send_gen) { + send_gen == READ_ONCE(cp->cp_send_gen)) { rds_stats_inc(s_send_lock_queue_raced); if (batch_count < send_batch_count) goto restart; @@ -829,7 +829,7 @@ static int rds_send_queue_rm(struct rds_sock *rs, struct rds_connection *conn, * throughput hits a certain threshold. */ if (rs->rs_snd_bytes >= rds_sk_sndbuf(rs) / 2) - __set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); + set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); list_add_tail(&rm->m_sock_item, &rs->rs_send_queue); set_bit(RDS_MSG_ON_SOCK, &rm->m_flags); diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 431404dbdad1..6b7ee71f40c6 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -592,7 +592,7 @@ static void rds_tcp_sysctl_reset(struct net *net) continue; /* reconnect with new parameters */ - rds_conn_path_drop(tc->t_cpath); + rds_conn_path_drop(tc->t_cpath, false); } spin_unlock_irq(&rds_tcp_conn_lock); } diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index cbe08a1fa4c7..46f74dad0e16 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -69,14 +69,14 @@ void rds_tcp_state_change(struct sock *sk) if (!IS_CANONICAL(cp->cp_conn->c_laddr, cp->cp_conn->c_faddr) && rds_conn_path_transition(cp, RDS_CONN_CONNECTING, RDS_CONN_ERROR)) { - rds_conn_path_drop(cp); + rds_conn_path_drop(cp, false); } else { rds_connect_path_complete(cp, RDS_CONN_CONNECTING); } break; case TCP_CLOSE_WAIT: case TCP_CLOSE: - rds_conn_path_drop(cp); + rds_conn_path_drop(cp, false); default: break; } diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index 0d8616aa5bad..dc860d1bb608 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -157,7 +157,7 @@ out: "returned %d, " "disconnecting and reconnecting\n", &conn->c_faddr, cp->cp_index, ret); - rds_conn_path_drop(cp); + rds_conn_path_drop(cp, false); } } } diff --git a/net/rds/threads.c b/net/rds/threads.c index 2852bc1d37d4..f121daa402c8 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -78,7 +78,7 @@ void rds_connect_path_complete(struct rds_conn_path *cp, int curr) "current state is %d\n", __func__, atomic_read(&cp->cp_state)); - rds_conn_path_drop(cp); + rds_conn_path_drop(cp, false); return; } diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index a2ad4482376f..fb17552fd292 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -337,6 +337,75 @@ void rxrpc_kernel_end_call(struct socket *sock, struct rxrpc_call *call) EXPORT_SYMBOL(rxrpc_kernel_end_call); /** + * rxrpc_kernel_check_call - Check a call's state + * @sock: The socket the call is on + * @call: The call to check + * @_compl: Where to store the completion state + * @_abort_code: Where to store any abort code + * + * Allow a kernel service to query the state of a call and find out the manner + * of its termination if it has completed. Returns -EINPROGRESS if the call is + * still going, 0 if the call finished successfully, -ECONNABORTED if the call + * was aborted and an appropriate error if the call failed in some other way. + */ +int rxrpc_kernel_check_call(struct socket *sock, struct rxrpc_call *call, + enum rxrpc_call_completion *_compl, u32 *_abort_code) +{ + if (call->state != RXRPC_CALL_COMPLETE) + return -EINPROGRESS; + smp_rmb(); + *_compl = call->completion; + *_abort_code = call->abort_code; + return call->error; +} +EXPORT_SYMBOL(rxrpc_kernel_check_call); + +/** + * rxrpc_kernel_retry_call - Allow a kernel service to retry a call + * @sock: The socket the call is on + * @call: The call to retry + * @srx: The address of the peer to contact + * @key: The security context to use (defaults to socket setting) + * + * Allow a kernel service to try resending a client call that failed due to a + * network error to a new address. The Tx queue is maintained intact, thereby + * relieving the need to re-encrypt any request data that has already been + * buffered. + */ +int rxrpc_kernel_retry_call(struct socket *sock, struct rxrpc_call *call, + struct sockaddr_rxrpc *srx, struct key *key) +{ + struct rxrpc_conn_parameters cp; + struct rxrpc_sock *rx = rxrpc_sk(sock->sk); + int ret; + + _enter("%d{%d}", call->debug_id, atomic_read(&call->usage)); + + if (!key) + key = rx->key; + if (key && !key->payload.data[0]) + key = NULL; /* a no-security key */ + + memset(&cp, 0, sizeof(cp)); + cp.local = rx->local; + cp.key = key; + cp.security_level = 0; + cp.exclusive = false; + cp.service_id = srx->srx_service; + + mutex_lock(&call->user_mutex); + + ret = rxrpc_prepare_call_for_retry(rx, call); + if (ret == 0) + ret = rxrpc_retry_client_call(rx, call, &cp, srx, GFP_KERNEL); + + mutex_unlock(&call->user_mutex); + _leave(" = %d", ret); + return ret; +} +EXPORT_SYMBOL(rxrpc_kernel_retry_call); + +/** * rxrpc_kernel_new_call_notification - Get notifications of new calls * @sock: The socket to intercept received messages on * @notify_new_call: Function to be called when new calls appear @@ -591,13 +660,13 @@ static int rxrpc_getsockopt(struct socket *sock, int level, int optname, char __user *optval, int __user *_optlen) { int optlen; - + if (level != SOL_RXRPC) return -EOPNOTSUPP; if (get_user(optlen, _optlen)) return -EFAULT; - + switch (optname) { case RXRPC_SUPPORTED_CMSG: if (optlen < sizeof(int)) @@ -606,7 +675,7 @@ static int rxrpc_getsockopt(struct socket *sock, int level, int optname, put_user(sizeof(int), _optlen)) return -EFAULT; return 0; - + default: return -EOPNOTSUPP; } diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 69b97339ff9d..ea5600b747cc 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -15,7 +15,7 @@ #include <net/netns/generic.h> #include <net/sock.h> #include <net/af_rxrpc.h> -#include <rxrpc/packet.h> +#include "protocol.h" #if 0 #define CHECK_SLAB_OKAY(X) \ @@ -445,6 +445,7 @@ enum rxrpc_call_flag { RXRPC_CALL_EXPOSED, /* The call was exposed to the world */ RXRPC_CALL_RX_LAST, /* Received the last packet (at rxtx_top) */ RXRPC_CALL_TX_LAST, /* Last packet in Tx buffer (at rxtx_top) */ + RXRPC_CALL_TX_LASTQ, /* Last packet has been queued */ RXRPC_CALL_SEND_PING, /* A ping will need to be sent */ RXRPC_CALL_PINGING, /* Ping in process */ RXRPC_CALL_RETRANS_TIMEOUT, /* Retransmission due to timeout occurred */ @@ -482,18 +483,6 @@ enum rxrpc_call_state { }; /* - * Call completion condition (state == RXRPC_CALL_COMPLETE). - */ -enum rxrpc_call_completion { - RXRPC_CALL_SUCCEEDED, /* - Normal termination */ - RXRPC_CALL_REMOTELY_ABORTED, /* - call aborted by peer */ - RXRPC_CALL_LOCALLY_ABORTED, /* - call aborted locally on error or close */ - RXRPC_CALL_LOCAL_ERROR, /* - call failed due to local error */ - RXRPC_CALL_NETWORK_ERROR, /* - call terminated by network error */ - NR__RXRPC_CALL_COMPLETIONS -}; - -/* * Call Tx congestion management modes. */ enum rxrpc_congest_mode { @@ -687,9 +676,15 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, struct sockaddr_rxrpc *, unsigned long, s64, gfp_t); +int rxrpc_retry_client_call(struct rxrpc_sock *, + struct rxrpc_call *, + struct rxrpc_conn_parameters *, + struct sockaddr_rxrpc *, + gfp_t); void rxrpc_incoming_call(struct rxrpc_sock *, struct rxrpc_call *, struct sk_buff *); void rxrpc_release_call(struct rxrpc_sock *, struct rxrpc_call *); +int rxrpc_prepare_call_for_retry(struct rxrpc_sock *, struct rxrpc_call *); void rxrpc_release_calls_on_socket(struct rxrpc_sock *); bool __rxrpc_queue_call(struct rxrpc_call *); bool rxrpc_queue_call(struct rxrpc_call *); @@ -830,7 +825,6 @@ void rxrpc_process_connection(struct work_struct *); */ extern unsigned int rxrpc_connection_expiry; -int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *); struct rxrpc_connection *rxrpc_alloc_connection(gfp_t); struct rxrpc_connection *rxrpc_find_connection_rcu(struct rxrpc_local *, struct sk_buff *); @@ -894,7 +888,7 @@ extern struct key_type key_type_rxrpc_s; int rxrpc_request_key(struct rxrpc_sock *, char __user *, int); int rxrpc_server_keyring(struct rxrpc_sock *, char __user *, int); -int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, +int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time64_t, u32); /* @@ -1060,7 +1054,8 @@ static inline void rxrpc_sysctl_exit(void) {} /* * utils.c */ -int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *); +int rxrpc_extract_addr_from_skb(struct rxrpc_local *, struct sockaddr_rxrpc *, + struct sk_buff *); static inline bool before(u32 seq1, u32 seq2) { diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index dd30d74824b0..cbd1701e813a 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -223,6 +223,7 @@ void rxrpc_discard_prealloc(struct rxrpc_sock *rx) tail = b->call_backlog_tail; while (CIRC_CNT(head, tail, size) > 0) { struct rxrpc_call *call = b->call_backlog[tail]; + call->socket = rx; if (rx->discard_new_call) { _debug("discard %lx", call->user_call_ID); rx->discard_new_call(call, call->user_call_ID); @@ -276,7 +277,7 @@ static struct rxrpc_call *rxrpc_alloc_incoming_call(struct rxrpc_sock *rx, * anticipation - and to save on stack space. */ xpeer = b->peer_backlog[peer_tail]; - if (rxrpc_extract_addr_from_skb(&xpeer->srx, skb) < 0) + if (rxrpc_extract_addr_from_skb(local, &xpeer->srx, skb) < 0) return NULL; peer = rxrpc_lookup_incoming_peer(local, xpeer); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index d7809a0620b4..fcdd6555a820 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -269,11 +269,6 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, trace_rxrpc_call(call, rxrpc_call_connected, atomic_read(&call->usage), here, NULL); - spin_lock_bh(&call->conn->params.peer->lock); - hlist_add_head(&call->error_link, - &call->conn->params.peer->error_targets); - spin_unlock_bh(&call->conn->params.peer->lock); - rxrpc_start_call_timer(call); _net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id); @@ -304,6 +299,48 @@ error: } /* + * Retry a call to a new address. It is expected that the Tx queue of the call + * will contain data previously packaged for an old call. + */ +int rxrpc_retry_client_call(struct rxrpc_sock *rx, + struct rxrpc_call *call, + struct rxrpc_conn_parameters *cp, + struct sockaddr_rxrpc *srx, + gfp_t gfp) +{ + const void *here = __builtin_return_address(0); + int ret; + + /* Set up or get a connection record and set the protocol parameters, + * including channel number and call ID. + */ + ret = rxrpc_connect_call(call, cp, srx, gfp); + if (ret < 0) + goto error; + + trace_rxrpc_call(call, rxrpc_call_connected, atomic_read(&call->usage), + here, NULL); + + rxrpc_start_call_timer(call); + + _net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id); + + if (!test_and_set_bit(RXRPC_CALL_EV_RESEND, &call->events)) + rxrpc_queue_call(call); + + _leave(" = 0"); + return 0; + +error: + rxrpc_set_call_completion(call, RXRPC_CALL_LOCAL_ERROR, + RX_CALL_DEAD, ret); + trace_rxrpc_call(call, rxrpc_call_error, atomic_read(&call->usage), + here, ERR_PTR(ret)); + _leave(" = %d", ret); + return ret; +} + +/* * Set up an incoming call. call->conn points to the connection. * This is called in BH context and isn't allowed to fail. */ @@ -471,6 +508,61 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) } /* + * Prepare a kernel service call for retry. + */ +int rxrpc_prepare_call_for_retry(struct rxrpc_sock *rx, struct rxrpc_call *call) +{ + const void *here = __builtin_return_address(0); + int i; + u8 last = 0; + + _enter("{%d,%d}", call->debug_id, atomic_read(&call->usage)); + + trace_rxrpc_call(call, rxrpc_call_release, atomic_read(&call->usage), + here, (const void *)call->flags); + + ASSERTCMP(call->state, ==, RXRPC_CALL_COMPLETE); + ASSERTCMP(call->completion, !=, RXRPC_CALL_REMOTELY_ABORTED); + ASSERTCMP(call->completion, !=, RXRPC_CALL_LOCALLY_ABORTED); + ASSERT(list_empty(&call->recvmsg_link)); + + del_timer_sync(&call->timer); + + _debug("RELEASE CALL %p (%d CONN %p)", call, call->debug_id, call->conn); + + if (call->conn) + rxrpc_disconnect_call(call); + + if (rxrpc_is_service_call(call) || + !call->tx_phase || + call->tx_hard_ack != 0 || + call->rx_hard_ack != 0 || + call->rx_top != 0) + return -EINVAL; + + call->state = RXRPC_CALL_UNINITIALISED; + call->completion = RXRPC_CALL_SUCCEEDED; + call->call_id = 0; + call->cid = 0; + call->cong_cwnd = 0; + call->cong_extra = 0; + call->cong_ssthresh = 0; + call->cong_mode = 0; + call->cong_dup_acks = 0; + call->cong_cumul_acks = 0; + call->acks_lowest_nak = 0; + + for (i = 0; i < RXRPC_RXTX_BUFF_SIZE; i++) { + last |= call->rxtx_annotations[i]; + call->rxtx_annotations[i] &= RXRPC_TX_ANNO_LAST; + call->rxtx_annotations[i] |= RXRPC_TX_ANNO_RETRANS; + } + + _leave(" = 0"); + return 0; +} + +/* * release all the calls associated with a socket */ void rxrpc_release_calls_on_socket(struct rxrpc_sock *rx) diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index eb2157680399..5f9624bd311c 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -555,7 +555,10 @@ static void rxrpc_activate_one_channel(struct rxrpc_connection *conn, trace_rxrpc_client(conn, channel, rxrpc_client_chan_activate); write_lock_bh(&call->state_lock); - call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; + if (!test_bit(RXRPC_CALL_TX_LASTQ, &call->flags)) + call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; + else + call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY; write_unlock_bh(&call->state_lock); rxrpc_see_call(call); @@ -688,15 +691,23 @@ int rxrpc_connect_call(struct rxrpc_call *call, ret = rxrpc_get_client_conn(call, cp, srx, gfp); if (ret < 0) - return ret; + goto out; rxrpc_animate_client_conn(rxnet, call->conn); rxrpc_activate_channels(call->conn); ret = rxrpc_wait_for_channel(call, gfp); - if (ret < 0) + if (ret < 0) { rxrpc_disconnect_client_call(call); + goto out; + } + + spin_lock_bh(&call->conn->params.peer->lock); + hlist_add_head(&call->error_link, + &call->conn->params.peer->error_targets); + spin_unlock_bh(&call->conn->params.peer->lock); +out: _leave(" = %d", ret); return ret; } diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 929b50d5afe8..fe575798592f 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -72,7 +72,7 @@ struct rxrpc_connection *rxrpc_find_connection_rcu(struct rxrpc_local *local, _enter(",%x", sp->hdr.cid & RXRPC_CIDMASK); - if (rxrpc_extract_addr_from_skb(&srx, skb) < 0) + if (rxrpc_extract_addr_from_skb(local, &srx, skb) < 0) goto not_found; k.epoch = sp->hdr.epoch; diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index e60fcd2a4a02..f6fcdb3130a1 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -50,12 +50,11 @@ struct rxrpc_connection *rxrpc_find_service_conn_rcu(struct rxrpc_peer *peer, else if (conn->proto.index_key > k.index_key) p = rcu_dereference_raw(p->rb_right); else - goto done; + break; conn = NULL; } } while (need_seqretry(&peer->service_conn_lock, seq)); -done: done_seqretry(&peer->service_conn_lock, seq); _leave(" = %d", conn ? conn->debug_id : -1); return conn; diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c index 54369225766e..e7f6b8823eb6 100644 --- a/net/rxrpc/key.c +++ b/net/rxrpc/key.c @@ -92,6 +92,7 @@ static int rxrpc_preparse_xdr_rxkad(struct key_preparsed_payload *prep, const __be32 *xdr, unsigned int toklen) { struct rxrpc_key_token *token, **pptoken; + time64_t expiry; size_t plen; u32 tktlen; @@ -158,8 +159,9 @@ static int rxrpc_preparse_xdr_rxkad(struct key_preparsed_payload *prep, pptoken = &(*pptoken)->next) continue; *pptoken = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; + expiry = rxrpc_u32_to_time64(token->kad->expiry); + if (expiry < prep->expiry) + prep->expiry = expiry; _leave(" = 0"); return 0; @@ -433,6 +435,7 @@ static int rxrpc_preparse_xdr_rxk5(struct key_preparsed_payload *prep, struct rxrpc_key_token *token, **pptoken; struct rxk5_key *rxk5; const __be32 *end_xdr = xdr + (toklen >> 2); + time64_t expiry; int ret; _enter(",{%x,%x,%x,%x},%u", @@ -533,8 +536,9 @@ static int rxrpc_preparse_xdr_rxk5(struct key_preparsed_payload *prep, pptoken = &(*pptoken)->next) continue; *pptoken = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; + expiry = rxrpc_u32_to_time64(token->k5->endtime); + if (expiry < prep->expiry) + prep->expiry = expiry; _leave(" = 0"); return 0; @@ -691,6 +695,7 @@ static int rxrpc_preparse(struct key_preparsed_payload *prep) { const struct rxrpc_key_data_v1 *v1; struct rxrpc_key_token *token, **pp; + time64_t expiry; size_t plen; u32 kver; int ret; @@ -777,8 +782,9 @@ static int rxrpc_preparse(struct key_preparsed_payload *prep) while (*pp) pp = &(*pp)->next; *pp = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; + expiry = rxrpc_u32_to_time64(token->kad->expiry); + if (expiry < prep->expiry) + prep->expiry = expiry; token = NULL; ret = 0; @@ -955,7 +961,7 @@ int rxrpc_server_keyring(struct rxrpc_sock *rx, char __user *optval, */ int rxrpc_get_server_data_key(struct rxrpc_connection *conn, const void *session_key, - time_t expiry, + time64_t expiry, u32 kvno) { const struct cred *cred = current_cred(); @@ -982,7 +988,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *conn, data.kver = 1; data.v1.security_index = RXRPC_SECURITY_RXKAD; data.v1.ticket_length = 0; - data.v1.expiry = expiry; + data.v1.expiry = rxrpc_time64_to_u32(expiry); data.v1.kvno = 0; memcpy(&data.v1.session_key, session_key, sizeof(data.v1.session_key)); diff --git a/net/rxrpc/local_event.c b/net/rxrpc/local_event.c index 540d3955c1bc..93b5d910b4a1 100644 --- a/net/rxrpc/local_event.c +++ b/net/rxrpc/local_event.c @@ -39,7 +39,7 @@ static void rxrpc_send_version_request(struct rxrpc_local *local, _enter(""); - if (rxrpc_extract_addr_from_skb(&srx, skb) < 0) + if (rxrpc_extract_addr_from_skb(local, &srx, skb) < 0) return; msg.msg_name = &srx.transport; diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 5bd2d0fa4a03..71e6f713fbe7 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -444,7 +444,7 @@ void rxrpc_reject_packets(struct rxrpc_local *local) rxrpc_see_skb(skb, rxrpc_skb_rx_seen); sp = rxrpc_skb(skb); - if (rxrpc_extract_addr_from_skb(&srx, skb) == 0) { + if (rxrpc_extract_addr_from_skb(local, &srx, skb) == 0) { msg.msg_namelen = srx.transport_len; code = htonl(skb->priority); diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 1ed9c0c2e94f..7f749505e699 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -37,6 +37,7 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, memset(&srx, 0, sizeof(srx)); srx.transport_type = local->srx.transport_type; + srx.transport_len = local->srx.transport_len; srx.transport.family = local->srx.transport.family; /* Can we see an ICMP4 packet on an ICMP6 listening socket? and vice @@ -45,7 +46,6 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, switch (srx.transport.family) { case AF_INET: srx.transport.sin.sin_port = serr->port; - srx.transport_len = sizeof(struct sockaddr_in); switch (serr->ee.ee_origin) { case SO_EE_ORIGIN_ICMP: _net("Rx ICMP"); @@ -69,7 +69,6 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, #ifdef CONFIG_AF_RXRPC_IPV6 case AF_INET6: srx.transport.sin6.sin6_port = serr->port; - srx.transport_len = sizeof(struct sockaddr_in6); switch (serr->ee.ee_origin) { case SO_EE_ORIGIN_ICMP6: _net("Rx ICMP6"); @@ -79,6 +78,9 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, break; case SO_EE_ORIGIN_ICMP: _net("Rx ICMP on v6 sock"); + srx.transport.sin6.sin6_addr.s6_addr32[0] = 0; + srx.transport.sin6.sin6_addr.s6_addr32[1] = 0; + srx.transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); memcpy(srx.transport.sin6.sin6_addr.s6_addr + 12, skb_network_header(skb) + serr->addr_offset, sizeof(struct in_addr)); diff --git a/net/rxrpc/protocol.h b/net/rxrpc/protocol.h new file mode 100644 index 000000000000..4bddcf3face3 --- /dev/null +++ b/net/rxrpc/protocol.h @@ -0,0 +1,190 @@ +/* packet.h: Rx packet layout and definitions + * + * Copyright (C) 2002, 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#ifndef _LINUX_RXRPC_PACKET_H +#define _LINUX_RXRPC_PACKET_H + +typedef u32 rxrpc_seq_t; /* Rx message sequence number */ +typedef u32 rxrpc_serial_t; /* Rx message serial number */ +typedef __be32 rxrpc_seq_net_t; /* on-the-wire Rx message sequence number */ +typedef __be32 rxrpc_serial_net_t; /* on-the-wire Rx message serial number */ + +/*****************************************************************************/ +/* + * on-the-wire Rx packet header + * - all multibyte fields should be in network byte order + */ +struct rxrpc_wire_header { + __be32 epoch; /* client boot timestamp */ +#define RXRPC_RANDOM_EPOCH 0x80000000 /* Random if set, date-based if not */ + + __be32 cid; /* connection and channel ID */ +#define RXRPC_MAXCALLS 4 /* max active calls per conn */ +#define RXRPC_CHANNELMASK (RXRPC_MAXCALLS-1) /* mask for channel ID */ +#define RXRPC_CIDMASK (~RXRPC_CHANNELMASK) /* mask for connection ID */ +#define RXRPC_CIDSHIFT ilog2(RXRPC_MAXCALLS) /* shift for connection ID */ +#define RXRPC_CID_INC (1 << RXRPC_CIDSHIFT) /* connection ID increment */ + + __be32 callNumber; /* call ID (0 for connection-level packets) */ + __be32 seq; /* sequence number of pkt in call stream */ + __be32 serial; /* serial number of pkt sent to network */ + + uint8_t type; /* packet type */ +#define RXRPC_PACKET_TYPE_DATA 1 /* data */ +#define RXRPC_PACKET_TYPE_ACK 2 /* ACK */ +#define RXRPC_PACKET_TYPE_BUSY 3 /* call reject */ +#define RXRPC_PACKET_TYPE_ABORT 4 /* call/connection abort */ +#define RXRPC_PACKET_TYPE_ACKALL 5 /* ACK all outstanding packets on call */ +#define RXRPC_PACKET_TYPE_CHALLENGE 6 /* connection security challenge (SRVR->CLNT) */ +#define RXRPC_PACKET_TYPE_RESPONSE 7 /* connection secutity response (CLNT->SRVR) */ +#define RXRPC_PACKET_TYPE_DEBUG 8 /* debug info request */ +#define RXRPC_PACKET_TYPE_VERSION 13 /* version string request */ +#define RXRPC_N_PACKET_TYPES 14 /* number of packet types (incl type 0) */ + + uint8_t flags; /* packet flags */ +#define RXRPC_CLIENT_INITIATED 0x01 /* signifies a packet generated by a client */ +#define RXRPC_REQUEST_ACK 0x02 /* request an unconditional ACK of this packet */ +#define RXRPC_LAST_PACKET 0x04 /* the last packet from this side for this call */ +#define RXRPC_MORE_PACKETS 0x08 /* more packets to come */ +#define RXRPC_JUMBO_PACKET 0x20 /* [DATA] this is a jumbo packet */ +#define RXRPC_SLOW_START_OK 0x20 /* [ACK] slow start supported */ + + uint8_t userStatus; /* app-layer defined status */ +#define RXRPC_USERSTATUS_SERVICE_UPGRADE 0x01 /* AuriStor service upgrade request */ + + uint8_t securityIndex; /* security protocol ID */ + union { + __be16 _rsvd; /* reserved */ + __be16 cksum; /* kerberos security checksum */ + }; + __be16 serviceId; /* service ID */ + +} __packed; + +#define RXRPC_SUPPORTED_PACKET_TYPES ( \ + (1 << RXRPC_PACKET_TYPE_DATA) | \ + (1 << RXRPC_PACKET_TYPE_ACK) | \ + (1 << RXRPC_PACKET_TYPE_BUSY) | \ + (1 << RXRPC_PACKET_TYPE_ABORT) | \ + (1 << RXRPC_PACKET_TYPE_ACKALL) | \ + (1 << RXRPC_PACKET_TYPE_CHALLENGE) | \ + (1 << RXRPC_PACKET_TYPE_RESPONSE) | \ + /*(1 << RXRPC_PACKET_TYPE_DEBUG) | */ \ + (1 << RXRPC_PACKET_TYPE_VERSION)) + +/*****************************************************************************/ +/* + * jumbo packet secondary header + * - can be mapped to read header by: + * - new_serial = serial + 1 + * - new_seq = seq + 1 + * - new_flags = j_flags + * - new__rsvd = j__rsvd + * - duplicating all other fields + */ +struct rxrpc_jumbo_header { + uint8_t flags; /* packet flags (as per rxrpc_header) */ + uint8_t pad; + union { + __be16 _rsvd; /* reserved */ + __be16 cksum; /* kerberos security checksum */ + }; +}; + +#define RXRPC_JUMBO_DATALEN 1412 /* non-terminal jumbo packet data length */ +#define RXRPC_JUMBO_SUBPKTLEN (RXRPC_JUMBO_DATALEN + sizeof(struct rxrpc_jumbo_header)) + +/*****************************************************************************/ +/* + * on-the-wire Rx ACK packet data payload + * - all multibyte fields should be in network byte order + */ +struct rxrpc_ackpacket { + __be16 bufferSpace; /* number of packet buffers available */ + __be16 maxSkew; /* diff between serno being ACK'd and highest serial no + * received */ + __be32 firstPacket; /* sequence no of first ACK'd packet in attached list */ + __be32 previousPacket; /* sequence no of previous packet received */ + __be32 serial; /* serial no of packet that prompted this ACK */ + + uint8_t reason; /* reason for ACK */ +#define RXRPC_ACK_REQUESTED 1 /* ACK was requested on packet */ +#define RXRPC_ACK_DUPLICATE 2 /* duplicate packet received */ +#define RXRPC_ACK_OUT_OF_SEQUENCE 3 /* out of sequence packet received */ +#define RXRPC_ACK_EXCEEDS_WINDOW 4 /* packet received beyond end of ACK window */ +#define RXRPC_ACK_NOSPACE 5 /* packet discarded due to lack of buffer space */ +#define RXRPC_ACK_PING 6 /* keep alive ACK */ +#define RXRPC_ACK_PING_RESPONSE 7 /* response to RXRPC_ACK_PING */ +#define RXRPC_ACK_DELAY 8 /* nothing happened since received packet */ +#define RXRPC_ACK_IDLE 9 /* ACK due to fully received ACK window */ +#define RXRPC_ACK__INVALID 10 /* Representation of invalid ACK reason */ + + uint8_t nAcks; /* number of ACKs */ +#define RXRPC_MAXACKS 255 + + uint8_t acks[0]; /* list of ACK/NAKs */ +#define RXRPC_ACK_TYPE_NACK 0 +#define RXRPC_ACK_TYPE_ACK 1 + +} __packed; + +/* Some ACKs refer to specific packets and some are general and can be updated. */ +#define RXRPC_ACK_UPDATEABLE ((1 << RXRPC_ACK_REQUESTED) | \ + (1 << RXRPC_ACK_PING_RESPONSE) | \ + (1 << RXRPC_ACK_DELAY) | \ + (1 << RXRPC_ACK_IDLE)) + + +/* + * ACK packets can have a further piece of information tagged on the end + */ +struct rxrpc_ackinfo { + __be32 rxMTU; /* maximum Rx MTU size (bytes) [AFS 3.3] */ + __be32 maxMTU; /* maximum interface MTU size (bytes) [AFS 3.3] */ + __be32 rwind; /* Rx window size (packets) [AFS 3.4] */ + __be32 jumbo_max; /* max packets to stick into a jumbo packet [AFS 3.5] */ +}; + +/*****************************************************************************/ +/* + * Kerberos security type-2 challenge packet + */ +struct rxkad_challenge { + __be32 version; /* version of this challenge type */ + __be32 nonce; /* encrypted random number */ + __be32 min_level; /* minimum security level */ + __be32 __padding; /* padding to 8-byte boundary */ +} __packed; + +/*****************************************************************************/ +/* + * Kerberos security type-2 response packet + */ +struct rxkad_response { + __be32 version; /* version of this response type */ + __be32 __pad; + + /* encrypted bit of the response */ + struct { + __be32 epoch; /* current epoch */ + __be32 cid; /* parent connection ID */ + __be32 checksum; /* checksum */ + __be32 securityIndex; /* security type */ + __be32 call_id[4]; /* encrypted call IDs */ + __be32 inc_nonce; /* challenge nonce + 1 */ + __be32 level; /* desired level */ + } encrypted; + + __be32 kvno; /* Kerberos key version number */ + __be32 ticket_len; /* Kerberos ticket length */ +} __packed; + +#endif /* _LINUX_RXRPC_PACKET_H */ diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 46d1a1f0b55b..c38b3a1de56c 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -634,8 +634,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) challenge.min_level = htonl(0); challenge.__padding = 0; - msg.msg_name = &conn->params.peer->srx.transport.sin; - msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin); + msg.msg_name = &conn->params.peer->srx.transport; + msg.msg_namelen = conn->params.peer->srx.transport_len; msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -689,8 +689,8 @@ static int rxkad_send_response(struct rxrpc_connection *conn, _enter(""); - msg.msg_name = &conn->params.peer->srx.transport.sin; - msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin); + msg.msg_name = &conn->params.peer->srx.transport; + msg.msg_namelen = conn->params.peer->srx.transport_len; msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -854,7 +854,7 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn, struct sk_buff *skb, void *ticket, size_t ticket_len, struct rxrpc_crypt *_session_key, - time_t *_expiry, + time64_t *_expiry, u32 *_abort_code) { struct skcipher_request *req; @@ -864,7 +864,7 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn, struct in_addr addr; unsigned int life; const char *eproto; - time_t issue, now; + time64_t issue, now; bool little_endian; int ret; u32 abort_code; @@ -960,15 +960,15 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn, if (little_endian) { __le32 stamp; memcpy(&stamp, p, 4); - issue = le32_to_cpu(stamp); + issue = rxrpc_u32_to_time64(le32_to_cpu(stamp)); } else { __be32 stamp; memcpy(&stamp, p, 4); - issue = be32_to_cpu(stamp); + issue = rxrpc_u32_to_time64(be32_to_cpu(stamp)); } p += 4; - now = get_seconds(); - _debug("KIV ISSUE: %lx [%lx]", issue, now); + now = ktime_get_real_seconds(); + _debug("KIV ISSUE: %llx [%llx]", issue, now); /* check the ticket is in date */ if (issue > now) { @@ -1053,7 +1053,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_crypt session_key; const char *eproto; - time_t expiry; + time64_t expiry; void *ticket; u32 abort_code, version, kvno, ticket_len, level; __be32 csum; diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index b0d2cda6ec0a..9ea6f972767e 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -61,7 +61,7 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx, call->cong_cwnd + call->cong_extra)) break; if (call->state >= RXRPC_CALL_COMPLETE) { - ret = -call->error; + ret = call->error; break; } if (signal_pending(current)) { @@ -101,11 +101,23 @@ static inline void rxrpc_instant_resend(struct rxrpc_call *call, int ix) } /* + * Notify the owner of the call that the transmit phase is ended and the last + * packet has been queued. + */ +static void rxrpc_notify_end_tx(struct rxrpc_sock *rx, struct rxrpc_call *call, + rxrpc_notify_end_tx_t notify_end_tx) +{ + if (notify_end_tx) + notify_end_tx(&rx->sk, call, call->user_call_ID); +} + +/* * Queue a DATA packet for transmission, set the resend timeout and send the * packet immediately */ -static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, - bool last) +static void rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call, + struct sk_buff *skb, bool last, + rxrpc_notify_end_tx_t notify_end_tx) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); rxrpc_seq_t seq = sp->hdr.seq; @@ -116,8 +128,10 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, ASSERTCMP(seq, ==, call->tx_top + 1); - if (last) + if (last) { annotation |= RXRPC_TX_ANNO_LAST; + set_bit(RXRPC_CALL_TX_LASTQ, &call->flags); + } /* We have to set the timestamp before queueing as the retransmit * algorithm can see the packet as soon as we queue it. @@ -141,6 +155,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, switch (call->state) { case RXRPC_CALL_CLIENT_SEND_REQUEST: call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY; + rxrpc_notify_end_tx(rx, call, notify_end_tx); break; case RXRPC_CALL_SERVER_ACK_REQUEST: call->state = RXRPC_CALL_SERVER_SEND_REPLY; @@ -153,6 +168,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, break; case RXRPC_CALL_SERVER_SEND_REPLY: call->state = RXRPC_CALL_SERVER_AWAIT_ACK; + rxrpc_notify_end_tx(rx, call, notify_end_tx); break; default: break; @@ -189,7 +205,8 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, */ static int rxrpc_send_data(struct rxrpc_sock *rx, struct rxrpc_call *call, - struct msghdr *msg, size_t len) + struct msghdr *msg, size_t len, + rxrpc_notify_end_tx_t notify_end_tx) { struct rxrpc_skb_priv *sp; struct sk_buff *skb; @@ -311,11 +328,6 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, call->tx_total_len -= copy; } - /* check for the far side aborting the call or a network error - * occurring */ - if (call->state == RXRPC_CALL_COMPLETE) - goto call_terminated; - /* add the packet to the send queue if it's now full */ if (sp->remain <= 0 || (msg_data_left(msg) == 0 && !more)) { @@ -350,9 +362,21 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, if (ret < 0) goto out; - rxrpc_queue_packet(call, skb, !msg_data_left(msg) && !more); + rxrpc_queue_packet(rx, call, skb, + !msg_data_left(msg) && !more, + notify_end_tx); skb = NULL; } + + /* Check for the far side aborting the call or a network error + * occurring. If this happens, save any packet that was under + * construction so that in the case of a network error, the + * call can be retried or redirected. + */ + if (call->state == RXRPC_CALL_COMPLETE) { + ret = call->error; + goto out; + } } while (msg_data_left(msg) > 0); success: @@ -362,11 +386,6 @@ out: _leave(" = %d", ret); return ret; -call_terminated: - rxrpc_free_skb(skb, rxrpc_skb_tx_freed); - _leave(" = %d", -call->error); - return -call->error; - maybe_error: if (copied) goto success; @@ -611,7 +630,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) /* Reply phase not begun or not complete for service call. */ ret = -EPROTO; } else { - ret = rxrpc_send_data(rx, call, msg, len); + ret = rxrpc_send_data(rx, call, msg, len, NULL); } mutex_unlock(&call->user_mutex); @@ -631,6 +650,7 @@ error_release_sock: * @call: The call to send data through * @msg: The data to send * @len: The amount of data to send + * @notify_end_tx: Notification that the last packet is queued. * * Allow a kernel service to send data on a call. The call must be in an state * appropriate to sending data. No control data should be supplied in @msg, @@ -638,7 +658,8 @@ error_release_sock: * more data to come, otherwise this data will end the transmission phase. */ int rxrpc_kernel_send_data(struct socket *sock, struct rxrpc_call *call, - struct msghdr *msg, size_t len) + struct msghdr *msg, size_t len, + rxrpc_notify_end_tx_t notify_end_tx) { int ret; @@ -656,11 +677,12 @@ int rxrpc_kernel_send_data(struct socket *sock, struct rxrpc_call *call, case RXRPC_CALL_CLIENT_SEND_REQUEST: case RXRPC_CALL_SERVER_ACK_REQUEST: case RXRPC_CALL_SERVER_SEND_REPLY: - ret = rxrpc_send_data(rxrpc_sk(sock->sk), call, msg, len); + ret = rxrpc_send_data(rxrpc_sk(sock->sk), call, msg, len, + notify_end_tx); break; case RXRPC_CALL_COMPLETE: read_lock_bh(&call->state_lock); - ret = -call->error; + ret = call->error; read_unlock_bh(&call->state_lock); break; default: diff --git a/net/rxrpc/utils.c b/net/rxrpc/utils.c index ff7af71c4b49..e801171fa351 100644 --- a/net/rxrpc/utils.c +++ b/net/rxrpc/utils.c @@ -17,17 +17,28 @@ /* * Fill out a peer address from a socket buffer containing a packet. */ -int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *srx, struct sk_buff *skb) +int rxrpc_extract_addr_from_skb(struct rxrpc_local *local, + struct sockaddr_rxrpc *srx, + struct sk_buff *skb) { memset(srx, 0, sizeof(*srx)); switch (ntohs(skb->protocol)) { case ETH_P_IP: - srx->transport_type = SOCK_DGRAM; - srx->transport_len = sizeof(srx->transport.sin); - srx->transport.sin.sin_family = AF_INET; - srx->transport.sin.sin_port = udp_hdr(skb)->source; - srx->transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + if (local->srx.transport.family == AF_INET6) { + srx->transport_type = SOCK_DGRAM; + srx->transport_len = sizeof(srx->transport.sin6); + srx->transport.sin6.sin6_family = AF_INET6; + srx->transport.sin6.sin6_port = udp_hdr(skb)->source; + srx->transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); + srx->transport.sin6.sin6_addr.s6_addr32[3] = ip_hdr(skb)->saddr; + } else { + srx->transport_type = SOCK_DGRAM; + srx->transport_len = sizeof(srx->transport.sin); + srx->transport.sin.sin_family = AF_INET; + srx->transport.sin.sin_port = udp_hdr(skb)->source; + srx->transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + } return 0; #ifdef CONFIG_AF_RXRPC_IPV6 diff --git a/net/sched/act_api.c b/net/sched/act_api.c index aed6cf2e9fd8..a306974e2fb4 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -70,11 +70,11 @@ static void free_tcf(struct rcu_head *head) kfree(p); } -static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *p) +static void tcf_idr_remove(struct tcf_idrinfo *idrinfo, struct tc_action *p) { - spin_lock_bh(&hinfo->lock); - hlist_del(&p->tcfa_head); - spin_unlock_bh(&hinfo->lock); + spin_lock_bh(&idrinfo->lock); + idr_remove_ext(&idrinfo->action_idr, p->tcfa_index); + spin_unlock_bh(&idrinfo->lock); gen_kill_estimator(&p->tcfa_rate_est); /* * gen_estimator est_timer() might access p->tcfa_lock @@ -83,7 +83,7 @@ static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *p) call_rcu(&p->tcfa_rcu, free_tcf); } -int __tcf_hash_release(struct tc_action *p, bool bind, bool strict) +int __tcf_idr_release(struct tc_action *p, bool bind, bool strict) { int ret = 0; @@ -97,55 +97,64 @@ int __tcf_hash_release(struct tc_action *p, bool bind, bool strict) if (p->tcfa_bindcnt <= 0 && p->tcfa_refcnt <= 0) { if (p->ops->cleanup) p->ops->cleanup(p, bind); - tcf_hash_destroy(p->hinfo, p); + tcf_idr_remove(p->idrinfo, p); ret = ACT_P_DELETED; } } return ret; } -EXPORT_SYMBOL(__tcf_hash_release); +EXPORT_SYMBOL(__tcf_idr_release); -static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, +static int tcf_dump_walker(struct tcf_idrinfo *idrinfo, struct sk_buff *skb, struct netlink_callback *cb) { - int err = 0, index = -1, i = 0, s_i = 0, n_i = 0; + int err = 0, index = -1, s_i = 0, n_i = 0; + u32 act_flags = cb->args[2]; + unsigned long jiffy_since = cb->args[3]; struct nlattr *nest; + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + unsigned long id = 1; - spin_lock_bh(&hinfo->lock); + spin_lock_bh(&idrinfo->lock); s_i = cb->args[0]; - for (i = 0; i < (hinfo->hmask + 1); i++) { - struct hlist_head *head; - struct tc_action *p; - - head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; - - hlist_for_each_entry_rcu(p, head, tcfa_head) { - index++; - if (index < s_i) - continue; - - nest = nla_nest_start(skb, n_i); - if (nest == NULL) - goto nla_put_failure; - err = tcf_action_dump_1(skb, p, 0, 0); - if (err < 0) { - index--; - nlmsg_trim(skb, nest); - goto done; - } - nla_nest_end(skb, nest); - n_i++; - if (n_i >= TCA_ACT_MAX_PRIO) - goto done; + idr_for_each_entry_ext(idr, p, id) { + index++; + if (index < s_i) + continue; + + if (jiffy_since && + time_after(jiffy_since, + (unsigned long)p->tcfa_tm.lastuse)) + continue; + + nest = nla_nest_start(skb, n_i); + if (!nest) + goto nla_put_failure; + err = tcf_action_dump_1(skb, p, 0, 0); + if (err < 0) { + index--; + nlmsg_trim(skb, nest); + goto done; } + nla_nest_end(skb, nest); + n_i++; + if (!(act_flags & TCA_FLAG_LARGE_DUMP_ON) && + n_i >= TCA_ACT_MAX_PRIO) + goto done; } done: - spin_unlock_bh(&hinfo->lock); - if (n_i) - cb->args[0] += n_i; + if (index >= 0) + cb->args[0] = index + 1; + + spin_unlock_bh(&idrinfo->lock); + if (n_i) { + if (act_flags & TCA_FLAG_LARGE_DUMP_ON) + cb->args[1] = n_i; + } return n_i; nla_put_failure: @@ -153,31 +162,29 @@ nla_put_failure: goto done; } -static int tcf_del_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, +static int tcf_del_walker(struct tcf_idrinfo *idrinfo, struct sk_buff *skb, const struct tc_action_ops *ops) { struct nlattr *nest; - int i = 0, n_i = 0; + int n_i = 0; int ret = -EINVAL; + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + unsigned long id = 1; nest = nla_nest_start(skb, 0); if (nest == NULL) goto nla_put_failure; if (nla_put_string(skb, TCA_KIND, ops->kind)) goto nla_put_failure; - for (i = 0; i < (hinfo->hmask + 1); i++) { - struct hlist_head *head; - struct hlist_node *n; - struct tc_action *p; - - head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; - hlist_for_each_entry_safe(p, n, head, tcfa_head) { - ret = __tcf_hash_release(p, false, true); - if (ret == ACT_P_DELETED) { - module_put(p->ops->owner); - n_i++; - } else if (ret < 0) - goto nla_put_failure; + + idr_for_each_entry_ext(idr, p, id) { + ret = __tcf_idr_release(p, false, true); + if (ret == ACT_P_DELETED) { + module_put(p->ops->owner); + n_i++; + } else if (ret < 0) { + goto nla_put_failure; } } if (nla_put_u32(skb, TCA_FCNT, n_i)) @@ -194,12 +201,12 @@ int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, struct netlink_callback *cb, int type, const struct tc_action_ops *ops) { - struct tcf_hashinfo *hinfo = tn->hinfo; + struct tcf_idrinfo *idrinfo = tn->idrinfo; if (type == RTM_DELACTION) { - return tcf_del_walker(hinfo, skb, ops); + return tcf_del_walker(idrinfo, skb, ops); } else if (type == RTM_GETACTION) { - return tcf_dump_walker(hinfo, skb, cb); + return tcf_dump_walker(idrinfo, skb, cb); } else { WARN(1, "tcf_generic_walker: unknown action %d\n", type); return -EINVAL; @@ -207,40 +214,21 @@ int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, } EXPORT_SYMBOL(tcf_generic_walker); -static struct tc_action *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo) +static struct tc_action *tcf_idr_lookup(u32 index, struct tcf_idrinfo *idrinfo) { struct tc_action *p = NULL; - struct hlist_head *head; - spin_lock_bh(&hinfo->lock); - head = &hinfo->htab[tcf_hash(index, hinfo->hmask)]; - hlist_for_each_entry_rcu(p, head, tcfa_head) - if (p->tcfa_index == index) - break; - spin_unlock_bh(&hinfo->lock); + spin_lock_bh(&idrinfo->lock); + p = idr_find_ext(&idrinfo->action_idr, index); + spin_unlock_bh(&idrinfo->lock); return p; } -u32 tcf_hash_new_index(struct tc_action_net *tn) -{ - struct tcf_hashinfo *hinfo = tn->hinfo; - u32 val = hinfo->index; - - do { - if (++val == 0) - val = 1; - } while (tcf_hash_lookup(val, hinfo)); - - hinfo->index = val; - return val; -} -EXPORT_SYMBOL(tcf_hash_new_index); - -int tcf_hash_search(struct tc_action_net *tn, struct tc_action **a, u32 index) +int tcf_idr_search(struct tc_action_net *tn, struct tc_action **a, u32 index) { - struct tcf_hashinfo *hinfo = tn->hinfo; - struct tc_action *p = tcf_hash_lookup(index, hinfo); + struct tcf_idrinfo *idrinfo = tn->idrinfo; + struct tc_action *p = tcf_idr_lookup(index, idrinfo); if (p) { *a = p; @@ -248,15 +236,15 @@ int tcf_hash_search(struct tc_action_net *tn, struct tc_action **a, u32 index) } return 0; } -EXPORT_SYMBOL(tcf_hash_search); +EXPORT_SYMBOL(tcf_idr_search); -bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action **a, - int bind) +bool tcf_idr_check(struct tc_action_net *tn, u32 index, struct tc_action **a, + int bind) { - struct tcf_hashinfo *hinfo = tn->hinfo; - struct tc_action *p = NULL; + struct tcf_idrinfo *idrinfo = tn->idrinfo; + struct tc_action *p = tcf_idr_lookup(index, idrinfo); - if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) { + if (index && p) { if (bind) p->tcfa_bindcnt++; p->tcfa_refcnt++; @@ -265,23 +253,25 @@ bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action **a, } return false; } -EXPORT_SYMBOL(tcf_hash_check); +EXPORT_SYMBOL(tcf_idr_check); -void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est) +void tcf_idr_cleanup(struct tc_action *a, struct nlattr *est) { if (est) gen_kill_estimator(&a->tcfa_rate_est); call_rcu(&a->tcfa_rcu, free_tcf); } -EXPORT_SYMBOL(tcf_hash_cleanup); +EXPORT_SYMBOL(tcf_idr_cleanup); -int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est, - struct tc_action **a, const struct tc_action_ops *ops, - int bind, bool cpustats) +int tcf_idr_create(struct tc_action_net *tn, u32 index, struct nlattr *est, + struct tc_action **a, const struct tc_action_ops *ops, + int bind, bool cpustats) { struct tc_action *p = kzalloc(ops->size, GFP_KERNEL); - struct tcf_hashinfo *hinfo = tn->hinfo; + struct tcf_idrinfo *idrinfo = tn->idrinfo; + struct idr *idr = &idrinfo->action_idr; int err = -ENOMEM; + unsigned long idr_index; if (unlikely(!p)) return -ENOMEM; @@ -304,8 +294,32 @@ err2: } } spin_lock_init(&p->tcfa_lock); - INIT_HLIST_NODE(&p->tcfa_head); - p->tcfa_index = index ? index : tcf_hash_new_index(tn); + /* user doesn't specify an index */ + if (!index) { + idr_preload(GFP_KERNEL); + spin_lock_bh(&idrinfo->lock); + err = idr_alloc_ext(idr, NULL, &idr_index, 1, 0, + GFP_ATOMIC); + spin_unlock_bh(&idrinfo->lock); + idr_preload_end(); + if (err) { +err3: + free_percpu(p->cpu_qstats); + goto err2; + } + p->tcfa_index = idr_index; + } else { + idr_preload(GFP_KERNEL); + spin_lock_bh(&idrinfo->lock); + err = idr_alloc_ext(idr, NULL, NULL, index, index + 1, + GFP_ATOMIC); + spin_unlock_bh(&idrinfo->lock); + idr_preload_end(); + if (err) + goto err3; + p->tcfa_index = index; + } + p->tcfa_tm.install = jiffies; p->tcfa_tm.lastuse = jiffies; p->tcfa_tm.firstuse = 0; @@ -314,52 +328,46 @@ err2: &p->tcfa_rate_est, &p->tcfa_lock, NULL, est); if (err) { - free_percpu(p->cpu_qstats); - goto err2; + goto err3; } } - p->hinfo = hinfo; + p->idrinfo = idrinfo; p->ops = ops; INIT_LIST_HEAD(&p->list); *a = p; return 0; } -EXPORT_SYMBOL(tcf_hash_create); +EXPORT_SYMBOL(tcf_idr_create); -void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a) +void tcf_idr_insert(struct tc_action_net *tn, struct tc_action *a) { - struct tcf_hashinfo *hinfo = tn->hinfo; - unsigned int h = tcf_hash(a->tcfa_index, hinfo->hmask); + struct tcf_idrinfo *idrinfo = tn->idrinfo; - spin_lock_bh(&hinfo->lock); - hlist_add_head(&a->tcfa_head, &hinfo->htab[h]); - spin_unlock_bh(&hinfo->lock); + spin_lock_bh(&idrinfo->lock); + idr_replace_ext(&idrinfo->action_idr, a, a->tcfa_index); + spin_unlock_bh(&idrinfo->lock); } -EXPORT_SYMBOL(tcf_hash_insert); +EXPORT_SYMBOL(tcf_idr_insert); -void tcf_hashinfo_destroy(const struct tc_action_ops *ops, - struct tcf_hashinfo *hinfo) +void tcf_idrinfo_destroy(const struct tc_action_ops *ops, + struct tcf_idrinfo *idrinfo) { - int i; - - for (i = 0; i < hinfo->hmask + 1; i++) { - struct tc_action *p; - struct hlist_node *n; - - hlist_for_each_entry_safe(p, n, &hinfo->htab[i], tcfa_head) { - int ret; + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + int ret; + unsigned long id = 1; - ret = __tcf_hash_release(p, false, true); - if (ret == ACT_P_DELETED) - module_put(ops->owner); - else if (ret < 0) - return; - } + idr_for_each_entry_ext(idr, p, id) { + ret = __tcf_idr_release(p, false, true); + if (ret == ACT_P_DELETED) + module_put(ops->owner); + else if (ret < 0) + return; } - kfree(hinfo->htab); + idr_destroy(&idrinfo->action_idr); } -EXPORT_SYMBOL(tcf_hashinfo_destroy); +EXPORT_SYMBOL(tcf_idrinfo_destroy); static LIST_HEAD(act_base); static DEFINE_RWLOCK(act_mod_lock); @@ -460,9 +468,10 @@ static struct tc_action_ops *tc_lookup_action(struct nlattr *kind) int tcf_action_exec(struct sk_buff *skb, struct tc_action **actions, int nr_actions, struct tcf_result *res) { - int ret = -1, i; u32 jmp_prgcnt = 0; u32 jmp_ttl = TCA_ACT_MAX_PRIO; /*matches actions per filter */ + int i; + int ret = TC_ACT_OK; if (skb_skip_tc_classify(skb)) return TC_ACT_OK; @@ -510,7 +519,7 @@ int tcf_action_destroy(struct list_head *actions, int bind) int ret = 0; list_for_each_entry_safe(a, tmp, actions, list) { - ret = __tcf_hash_release(a, bind, true); + ret = __tcf_idr_release(a, bind, true); if (ret == ACT_P_DELETED) module_put(a->ops->owner); else if (ret < 0) @@ -835,7 +844,7 @@ out_nlmsg_trim: } static int -act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, +tcf_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, struct list_head *actions, int event) { struct sk_buff *skb; @@ -1018,7 +1027,7 @@ tca_action_gd(struct net *net, struct nlattr *nla, struct nlmsghdr *n, } if (event == RTM_GETACTION) - ret = act_get_notify(net, portid, n, &actions, event); + ret = tcf_get_notify(net, portid, n, &actions, event); else { /* delete */ ret = tcf_del_notify(net, n, &actions, portid); if (ret) @@ -1068,11 +1077,18 @@ static int tcf_action_add(struct net *net, struct nlattr *nla, return tcf_add_notify(net, n, &actions, portid); } +static u32 tcaa_root_flags_allowed = TCA_FLAG_LARGE_DUMP_ON; +static const struct nla_policy tcaa_policy[TCA_ROOT_MAX + 1] = { + [TCA_ROOT_FLAGS] = { .type = NLA_BITFIELD32, + .validation_data = &tcaa_root_flags_allowed }, + [TCA_ROOT_TIME_DELTA] = { .type = NLA_U32 }, +}; + static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); - struct nlattr *tca[TCA_ACT_MAX + 1]; + struct nlattr *tca[TCA_ROOT_MAX + 1]; u32 portid = skb ? NETLINK_CB(skb).portid : 0; int ret = 0, ovr = 0; @@ -1080,7 +1096,7 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; - ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL, + ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ROOT_MAX, NULL, extack); if (ret < 0) return ret; @@ -1121,16 +1137,12 @@ replay: return ret; } -static struct nlattr *find_dump_kind(const struct nlmsghdr *n) +static struct nlattr *find_dump_kind(struct nlattr **nla) { struct nlattr *tb1, *tb2[TCA_ACT_MAX + 1]; struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; - struct nlattr *nla[TCAA_MAX + 1]; struct nlattr *kind; - if (nlmsg_parse(n, sizeof(struct tcamsg), nla, TCAA_MAX, - NULL, NULL) < 0) - return NULL; tb1 = nla[TCA_ACT_TAB]; if (tb1 == NULL) return NULL; @@ -1157,8 +1169,20 @@ static int tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) struct tc_action_ops *a_o; int ret = 0; struct tcamsg *t = (struct tcamsg *) nlmsg_data(cb->nlh); - struct nlattr *kind = find_dump_kind(cb->nlh); + struct nlattr *tb[TCA_ROOT_MAX + 1]; + struct nlattr *count_attr = NULL; + unsigned long jiffy_since = 0; + struct nlattr *kind = NULL; + struct nla_bitfield32 bf; + u32 msecs_since = 0; + u32 act_count = 0; + + ret = nlmsg_parse(cb->nlh, sizeof(struct tcamsg), tb, TCA_ROOT_MAX, + tcaa_policy, NULL); + if (ret < 0) + return ret; + kind = find_dump_kind(tb); if (kind == NULL) { pr_info("tc_dump_action: action bad kind\n"); return 0; @@ -1168,14 +1192,32 @@ static int tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) if (a_o == NULL) return 0; + cb->args[2] = 0; + if (tb[TCA_ROOT_FLAGS]) { + bf = nla_get_bitfield32(tb[TCA_ROOT_FLAGS]); + cb->args[2] = bf.value; + } + + if (tb[TCA_ROOT_TIME_DELTA]) { + msecs_since = nla_get_u32(tb[TCA_ROOT_TIME_DELTA]); + } + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, cb->nlh->nlmsg_type, sizeof(*t), 0); if (!nlh) goto out_module_put; + + if (msecs_since) + jiffy_since = jiffies - msecs_to_jiffies(msecs_since); + t = nlmsg_data(nlh); t->tca_family = AF_UNSPEC; t->tca__pad1 = 0; t->tca__pad2 = 0; + cb->args[3] = jiffy_since; + count_attr = nla_reserve(skb, TCA_ROOT_COUNT, sizeof(u32)); + if (!count_attr) + goto out_module_put; nest = nla_nest_start(skb, TCA_ACT_TAB); if (nest == NULL) @@ -1188,6 +1230,9 @@ static int tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) if (ret > 0) { nla_nest_end(skb, nest); ret = skb->len; + act_count = cb->args[1]; + memcpy(nla_data(count_attr), &act_count, sizeof(u32)); + cb->args[1] = 0; } else nlmsg_trim(skb, b); @@ -1205,10 +1250,10 @@ out_module_put: static int __init tc_action_init(void) { - rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETACTION, tc_ctl_action, tc_dump_action, - NULL); + 0); return 0; } diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index 9afe1337cfd1..c0c707eb2c96 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -21,7 +21,6 @@ #include <linux/tc_act/tc_bpf.h> #include <net/tc_act/tc_bpf.h> -#define BPF_TAB_MASK 15 #define ACT_BPF_NAME_LEN 256 struct tcf_bpf_cfg { @@ -295,9 +294,9 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_ACT_BPF_PARMS]); - if (!tcf_hash_check(tn, parm->index, act, bind)) { - ret = tcf_hash_create(tn, parm->index, est, act, - &act_bpf_ops, bind, true); + if (!tcf_idr_check(tn, parm->index, act, bind)) { + ret = tcf_idr_create(tn, parm->index, est, act, + &act_bpf_ops, bind, true); if (ret < 0) return ret; @@ -307,7 +306,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, if (bind) return 0; - tcf_hash_release(*act, bind); + tcf_idr_release(*act, bind); if (!replace) return -EEXIST; } @@ -343,7 +342,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, rcu_assign_pointer(prog->filter, cfg.filter); if (res == ACT_P_CREATED) { - tcf_hash_insert(tn, *act); + tcf_idr_insert(tn, *act); } else { /* make sure the program being replaced is no longer executing */ synchronize_rcu(); @@ -353,7 +352,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, return res; out: if (res == ACT_P_CREATED) - tcf_hash_cleanup(*act, est); + tcf_idr_cleanup(*act, est); return ret; } @@ -379,7 +378,7 @@ static int tcf_bpf_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, bpf_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_bpf_ops __read_mostly = { @@ -399,7 +398,7 @@ 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, BPF_TAB_MASK); + return tc_action_net_init(tn, &act_bpf_ops); } static void __net_exit bpf_exit_net(struct net *net) diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 2155bc6c6a1e..10b7a8855a6c 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -28,8 +28,6 @@ #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_zones.h> -#define CONNMARK_TAB_MASK 3 - static unsigned int connmark_net_id; static struct tc_action_ops act_connmark_ops; @@ -119,9 +117,9 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_CONNMARK_PARMS]); - if (!tcf_hash_check(tn, parm->index, a, bind)) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_connmark_ops, bind, false); + if (!tcf_idr_check(tn, parm->index, a, bind)) { + ret = tcf_idr_create(tn, parm->index, est, a, + &act_connmark_ops, bind, false); if (ret) return ret; @@ -130,13 +128,13 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla, ci->net = net; ci->zone = parm->zone; - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); ret = ACT_P_CREATED; } else { ci = to_connmark(*a); if (bind) return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; /* replacing action and zone */ @@ -189,7 +187,7 @@ static int tcf_connmark_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, connmark_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_connmark_ops = { @@ -208,7 +206,7 @@ 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, CONNMARK_TAB_MASK); + return tc_action_net_init(tn, &act_connmark_ops); } static void __net_exit connmark_exit_net(struct net *net) diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 3317a2f579da..1c40caadcff9 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -37,8 +37,6 @@ #include <linux/tc_act/tc_csum.h> #include <net/tc_act/tc_csum.h> -#define CSUM_TAB_MASK 15 - static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = { [TCA_CSUM_PARMS] = { .len = sizeof(struct tc_csum), }, }; @@ -67,16 +65,16 @@ static int tcf_csum_init(struct net *net, struct nlattr *nla, return -EINVAL; parm = nla_data(tb[TCA_CSUM_PARMS]); - if (!tcf_hash_check(tn, parm->index, a, bind)) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_csum_ops, bind, false); + if (!tcf_idr_check(tn, parm->index, a, bind)) { + ret = tcf_idr_create(tn, parm->index, est, a, + &act_csum_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -88,7 +86,7 @@ static int tcf_csum_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -231,9 +229,6 @@ static int tcf_csum_ipv4_udp(struct sk_buff *skb, unsigned int ihl, const struct iphdr *iph; u16 ul; - if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP) - return 1; - /* * Support both UDP and UDPLITE checksum algorithms, Don't use * udph->len to get the real length without any protocol check, @@ -287,9 +282,6 @@ static int tcf_csum_ipv6_udp(struct sk_buff *skb, unsigned int ihl, const struct ipv6hdr *ip6h; u16 ul; - if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP) - return 1; - /* * Support both UDP and UDPLITE checksum algorithms, Don't use * udph->len to get the real length without any protocol check, @@ -615,7 +607,7 @@ static int tcf_csum_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, csum_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_csum_ops = { @@ -634,7 +626,7 @@ 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, CSUM_TAB_MASK); + return tc_action_net_init(tn, &act_csum_ops); } static void __net_exit csum_exit_net(struct net *net) diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 99afe8b1f1fb..e29a48ef7fc3 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -23,8 +23,6 @@ #include <linux/tc_act/tc_gact.h> #include <net/tc_act/tc_gact.h> -#define GACT_TAB_MASK 15 - static unsigned int gact_net_id; static struct tc_action_ops act_gact_ops; @@ -92,16 +90,16 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla, } #endif - if (!tcf_hash_check(tn, parm->index, a, bind)) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_gact_ops, bind, true); + if (!tcf_idr_check(tn, parm->index, a, bind)) { + ret = tcf_idr_create(tn, parm->index, est, a, + &act_gact_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -122,7 +120,7 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla, } #endif if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -214,7 +212,7 @@ static int tcf_gact_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, gact_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_gact_ops = { @@ -234,7 +232,7 @@ 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, GACT_TAB_MASK); + return tc_action_net_init(tn, &act_gact_ops); } static void __net_exit gact_exit_net(struct net *net) diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index c5dec308b8b1..8ccd35825b6b 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -34,8 +34,6 @@ #include <linux/etherdevice.h> #include <net/ife.h> -#define IFE_TAB_MASK 15 - static unsigned int ife_net_id; static int max_metacnt = IFE_META_MAX + 1; static struct tc_action_ops act_ife_ops; @@ -435,8 +433,8 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, struct nlattr *tb[TCA_IFE_MAX + 1]; struct nlattr *tb2[IFE_META_MAX + 1]; struct tcf_ife_info *ife; + u16 ife_type = ETH_P_IFE; struct tc_ife *parm; - u16 ife_type = 0; u8 *daddr = NULL; u8 *saddr = NULL; bool exists = false; @@ -452,30 +450,18 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_IFE_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; - if (parm->flags & IFE_ENCODE) { - /* Until we get issued the ethertype, we cant have - * a default.. - **/ - if (!tb[TCA_IFE_TYPE]) { - if (exists) - tcf_hash_release(*a, bind); - pr_info("You MUST pass etherype for encoding\n"); - return -EINVAL; - } - } - if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, &act_ife_ops, - bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, &act_ife_ops, + bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -484,7 +470,8 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, ife->flags = parm->flags; if (parm->flags & IFE_ENCODE) { - ife_type = nla_get_u16(tb[TCA_IFE_TYPE]); + if (tb[TCA_IFE_TYPE]) + ife_type = nla_get_u16(tb[TCA_IFE_TYPE]); if (tb[TCA_IFE_DMAC]) daddr = nla_data(tb[TCA_IFE_DMAC]); if (tb[TCA_IFE_SMAC]) @@ -518,7 +505,7 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, if (err) { metadata_parse_err: if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (ret == ACT_P_CREATED) _tcf_ife_cleanup(*a, bind); @@ -552,7 +539,7 @@ metadata_parse_err: spin_unlock_bh(&ife->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -811,7 +798,7 @@ static int tcf_ife_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, ife_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_ife_ops = { @@ -831,7 +818,7 @@ 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, IFE_TAB_MASK); + return tc_action_net_init(tn, &act_ife_ops); } static void __net_exit ife_exit_net(struct net *net) diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 36f0ced9e60c..d9e399a7e3d5 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -28,19 +28,18 @@ #include <linux/netfilter_ipv4/ip_tables.h> -#define IPT_TAB_MASK 15 - static unsigned int ipt_net_id; static struct tc_action_ops act_ipt_ops; static unsigned int xt_net_id; static struct tc_action_ops act_xt_ops; -static int ipt_init_target(struct xt_entry_target *t, char *table, - unsigned int hook) +static int ipt_init_target(struct net *net, struct xt_entry_target *t, + char *table, unsigned int hook) { struct xt_tgchk_param par; struct xt_target *target; + struct ipt_entry e = {}; int ret = 0; target = xt_request_find_target(AF_INET, t->u.user.name, @@ -49,8 +48,10 @@ static int ipt_init_target(struct xt_entry_target *t, char *table, return PTR_ERR(target); t->u.kernel.target = target; + memset(&par, 0, sizeof(par)); + par.net = net; par.table = table; - par.entryinfo = NULL; + par.entryinfo = &e; par.target = target; par.targinfo = t->data; par.hook_mask = hook; @@ -91,10 +92,11 @@ static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = { [TCA_IPT_TARG] = { .len = sizeof(struct xt_entry_target) }, }; -static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, +static int __tcf_ipt_init(struct net *net, unsigned int id, struct nlattr *nla, struct nlattr *est, struct tc_action **a, const struct tc_action_ops *ops, int ovr, int bind) { + struct tc_action_net *tn = net_generic(net, id); struct nlattr *tb[TCA_IPT_MAX + 1]; struct tcf_ipt *ipt; struct xt_entry_target *td, *t; @@ -114,33 +116,33 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, if (tb[TCA_IPT_INDEX] != NULL) index = nla_get_u32(tb[TCA_IPT_INDEX]); - exists = tcf_hash_check(tn, index, a, bind); + exists = tcf_idr_check(tn, index, a, bind); if (exists && bind) return 0; if (tb[TCA_IPT_HOOK] == NULL || tb[TCA_IPT_TARG] == NULL) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]); if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } if (!exists) { - ret = tcf_hash_create(tn, index, est, a, ops, bind, - false); + ret = tcf_idr_create(tn, index, est, a, ops, bind, + false); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; @@ -159,7 +161,7 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, if (unlikely(!t)) goto err2; - err = ipt_init_target(t, tname, hook); + err = ipt_init_target(net, t, tname, hook); if (err < 0) goto err3; @@ -176,7 +178,7 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, ipt->tcfi_hook = hook; spin_unlock_bh(&ipt->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; err3: @@ -185,7 +187,7 @@ err2: kfree(tname); err1: if (ret == ACT_P_CREATED) - tcf_hash_cleanup(*a, est); + tcf_idr_cleanup(*a, est); return err; } @@ -193,18 +195,16 @@ static int tcf_ipt_init(struct net *net, struct nlattr *nla, struct nlattr *est, struct tc_action **a, int ovr, int bind) { - struct tc_action_net *tn = net_generic(net, ipt_net_id); - - return __tcf_ipt_init(tn, nla, est, a, &act_ipt_ops, ovr, bind); + return __tcf_ipt_init(net, ipt_net_id, nla, est, a, &act_ipt_ops, ovr, + bind); } static int tcf_xt_init(struct net *net, struct nlattr *nla, struct nlattr *est, struct tc_action **a, int ovr, int bind) { - struct tc_action_net *tn = net_generic(net, xt_net_id); - - return __tcf_ipt_init(tn, nla, est, a, &act_xt_ops, ovr, bind); + return __tcf_ipt_init(net, xt_net_id, nla, est, a, &act_xt_ops, ovr, + bind); } static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, @@ -314,7 +314,7 @@ static int tcf_ipt_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, ipt_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_ipt_ops = { @@ -334,7 +334,7 @@ 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, IPT_TAB_MASK); + return tc_action_net_init(tn, &act_ipt_ops); } static void __net_exit ipt_exit_net(struct net *net) @@ -364,7 +364,7 @@ static int tcf_xt_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, xt_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_xt_ops = { @@ -384,7 +384,7 @@ 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, IPT_TAB_MASK); + return tc_action_net_init(tn, &act_xt_ops); } static void __net_exit xt_exit_net(struct net *net) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 1b5549ababd4..416627c66f08 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -28,7 +28,6 @@ #include <linux/tc_act/tc_mirred.h> #include <net/tc_act/tc_mirred.h> -#define MIRRED_TAB_MASK 7 static LIST_HEAD(mirred_list); static DEFINE_SPINLOCK(mirred_list_lock); @@ -94,7 +93,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, return -EINVAL; parm = nla_data(tb[TCA_MIRRED_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; @@ -106,14 +105,14 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, break; default: if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } if (parm->ifindex) { dev = __dev_get_by_index(net, parm->ifindex); if (dev == NULL) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -ENODEV; } mac_header_xmit = dev_is_mac_header_xmit(dev); @@ -124,13 +123,13 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, if (!exists) { if (dev == NULL) return -EINVAL; - ret = tcf_hash_create(tn, parm->index, est, a, - &act_mirred_ops, bind, true); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_mirred_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -152,7 +151,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, spin_lock_bh(&mirred_list_lock); list_add(&m->tcfm_list, &mirred_list); spin_unlock_bh(&mirred_list_lock); - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); } return ret; @@ -283,7 +282,7 @@ static int tcf_mirred_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, mirred_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static int mirred_device_event(struct notifier_block *unused, @@ -344,7 +343,7 @@ 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, MIRRED_TAB_MASK); + return tc_action_net_init(tn, &act_mirred_ops); } static void __net_exit mirred_exit_net(struct net *net) diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 9016ab8a0649..c365d01b99c8 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -29,8 +29,6 @@ #include <net/udp.h> -#define NAT_TAB_MASK 15 - static unsigned int nat_net_id; static struct tc_action_ops act_nat_ops; @@ -58,16 +56,16 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, return -EINVAL; parm = nla_data(tb[TCA_NAT_PARMS]); - if (!tcf_hash_check(tn, parm->index, a, bind)) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_nat_ops, bind, false); + if (!tcf_idr_check(tn, parm->index, a, bind)) { + ret = tcf_idr_create(tn, parm->index, est, a, + &act_nat_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind) return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -83,7 +81,7 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -290,7 +288,7 @@ static int tcf_nat_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, nat_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_nat_ops = { @@ -309,7 +307,7 @@ 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, NAT_TAB_MASK); + return tc_action_net_init(tn, &act_nat_ops); } static void __net_exit nat_exit_net(struct net *net) diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 7dc5892671c8..491fe5deb09e 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -24,8 +24,6 @@ #include <net/tc_act/tc_pedit.h> #include <uapi/linux/tc_act/tc_pedit.h> -#define PEDIT_TAB_MASK 15 - static unsigned int pedit_net_id; static struct tc_action_ops act_pedit_ops; @@ -168,17 +166,17 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, if (IS_ERR(keys_ex)) return PTR_ERR(keys_ex); - if (!tcf_hash_check(tn, parm->index, a, bind)) { + if (!tcf_idr_check(tn, parm->index, a, bind)) { if (!parm->nkeys) return -EINVAL; - ret = tcf_hash_create(tn, parm->index, est, a, - &act_pedit_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_pedit_ops, bind, false); if (ret) return ret; p = to_pedit(*a); keys = kmalloc(ksize, GFP_KERNEL); if (keys == NULL) { - tcf_hash_cleanup(*a, est); + tcf_idr_cleanup(*a, est); kfree(keys_ex); return -ENOMEM; } @@ -186,7 +184,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, } else { if (bind) return 0; - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; p = to_pedit(*a); @@ -214,7 +212,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -432,7 +430,7 @@ static int tcf_pedit_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, pedit_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_pedit_ops = { @@ -452,7 +450,7 @@ 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, PEDIT_TAB_MASK); + return tc_action_net_init(tn, &act_pedit_ops); } static void __net_exit pedit_exit_net(struct net *net) diff --git a/net/sched/act_police.c b/net/sched/act_police.c index b062bc80c7cb..3bb2ebf9e9ae 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -40,8 +40,6 @@ struct tcf_police { #define to_police(pc) ((struct tcf_police *)pc) -#define POL_TAB_MASK 15 - /* old policer structure from before tc actions */ struct tc_police_compat { u32 index; @@ -101,18 +99,18 @@ static int tcf_act_police_init(struct net *net, struct nlattr *nla, return -EINVAL; parm = nla_data(tb[TCA_POLICE_TBF]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; if (!exists) { - ret = tcf_hash_create(tn, parm->index, NULL, a, - &act_police_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, NULL, a, + &act_police_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -188,7 +186,7 @@ static int tcf_act_police_init(struct net *net, struct nlattr *nla, return ret; police->tcfp_t_c = ktime_get_ns(); - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; @@ -196,7 +194,7 @@ failure: qdisc_put_rtab(P_tab); qdisc_put_rtab(R_tab); if (ret == ACT_P_CREATED) - tcf_hash_cleanup(*a, est); + tcf_idr_cleanup(*a, est); return err; } @@ -310,7 +308,7 @@ static int tcf_police_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, police_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } MODULE_AUTHOR("Alexey Kuznetsov"); @@ -333,7 +331,7 @@ 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, POL_TAB_MASK); + return tc_action_net_init(tn, &act_police_ops); } static void __net_exit police_exit_net(struct net *net) diff --git a/net/sched/act_sample.c b/net/sched/act_sample.c index 59d6645a4007..ec986ae52808 100644 --- a/net/sched/act_sample.c +++ b/net/sched/act_sample.c @@ -25,7 +25,6 @@ #include <linux/if_arp.h> -#define SAMPLE_TAB_MASK 7 static unsigned int sample_net_id; static struct tc_action_ops act_sample_ops; @@ -59,18 +58,18 @@ static int tcf_sample_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_SAMPLE_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_sample_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_sample_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -82,7 +81,7 @@ static int tcf_sample_init(struct net *net, struct nlattr *nla, psample_group = psample_group_get(net, s->psample_group_num); if (!psample_group) { if (ret == ACT_P_CREATED) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -ENOMEM; } RCU_INIT_POINTER(s->psample_group, psample_group); @@ -93,7 +92,7 @@ static int tcf_sample_init(struct net *net, struct nlattr *nla, } if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -221,7 +220,7 @@ static int tcf_sample_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, sample_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_sample_ops = { @@ -241,7 +240,7 @@ 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, SAMPLE_TAB_MASK); + return tc_action_net_init(tn, &act_sample_ops); } static void __net_exit sample_exit_net(struct net *net) diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 43605e7ce051..e7b57e5071a3 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -24,8 +24,6 @@ #include <linux/tc_act/tc_defact.h> #include <net/tc_act/tc_defact.h> -#define SIMP_TAB_MASK 7 - static unsigned int simp_net_id; static struct tc_action_ops act_simp_ops; @@ -102,28 +100,28 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, return -EINVAL; parm = nla_data(tb[TCA_DEF_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; if (tb[TCA_DEF_DATA] == NULL) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } defdata = nla_data(tb[TCA_DEF_DATA]); if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_simp_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_simp_ops, bind, false); if (ret) return ret; d = to_defact(*a); ret = alloc_defdata(d, defdata); if (ret < 0) { - tcf_hash_cleanup(*a, est); + tcf_idr_cleanup(*a, est); return ret; } d->tcf_action = parm->action; @@ -131,7 +129,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, } else { d = to_defact(*a); - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; @@ -139,7 +137,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, } if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -183,7 +181,7 @@ static int tcf_simp_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, simp_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_simp_ops = { @@ -203,7 +201,7 @@ 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, SIMP_TAB_MASK); + return tc_action_net_init(tn, &act_simp_ops); } static void __net_exit simp_exit_net(struct net *net) diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 6b3e65d7de0c..59949d61f20d 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -27,8 +27,6 @@ #include <linux/tc_act/tc_skbedit.h> #include <net/tc_act/tc_skbedit.h> -#define SKBEDIT_TAB_MASK 15 - static unsigned int skbedit_net_id; static struct tc_action_ops act_skbedit_ops; @@ -118,18 +116,18 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_SKBEDIT_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; if (!flags) { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_skbedit_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_skbedit_ops, bind, false); if (ret) return ret; @@ -137,7 +135,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, ret = ACT_P_CREATED; } else { d = to_skbedit(*a); - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -163,7 +161,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&d->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -221,7 +219,7 @@ static int tcf_skbedit_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, skbedit_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_skbedit_ops = { @@ -240,7 +238,7 @@ 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, SKBEDIT_TAB_MASK); + return tc_action_net_init(tn, &act_skbedit_ops); } static void __net_exit skbedit_exit_net(struct net *net) diff --git a/net/sched/act_skbmod.c b/net/sched/act_skbmod.c index a73c4bbcada2..b642ad3d39dd 100644 --- a/net/sched/act_skbmod.c +++ b/net/sched/act_skbmod.c @@ -20,8 +20,6 @@ #include <linux/tc_act/tc_skbmod.h> #include <net/tc_act/tc_skbmod.h> -#define SKBMOD_TAB_MASK 15 - static unsigned int skbmod_net_id; static struct tc_action_ops act_skbmod_ops; @@ -129,7 +127,7 @@ static int tcf_skbmod_init(struct net *net, struct nlattr *nla, if (parm->flags & SKBMOD_F_SWAPMAC) lflags = SKBMOD_F_SWAPMAC; - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; @@ -137,14 +135,14 @@ static int tcf_skbmod_init(struct net *net, struct nlattr *nla, return -EINVAL; if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_skbmod_ops, bind, true); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_skbmod_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -155,7 +153,7 @@ static int tcf_skbmod_init(struct net *net, struct nlattr *nla, p = kzalloc(sizeof(struct tcf_skbmod_params), GFP_KERNEL); if (unlikely(!p)) { if (ovr) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -ENOMEM; } @@ -182,7 +180,7 @@ static int tcf_skbmod_init(struct net *net, struct nlattr *nla, kfree_rcu(p_old, rcu); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -245,7 +243,7 @@ static int tcf_skbmod_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, skbmod_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_skbmod_ops = { @@ -265,7 +263,7 @@ 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, SKBMOD_TAB_MASK); + return tc_action_net_init(tn, &act_skbmod_ops); } static void __net_exit skbmod_exit_net(struct net *net) diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c index fd7e75679c69..30c96274c638 100644 --- a/net/sched/act_tunnel_key.c +++ b/net/sched/act_tunnel_key.c @@ -20,8 +20,6 @@ #include <linux/tc_act/tc_tunnel_key.h> #include <net/tc_act/tc_tunnel_key.h> -#define TUNNEL_KEY_TAB_MASK 15 - static unsigned int tunnel_key_net_id; static struct tc_action_ops act_tunnel_key_ops; @@ -100,7 +98,7 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, return -EINVAL; parm = nla_data(tb[TCA_TUNNEL_KEY_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; @@ -159,14 +157,14 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, } if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_tunnel_key_ops, bind, true); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_tunnel_key_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -177,7 +175,7 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, params_new = kzalloc(sizeof(*params_new), GFP_KERNEL); if (unlikely(!params_new)) { if (ret == ACT_P_CREATED) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -ENOMEM; } @@ -193,13 +191,13 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, kfree_rcu(params_old, rcu); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; err_out: if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return ret; } @@ -304,7 +302,7 @@ static int tunnel_key_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_tunnel_key_ops = { @@ -324,7 +322,7 @@ 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, TUNNEL_KEY_TAB_MASK); + return tc_action_net_init(tn, &act_tunnel_key_ops); } static void __net_exit tunnel_key_exit_net(struct net *net) diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index 13ba3a89f675..16eb067a8d8f 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -19,8 +19,6 @@ #include <linux/tc_act/tc_vlan.h> #include <net/tc_act/tc_vlan.h> -#define VLAN_TAB_MASK 15 - static unsigned int vlan_net_id; static struct tc_action_ops act_vlan_ops; @@ -128,7 +126,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, if (!tb[TCA_VLAN_PARMS]) return -EINVAL; parm = nla_data(tb[TCA_VLAN_PARMS]); - exists = tcf_hash_check(tn, parm->index, a, bind); + exists = tcf_idr_check(tn, parm->index, a, bind); if (exists && bind) return 0; @@ -139,13 +137,13 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, case TCA_VLAN_ACT_MODIFY: if (!tb[TCA_VLAN_PUSH_VLAN_ID]) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } push_vid = nla_get_u16(tb[TCA_VLAN_PUSH_VLAN_ID]); if (push_vid >= VLAN_VID_MASK) { if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -ERANGE; } @@ -167,20 +165,20 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, break; default: if (exists) - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); return -EINVAL; } action = parm->v_action; if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, - &act_vlan_ops, bind, false); + ret = tcf_idr_create(tn, parm->index, est, a, + &act_vlan_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(*a, bind); + tcf_idr_release(*a, bind); if (!ovr) return -EEXIST; } @@ -199,7 +197,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&v->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, *a); + tcf_idr_insert(tn, *a); return ret; } @@ -252,7 +250,7 @@ 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); - return tcf_hash_search(tn, a, index); + return tcf_idr_search(tn, a, index); } static struct tc_action_ops act_vlan_ops = { @@ -271,7 +269,7 @@ 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, VLAN_TAB_MASK); + return tc_action_net_init(tn, &act_vlan_ops); } static void __net_exit vlan_exit_net(struct net *net) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 39da0c5801c9..ea6c65fd5fc5 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -100,21 +100,6 @@ int unregister_tcf_proto_ops(struct tcf_proto_ops *ops) } EXPORT_SYMBOL(unregister_tcf_proto_ops); -static int tfilter_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, struct tcf_proto *tp, - unsigned long fh, int event, bool unicast); - -static void tfilter_notify_chain(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, - struct tcf_chain *chain, int event) -{ - struct tcf_proto *tp; - - for (tp = rtnl_dereference(chain->filter_chain); - tp; tp = rtnl_dereference(tp->next)) - tfilter_notify(net, oskb, n, tp, 0, event, false); -} - /* Select new prio value from the range, managed by kernel. */ static inline u32 tcf_auto_prio(struct tcf_proto *tp) @@ -205,7 +190,7 @@ static void tcf_chain_flush(struct tcf_chain *chain) { struct tcf_proto *tp; - if (*chain->p_filter_chain) + if (chain->p_filter_chain) RCU_INIT_POINTER(*chain->p_filter_chain, NULL); while ((tp = rtnl_dereference(chain->filter_chain)) != NULL) { RCU_INIT_POINTER(chain->filter_chain, tp->next); @@ -215,9 +200,15 @@ static void tcf_chain_flush(struct tcf_chain *chain) static void tcf_chain_destroy(struct tcf_chain *chain) { - list_del(&chain->list); - tcf_chain_flush(chain); - kfree(chain); + /* May be already removed from the list by the previous call. */ + if (!list_empty(&chain->list)) + list_del_init(&chain->list); + + /* There might still be a reference held when we got here from + * tcf_block_put. Wait for the user to drop reference before free. + */ + if (!chain->refcnt) + kfree(chain); } struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, @@ -288,8 +279,10 @@ void tcf_block_put(struct tcf_block *block) if (!block) return; - list_for_each_entry_safe(chain, tmp, &block->chain_list, list) + list_for_each_entry_safe(chain, tmp, &block->chain_list, list) { + tcf_chain_flush(chain); tcf_chain_destroy(chain); + } kfree(block); } EXPORT_SYMBOL(tcf_block_put); @@ -407,6 +400,109 @@ static struct tcf_proto *tcf_chain_tp_find(struct tcf_chain *chain, return tp; } +static int tcf_fill_node(struct net *net, struct sk_buff *skb, + struct tcf_proto *tp, void *fh, u32 portid, + u32 seq, u16 flags, int event) +{ + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_ifindex = qdisc_dev(tp->q)->ifindex; + tcm->tcm_parent = tp->classid; + tcm->tcm_info = TC_H_MAKE(tp->prio, tp->protocol); + if (nla_put_string(skb, TCA_KIND, tp->ops->kind)) + goto nla_put_failure; + if (nla_put_u32(skb, TCA_CHAIN, tp->chain->index)) + goto nla_put_failure; + if (!fh) { + tcm->tcm_handle = 0; + } else { + if (tp->ops->dump && tp->ops->dump(net, tp, fh, skb, tcm) < 0) + goto nla_put_failure; + } + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tfilter_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct tcf_proto *tp, + void *fh, int event, bool unicast) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tcf_fill_node(net, skb, tp, fh, portid, n->nlmsg_seq, + n->nlmsg_flags, event) <= 0) { + kfree_skb(skb); + return -EINVAL; + } + + if (unicast) + return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +static int tfilter_del_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct tcf_proto *tp, + void *fh, bool unicast, bool *last) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + int err; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tcf_fill_node(net, skb, tp, fh, portid, n->nlmsg_seq, + n->nlmsg_flags, RTM_DELTFILTER) <= 0) { + kfree_skb(skb); + return -EINVAL; + } + + err = tp->ops->delete(tp, fh, last); + if (err) { + kfree_skb(skb); + return err; + } + + if (unicast) + return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +static void tfilter_notify_chain(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, + struct tcf_chain *chain, int event) +{ + struct tcf_proto *tp; + + for (tp = rtnl_dereference(chain->filter_chain); + tp; tp = rtnl_dereference(tp->next)) + tfilter_notify(net, oskb, n, tp, 0, event, false); +} + /* Add/change/delete/get a filter node */ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, @@ -428,7 +524,7 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, struct tcf_proto *tp; const struct Qdisc_class_ops *cops; unsigned long cl; - unsigned long fh; + void *fh; int err; int tp_created; @@ -498,7 +594,7 @@ replay: /* Do we search for filter, attached to class? */ if (TC_H_MIN(parent)) { - cl = cops->get(q, parent); + cl = cops->find(q, parent); if (cl == 0) return -ENOENT; } @@ -567,7 +663,7 @@ replay: fh = tp->ops->get(tp, t->tcm_handle); - if (fh == 0) { + if (!fh) { if (n->nlmsg_type == RTM_DELTFILTER && t->tcm_handle == 0) { tcf_chain_tp_remove(chain, &chain_info, tp); tfilter_notify(net, skb, n, tp, fh, @@ -595,11 +691,10 @@ replay: } break; case RTM_DELTFILTER: - err = tp->ops->delete(tp, fh, &last); + err = tfilter_del_notify(net, skb, n, tp, fh, false, + &last); if (err) goto errout; - tfilter_notify(net, skb, n, tp, t->tcm_handle, - RTM_DELTFILTER, false); if (last) { tcf_chain_tp_remove(chain, &chain_info, tp); tcf_proto_destroy(tp); @@ -629,83 +724,19 @@ replay: errout: if (chain) tcf_chain_put(chain); - if (cl) - cops->put(q, cl); if (err == -EAGAIN) /* Replay the request. */ goto replay; return err; } -static int tcf_fill_node(struct net *net, struct sk_buff *skb, - struct tcf_proto *tp, unsigned long fh, u32 portid, - u32 seq, u16 flags, int event) -{ - struct tcmsg *tcm; - struct nlmsghdr *nlh; - unsigned char *b = skb_tail_pointer(skb); - - nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); - if (!nlh) - goto out_nlmsg_trim; - tcm = nlmsg_data(nlh); - tcm->tcm_family = AF_UNSPEC; - tcm->tcm__pad1 = 0; - tcm->tcm__pad2 = 0; - tcm->tcm_ifindex = qdisc_dev(tp->q)->ifindex; - tcm->tcm_parent = tp->classid; - tcm->tcm_info = TC_H_MAKE(tp->prio, tp->protocol); - if (nla_put_string(skb, TCA_KIND, tp->ops->kind)) - goto nla_put_failure; - if (nla_put_u32(skb, TCA_CHAIN, tp->chain->index)) - goto nla_put_failure; - tcm->tcm_handle = fh; - if (RTM_DELTFILTER != event) { - tcm->tcm_handle = 0; - if (tp->ops->dump && tp->ops->dump(net, tp, fh, skb, tcm) < 0) - goto nla_put_failure; - } - nlh->nlmsg_len = skb_tail_pointer(skb) - b; - return skb->len; - -out_nlmsg_trim: -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int tfilter_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, struct tcf_proto *tp, - unsigned long fh, int event, bool unicast) -{ - struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - - skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); - if (!skb) - return -ENOBUFS; - - if (tcf_fill_node(net, skb, tp, fh, portid, n->nlmsg_seq, - n->nlmsg_flags, event) <= 0) { - kfree_skb(skb); - return -EINVAL; - } - - if (unicast) - return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); - - return rtnetlink_send(skb, net, portid, RTNLGRP_TC, - n->nlmsg_flags & NLM_F_ECHO); -} - struct tcf_dump_args { struct tcf_walker w; struct sk_buff *skb; struct netlink_callback *cb; }; -static int tcf_node_dump(struct tcf_proto *tp, unsigned long n, - struct tcf_walker *arg) +static int tcf_node_dump(struct tcf_proto *tp, void *n, struct tcf_walker *arg) { struct tcf_dump_args *a = (void *)arg; struct net *net = sock_net(a->skb->sk); @@ -797,17 +828,17 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) goto out; cops = q->ops->cl_ops; if (!cops) - goto errout; + goto out; if (!cops->tcf_block) - goto errout; + goto out; if (TC_H_MIN(tcm->tcm_parent)) { - cl = cops->get(q, tcm->tcm_parent); + cl = cops->find(q, tcm->tcm_parent); if (cl == 0) - goto errout; + goto out; } block = cops->tcf_block(q, cl); if (!block) - goto errout; + goto out; index_start = cb->args[0]; index = 0; @@ -822,9 +853,6 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) cb->args[0] = index; -errout: - if (cl) - cops->put(q, cl); out: return skb->len; } @@ -883,18 +911,12 @@ int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, } EXPORT_SYMBOL(tcf_exts_validate); -void tcf_exts_change(struct tcf_proto *tp, struct tcf_exts *dst, - struct tcf_exts *src) +void tcf_exts_change(struct tcf_exts *dst, struct tcf_exts *src) { #ifdef CONFIG_NET_CLS_ACT struct tcf_exts old = *dst; - tcf_tree_lock(tp); - dst->nr_actions = src->nr_actions; - dst->actions = src->actions; - dst->type = src->type; - tcf_tree_unlock(tp); - + *dst = *src; tcf_exts_destroy(&old); #endif } @@ -915,7 +937,7 @@ int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts) #ifdef CONFIG_NET_CLS_ACT struct nlattr *nest; - if (exts->action && exts->nr_actions) { + if (exts->action && tcf_exts_has_actions(exts)) { /* * again for backward compatible mode - we want * to work with both old and new modes of entering @@ -972,7 +994,7 @@ int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, const struct tc_action *a; LIST_HEAD(actions); - if (tc_no_actions(exts)) + if (!tcf_exts_has_actions(exts)) return -EINVAL; tcf_exts_to_list(exts, &actions); @@ -991,10 +1013,10 @@ EXPORT_SYMBOL(tcf_exts_get_dev); static int __init tc_filter_init(void) { - rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_ctl_tfilter, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_ctl_tfilter, NULL, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_ctl_tfilter, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_ctl_tfilter, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETTFILTER, tc_ctl_tfilter, - tc_dump_tfilter, NULL); + tc_dump_tfilter, 0); return 0; } diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c index c4fd63a068f9..d89ebafd2239 100644 --- a/net/sched/cls_basic.c +++ b/net/sched/cls_basic.c @@ -56,20 +56,18 @@ static int basic_classify(struct sk_buff *skb, const struct tcf_proto *tp, return -1; } -static unsigned long basic_get(struct tcf_proto *tp, u32 handle) +static void *basic_get(struct tcf_proto *tp, u32 handle) { - unsigned long l = 0UL; struct basic_head *head = rtnl_dereference(tp->root); struct basic_filter *f; list_for_each_entry(f, &head->flist, link) { if (f->handle == handle) { - l = (unsigned long) f; - break; + return f; } } - return l; + return NULL; } static int basic_init(struct tcf_proto *tp) @@ -106,10 +104,10 @@ static void basic_destroy(struct tcf_proto *tp) kfree_rcu(head, rcu); } -static int basic_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int basic_delete(struct tcf_proto *tp, void *arg, bool *last) { struct basic_head *head = rtnl_dereference(tp->root); - struct basic_filter *f = (struct basic_filter *) arg; + struct basic_filter *f = arg; list_del_rcu(&f->link); tcf_unbind_filter(tp, &f->res); @@ -129,38 +127,27 @@ static int basic_set_parms(struct net *net, struct tcf_proto *tp, struct nlattr *est, bool ovr) { int err; - struct tcf_exts e; - struct tcf_ematch_tree t; - err = tcf_exts_init(&e, TCA_BASIC_ACT, TCA_BASIC_POLICE); + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr); if (err < 0) return err; - err = tcf_exts_validate(net, tp, tb, est, &e, ovr); - if (err < 0) - goto errout; - err = tcf_em_tree_validate(tp, tb[TCA_BASIC_EMATCHES], &t); + err = tcf_em_tree_validate(tp, tb[TCA_BASIC_EMATCHES], &f->ematches); if (err < 0) - goto errout; + return err; if (tb[TCA_BASIC_CLASSID]) { f->res.classid = nla_get_u32(tb[TCA_BASIC_CLASSID]); tcf_bind_filter(tp, &f->res, base); } - tcf_exts_change(tp, &f->exts, &e); - tcf_em_tree_change(tp, &f->ematches, &t); f->tp = tp; - return 0; -errout: - tcf_exts_destroy(&e); - return err; } static int basic_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg, bool ovr) + struct nlattr **tca, void **arg, bool ovr) { int err; struct basic_head *head = rtnl_dereference(tp->root); @@ -213,7 +200,7 @@ static int basic_change(struct net *net, struct sk_buff *in_skb, if (err < 0) goto errout; - *arg = (unsigned long)fnew; + *arg = fnew; if (fold) { list_replace_rcu(&fold->link, &fnew->link); @@ -239,7 +226,7 @@ static void basic_walk(struct tcf_proto *tp, struct tcf_walker *arg) if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; break; } @@ -248,10 +235,18 @@ skip: } } -static int basic_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static void basic_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct basic_filter *f = fh; + + if (f && f->res.classid == classid) + f->res.class = cl; +} + +static int basic_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct basic_filter *f = (struct basic_filter *) fh; + struct basic_filter *f = fh; struct nlattr *nest; if (f == NULL) @@ -293,6 +288,7 @@ static struct tcf_proto_ops cls_basic_ops __read_mostly = { .delete = basic_delete, .walk = basic_walk, .dump = basic_dump, + .bind_class = basic_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index f57bd531ba98..520c5027646a 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -147,24 +147,18 @@ static int cls_bpf_offload_cmd(struct tcf_proto *tp, struct cls_bpf_prog *prog, enum tc_clsbpf_command cmd) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_bpf_offload bpf_offload = {}; - struct tc_to_netdev offload; + struct tc_cls_bpf_offload cls_bpf = {}; int err; - offload.type = TC_SETUP_CLSBPF; - offload.cls_bpf = &bpf_offload; - - bpf_offload.command = cmd; - bpf_offload.exts = &prog->exts; - bpf_offload.prog = prog->filter; - bpf_offload.name = prog->bpf_name; - bpf_offload.exts_integrated = prog->exts_integrated; - bpf_offload.gen_flags = prog->gen_flags; - - err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, - tp->protocol, &offload); + tc_cls_common_offload_init(&cls_bpf.common, tp); + cls_bpf.command = cmd; + cls_bpf.exts = &prog->exts; + cls_bpf.prog = prog->filter; + cls_bpf.name = prog->bpf_name; + cls_bpf.exts_integrated = prog->exts_integrated; + cls_bpf.gen_flags = prog->gen_flags; + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSBPF, &cls_bpf); if (!err && (cmd == TC_CLSBPF_ADD || cmd == TC_CLSBPF_REPLACE)) prog->gen_flags |= TCA_CLS_FLAGS_IN_HW; @@ -184,7 +178,7 @@ static int cls_bpf_offload(struct tcf_proto *tp, struct cls_bpf_prog *prog, (oldprog && tc_skip_sw(oldprog->gen_flags)); if (oldprog && oldprog->offloaded) { - if (tc_should_offload(dev, tp, prog->gen_flags)) { + if (tc_should_offload(dev, prog->gen_flags)) { cmd = TC_CLSBPF_REPLACE; } else if (!tc_skip_sw(prog->gen_flags)) { obj = oldprog; @@ -193,7 +187,7 @@ static int cls_bpf_offload(struct tcf_proto *tp, struct cls_bpf_prog *prog, return -EINVAL; } } else { - if (!tc_should_offload(dev, tp, prog->gen_flags)) + if (!tc_should_offload(dev, prog->gen_flags)) return skip_sw ? -EINVAL : 0; cmd = TC_CLSBPF_ADD; } @@ -276,11 +270,11 @@ static void __cls_bpf_delete(struct tcf_proto *tp, struct cls_bpf_prog *prog) call_rcu(&prog->rcu, cls_bpf_delete_prog_rcu); } -static int cls_bpf_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int cls_bpf_delete(struct tcf_proto *tp, void *arg, bool *last) { struct cls_bpf_head *head = rtnl_dereference(tp->root); - __cls_bpf_delete(tp, (struct cls_bpf_prog *) arg); + __cls_bpf_delete(tp, arg); *last = list_empty(&head->plist); return 0; } @@ -296,20 +290,17 @@ static void cls_bpf_destroy(struct tcf_proto *tp) kfree_rcu(head, rcu); } -static unsigned long cls_bpf_get(struct tcf_proto *tp, u32 handle) +static void *cls_bpf_get(struct tcf_proto *tp, u32 handle) { struct cls_bpf_head *head = rtnl_dereference(tp->root); struct cls_bpf_prog *prog; - unsigned long ret = 0UL; list_for_each_entry(prog, &head->plist, link) { - if (prog->handle == handle) { - ret = (unsigned long) prog; - break; - } + if (prog->handle == handle) + return prog; } - return ret; + return NULL; } static int cls_bpf_prog_from_ops(struct nlattr **tb, struct cls_bpf_prog *prog) @@ -382,13 +373,11 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog, return 0; } -static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, - struct cls_bpf_prog *prog, - unsigned long base, struct nlattr **tb, - struct nlattr *est, bool ovr) +static int cls_bpf_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_bpf_prog *prog, unsigned long base, + struct nlattr **tb, struct nlattr *est, bool ovr) { bool is_bpf, is_ebpf, have_exts = false; - struct tcf_exts exts; u32 gen_flags = 0; int ret; @@ -397,30 +386,23 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, if ((!is_bpf && !is_ebpf) || (is_bpf && is_ebpf)) return -EINVAL; - ret = tcf_exts_init(&exts, TCA_BPF_ACT, TCA_BPF_POLICE); + ret = tcf_exts_validate(net, tp, tb, est, &prog->exts, ovr); if (ret < 0) return ret; - ret = tcf_exts_validate(net, tp, tb, est, &exts, ovr); - if (ret < 0) - goto errout; if (tb[TCA_BPF_FLAGS]) { u32 bpf_flags = nla_get_u32(tb[TCA_BPF_FLAGS]); - if (bpf_flags & ~TCA_BPF_FLAG_ACT_DIRECT) { - ret = -EINVAL; - goto errout; - } + if (bpf_flags & ~TCA_BPF_FLAG_ACT_DIRECT) + return -EINVAL; have_exts = bpf_flags & TCA_BPF_FLAG_ACT_DIRECT; } if (tb[TCA_BPF_FLAGS_GEN]) { gen_flags = nla_get_u32(tb[TCA_BPF_FLAGS_GEN]); if (gen_flags & ~CLS_BPF_SUPPORTED_GEN_FLAGS || - !tc_flags_valid(gen_flags)) { - ret = -EINVAL; - goto errout; - } + !tc_flags_valid(gen_flags)) + return -EINVAL; } prog->exts_integrated = have_exts; @@ -429,19 +411,14 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, ret = is_bpf ? cls_bpf_prog_from_ops(tb, prog) : cls_bpf_prog_from_efd(tb, prog, tp); if (ret < 0) - goto errout; + return ret; if (tb[TCA_BPF_CLASSID]) { prog->res.classid = nla_get_u32(tb[TCA_BPF_CLASSID]); tcf_bind_filter(tp, &prog->res, base); } - tcf_exts_change(tp, &prog->exts, &exts); return 0; - -errout: - tcf_exts_destroy(&exts); - return ret; } static u32 cls_bpf_grab_new_handle(struct tcf_proto *tp, @@ -468,10 +445,10 @@ static u32 cls_bpf_grab_new_handle(struct tcf_proto *tp, static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct cls_bpf_head *head = rtnl_dereference(tp->root); - struct cls_bpf_prog *oldprog = (struct cls_bpf_prog *) *arg; + struct cls_bpf_prog *oldprog = *arg; struct nlattr *tb[TCA_BPF_MAX + 1]; struct cls_bpf_prog *prog; int ret; @@ -508,8 +485,7 @@ static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, goto errout; } - ret = cls_bpf_modify_existing(net, tp, prog, base, tb, tca[TCA_RATE], - ovr); + ret = cls_bpf_set_parms(net, tp, prog, base, tb, tca[TCA_RATE], ovr); if (ret < 0) goto errout; @@ -530,7 +506,7 @@ static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, list_add_rcu(&prog->link, &head->plist); } - *arg = (unsigned long) prog; + *arg = prog; return 0; errout: @@ -578,10 +554,10 @@ static int cls_bpf_dump_ebpf_info(const struct cls_bpf_prog *prog, return 0; } -static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *tm) { - struct cls_bpf_prog *prog = (struct cls_bpf_prog *) fh; + struct cls_bpf_prog *prog = fh; struct nlattr *nest; u32 bpf_flags = 0; int ret; @@ -631,6 +607,14 @@ nla_put_failure: return -1; } +static void cls_bpf_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct cls_bpf_prog *prog = fh; + + if (prog && prog->res.classid == classid) + prog->res.class = cl; +} + static void cls_bpf_walk(struct tcf_proto *tp, struct tcf_walker *arg) { struct cls_bpf_head *head = rtnl_dereference(tp->root); @@ -639,7 +623,7 @@ static void cls_bpf_walk(struct tcf_proto *tp, struct tcf_walker *arg) list_for_each_entry(prog, &head->plist, link) { if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) prog, arg) < 0) { + if (arg->fn(tp, prog, arg) < 0) { arg->stop = 1; break; } @@ -659,6 +643,7 @@ static struct tcf_proto_ops cls_bpf_ops __read_mostly = { .delete = cls_bpf_delete, .walk = cls_bpf_walk, .dump = cls_bpf_dump, + .bind_class = cls_bpf_bind_class, }; static int __init cls_bpf_init_mod(void) diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c index 12ce547eea04..d48452f87975 100644 --- a/net/sched/cls_cgroup.c +++ b/net/sched/cls_cgroup.c @@ -43,9 +43,9 @@ static int cls_cgroup_classify(struct sk_buff *skb, const struct tcf_proto *tp, return tcf_exts_exec(skb, &head->exts, res); } -static unsigned long cls_cgroup_get(struct tcf_proto *tp, u32 handle) +static void *cls_cgroup_get(struct tcf_proto *tp, u32 handle) { - return 0UL; + return NULL; } static int cls_cgroup_init(struct tcf_proto *tp) @@ -71,13 +71,11 @@ static void cls_cgroup_destroy_rcu(struct rcu_head *root) static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct nlattr *tb[TCA_CGROUP_MAX + 1]; struct cls_cgroup_head *head = rtnl_dereference(tp->root); struct cls_cgroup_head *new; - struct tcf_ematch_tree t; - struct tcf_exts e; int err; if (!tca[TCA_OPTIONS]) @@ -103,23 +101,13 @@ static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb, if (err < 0) goto errout; - err = tcf_exts_init(&e, TCA_CGROUP_ACT, TCA_CGROUP_POLICE); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &new->exts, ovr); if (err < 0) goto errout; - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); - if (err < 0) { - tcf_exts_destroy(&e); - goto errout; - } - err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &t); - if (err < 0) { - tcf_exts_destroy(&e); + err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &new->ematches); + if (err < 0) goto errout; - } - - tcf_exts_change(tp, &new->exts, &e); - tcf_em_tree_change(tp, &new->ematches, &t); rcu_assign_pointer(tp->root, new); if (head) @@ -140,7 +128,7 @@ static void cls_cgroup_destroy(struct tcf_proto *tp) call_rcu(&head->rcu, cls_cgroup_destroy_rcu); } -static int cls_cgroup_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int cls_cgroup_delete(struct tcf_proto *tp, void *arg, bool *last) { return -EOPNOTSUPP; } @@ -152,7 +140,7 @@ static void cls_cgroup_walk(struct tcf_proto *tp, struct tcf_walker *arg) if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) head, arg) < 0) { + if (arg->fn(tp, head, arg) < 0) { arg->stop = 1; return; } @@ -160,7 +148,7 @@ skip: arg->count++; } -static int cls_cgroup_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int cls_cgroup_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { struct cls_cgroup_head *head = rtnl_dereference(tp->root); diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index 3065752b9cda..2a3a60ec5b86 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -382,14 +382,12 @@ static void flow_destroy_filter(struct rcu_head *head) static int flow_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct flow_head *head = rtnl_dereference(tp->root); struct flow_filter *fold, *fnew; struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_FLOW_MAX + 1]; - struct tcf_exts e; - struct tcf_ematch_tree t; unsigned int nkeys = 0; unsigned int perturb_period = 0; u32 baseclass = 0; @@ -425,31 +423,27 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, return -EOPNOTSUPP; } - err = tcf_exts_init(&e, TCA_FLOW_ACT, TCA_FLOW_POLICE); - if (err < 0) - goto err1; - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); - if (err < 0) - goto err1; + fnew = kzalloc(sizeof(*fnew), GFP_KERNEL); + if (!fnew) + return -ENOBUFS; - err = tcf_em_tree_validate(tp, tb[TCA_FLOW_EMATCHES], &t); + err = tcf_em_tree_validate(tp, tb[TCA_FLOW_EMATCHES], &fnew->ematches); if (err < 0) goto err1; - err = -ENOBUFS; - fnew = kzalloc(sizeof(*fnew), GFP_KERNEL); - if (!fnew) + err = tcf_exts_init(&fnew->exts, TCA_FLOW_ACT, TCA_FLOW_POLICE); + if (err < 0) goto err2; - err = tcf_exts_init(&fnew->exts, TCA_FLOW_ACT, TCA_FLOW_POLICE); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &fnew->exts, ovr); if (err < 0) - goto err3; + goto err2; - fold = (struct flow_filter *)*arg; + fold = *arg; if (fold) { err = -EINVAL; if (fold->handle != handle && handle) - goto err3; + goto err2; /* Copy fold into fnew */ fnew->tp = fold->tp; @@ -469,31 +463,31 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, if (tb[TCA_FLOW_MODE]) mode = nla_get_u32(tb[TCA_FLOW_MODE]); if (mode != FLOW_MODE_HASH && nkeys > 1) - goto err3; + goto err2; if (mode == FLOW_MODE_HASH) perturb_period = fold->perturb_period; if (tb[TCA_FLOW_PERTURB]) { if (mode != FLOW_MODE_HASH) - goto err3; + goto err2; perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ; } } else { err = -EINVAL; if (!handle) - goto err3; + goto err2; if (!tb[TCA_FLOW_KEYS]) - goto err3; + goto err2; mode = FLOW_MODE_MAP; if (tb[TCA_FLOW_MODE]) mode = nla_get_u32(tb[TCA_FLOW_MODE]); if (mode != FLOW_MODE_HASH && nkeys > 1) - goto err3; + goto err2; if (tb[TCA_FLOW_PERTURB]) { if (mode != FLOW_MODE_HASH) - goto err3; + goto err2; perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ; } @@ -511,9 +505,6 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, setup_deferrable_timer(&fnew->perturb_timer, flow_perturbation, (unsigned long)fnew); - tcf_exts_change(tp, &fnew->exts, &e); - tcf_em_tree_change(tp, &fnew->ematches, &t); - netif_keep_dst(qdisc_dev(tp->q)); if (tb[TCA_FLOW_KEYS]) { @@ -541,31 +532,29 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, if (perturb_period) mod_timer(&fnew->perturb_timer, jiffies + perturb_period); - if (*arg == 0) + if (!*arg) list_add_tail_rcu(&fnew->list, &head->filters); else list_replace_rcu(&fold->list, &fnew->list); - *arg = (unsigned long)fnew; + *arg = fnew; if (fold) call_rcu(&fold->rcu, flow_destroy_filter); return 0; -err3: - tcf_exts_destroy(&fnew->exts); err2: - tcf_em_tree_destroy(&t); - kfree(fnew); + tcf_exts_destroy(&fnew->exts); + tcf_em_tree_destroy(&fnew->ematches); err1: - tcf_exts_destroy(&e); + kfree(fnew); return err; } -static int flow_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int flow_delete(struct tcf_proto *tp, void *arg, bool *last) { struct flow_head *head = rtnl_dereference(tp->root); - struct flow_filter *f = (struct flow_filter *)arg; + struct flow_filter *f = arg; list_del_rcu(&f->list); call_rcu(&f->rcu, flow_destroy_filter); @@ -597,21 +586,21 @@ static void flow_destroy(struct tcf_proto *tp) kfree_rcu(head, rcu); } -static unsigned long flow_get(struct tcf_proto *tp, u32 handle) +static void *flow_get(struct tcf_proto *tp, u32 handle) { struct flow_head *head = rtnl_dereference(tp->root); struct flow_filter *f; list_for_each_entry(f, &head->filters, list) if (f->handle == handle) - return (unsigned long)f; - return 0; + return f; + return NULL; } -static int flow_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int flow_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct flow_filter *f = (struct flow_filter *)fh; + struct flow_filter *f = fh; struct nlattr *nest; if (f == NULL) @@ -677,7 +666,7 @@ static void flow_walk(struct tcf_proto *tp, struct tcf_walker *arg) list_for_each_entry(f, &head->filters, list) { if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long)f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; break; } diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 7832eb93379b..1a267e77c6de 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -68,7 +68,6 @@ struct cls_fl_head { struct rhashtable ht; struct fl_flow_mask mask; struct flow_dissector dissector; - u32 hgen; bool mask_assigned; struct list_head filters; struct rhashtable_params ht_params; @@ -76,6 +75,7 @@ struct cls_fl_head { struct work_struct work; struct rcu_head rcu; }; + struct idr handle_idr; }; struct cls_fl_filter { @@ -88,7 +88,6 @@ struct cls_fl_filter { u32 handle; u32 flags; struct rcu_head rcu; - struct tc_to_netdev tc; struct net_device *hw_dev; }; @@ -211,6 +210,7 @@ static int fl_init(struct tcf_proto *tp) INIT_LIST_HEAD_RCU(&head->filters); rcu_assign_pointer(tp->root, head); + idr_init(&head->handle_idr); return 0; } @@ -225,22 +225,17 @@ static void fl_destroy_filter(struct rcu_head *head) static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f) { - struct tc_cls_flower_offload offload = {0}; + struct tc_cls_flower_offload cls_flower = {}; struct net_device *dev = f->hw_dev; - struct tc_to_netdev *tc = &f->tc; - if (!tc_can_offload(dev, tp)) + if (!tc_can_offload(dev)) return; - offload.command = TC_CLSFLOWER_DESTROY; - offload.prio = tp->prio; - offload.cookie = (unsigned long)f; + tc_cls_common_offload_init(&cls_flower.common, tp); + cls_flower.command = TC_CLSFLOWER_DESTROY; + cls_flower.cookie = (unsigned long) f; - tc->type = TC_SETUP_CLSFLOWER; - tc->cls_flower = &offload; - - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->chain->index, - tp->protocol, tc); + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, &cls_flower); } static int fl_hw_replace_filter(struct tcf_proto *tp, @@ -249,35 +244,31 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, struct cls_fl_filter *f) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_flower_offload offload = {0}; - struct tc_to_netdev *tc = &f->tc; + struct tc_cls_flower_offload cls_flower = {}; int err; - if (!tc_can_offload(dev, tp)) { + if (!tc_can_offload(dev)) { if (tcf_exts_get_dev(dev, &f->exts, &f->hw_dev) || - (f->hw_dev && !tc_can_offload(f->hw_dev, tp))) { + (f->hw_dev && !tc_can_offload(f->hw_dev))) { f->hw_dev = dev; return tc_skip_sw(f->flags) ? -EINVAL : 0; } dev = f->hw_dev; - tc->egress_dev = true; + cls_flower.egress_dev = true; } else { f->hw_dev = dev; } - offload.command = TC_CLSFLOWER_REPLACE; - offload.prio = tp->prio; - offload.cookie = (unsigned long)f; - offload.dissector = dissector; - offload.mask = mask; - offload.key = &f->mkey; - offload.exts = &f->exts; - - tc->type = TC_SETUP_CLSFLOWER; - tc->cls_flower = &offload; + tc_cls_common_offload_init(&cls_flower.common, tp); + cls_flower.command = TC_CLSFLOWER_REPLACE; + cls_flower.cookie = (unsigned long) f; + cls_flower.dissector = dissector; + cls_flower.mask = mask; + cls_flower.key = &f->mkey; + cls_flower.exts = &f->exts; - err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, tc); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, + &cls_flower); if (!err) f->flags |= TCA_CLS_FLAGS_IN_HW; @@ -288,27 +279,26 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) { - struct tc_cls_flower_offload offload = {0}; + struct tc_cls_flower_offload cls_flower = {}; struct net_device *dev = f->hw_dev; - struct tc_to_netdev *tc = &f->tc; - if (!tc_can_offload(dev, tp)) + if (!tc_can_offload(dev)) return; - offload.command = TC_CLSFLOWER_STATS; - offload.prio = tp->prio; - offload.cookie = (unsigned long)f; - offload.exts = &f->exts; - - tc->type = TC_SETUP_CLSFLOWER; - tc->cls_flower = &offload; + tc_cls_common_offload_init(&cls_flower.common, tp); + cls_flower.command = TC_CLSFLOWER_STATS; + cls_flower.cookie = (unsigned long) f; + cls_flower.exts = &f->exts; - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, tc); + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, + &cls_flower); } static void __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f) { + struct cls_fl_head *head = rtnl_dereference(tp->root); + + idr_remove_ext(&head->handle_idr, f->handle); list_del_rcu(&f->list); if (!tc_skip_hw(f->flags)) fl_hw_destroy_filter(tp, f); @@ -341,20 +331,17 @@ static void fl_destroy(struct tcf_proto *tp) list_for_each_entry_safe(f, next, &head->filters, list) __fl_delete(tp, f); + idr_destroy(&head->handle_idr); __module_get(THIS_MODULE); call_rcu(&head->rcu, fl_destroy_rcu); } -static unsigned long fl_get(struct tcf_proto *tp, u32 handle) +static void *fl_get(struct tcf_proto *tp, u32 handle) { struct cls_fl_head *head = rtnl_dereference(tp->root); - struct cls_fl_filter *f; - list_for_each_entry(f, &head->filters, list) - if (f->handle == handle) - return (unsigned long) f; - return 0; + return idr_find_ext(&head->handle_idr, handle); } static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = { @@ -852,15 +839,11 @@ static int fl_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct nlattr **tb, struct nlattr *est, bool ovr) { - struct tcf_exts e; int err; - err = tcf_exts_init(&e, TCA_FLOWER_ACT, 0); + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr); if (err < 0) return err; - err = tcf_exts_validate(net, tp, tb, est, &e, ovr); - if (err < 0) - goto errout; if (tb[TCA_FLOWER_CLASSID]) { f->res.classid = nla_get_u32(tb[TCA_FLOWER_CLASSID]); @@ -869,50 +852,25 @@ static int fl_set_parms(struct net *net, struct tcf_proto *tp, err = fl_set_key(net, tb, &f->key, &mask->key); if (err) - goto errout; + return err; fl_mask_update_range(mask); fl_set_masked_key(&f->mkey, &f->key, mask); - tcf_exts_change(tp, &f->exts, &e); - return 0; -errout: - tcf_exts_destroy(&e); - return err; -} - -static u32 fl_grab_new_handle(struct tcf_proto *tp, - struct cls_fl_head *head) -{ - unsigned int i = 0x80000000; - u32 handle; - - do { - if (++head->hgen == 0x7FFFFFFF) - head->hgen = 1; - } while (--i > 0 && fl_get(tp, head->hgen)); - - if (unlikely(i == 0)) { - pr_err("Insufficient number of handles\n"); - handle = 0; - } else { - handle = head->hgen; - } - - return handle; } static int fl_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct cls_fl_head *head = rtnl_dereference(tp->root); - struct cls_fl_filter *fold = (struct cls_fl_filter *) *arg; + struct cls_fl_filter *fold = *arg; struct cls_fl_filter *fnew; struct nlattr **tb; struct fl_flow_mask mask = {}; + unsigned long idr_index; int err; if (!tca[TCA_OPTIONS]) @@ -943,13 +901,21 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, goto errout; if (!handle) { - handle = fl_grab_new_handle(tp, head); - if (!handle) { - err = -EINVAL; + err = idr_alloc_ext(&head->handle_idr, fnew, &idr_index, + 1, 0x80000000, GFP_KERNEL); + if (err) goto errout; - } + fnew->handle = idr_index; + } + + /* user specifies a handle and it doesn't exist */ + if (handle && !fold) { + err = idr_alloc_ext(&head->handle_idr, fnew, &idr_index, + handle, handle + 1, GFP_KERNEL); + if (err) + goto errout; + fnew->handle = idr_index; } - fnew->handle = handle; if (tb[TCA_FLOWER_FLAGS]) { fnew->flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]); @@ -1000,9 +966,11 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, fl_hw_destroy_filter(tp, fold); } - *arg = (unsigned long) fnew; + *arg = fnew; if (fold) { + fnew->handle = handle; + idr_replace_ext(&head->handle_idr, fnew, fnew->handle); list_replace_rcu(&fold->list, &fnew->list); tcf_unbind_filter(tp, &fold->res); call_rcu(&fold->rcu, fl_destroy_filter); @@ -1021,10 +989,10 @@ errout_tb: return err; } -static int fl_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int fl_delete(struct tcf_proto *tp, void *arg, bool *last) { struct cls_fl_head *head = rtnl_dereference(tp->root); - struct cls_fl_filter *f = (struct cls_fl_filter *) arg; + struct cls_fl_filter *f = arg; if (!tc_skip_sw(f->flags)) rhashtable_remove_fast(&head->ht, &f->ht_node, @@ -1042,7 +1010,7 @@ static void fl_walk(struct tcf_proto *tp, struct tcf_walker *arg) list_for_each_entry_rcu(f, &head->filters, list) { if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; break; } @@ -1177,11 +1145,11 @@ static int fl_dump_key_flags(struct sk_buff *skb, u32 flags_key, u32 flags_mask) return nla_put(skb, TCA_FLOWER_KEY_FLAGS_MASK, 4, &_mask); } -static int fl_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int fl_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { struct cls_fl_head *head = rtnl_dereference(tp->root); - struct cls_fl_filter *f = (struct cls_fl_filter *) fh; + struct cls_fl_filter *f = fh; struct nlattr *nest; struct fl_flow_key *key, *mask; @@ -1383,6 +1351,14 @@ nla_put_failure: return -1; } +static void fl_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct cls_fl_filter *f = fh; + + if (f && f->res.classid == classid) + f->res.class = cl; +} + static struct tcf_proto_ops cls_fl_ops __read_mostly = { .kind = "flower", .classify = fl_classify, @@ -1393,6 +1369,7 @@ static struct tcf_proto_ops cls_fl_ops __read_mostly = { .delete = fl_delete, .walk = fl_walk, .dump = fl_dump, + .bind_class = fl_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index d3885362e017..941245ad07fd 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -95,20 +95,20 @@ static int fw_classify(struct sk_buff *skb, const struct tcf_proto *tp, return -1; } -static unsigned long fw_get(struct tcf_proto *tp, u32 handle) +static void *fw_get(struct tcf_proto *tp, u32 handle) { struct fw_head *head = rtnl_dereference(tp->root); struct fw_filter *f; if (head == NULL) - return 0; + return NULL; f = rtnl_dereference(head->ht[fw_hash(handle)]); for (; f; f = rtnl_dereference(f->next)) { if (f->id == handle) - return (unsigned long)f; + return f; } - return 0; + return NULL; } static int fw_init(struct tcf_proto *tp) @@ -147,10 +147,10 @@ static void fw_destroy(struct tcf_proto *tp) kfree_rcu(head, rcu); } -static int fw_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int fw_delete(struct tcf_proto *tp, void *arg, bool *last) { struct fw_head *head = rtnl_dereference(tp->root); - struct fw_filter *f = (struct fw_filter *)arg; + struct fw_filter *f = arg; struct fw_filter __rcu **fp; struct fw_filter *pfp; int ret = -EINVAL; @@ -190,22 +190,17 @@ static const struct nla_policy fw_policy[TCA_FW_MAX + 1] = { [TCA_FW_MASK] = { .type = NLA_U32 }, }; -static int -fw_change_attrs(struct net *net, struct tcf_proto *tp, struct fw_filter *f, - struct nlattr **tb, struct nlattr **tca, unsigned long base, - bool ovr) +static int fw_set_parms(struct net *net, struct tcf_proto *tp, + struct fw_filter *f, struct nlattr **tb, + struct nlattr **tca, unsigned long base, bool ovr) { struct fw_head *head = rtnl_dereference(tp->root); - struct tcf_exts e; u32 mask; int err; - err = tcf_exts_init(&e, TCA_FW_ACT, TCA_FW_POLICE); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &f->exts, ovr); if (err < 0) return err; - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); - if (err < 0) - goto errout; if (tb[TCA_FW_CLASSID]) { f->res.classid = nla_get_u32(tb[TCA_FW_CLASSID]); @@ -216,10 +211,8 @@ fw_change_attrs(struct net *net, struct tcf_proto *tp, struct fw_filter *f, if (tb[TCA_FW_INDEV]) { int ret; ret = tcf_change_indev(net, tb[TCA_FW_INDEV]); - if (ret < 0) { - err = ret; - goto errout; - } + if (ret < 0) + return ret; f->ifindex = ret; } #endif /* CONFIG_NET_CLS_IND */ @@ -228,25 +221,20 @@ fw_change_attrs(struct net *net, struct tcf_proto *tp, struct fw_filter *f, if (tb[TCA_FW_MASK]) { mask = nla_get_u32(tb[TCA_FW_MASK]); if (mask != head->mask) - goto errout; + return err; } else if (head->mask != 0xFFFFFFFF) - goto errout; - - tcf_exts_change(tp, &f->exts, &e); + return err; return 0; -errout: - tcf_exts_destroy(&e); - return err; } static int fw_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, - u32 handle, struct nlattr **tca, unsigned long *arg, + u32 handle, struct nlattr **tca, void **arg, bool ovr) { struct fw_head *head = rtnl_dereference(tp->root); - struct fw_filter *f = (struct fw_filter *) *arg; + struct fw_filter *f = *arg; struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_FW_MAX + 1]; int err; @@ -282,7 +270,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, return err; } - err = fw_change_attrs(net, tp, fnew, tb, tca, base, ovr); + err = fw_set_parms(net, tp, fnew, tb, tca, base, ovr); if (err < 0) { tcf_exts_destroy(&fnew->exts); kfree(fnew); @@ -300,7 +288,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, tcf_unbind_filter(tp, &f->res); call_rcu(&f->rcu, fw_delete_filter); - *arg = (unsigned long)fnew; + *arg = fnew; return err; } @@ -330,14 +318,14 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, f->id = handle; f->tp = tp; - err = fw_change_attrs(net, tp, f, tb, tca, base, ovr); + err = fw_set_parms(net, tp, f, tb, tca, base, ovr); if (err < 0) goto errout; RCU_INIT_POINTER(f->next, head->ht[fw_hash(handle)]); rcu_assign_pointer(head->ht[fw_hash(handle)], f); - *arg = (unsigned long)f; + *arg = f; return 0; errout: @@ -366,7 +354,7 @@ static void fw_walk(struct tcf_proto *tp, struct tcf_walker *arg) arg->count++; continue; } - if (arg->fn(tp, (unsigned long)f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; return; } @@ -375,11 +363,11 @@ static void fw_walk(struct tcf_proto *tp, struct tcf_walker *arg) } } -static int fw_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int fw_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { struct fw_head *head = rtnl_dereference(tp->root); - struct fw_filter *f = (struct fw_filter *)fh; + struct fw_filter *f = fh; struct nlattr *nest; if (f == NULL) @@ -387,7 +375,7 @@ static int fw_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, t->tcm_handle = f->id; - if (!f->res.classid && !tcf_exts_is_available(&f->exts)) + if (!f->res.classid && !tcf_exts_has_actions(&f->exts)) return skb->len; nest = nla_nest_start(skb, TCA_OPTIONS); @@ -424,6 +412,14 @@ nla_put_failure: return -1; } +static void fw_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct fw_filter *f = fh; + + if (f && f->res.classid == classid) + f->res.class = cl; +} + static struct tcf_proto_ops cls_fw_ops __read_mostly = { .kind = "fw", .classify = fw_classify, @@ -434,6 +430,7 @@ static struct tcf_proto_ops cls_fw_ops __read_mostly = { .delete = fw_delete, .walk = fw_walk, .dump = fw_dump, + .bind_class = fw_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index 9dc26c32cf32..21cc45caf842 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -54,19 +54,16 @@ static int mall_replace_hw_filter(struct tcf_proto *tp, unsigned long cookie) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_to_netdev offload; - struct tc_cls_matchall_offload mall_offload = {0}; + struct tc_cls_matchall_offload cls_mall = {}; int err; - offload.type = TC_SETUP_MATCHALL; - offload.cls_mall = &mall_offload; - offload.cls_mall->command = TC_CLSMATCHALL_REPLACE; - offload.cls_mall->exts = &head->exts; - offload.cls_mall->cookie = cookie; + tc_cls_common_offload_init(&cls_mall.common, tp); + cls_mall.command = TC_CLSMATCHALL_REPLACE; + cls_mall.exts = &head->exts; + cls_mall.cookie = cookie; - err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, - tp->protocol, &offload); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSMATCHALL, + &cls_mall); if (!err) head->flags |= TCA_CLS_FLAGS_IN_HW; @@ -78,17 +75,13 @@ static void mall_destroy_hw_filter(struct tcf_proto *tp, unsigned long cookie) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_to_netdev offload; - struct tc_cls_matchall_offload mall_offload = {0}; + struct tc_cls_matchall_offload cls_mall = {}; - offload.type = TC_SETUP_MATCHALL; - offload.cls_mall = &mall_offload; - offload.cls_mall->command = TC_CLSMATCHALL_DESTROY; - offload.cls_mall->exts = NULL; - offload.cls_mall->cookie = cookie; + tc_cls_common_offload_init(&cls_mall.common, tp); + cls_mall.command = TC_CLSMATCHALL_DESTROY; + cls_mall.cookie = cookie; - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->chain->index, - tp->protocol, &offload); + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSMATCHALL, &cls_mall); } static void mall_destroy(struct tcf_proto *tp) @@ -99,15 +92,15 @@ static void mall_destroy(struct tcf_proto *tp) if (!head) return; - if (tc_should_offload(dev, tp, head->flags)) + if (tc_should_offload(dev, head->flags)) mall_destroy_hw_filter(tp, head, (unsigned long) head); call_rcu(&head->rcu, mall_destroy_rcu); } -static unsigned long mall_get(struct tcf_proto *tp, u32 handle) +static void *mall_get(struct tcf_proto *tp, u32 handle) { - return 0UL; + return NULL; } static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { @@ -120,33 +113,23 @@ static int mall_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct nlattr **tb, struct nlattr *est, bool ovr) { - struct tcf_exts e; int err; - err = tcf_exts_init(&e, TCA_MATCHALL_ACT, 0); - if (err) - return err; - err = tcf_exts_validate(net, tp, tb, est, &e, ovr); + err = tcf_exts_validate(net, tp, tb, est, &head->exts, ovr); if (err < 0) - goto errout; + return err; if (tb[TCA_MATCHALL_CLASSID]) { head->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); tcf_bind_filter(tp, &head->res, base); } - - tcf_exts_change(tp, &head->exts, &e); - return 0; -errout: - tcf_exts_destroy(&e); - return err; } static int mall_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct cls_mall_head *head = rtnl_dereference(tp->root); struct net_device *dev = tp->q->dev_queue->dev; @@ -189,7 +172,7 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (err) goto err_set_parms; - if (tc_should_offload(dev, tp, flags)) { + if (tc_should_offload(dev, flags)) { err = mall_replace_hw_filter(tp, new, (unsigned long) new); if (err) { if (tc_skip_sw(flags)) @@ -202,7 +185,7 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (!tc_in_hw(new->flags)) new->flags |= TCA_CLS_FLAGS_NOT_IN_HW; - *arg = (unsigned long) head; + *arg = head; rcu_assign_pointer(tp->root, new); return 0; @@ -214,7 +197,7 @@ err_exts_init: return err; } -static int mall_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int mall_delete(struct tcf_proto *tp, void *arg, bool *last) { return -EOPNOTSUPP; } @@ -225,16 +208,16 @@ static void mall_walk(struct tcf_proto *tp, struct tcf_walker *arg) if (arg->count < arg->skip) goto skip; - if (arg->fn(tp, (unsigned long) head, arg) < 0) + if (arg->fn(tp, head, arg) < 0) arg->stop = 1; skip: arg->count++; } -static int mall_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int mall_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct cls_mall_head *head = (struct cls_mall_head *) fh; + struct cls_mall_head *head = fh; struct nlattr *nest; if (!head) @@ -268,6 +251,14 @@ nla_put_failure: return -1; } +static void mall_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct cls_mall_head *head = fh; + + if (head && head->res.classid == classid) + head->res.class = cl; +} + static struct tcf_proto_ops cls_mall_ops __read_mostly = { .kind = "matchall", .classify = mall_classify, @@ -278,6 +269,7 @@ static struct tcf_proto_ops cls_mall_ops __read_mostly = { .delete = mall_delete, .walk = mall_walk, .dump = mall_dump, + .bind_class = mall_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c index d63d5502ee02..9ddde65915d2 100644 --- a/net/sched/cls_route.c +++ b/net/sched/cls_route.c @@ -113,7 +113,7 @@ static inline int route4_hash_wild(void) #define ROUTE4_APPLY_RESULT() \ { \ *res = f->res; \ - if (tcf_exts_is_available(&f->exts)) { \ + if (tcf_exts_has_actions(&f->exts)) { \ int r = tcf_exts_exec(skb, &f->exts, res); \ if (r < 0) { \ dont_cache = 1; \ @@ -216,7 +216,7 @@ static inline u32 from_hash(u32 id) return 16 + (id & 0xF); } -static unsigned long route4_get(struct tcf_proto *tp, u32 handle) +static void *route4_get(struct tcf_proto *tp, u32 handle) { struct route4_head *head = rtnl_dereference(tp->root); struct route4_bucket *b; @@ -225,11 +225,11 @@ static unsigned long route4_get(struct tcf_proto *tp, u32 handle) h1 = to_hash(handle); if (h1 > 256) - return 0; + return NULL; h2 = from_hash(handle >> 16); if (h2 > 32) - return 0; + return NULL; b = rtnl_dereference(head->table[h1]); if (b) { @@ -237,9 +237,9 @@ static unsigned long route4_get(struct tcf_proto *tp, u32 handle) f; f = rtnl_dereference(f->next)) if (f->handle == handle) - return (unsigned long)f; + return f; } - return 0; + return NULL; } static int route4_init(struct tcf_proto *tp) @@ -294,10 +294,10 @@ static void route4_destroy(struct tcf_proto *tp) kfree_rcu(head, rcu); } -static int route4_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int route4_delete(struct tcf_proto *tp, void *arg, bool *last) { struct route4_head *head = rtnl_dereference(tp->root); - struct route4_filter *f = (struct route4_filter *)arg; + struct route4_filter *f = arg; struct route4_filter __rcu **fp; struct route4_filter *nf; struct route4_bucket *b; @@ -372,37 +372,32 @@ static int route4_set_parms(struct net *net, struct tcf_proto *tp, struct route4_filter *fp; unsigned int h1; struct route4_bucket *b; - struct tcf_exts e; int err; - err = tcf_exts_init(&e, TCA_ROUTE4_ACT, TCA_ROUTE4_POLICE); + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr); if (err < 0) return err; - err = tcf_exts_validate(net, tp, tb, est, &e, ovr); - if (err < 0) - goto errout; - err = -EINVAL; if (tb[TCA_ROUTE4_TO]) { if (new && handle & 0x8000) - goto errout; + return -EINVAL; to = nla_get_u32(tb[TCA_ROUTE4_TO]); if (to > 0xFF) - goto errout; + return -EINVAL; nhandle = to; } if (tb[TCA_ROUTE4_FROM]) { if (tb[TCA_ROUTE4_IIF]) - goto errout; + return -EINVAL; id = nla_get_u32(tb[TCA_ROUTE4_FROM]); if (id > 0xFF) - goto errout; + return -EINVAL; nhandle |= id << 16; } else if (tb[TCA_ROUTE4_IIF]) { id = nla_get_u32(tb[TCA_ROUTE4_IIF]); if (id > 0x7FFF) - goto errout; + return -EINVAL; nhandle |= (id | 0x8000) << 16; } else nhandle |= 0xFFFF << 16; @@ -410,27 +405,25 @@ static int route4_set_parms(struct net *net, struct tcf_proto *tp, if (handle && new) { nhandle |= handle & 0x7F00; if (nhandle != handle) - goto errout; + return -EINVAL; } h1 = to_hash(nhandle); b = rtnl_dereference(head->table[h1]); if (!b) { - err = -ENOBUFS; b = kzalloc(sizeof(struct route4_bucket), GFP_KERNEL); if (b == NULL) - goto errout; + return -ENOBUFS; rcu_assign_pointer(head->table[h1], b); } else { unsigned int h2 = from_hash(nhandle >> 16); - err = -EEXIST; for (fp = rtnl_dereference(b->ht[h2]); fp; fp = rtnl_dereference(fp->next)) if (fp->handle == f->handle) - goto errout; + return -EEXIST; } if (tb[TCA_ROUTE4_TO]) @@ -450,17 +443,12 @@ static int route4_set_parms(struct net *net, struct tcf_proto *tp, tcf_bind_filter(tp, &f->res, base); } - tcf_exts_change(tp, &f->exts, &e); - return 0; -errout: - tcf_exts_destroy(&e); - return err; } static int route4_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg, bool ovr) + struct nlattr **tca, void **arg, bool ovr) { struct route4_head *head = rtnl_dereference(tp->root); struct route4_filter __rcu **fp; @@ -479,7 +467,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb, if (err < 0) return err; - fold = (struct route4_filter *)*arg; + fold = *arg; if (fold && handle && fold->handle != handle) return -EINVAL; @@ -537,7 +525,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb, } route4_reset_fastmap(head); - *arg = (unsigned long)f; + *arg = f; if (fold) { tcf_unbind_filter(tp, &fold->res); call_rcu(&fold->rcu, route4_delete_filter); @@ -576,7 +564,7 @@ static void route4_walk(struct tcf_proto *tp, struct tcf_walker *arg) arg->count++; continue; } - if (arg->fn(tp, (unsigned long)f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; return; } @@ -587,10 +575,10 @@ static void route4_walk(struct tcf_proto *tp, struct tcf_walker *arg) } } -static int route4_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int route4_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct route4_filter *f = (struct route4_filter *)fh; + struct route4_filter *f = fh; struct nlattr *nest; u32 id; @@ -636,6 +624,14 @@ nla_put_failure: return -1; } +static void route4_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct route4_filter *f = fh; + + if (f && f->res.classid == classid) + f->res.class = cl; +} + static struct tcf_proto_ops cls_route4_ops __read_mostly = { .kind = "route", .classify = route4_classify, @@ -646,6 +642,7 @@ static struct tcf_proto_ops cls_route4_ops __read_mostly = { .delete = route4_delete, .walk = route4_walk, .dump = route4_dump, + .bind_class = route4_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h index 0d9d07798699..98c05db85bcb 100644 --- a/net/sched/cls_rsvp.h +++ b/net/sched/cls_rsvp.h @@ -248,7 +248,7 @@ static void rsvp_replace(struct tcf_proto *tp, struct rsvp_filter *n, u32 h) BUG_ON(1); } -static unsigned long rsvp_get(struct tcf_proto *tp, u32 handle) +static void *rsvp_get(struct tcf_proto *tp, u32 handle) { struct rsvp_head *head = rtnl_dereference(tp->root); struct rsvp_session *s; @@ -257,17 +257,17 @@ static unsigned long rsvp_get(struct tcf_proto *tp, u32 handle) unsigned int h2 = (handle >> 8) & 0xFF; if (h2 > 16) - return 0; + return NULL; for (s = rtnl_dereference(head->ht[h1]); s; s = rtnl_dereference(s->next)) { for (f = rtnl_dereference(s->ht[h2]); f; f = rtnl_dereference(f->next)) { if (f->handle == handle) - return (unsigned long)f; + return f; } } - return 0; + return NULL; } static int rsvp_init(struct tcf_proto *tp) @@ -328,10 +328,10 @@ static void rsvp_destroy(struct tcf_proto *tp) kfree_rcu(data, rcu); } -static int rsvp_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int rsvp_delete(struct tcf_proto *tp, void *arg, bool *last) { struct rsvp_head *head = rtnl_dereference(tp->root); - struct rsvp_filter *nfp, *f = (struct rsvp_filter *)arg; + struct rsvp_filter *nfp, *f = arg; struct rsvp_filter __rcu **fp; unsigned int h = f->handle; struct rsvp_session __rcu **sp; @@ -464,7 +464,7 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg, bool ovr) + void **arg, bool ovr) { struct rsvp_head *data = rtnl_dereference(tp->root); struct rsvp_filter *f, *nfp; @@ -493,7 +493,7 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb, if (err < 0) goto errout2; - f = (struct rsvp_filter *)*arg; + f = *arg; if (f) { /* Node exists: adjust only classid */ struct rsvp_filter *n; @@ -518,7 +518,7 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb, tcf_bind_filter(tp, &n->res, base); } - tcf_exts_change(tp, &n->exts, &e); + tcf_exts_change(&n->exts, &e); rsvp_replace(tp, n, handle); return 0; } @@ -591,7 +591,7 @@ insert: if (f->tunnelhdr == 0) tcf_bind_filter(tp, &f->res, base); - tcf_exts_change(tp, &f->exts, &e); + tcf_exts_change(&f->exts, &e); fp = &s->ht[h2]; for (nfp = rtnl_dereference(*fp); nfp; @@ -604,7 +604,7 @@ insert: RCU_INIT_POINTER(f->next, nfp); rcu_assign_pointer(*fp, f); - *arg = (unsigned long)f; + *arg = f; return 0; } } @@ -663,7 +663,7 @@ static void rsvp_walk(struct tcf_proto *tp, struct tcf_walker *arg) arg->count++; continue; } - if (arg->fn(tp, (unsigned long)f, arg) < 0) { + if (arg->fn(tp, f, arg) < 0) { arg->stop = 1; return; } @@ -674,10 +674,10 @@ static void rsvp_walk(struct tcf_proto *tp, struct tcf_walker *arg) } } -static int rsvp_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int rsvp_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct rsvp_filter *f = (struct rsvp_filter *)fh; + struct rsvp_filter *f = fh; struct rsvp_session *s; struct nlattr *nest; struct tc_rsvp_pinfo pinfo; @@ -723,6 +723,14 @@ nla_put_failure: return -1; } +static void rsvp_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct rsvp_filter *f = fh; + + if (f && f->res.classid == classid) + f->res.class = cl; +} + static struct tcf_proto_ops RSVP_OPS __read_mostly = { .kind = RSVP_ID, .classify = rsvp_classify, @@ -733,6 +741,7 @@ static struct tcf_proto_ops RSVP_OPS __read_mostly = { .delete = rsvp_delete, .walk = rsvp_walk, .dump = rsvp_dump, + .bind_class = rsvp_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c index 8a8a58357c39..14a7e08b2fa9 100644 --- a/net/sched/cls_tcindex.c +++ b/net/sched/cls_tcindex.c @@ -52,7 +52,7 @@ struct tcindex_data { static inline int tcindex_filter_is_set(struct tcindex_filter_result *r) { - return tcf_exts_is_predicative(&r->exts) || r->res.classid; + return tcf_exts_has_actions(&r->exts) || r->res.classid; } static struct tcindex_filter_result *tcindex_lookup(struct tcindex_data *p, @@ -104,16 +104,16 @@ static int tcindex_classify(struct sk_buff *skb, const struct tcf_proto *tp, } -static unsigned long tcindex_get(struct tcf_proto *tp, u32 handle) +static void *tcindex_get(struct tcf_proto *tp, u32 handle) { struct tcindex_data *p = rtnl_dereference(tp->root); struct tcindex_filter_result *r; pr_debug("tcindex_get(tp %p,handle 0x%08x)\n", tp, handle); if (p->perfect && handle >= p->alloc_hash) - return 0; + return NULL; r = tcindex_lookup(p, handle); - return r && tcindex_filter_is_set(r) ? (unsigned long) r : 0UL; + return r && tcindex_filter_is_set(r) ? r : NULL; } static int tcindex_init(struct tcf_proto *tp) @@ -150,14 +150,14 @@ static void tcindex_destroy_fexts(struct rcu_head *head) kfree(f); } -static int tcindex_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int tcindex_delete(struct tcf_proto *tp, void *arg, bool *last) { struct tcindex_data *p = rtnl_dereference(tp->root); - struct tcindex_filter_result *r = (struct tcindex_filter_result *) arg; + struct tcindex_filter_result *r = arg; struct tcindex_filter __rcu **walk; struct tcindex_filter *f = NULL; - pr_debug("tcindex_delete(tp %p,arg 0x%lx),p %p\n", tp, arg, p); + pr_debug("tcindex_delete(tp %p,arg %p),p %p\n", tp, arg, p); if (p->perfect) { if (!r->res.class) return -ENOENT; @@ -192,8 +192,7 @@ found: } static int tcindex_destroy_element(struct tcf_proto *tp, - unsigned long arg, - struct tcf_walker *walker) + void *arg, struct tcf_walker *walker) { bool last; @@ -419,9 +418,9 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, } if (old_r) - tcf_exts_change(tp, &r->exts, &e); + tcf_exts_change(&r->exts, &e); else - tcf_exts_change(tp, &cr.exts, &e); + tcf_exts_change(&cr.exts, &e); if (old_r && old_r != r) { err = tcindex_filter_result_init(old_r); @@ -439,7 +438,7 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct tcindex_filter *nfp; struct tcindex_filter __rcu **fp; - tcf_exts_change(tp, &f->result.exts, &r->exts); + tcf_exts_change(&f->result.exts, &r->exts); fp = cp->h + (handle % cp->hash); for (nfp = rtnl_dereference(*fp); @@ -471,17 +470,17 @@ errout: static int tcindex_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg, bool ovr) + struct nlattr **tca, void **arg, bool ovr) { struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_TCINDEX_MAX + 1]; struct tcindex_data *p = rtnl_dereference(tp->root); - struct tcindex_filter_result *r = (struct tcindex_filter_result *) *arg; + struct tcindex_filter_result *r = *arg; int err; pr_debug("tcindex_change(tp %p,handle 0x%08x,tca %p,arg %p),opt %p," - "p %p,r %p,*arg 0x%lx\n", - tp, handle, tca, arg, opt, p, r, arg ? *arg : 0L); + "p %p,r %p,*arg %p\n", + tp, handle, tca, arg, opt, p, r, arg ? *arg : NULL); if (!opt) return 0; @@ -506,9 +505,7 @@ static void tcindex_walk(struct tcf_proto *tp, struct tcf_walker *walker) if (!p->perfect[i].res.class) continue; if (walker->count >= walker->skip) { - if (walker->fn(tp, - (unsigned long) (p->perfect+i), walker) - < 0) { + if (walker->fn(tp, p->perfect + i, walker) < 0) { walker->stop = 1; return; } @@ -522,8 +519,7 @@ static void tcindex_walk(struct tcf_proto *tp, struct tcf_walker *walker) for (f = rtnl_dereference(p->h[i]); f; f = next) { next = rtnl_dereference(f->next); if (walker->count >= walker->skip) { - if (walker->fn(tp, (unsigned long) &f->result, - walker) < 0) { + if (walker->fn(tp, &f->result, walker) < 0) { walker->stop = 1; return; } @@ -548,14 +544,14 @@ static void tcindex_destroy(struct tcf_proto *tp) } -static int tcindex_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static int tcindex_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { struct tcindex_data *p = rtnl_dereference(tp->root); - struct tcindex_filter_result *r = (struct tcindex_filter_result *) fh; + struct tcindex_filter_result *r = fh; struct nlattr *nest; - pr_debug("tcindex_dump(tp %p,fh 0x%lx,skb %p,t %p),p %p,r %p\n", + pr_debug("tcindex_dump(tp %p,fh %p,skb %p,t %p),p %p,r %p\n", tp, fh, skb, t, p, r); pr_debug("p->perfect %p p->h %p\n", p->perfect, p->h); @@ -610,6 +606,14 @@ nla_put_failure: return -1; } +static void tcindex_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct tcindex_filter_result *r = fh; + + if (r && r->res.classid == classid) + r->res.class = cl; +} + static struct tcf_proto_ops cls_tcindex_ops __read_mostly = { .kind = "tcindex", .classify = tcindex_classify, @@ -620,6 +624,7 @@ static struct tcf_proto_ops cls_tcindex_ops __read_mostly = { .delete = tcindex_delete, .walk = tcindex_walk, .dump = tcindex_dump, + .bind_class = tcindex_bind_class, .owner = THIS_MODULE, }; diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 2d01195153e6..10b8d851fc6b 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -40,6 +40,8 @@ #include <linux/rtnetlink.h> #include <linux/skbuff.h> #include <linux/bitmap.h> +#include <linux/netdevice.h> +#include <linux/hash.h> #include <net/netlink.h> #include <net/act_api.h> #include <net/pkt_cls.h> @@ -92,6 +94,7 @@ struct tc_u_common { struct Qdisc *q; int refcnt; u32 hgenerator; + struct hlist_node hnode; struct rcu_head rcu; }; @@ -289,7 +292,7 @@ out: } -static unsigned long u32_get(struct tcf_proto *tp, u32 handle) +static void *u32_get(struct tcf_proto *tp, u32 handle) { struct tc_u_hnode *ht; struct tc_u_common *tp_c = tp->data; @@ -300,12 +303,12 @@ static unsigned long u32_get(struct tcf_proto *tp, u32 handle) ht = u32_lookup_ht(tp_c, TC_U32_HTID(handle)); if (!ht) - return 0; + return NULL; if (TC_U32_KEY(handle) == 0) - return (unsigned long)ht; + return ht; - return (unsigned long)u32_lookup_key(ht, handle); + return u32_lookup_key(ht, handle); } static u32 gen_new_htid(struct tc_u_common *tp_c) @@ -323,12 +326,40 @@ static u32 gen_new_htid(struct tc_u_common *tp_c) return i > 0 ? (tp_c->hgenerator|0x800)<<20 : 0; } +static struct hlist_head *tc_u_common_hash; + +#define U32_HASH_SHIFT 10 +#define U32_HASH_SIZE (1 << U32_HASH_SHIFT) + +static unsigned int tc_u_hash(const struct tcf_proto *tp) +{ + struct net_device *dev = tp->q->dev_queue->dev; + u32 qhandle = tp->q->handle; + int ifindex = dev->ifindex; + + return hash_64((u64)ifindex << 32 | qhandle, U32_HASH_SHIFT); +} + +static struct tc_u_common *tc_u_common_find(const struct tcf_proto *tp) +{ + struct tc_u_common *tc; + unsigned int h; + + h = tc_u_hash(tp); + hlist_for_each_entry(tc, &tc_u_common_hash[h], hnode) { + if (tc->q == tp->q) + return tc; + } + return NULL; +} + static int u32_init(struct tcf_proto *tp) { struct tc_u_hnode *root_ht; struct tc_u_common *tp_c; + unsigned int h; - tp_c = tp->q->u32_node; + tp_c = tc_u_common_find(tp); root_ht = kzalloc(sizeof(*root_ht), GFP_KERNEL); if (root_ht == NULL) @@ -345,7 +376,10 @@ static int u32_init(struct tcf_proto *tp) return -ENOBUFS; } tp_c->q = tp->q; - tp->q->u32_node = tp_c; + INIT_HLIST_NODE(&tp_c->hnode); + + h = tc_u_hash(tp); + hlist_add_head(&tp_c->hnode, &tc_u_common_hash[h]); } tp_c->refcnt++; @@ -431,43 +465,35 @@ static int u32_delete_key(struct tcf_proto *tp, struct tc_u_knode *key) static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_u32_offload u32_offload = {0}; - struct tc_to_netdev offload; - - offload.type = TC_SETUP_CLSU32; - offload.cls_u32 = &u32_offload; - - if (tc_should_offload(dev, tp, 0)) { - offload.cls_u32->command = TC_CLSU32_DELETE_KNODE; - offload.cls_u32->knode.handle = handle; - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, - &offload); - } + struct tc_cls_u32_offload cls_u32 = {}; + + if (!tc_should_offload(dev, 0)) + return; + + tc_cls_common_offload_init(&cls_u32.common, tp); + cls_u32.command = TC_CLSU32_DELETE_KNODE; + cls_u32.knode.handle = handle; + + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSU32, &cls_u32); } static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, u32 flags) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_u32_offload u32_offload = {0}; - struct tc_to_netdev offload; + struct tc_cls_u32_offload cls_u32 = {}; int err; - if (!tc_should_offload(dev, tp, flags)) + if (!tc_should_offload(dev, flags)) return tc_skip_sw(flags) ? -EINVAL : 0; - offload.type = TC_SETUP_CLSU32; - offload.cls_u32 = &u32_offload; - - offload.cls_u32->command = TC_CLSU32_NEW_HNODE; - offload.cls_u32->hnode.divisor = h->divisor; - offload.cls_u32->hnode.handle = h->handle; - offload.cls_u32->hnode.prio = h->prio; + tc_cls_common_offload_init(&cls_u32.common, tp); + cls_u32.command = TC_CLSU32_NEW_HNODE; + cls_u32.hnode.divisor = h->divisor; + cls_u32.hnode.handle = h->handle; + cls_u32.hnode.prio = h->prio; - err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, - &offload); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSU32, &cls_u32); if (tc_skip_sw(flags)) return err; @@ -477,56 +503,47 @@ static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_u32_offload u32_offload = {0}; - struct tc_to_netdev offload; + struct tc_cls_u32_offload cls_u32 = {}; - offload.type = TC_SETUP_CLSU32; - offload.cls_u32 = &u32_offload; + if (!tc_should_offload(dev, 0)) + return; - if (tc_should_offload(dev, tp, 0)) { - offload.cls_u32->command = TC_CLSU32_DELETE_HNODE; - offload.cls_u32->hnode.divisor = h->divisor; - offload.cls_u32->hnode.handle = h->handle; - offload.cls_u32->hnode.prio = h->prio; + tc_cls_common_offload_init(&cls_u32.common, tp); + cls_u32.command = TC_CLSU32_DELETE_HNODE; + cls_u32.hnode.divisor = h->divisor; + cls_u32.hnode.handle = h->handle; + cls_u32.hnode.prio = h->prio; - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, - &offload); - } + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSU32, &cls_u32); } static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, u32 flags) { struct net_device *dev = tp->q->dev_queue->dev; - struct tc_cls_u32_offload u32_offload = {0}; - struct tc_to_netdev offload; + struct tc_cls_u32_offload cls_u32 = {}; int err; - offload.type = TC_SETUP_CLSU32; - offload.cls_u32 = &u32_offload; - - if (!tc_should_offload(dev, tp, flags)) + if (!tc_should_offload(dev, flags)) return tc_skip_sw(flags) ? -EINVAL : 0; - offload.cls_u32->command = TC_CLSU32_REPLACE_KNODE; - offload.cls_u32->knode.handle = n->handle; - offload.cls_u32->knode.fshift = n->fshift; + tc_cls_common_offload_init(&cls_u32.common, tp); + cls_u32.command = TC_CLSU32_REPLACE_KNODE; + cls_u32.knode.handle = n->handle; + cls_u32.knode.fshift = n->fshift; #ifdef CONFIG_CLS_U32_MARK - offload.cls_u32->knode.val = n->val; - offload.cls_u32->knode.mask = n->mask; + cls_u32.knode.val = n->val; + cls_u32.knode.mask = n->mask; #else - offload.cls_u32->knode.val = 0; - offload.cls_u32->knode.mask = 0; + cls_u32.knode.val = 0; + cls_u32.knode.mask = 0; #endif - offload.cls_u32->knode.sel = &n->sel; - offload.cls_u32->knode.exts = &n->exts; + cls_u32.knode.sel = &n->sel; + cls_u32.knode.exts = &n->exts; if (n->ht_down) - offload.cls_u32->knode.link_handle = n->ht_down->handle; + cls_u32.knode.link_handle = n->ht_down->handle; - err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, - tp->chain->index, tp->protocol, - &offload); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSU32, &cls_u32); if (!err) n->flags |= TCA_CLS_FLAGS_IN_HW; @@ -602,7 +619,7 @@ static void u32_destroy(struct tcf_proto *tp) if (--tp_c->refcnt == 0) { struct tc_u_hnode *ht; - tp->q->u32_node = NULL; + hlist_del(&tp_c->hnode); for (ht = rtnl_dereference(tp_c->hlist); ht; @@ -622,9 +639,9 @@ static void u32_destroy(struct tcf_proto *tp) tp->data = NULL; } -static int u32_delete(struct tcf_proto *tp, unsigned long arg, bool *last) +static int u32_delete(struct tcf_proto *tp, void *arg, bool *last) { - struct tc_u_hnode *ht = (struct tc_u_hnode *)arg; + struct tc_u_hnode *ht = arg; struct tc_u_hnode *root_ht = rtnl_dereference(tp->root); struct tc_u_common *tp_c = tp->data; int ret = 0; @@ -723,29 +740,24 @@ static int u32_set_parms(struct net *net, struct tcf_proto *tp, struct tc_u_knode *n, struct nlattr **tb, struct nlattr *est, bool ovr) { - struct tcf_exts e; int err; - err = tcf_exts_init(&e, TCA_U32_ACT, TCA_U32_POLICE); + err = tcf_exts_validate(net, tp, tb, est, &n->exts, ovr); if (err < 0) return err; - err = tcf_exts_validate(net, tp, tb, est, &e, ovr); - if (err < 0) - goto errout; - err = -EINVAL; if (tb[TCA_U32_LINK]) { u32 handle = nla_get_u32(tb[TCA_U32_LINK]); struct tc_u_hnode *ht_down = NULL, *ht_old; if (TC_U32_KEY(handle)) - goto errout; + return -EINVAL; if (handle) { ht_down = u32_lookup_ht(ht->tp_c, handle); if (ht_down == NULL) - goto errout; + return -EINVAL; ht_down->refcnt++; } @@ -765,16 +777,11 @@ static int u32_set_parms(struct net *net, struct tcf_proto *tp, int ret; ret = tcf_change_indev(net, tb[TCA_U32_INDEV]); if (ret < 0) - goto errout; + return -EINVAL; n->ifindex = ret; } #endif - tcf_exts_change(tp, &n->exts, &e); - return 0; -errout: - tcf_exts_destroy(&e); - return err; } static void u32_replace_knode(struct tcf_proto *tp, struct tc_u_common *tp_c, @@ -858,7 +865,7 @@ static struct tc_u_knode *u32_init_knode(struct tcf_proto *tp, static int u32_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg, bool ovr) + struct nlattr **tca, void **arg, bool ovr) { struct tc_u_common *tp_c = tp->data; struct tc_u_hnode *ht; @@ -885,7 +892,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, return -EINVAL; } - n = (struct tc_u_knode *)*arg; + n = *arg; if (n) { struct tc_u_knode *new; @@ -952,7 +959,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, RCU_INIT_POINTER(ht->next, tp_c->hlist); rcu_assign_pointer(tp_c->hlist, ht); - *arg = (unsigned long)ht; + *arg = ht; return 0; } @@ -1047,7 +1054,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, RCU_INIT_POINTER(n->next, pins); rcu_assign_pointer(*ins, n); - *arg = (unsigned long)n; + *arg = n; return 0; } @@ -1081,7 +1088,7 @@ static void u32_walk(struct tcf_proto *tp, struct tcf_walker *arg) if (ht->prio != tp->prio) continue; if (arg->count >= arg->skip) { - if (arg->fn(tp, (unsigned long)ht, arg) < 0) { + if (arg->fn(tp, ht, arg) < 0) { arg->stop = 1; return; } @@ -1095,7 +1102,7 @@ static void u32_walk(struct tcf_proto *tp, struct tcf_walker *arg) arg->count++; continue; } - if (arg->fn(tp, (unsigned long)n, arg) < 0) { + if (arg->fn(tp, n, arg) < 0) { arg->stop = 1; return; } @@ -1105,10 +1112,18 @@ static void u32_walk(struct tcf_proto *tp, struct tcf_walker *arg) } } -static int u32_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, +static void u32_bind_class(void *fh, u32 classid, unsigned long cl) +{ + struct tc_u_knode *n = fh; + + if (n && n->res.classid == classid) + n->res.class = cl; +} + +static int u32_dump(struct net *net, struct tcf_proto *tp, void *fh, struct sk_buff *skb, struct tcmsg *t) { - struct tc_u_knode *n = (struct tc_u_knode *)fh; + struct tc_u_knode *n = fh; struct tc_u_hnode *ht_up, *ht_down; struct nlattr *nest; @@ -1122,7 +1137,7 @@ static int u32_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, goto nla_put_failure; if (TC_U32_KEY(n->handle) == 0) { - struct tc_u_hnode *ht = (struct tc_u_hnode *)fh; + struct tc_u_hnode *ht = fh; u32 divisor = ht->divisor + 1; if (nla_put_u32(skb, TCA_U32_DIVISOR, divisor)) @@ -1235,11 +1250,14 @@ static struct tcf_proto_ops cls_u32_ops __read_mostly = { .delete = u32_delete, .walk = u32_walk, .dump = u32_dump, + .bind_class = u32_bind_class, .owner = THIS_MODULE, }; static int __init init_u32(void) { + int i, ret; + pr_info("u32 classifier\n"); #ifdef CONFIG_CLS_U32_PERF pr_info(" Performance counters on\n"); @@ -1250,12 +1268,25 @@ static int __init init_u32(void) #ifdef CONFIG_NET_CLS_ACT pr_info(" Actions configured\n"); #endif - return register_tcf_proto_ops(&cls_u32_ops); + tc_u_common_hash = kvmalloc_array(U32_HASH_SIZE, + sizeof(struct hlist_head), + GFP_KERNEL); + if (!tc_u_common_hash) + return -ENOMEM; + + for (i = 0; i < U32_HASH_SIZE; i++) + INIT_HLIST_HEAD(&tc_u_common_hash[i]); + + ret = register_tcf_proto_ops(&cls_u32_ops); + if (ret) + kvfree(tc_u_common_hash); + return ret; } static void __exit exit_u32(void) { unregister_tcf_proto_ops(&cls_u32_ops); + kvfree(tc_u_common_hash); } module_init(init_u32) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index bd24a550e0f9..c6deb74e3d2f 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -35,13 +35,7 @@ #include <net/sock.h> #include <net/netlink.h> #include <net/pkt_sched.h> - -static int qdisc_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, u32 clid, - struct Qdisc *old, struct Qdisc *new); -static int tclass_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, struct Qdisc *q, - unsigned long cl, int event); +#include <net/pkt_cls.h> /* @@ -160,7 +154,7 @@ int register_qdisc(struct Qdisc_ops *qops) if (qops->cl_ops) { const struct Qdisc_class_ops *cops = qops->cl_ops; - if (!(cops->get && cops->put && cops->walk && cops->leaf)) + if (!(cops->find && cops->walk && cops->leaf)) goto out_einval; if (cops->tcf_block && !(cops->bind_tcf && cops->unbind_tcf)) @@ -286,9 +280,6 @@ static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) void qdisc_hash_add(struct Qdisc *q, bool invisible) { if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { - struct Qdisc *root = qdisc_dev(q)->qdisc; - - WARN_ON_ONCE(root == &noop_qdisc); ASSERT_RTNL(); hash_add_rcu(qdisc_dev(q)->qdisc_hash, &q->hash, q->handle); if (invisible) @@ -330,12 +321,11 @@ static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) if (cops == NULL) return NULL; - cl = cops->get(p, classid); + cl = cops->find(p, classid); if (cl == 0) return NULL; leaf = cops->leaf(p, cl); - cops->put(p, cl); return leaf; } @@ -624,14 +614,10 @@ EXPORT_SYMBOL(qdisc_watchdog_cancel); static struct hlist_head *qdisc_class_hash_alloc(unsigned int n) { - unsigned int size = n * sizeof(struct hlist_head), i; struct hlist_head *h; + unsigned int i; - if (size <= PAGE_SIZE) - h = kmalloc(size, GFP_KERNEL); - else - h = (struct hlist_head *) - __get_free_pages(GFP_KERNEL, get_order(size)); + h = kvmalloc_array(n, sizeof(struct hlist_head), GFP_KERNEL); if (h != NULL) { for (i = 0; i < n; i++) @@ -640,16 +626,6 @@ static struct hlist_head *qdisc_class_hash_alloc(unsigned int n) return h; } -static void qdisc_class_hash_free(struct hlist_head *h, unsigned int n) -{ - unsigned int size = n * sizeof(struct hlist_head); - - if (size <= PAGE_SIZE) - kfree(h); - else - free_pages((unsigned long)h, get_order(size)); -} - void qdisc_class_hash_grow(struct Qdisc *sch, struct Qdisc_class_hash *clhash) { struct Qdisc_class_common *cl; @@ -682,7 +658,7 @@ void qdisc_class_hash_grow(struct Qdisc *sch, struct Qdisc_class_hash *clhash) clhash->hashmask = nmask; sch_tree_unlock(sch); - qdisc_class_hash_free(ohash, osize); + kvfree(ohash); } EXPORT_SYMBOL(qdisc_class_hash_grow); @@ -702,7 +678,7 @@ EXPORT_SYMBOL(qdisc_class_hash_init); void qdisc_class_hash_destroy(struct Qdisc_class_hash *clhash) { - qdisc_class_hash_free(clhash->hash, clhash->hashsize); + kvfree(clhash->hash); } EXPORT_SYMBOL(qdisc_class_hash_destroy); @@ -752,6 +728,7 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n, const struct Qdisc_class_ops *cops; unsigned long cl; u32 parentid; + bool notify; int drops; if (n == 0 && len == 0) @@ -764,6 +741,13 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n, if (sch->flags & TCQ_F_NOPARENT) break; + /* Notify parent qdisc only if child qdisc becomes empty. + * + * If child was empty even before update then backlog + * counter is screwed and we skip notification because + * parent class is already passive. + */ + notify = !sch->q.qlen && !WARN_ON_ONCE(!n); /* TODO: perform the search on a per txq basis */ sch = qdisc_lookup(qdisc_dev(sch), TC_H_MAJ(parentid)); if (sch == NULL) { @@ -771,10 +755,9 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n, break; } cops = sch->ops->cl_ops; - if (cops->qlen_notify) { - cl = cops->get(sch, parentid); + if (notify && cops->qlen_notify) { + cl = cops->find(sch, parentid); cops->qlen_notify(sch, cl); - cops->put(sch, cl); } sch->q.qlen -= n; sch->qstats.backlog -= len; @@ -784,6 +767,111 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n, } EXPORT_SYMBOL(qdisc_tree_reduce_backlog); +static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, + u32 portid, u32 seq, u16 flags, int event) +{ + struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL; + struct gnet_stats_queue __percpu *cpu_qstats = NULL; + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct gnet_dump d; + struct qdisc_size_table *stab; + __u32 qlen; + + cond_resched(); + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_ifindex = qdisc_dev(q)->ifindex; + tcm->tcm_parent = clid; + tcm->tcm_handle = q->handle; + tcm->tcm_info = refcount_read(&q->refcnt); + if (nla_put_string(skb, TCA_KIND, q->ops->id)) + goto nla_put_failure; + if (q->ops->dump && q->ops->dump(q, skb) < 0) + goto nla_put_failure; + qlen = q->q.qlen; + + stab = rtnl_dereference(q->stab); + if (stab && qdisc_dump_stab(skb, stab) < 0) + goto nla_put_failure; + + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + NULL, &d, TCA_PAD) < 0) + goto nla_put_failure; + + if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) + goto nla_put_failure; + + if (qdisc_is_percpu_stats(q)) { + cpu_bstats = q->cpu_bstats; + cpu_qstats = q->cpu_qstats; + } + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(q), + &d, cpu_bstats, &q->bstats) < 0 || + gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 || + gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0) + goto nla_put_failure; + + if (gnet_stats_finish_copy(&d) < 0) + goto nla_put_failure; + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static bool tc_qdisc_dump_ignore(struct Qdisc *q, bool dump_invisible) +{ + if (q->flags & TCQ_F_BUILTIN) + return true; + if ((q->flags & TCQ_F_INVISIBLE) && !dump_invisible) + return true; + + return false; +} + +static int qdisc_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, u32 clid, + struct Qdisc *old, struct Qdisc *new) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (old && !tc_qdisc_dump_ignore(old, false)) { + if (tc_fill_qdisc(skb, old, clid, portid, n->nlmsg_seq, + 0, RTM_DELQDISC) < 0) + goto err_out; + } + if (new && !tc_qdisc_dump_ignore(new, false)) { + if (tc_fill_qdisc(skb, new, clid, portid, n->nlmsg_seq, + old ? NLM_F_REPLACE : 0, RTM_NEWQDISC) < 0) + goto err_out; + } + + if (skb->len) + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + +err_out: + kfree_skb(skb); + return -EINVAL; +} + static void notify_and_destroy(struct net *net, struct sk_buff *skb, struct nlmsghdr *n, u32 clid, struct Qdisc *old, struct Qdisc *new) @@ -839,7 +927,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, old = dev_graft_qdisc(dev_queue, new); if (new && i > 0) - refcount_inc(&new->refcnt); + qdisc_refcount_inc(new); if (!ingress) qdisc_destroy(old); @@ -850,7 +938,7 @@ skip: notify_and_destroy(net, skb, n, classid, dev->qdisc, new); if (new && !new->ops->attach) - refcount_inc(&new->refcnt); + qdisc_refcount_inc(new); dev->qdisc = new ? : &noop_qdisc; if (new && new->ops->attach) @@ -866,11 +954,11 @@ skip: err = -EOPNOTSUPP; if (cops && cops->graft) { - unsigned long cl = cops->get(parent, classid); - if (cl) { + unsigned long cl = cops->find(parent, classid); + + if (cl) err = cops->graft(parent, cl, new, &old); - cops->put(parent, cl); - } else + else err = -ENOENT; } if (!err) @@ -1259,7 +1347,7 @@ replay: if (q == p || (p && check_loop(q, p, 0))) return -ELOOP; - refcount_inc(&q->refcnt); + qdisc_refcount_inc(q); goto graft; } else { if (!q) @@ -1351,111 +1439,6 @@ graft: return 0; } -static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, - u32 portid, u32 seq, u16 flags, int event) -{ - struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL; - struct gnet_stats_queue __percpu *cpu_qstats = NULL; - struct tcmsg *tcm; - struct nlmsghdr *nlh; - unsigned char *b = skb_tail_pointer(skb); - struct gnet_dump d; - struct qdisc_size_table *stab; - __u32 qlen; - - cond_resched(); - nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); - if (!nlh) - goto out_nlmsg_trim; - tcm = nlmsg_data(nlh); - tcm->tcm_family = AF_UNSPEC; - tcm->tcm__pad1 = 0; - tcm->tcm__pad2 = 0; - tcm->tcm_ifindex = qdisc_dev(q)->ifindex; - tcm->tcm_parent = clid; - tcm->tcm_handle = q->handle; - tcm->tcm_info = refcount_read(&q->refcnt); - if (nla_put_string(skb, TCA_KIND, q->ops->id)) - goto nla_put_failure; - if (q->ops->dump && q->ops->dump(q, skb) < 0) - goto nla_put_failure; - qlen = q->q.qlen; - - stab = rtnl_dereference(q->stab); - if (stab && qdisc_dump_stab(skb, stab) < 0) - goto nla_put_failure; - - if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, - NULL, &d, TCA_PAD) < 0) - goto nla_put_failure; - - if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) - goto nla_put_failure; - - if (qdisc_is_percpu_stats(q)) { - cpu_bstats = q->cpu_bstats; - cpu_qstats = q->cpu_qstats; - } - - if (gnet_stats_copy_basic(qdisc_root_sleeping_running(q), - &d, cpu_bstats, &q->bstats) < 0 || - gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 || - gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0) - goto nla_put_failure; - - if (gnet_stats_finish_copy(&d) < 0) - goto nla_put_failure; - - nlh->nlmsg_len = skb_tail_pointer(skb) - b; - return skb->len; - -out_nlmsg_trim: -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static bool tc_qdisc_dump_ignore(struct Qdisc *q, bool dump_invisible) -{ - if (q->flags & TCQ_F_BUILTIN) - return true; - if ((q->flags & TCQ_F_INVISIBLE) && !dump_invisible) - return true; - - return false; -} - -static int qdisc_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, u32 clid, - struct Qdisc *old, struct Qdisc *new) -{ - struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - - skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); - if (!skb) - return -ENOBUFS; - - if (old && !tc_qdisc_dump_ignore(old, false)) { - if (tc_fill_qdisc(skb, old, clid, portid, n->nlmsg_seq, - 0, RTM_DELQDISC) < 0) - goto err_out; - } - if (new && !tc_qdisc_dump_ignore(new, false)) { - if (tc_fill_qdisc(skb, new, clid, portid, n->nlmsg_seq, - old ? NLM_F_REPLACE : 0, RTM_NEWQDISC) < 0) - goto err_out; - } - - if (skb->len) - return rtnetlink_send(skb, net, portid, RTNLGRP_TC, - n->nlmsg_flags & NLM_F_ECHO); - -err_out: - kfree_skb(skb); - return -EINVAL; -} - static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb, struct netlink_callback *cb, int *q_idx_p, int s_q_idx, bool recur, @@ -1568,7 +1551,161 @@ done: * Traffic classes manipulation. * ************************************************/ +static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, + unsigned long cl, + u32 portid, u32 seq, u16 flags, int event) +{ + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct gnet_dump d; + const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops; + + cond_resched(); + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_ifindex = qdisc_dev(q)->ifindex; + tcm->tcm_parent = q->handle; + tcm->tcm_handle = q->handle; + tcm->tcm_info = 0; + if (nla_put_string(skb, TCA_KIND, q->ops->id)) + goto nla_put_failure; + if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0) + goto nla_put_failure; + + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + NULL, &d, TCA_PAD) < 0) + goto nla_put_failure; + + if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) + goto nla_put_failure; + + if (gnet_stats_finish_copy(&d) < 0) + goto nla_put_failure; + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tclass_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct Qdisc *q, + unsigned long cl, int event) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, event) < 0) { + kfree_skb(skb); + return -EINVAL; + } + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +static int tclass_del_notify(struct net *net, + const struct Qdisc_class_ops *cops, + struct sk_buff *oskb, struct nlmsghdr *n, + struct Qdisc *q, unsigned long cl) +{ + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + struct sk_buff *skb; + int err = 0; + + if (!cops->delete) + return -EOPNOTSUPP; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, + RTM_DELTCLASS) < 0) { + kfree_skb(skb); + return -EINVAL; + } + + err = cops->delete(q, cl); + if (err) { + kfree_skb(skb); + return err; + } + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +#ifdef CONFIG_NET_CLS + +struct tcf_bind_args { + struct tcf_walker w; + u32 classid; + unsigned long cl; +}; + +static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg) +{ + struct tcf_bind_args *a = (void *)arg; + + if (tp->ops->bind_class) { + tcf_tree_lock(tp); + tp->ops->bind_class(n, a->classid, a->cl); + tcf_tree_unlock(tp); + } + return 0; +} + +static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, + unsigned long new_cl) +{ + const struct Qdisc_class_ops *cops = q->ops->cl_ops; + struct tcf_block *block; + struct tcf_chain *chain; + unsigned long cl; + + cl = cops->find(q, portid); + if (!cl) + return; + block = cops->tcf_block(q, cl); + if (!block) + return; + list_for_each_entry(chain, &block->chain_list, list) { + struct tcf_proto *tp; + + for (tp = rtnl_dereference(chain->filter_chain); + tp; tp = rtnl_dereference(tp->next)) { + struct tcf_bind_args arg = {}; + + arg.w.fn = tcf_node_bind; + arg.classid = clid; + arg.cl = new_cl; + tp->ops->walk(tp, &arg.w); + } + } +} + +#else + +static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, + unsigned long new_cl) +{ +} + +#endif static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack) @@ -1659,7 +1796,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, clid = TC_H_MAKE(qid, clid); if (clid) - cl = cops->get(q, clid); + cl = cops->find(q, clid); if (cl == 0) { err = -ENOENT; @@ -1674,12 +1811,9 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, goto out; break; case RTM_DELTCLASS: - err = -EOPNOTSUPP; - if (cops->delete) - err = cops->delete(q, cl); - if (err == 0) - tclass_notify(net, skb, n, q, cl, - RTM_DELTCLASS); + err = tclass_del_notify(net, cops, skb, n, q, cl); + /* Unbind the class with flilters with 0 */ + tc_bind_tclass(q, portid, clid, 0); goto out; case RTM_GETTCLASS: err = tclass_notify(net, skb, n, q, cl, RTM_NEWTCLASS); @@ -1694,83 +1828,16 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, err = -EOPNOTSUPP; if (cops->change) err = cops->change(q, clid, portid, tca, &new_cl); - if (err == 0) + if (err == 0) { tclass_notify(net, skb, n, q, new_cl, RTM_NEWTCLASS); - + /* We just create a new class, need to do reverse binding. */ + if (cl != new_cl) + tc_bind_tclass(q, portid, clid, new_cl); + } out: - if (cl) - cops->put(q, cl); - return err; } - -static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, - unsigned long cl, - u32 portid, u32 seq, u16 flags, int event) -{ - struct tcmsg *tcm; - struct nlmsghdr *nlh; - unsigned char *b = skb_tail_pointer(skb); - struct gnet_dump d; - const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops; - - cond_resched(); - nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); - if (!nlh) - goto out_nlmsg_trim; - tcm = nlmsg_data(nlh); - tcm->tcm_family = AF_UNSPEC; - tcm->tcm__pad1 = 0; - tcm->tcm__pad2 = 0; - tcm->tcm_ifindex = qdisc_dev(q)->ifindex; - tcm->tcm_parent = q->handle; - tcm->tcm_handle = q->handle; - tcm->tcm_info = 0; - if (nla_put_string(skb, TCA_KIND, q->ops->id)) - goto nla_put_failure; - if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0) - goto nla_put_failure; - - if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, - NULL, &d, TCA_PAD) < 0) - goto nla_put_failure; - - if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) - goto nla_put_failure; - - if (gnet_stats_finish_copy(&d) < 0) - goto nla_put_failure; - - nlh->nlmsg_len = skb_tail_pointer(skb) - b; - return skb->len; - -out_nlmsg_trim: -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - -static int tclass_notify(struct net *net, struct sk_buff *oskb, - struct nlmsghdr *n, struct Qdisc *q, - unsigned long cl, int event) -{ - struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - - skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); - if (!skb) - return -ENOBUFS; - - if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, event) < 0) { - kfree_skb(skb); - return -EINVAL; - } - - return rtnetlink_send(skb, net, portid, RTNLGRP_TC, - n->nlmsg_flags & NLM_F_ECHO); -} - struct qdisc_dump_args { struct qdisc_walker w; struct sk_buff *skb; @@ -1952,14 +2019,14 @@ static int __init pktsched_init(void) register_qdisc(&mq_qdisc_ops); register_qdisc(&noqueue_qdisc_ops); - rtnl_register(PF_UNSPEC, RTM_NEWQDISC, tc_modify_qdisc, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELQDISC, tc_get_qdisc, NULL, NULL); + rtnl_register(PF_UNSPEC, RTM_NEWQDISC, tc_modify_qdisc, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELQDISC, tc_get_qdisc, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETQDISC, tc_get_qdisc, tc_dump_qdisc, - NULL); - rtnl_register(PF_UNSPEC, RTM_NEWTCLASS, tc_ctl_tclass, NULL, NULL); - rtnl_register(PF_UNSPEC, RTM_DELTCLASS, tc_ctl_tclass, NULL, NULL); + 0); + rtnl_register(PF_UNSPEC, RTM_NEWTCLASS, tc_ctl_tclass, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELTCLASS, tc_ctl_tclass, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETTCLASS, tc_ctl_tclass, tc_dump_tclass, - NULL); + 0); return 0; } diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 572fe2584e48..c5fcdf1a58a0 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -41,6 +41,7 @@ #define VCC2FLOW(vcc) ((struct atm_flow_data *) ((vcc)->user_back)) struct atm_flow_data { + struct Qdisc_class_common common; struct Qdisc *q; /* FIFO, TBF, etc. */ struct tcf_proto __rcu *filter_list; struct tcf_block *block; @@ -49,7 +50,6 @@ struct atm_flow_data { struct sk_buff *skb); /* chaining */ struct atm_qdisc_data *parent; /* parent qdisc */ struct socket *sock; /* for closing */ - u32 classid; /* x:y type ID */ int ref; /* reference count */ struct gnet_stats_basic_packed bstats; struct gnet_stats_queue qstats; @@ -75,7 +75,7 @@ static inline struct atm_flow_data *lookup_flow(struct Qdisc *sch, u32 classid) struct atm_flow_data *flow; list_for_each_entry(flow, &p->flows, list) { - if (flow->classid == classid) + if (flow->common.classid == classid) return flow; } return NULL; @@ -108,23 +108,29 @@ static struct Qdisc *atm_tc_leaf(struct Qdisc *sch, unsigned long cl) return flow ? flow->q : NULL; } -static unsigned long atm_tc_get(struct Qdisc *sch, u32 classid) +static unsigned long atm_tc_find(struct Qdisc *sch, u32 classid) { struct atm_qdisc_data *p __maybe_unused = qdisc_priv(sch); struct atm_flow_data *flow; - pr_debug("atm_tc_get(sch %p,[qdisc %p],classid %x)\n", sch, p, classid); + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", __func__, sch, p, classid); flow = lookup_flow(sch, classid); - if (flow) - flow->ref++; - pr_debug("atm_tc_get: flow %p\n", flow); + pr_debug("%s: flow %p\n", __func__, flow); return (unsigned long)flow; } static unsigned long atm_tc_bind_filter(struct Qdisc *sch, unsigned long parent, u32 classid) { - return atm_tc_get(sch, classid); + struct atm_qdisc_data *p __maybe_unused = qdisc_priv(sch); + struct atm_flow_data *flow; + + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", __func__, sch, p, classid); + flow = lookup_flow(sch, classid); + if (flow) + flow->ref++; + pr_debug("%s: flow %p\n", __func__, flow); + return (unsigned long)flow; } /* @@ -234,7 +240,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, excess = NULL; else { excess = (struct atm_flow_data *) - atm_tc_get(sch, nla_get_u32(tb[TCA_ATM_EXCESS])); + atm_tc_find(sch, nla_get_u32(tb[TCA_ATM_EXCESS])); if (!excess) return -ENOENT; } @@ -262,10 +268,9 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, for (i = 1; i < 0x8000; i++) { classid = TC_H_MAKE(sch->handle, 0x8000 | i); - cl = atm_tc_get(sch, classid); + cl = atm_tc_find(sch, classid); if (!cl) break; - atm_tc_put(sch, cl); } } pr_debug("atm_tc_change: new id %x\n", classid); @@ -293,7 +298,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, flow->old_pop = flow->vcc->pop; flow->parent = p; flow->vcc->pop = sch_atm_pop; - flow->classid = classid; + flow->common.classid = classid; flow->ref = 1; flow->excess = excess; list_add(&flow->list, &p->link.list); @@ -305,8 +310,6 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, *arg = (unsigned long)flow; return 0; err_out: - if (excess) - atm_tc_put(sch, (unsigned long)excess); sockfd_put(sock); return error; } @@ -377,7 +380,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, result = TC_ACT_OK; /* be nice to gcc */ flow = NULL; if (TC_H_MAJ(skb->priority) != sch->handle || - !(flow = (struct atm_flow_data *)atm_tc_get(sch, skb->priority))) { + !(flow = (struct atm_flow_data *)atm_tc_find(sch, skb->priority))) { struct tcf_proto *fl; list_for_each_entry(flow, &p->flows, list) { @@ -549,7 +552,7 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt) p->link.vcc = NULL; p->link.sock = NULL; - p->link.classid = sch->handle; + p->link.common.classid = sch->handle; p->link.ref = 1; tasklet_init(&p->task, sch_atm_dequeue, (unsigned long)sch); return 0; @@ -572,8 +575,10 @@ static void atm_tc_destroy(struct Qdisc *sch) struct atm_flow_data *flow, *tmp; pr_debug("atm_tc_destroy(sch %p,[qdisc %p])\n", sch, p); - list_for_each_entry(flow, &p->flows, list) + list_for_each_entry(flow, &p->flows, list) { tcf_block_put(flow->block); + flow->block = NULL; + } list_for_each_entry_safe(flow, tmp, &p->flows, list) { if (flow->ref > 1) @@ -594,7 +599,7 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl, sch, p, flow, skb, tcm); if (list_empty(&flow->list)) return -EINVAL; - tcm->tcm_handle = flow->classid; + tcm->tcm_handle = flow->common.classid; tcm->tcm_info = flow->q->handle; nest = nla_nest_start(skb, TCA_OPTIONS); @@ -619,7 +624,7 @@ static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl, goto nla_put_failure; } if (flow->excess) { - if (nla_put_u32(skb, TCA_ATM_EXCESS, flow->classid)) + if (nla_put_u32(skb, TCA_ATM_EXCESS, flow->common.classid)) goto nla_put_failure; } else { if (nla_put_u32(skb, TCA_ATM_EXCESS, 0)) @@ -653,8 +658,7 @@ static int atm_tc_dump(struct Qdisc *sch, struct sk_buff *skb) static const struct Qdisc_class_ops atm_class_ops = { .graft = atm_tc_graft, .leaf = atm_tc_leaf, - .get = atm_tc_get, - .put = atm_tc_put, + .find = atm_tc_find, .change = atm_tc_change, .delete = atm_tc_delete, .walk = atm_tc_walk, diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 481036f6b54e..dcef97fa8047 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -129,7 +129,6 @@ struct cbq_class { struct tcf_proto __rcu *filter_list; struct tcf_block *block; - int refcnt; int filters; struct cbq_class *defaults[TC_PRIO_MAX + 1]; @@ -1139,6 +1138,13 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) struct tc_ratespec *r; int err; + qdisc_watchdog_init(&q->watchdog, sch); + hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); + q->delay_timer.function = cbq_undelay; + + if (!opt) + return -EINVAL; + err = nla_parse_nested(tb, TCA_CBQ_MAX, opt, cbq_policy, NULL); if (err < 0) return err; @@ -1155,7 +1161,6 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) if (err < 0) goto put_rtab; - q->link.refcnt = 1; q->link.sibling = &q->link; q->link.common.classid = sch->handle; q->link.qdisc = sch; @@ -1177,9 +1182,6 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) q->link.avpkt = q->link.allot/2; q->link.minidle = -0x7FFFFFFF; - qdisc_watchdog_init(&q->watchdog, sch); - hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); - q->delay_timer.function = cbq_undelay; q->toplevel = TC_CBQ_MAXLEVEL; q->now = psched_get_time(); @@ -1385,20 +1387,14 @@ static void cbq_qlen_notify(struct Qdisc *sch, unsigned long arg) { struct cbq_class *cl = (struct cbq_class *)arg; - if (cl->q->q.qlen == 0) - cbq_deactivate_class(cl); + cbq_deactivate_class(cl); } -static unsigned long cbq_get(struct Qdisc *sch, u32 classid) +static unsigned long cbq_find(struct Qdisc *sch, u32 classid) { struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = cbq_class_lookup(q, classid); - if (cl) { - cl->refcnt++; - return (unsigned long)cl; - } - return 0; + return (unsigned long)cbq_class_lookup(q, classid); } static void cbq_destroy_class(struct Qdisc *sch, struct cbq_class *cl) @@ -1431,8 +1427,10 @@ static void cbq_destroy(struct Qdisc *sch) * be bound to classes which have been destroyed already. --TGR '04 */ for (h = 0; h < q->clhash.hashsize; h++) { - hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) + hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { tcf_block_put(cl->block); + cl->block = NULL; + } } for (h = 0; h < q->clhash.hashsize; h++) { hlist_for_each_entry_safe(cl, next, &q->clhash.hash[h], @@ -1442,25 +1440,6 @@ static void cbq_destroy(struct Qdisc *sch) qdisc_class_hash_destroy(&q->clhash); } -static void cbq_put(struct Qdisc *sch, unsigned long arg) -{ - struct cbq_class *cl = (struct cbq_class *)arg; - - if (--cl->refcnt == 0) { -#ifdef CONFIG_NET_CLS_ACT - spinlock_t *root_lock = qdisc_root_sleeping_lock(sch); - struct cbq_sched_data *q = qdisc_priv(sch); - - spin_lock_bh(root_lock); - if (q->rx_class == cl) - q->rx_class = NULL; - spin_unlock_bh(root_lock); -#endif - - cbq_destroy_class(sch, cl); - } -} - static int cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **tca, unsigned long *arg) @@ -1607,7 +1586,6 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cl->R_tab = rtab; rtab = NULL; - cl->refcnt = 1; cl->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid); if (!cl->q) cl->q = &noop_qdisc; @@ -1688,12 +1666,7 @@ static int cbq_delete(struct Qdisc *sch, unsigned long arg) cbq_rmprio(q, cl); sch_tree_unlock(sch); - BUG_ON(--cl->refcnt == 0); - /* - * This shouldn't happen: we "hold" one cops->get() when called - * from tc_ctl_tclass; the destroy method is done from cops->put(). - */ - + cbq_destroy_class(sch, cl); return 0; } @@ -1759,8 +1732,7 @@ static const struct Qdisc_class_ops cbq_class_ops = { .graft = cbq_graft, .leaf = cbq_leaf, .qlen_notify = cbq_qlen_notify, - .get = cbq_get, - .put = cbq_put, + .find = cbq_find, .change = cbq_change_class, .delete = cbq_delete, .walk = cbq_walk, diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index a413dc1c2098..2d0e8d4bdc29 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -20,7 +20,6 @@ struct drr_class { struct Qdisc_class_common common; - unsigned int refcnt; unsigned int filter_cnt; struct gnet_stats_basic_packed bstats; @@ -111,7 +110,6 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (cl == NULL) return -ENOBUFS; - cl->refcnt = 1; cl->common.classid = classid; cl->quantum = quantum; cl->qdisc = qdisc_create_dflt(sch->dev_queue, @@ -163,32 +161,15 @@ static int drr_delete_class(struct Qdisc *sch, unsigned long arg) drr_purge_queue(cl); qdisc_class_hash_remove(&q->clhash, &cl->common); - BUG_ON(--cl->refcnt == 0); - /* - * This shouldn't happen: we "hold" one cops->get() when called - * from tc_ctl_tclass; the destroy method is done from cops->put(). - */ - sch_tree_unlock(sch); - return 0; -} - -static unsigned long drr_get_class(struct Qdisc *sch, u32 classid) -{ - struct drr_class *cl = drr_find_class(sch, classid); - if (cl != NULL) - cl->refcnt++; - - return (unsigned long)cl; + drr_destroy_class(sch, cl); + return 0; } -static void drr_put_class(struct Qdisc *sch, unsigned long arg) +static unsigned long drr_search_class(struct Qdisc *sch, u32 classid) { - struct drr_class *cl = (struct drr_class *)arg; - - if (--cl->refcnt == 0) - drr_destroy_class(sch, cl); + return (unsigned long)drr_find_class(sch, classid); } static struct tcf_block *drr_tcf_block(struct Qdisc *sch, unsigned long cl) @@ -246,8 +227,7 @@ static void drr_qlen_notify(struct Qdisc *csh, unsigned long arg) { struct drr_class *cl = (struct drr_class *)arg; - if (cl->qdisc->q.qlen == 0) - list_del(&cl->alist); + list_del(&cl->alist); } static int drr_dump_class(struct Qdisc *sch, unsigned long arg, @@ -479,8 +459,7 @@ static void drr_destroy_qdisc(struct Qdisc *sch) static const struct Qdisc_class_ops drr_class_ops = { .change = drr_change_class, .delete = drr_delete_class, - .get = drr_get_class, - .put = drr_put_class, + .find = drr_search_class, .tcf_block = drr_tcf_block, .bind_tcf = drr_bind_tcf, .unbind_tcf = drr_unbind_tcf, diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 6d94fcc3592a..2836c80c7aa5 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -85,21 +85,21 @@ static struct Qdisc *dsmark_leaf(struct Qdisc *sch, unsigned long arg) return p->q; } -static unsigned long dsmark_get(struct Qdisc *sch, u32 classid) +static unsigned long dsmark_find(struct Qdisc *sch, u32 classid) { - pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", - __func__, sch, qdisc_priv(sch), classid); - return TC_H_MIN(classid) + 1; } static unsigned long dsmark_bind_filter(struct Qdisc *sch, unsigned long parent, u32 classid) { - return dsmark_get(sch, classid); + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", + __func__, sch, qdisc_priv(sch), classid); + + return dsmark_find(sch, classid); } -static void dsmark_put(struct Qdisc *sch, unsigned long cl) +static void dsmark_unbind_filter(struct Qdisc *sch, unsigned long cl) { } @@ -469,14 +469,13 @@ nla_put_failure: static const struct Qdisc_class_ops dsmark_class_ops = { .graft = dsmark_graft, .leaf = dsmark_leaf, - .get = dsmark_get, - .put = dsmark_put, + .find = dsmark_find, .change = dsmark_change, .delete = dsmark_delete, .walk = dsmark_walk, .tcf_block = dsmark_tcf_block, .bind_tcf = dsmark_bind_filter, - .unbind_tcf = dsmark_put, + .unbind_tcf = dsmark_unbind_filter, .dump = dsmark_dump_class, }; diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 337f2d6d81e4..de3b57ceca7b 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -491,10 +491,8 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt) if (!q->flows) return -ENOMEM; q->backlogs = kvzalloc(q->flows_cnt * sizeof(u32), GFP_KERNEL); - if (!q->backlogs) { - kvfree(q->flows); + if (!q->backlogs) return -ENOMEM; - } for (i = 0; i < q->flows_cnt; i++) { struct fq_codel_flow *flow = q->flows + i; @@ -579,7 +577,7 @@ static struct Qdisc *fq_codel_leaf(struct Qdisc *sch, unsigned long arg) return NULL; } -static unsigned long fq_codel_get(struct Qdisc *sch, u32 classid) +static unsigned long fq_codel_find(struct Qdisc *sch, u32 classid) { return 0; } @@ -592,7 +590,7 @@ static unsigned long fq_codel_bind(struct Qdisc *sch, unsigned long parent, return 0; } -static void fq_codel_put(struct Qdisc *q, unsigned long cl) +static void fq_codel_unbind(struct Qdisc *q, unsigned long cl) { } @@ -683,11 +681,10 @@ static void fq_codel_walk(struct Qdisc *sch, struct qdisc_walker *arg) static const struct Qdisc_class_ops fq_codel_class_ops = { .leaf = fq_codel_leaf, - .get = fq_codel_get, - .put = fq_codel_put, + .find = fq_codel_find, .tcf_block = fq_codel_tcf_block, .bind_tcf = fq_codel_bind, - .unbind_tcf = fq_codel_put, + .unbind_tcf = fq_codel_unbind, .dump = fq_codel_dump_class, .dump_stats = fq_codel_dump_class_stats, .walk = fq_codel_walk, diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 57ba406f1437..92237e75dbbc 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -29,6 +29,7 @@ #include <net/sch_generic.h> #include <net/pkt_sched.h> #include <net/dst.h> +#include <trace/events/qdisc.h> /* Qdisc to use by default */ const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops; @@ -126,7 +127,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, q->q.qlen--; } else skb = NULL; - return skb; + goto trace; } *validate = true; skb = q->skb_bad_txq; @@ -139,7 +140,8 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, q->q.qlen--; goto bulk; } - return NULL; + skb = NULL; + goto trace; } if (!(q->flags & TCQ_F_ONETXQUEUE) || !netif_xmit_frozen_or_stopped(txq)) @@ -151,6 +153,8 @@ bulk: else try_bulk_dequeue_skb_slow(q, skb, packets); } +trace: + trace_qdisc_dequeue(q, txq, *packets, skb); return skb; } @@ -785,7 +789,7 @@ static void attach_default_qdiscs(struct net_device *dev) dev->priv_flags & IFF_NO_QUEUE) { netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL); dev->qdisc = txq->qdisc_sleeping; - refcount_inc(&dev->qdisc->refcnt); + qdisc_refcount_inc(dev->qdisc); } else { qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT); if (qdisc) { diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index b52f74610dc7..daaf214e5201 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -110,7 +110,6 @@ enum hfsc_class_flags { struct hfsc_class { struct Qdisc_class_common cl_common; - unsigned int refcnt; /* usage count */ struct gnet_stats_basic_packed bstats; struct gnet_stats_queue qstats; @@ -829,28 +828,6 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) } } -static void -set_active(struct hfsc_class *cl, unsigned int len) -{ - if (cl->cl_flags & HFSC_RSC) - init_ed(cl, len); - if (cl->cl_flags & HFSC_FSC) - init_vf(cl, len); - -} - -static void -set_passive(struct hfsc_class *cl) -{ - if (cl->cl_flags & HFSC_RSC) - eltree_remove(cl); - - /* - * vttree is now handled in update_vf() so that update_vf(cl, 0, 0) - * needs to be called explicitly to remove a class from vttree. - */ -} - static unsigned int qdisc_peek_len(struct Qdisc *sch) { @@ -1067,7 +1044,6 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, hfsc_change_usc(cl, usc, 0); cl->cl_common.classid = classid; - cl->refcnt = 1; cl->sched = q; cl->cl_parent = parent; cl->qdisc = qdisc_create_dflt(sch->dev_queue, @@ -1123,13 +1099,9 @@ hfsc_delete_class(struct Qdisc *sch, unsigned long arg) hfsc_purge_queue(sch, cl); qdisc_class_hash_remove(&q->clhash, &cl->cl_common); - BUG_ON(--cl->refcnt == 0); - /* - * This shouldn't happen: we "hold" one cops->get() when called - * from tc_ctl_tclass; the destroy method is done from cops->put(). - */ - sch_tree_unlock(sch); + + hfsc_destroy_class(sch, cl); return 0; } @@ -1221,30 +1193,18 @@ hfsc_qlen_notify(struct Qdisc *sch, unsigned long arg) { struct hfsc_class *cl = (struct hfsc_class *)arg; - if (cl->qdisc->q.qlen == 0) { - update_vf(cl, 0, 0); - set_passive(cl); - } + /* vttree is now handled in update_vf() so that update_vf(cl, 0, 0) + * needs to be called explicitly to remove a class from vttree. + */ + update_vf(cl, 0, 0); + if (cl->cl_flags & HFSC_RSC) + eltree_remove(cl); } static unsigned long -hfsc_get_class(struct Qdisc *sch, u32 classid) -{ - struct hfsc_class *cl = hfsc_find_class(classid, sch); - - if (cl != NULL) - cl->refcnt++; - - return (unsigned long)cl; -} - -static void -hfsc_put_class(struct Qdisc *sch, unsigned long arg) +hfsc_search_class(struct Qdisc *sch, u32 classid) { - struct hfsc_class *cl = (struct hfsc_class *)arg; - - if (--cl->refcnt == 0) - hfsc_destroy_class(sch, cl); + return (unsigned long)hfsc_find_class(classid, sch); } static unsigned long @@ -1418,6 +1378,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) struct tc_hfsc_qopt *qopt; int err; + qdisc_watchdog_init(&q->watchdog, sch); + if (opt == NULL || nla_len(opt) < sizeof(*qopt)) return -EINVAL; qopt = nla_data(opt); @@ -1428,8 +1390,11 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) return err; q->eligible = RB_ROOT; + err = tcf_block_get(&q->root.block, &q->root.filter_list); + if (err) + return err; + q->root.cl_common.classid = sch->handle; - q->root.refcnt = 1; q->root.sched = q; q->root.qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle); @@ -1444,8 +1409,6 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) qdisc_class_hash_insert(&q->clhash, &q->root.cl_common); qdisc_class_hash_grow(sch, &q->clhash); - qdisc_watchdog_init(&q->watchdog, sch); - return 0; } @@ -1522,8 +1485,10 @@ hfsc_destroy_qdisc(struct Qdisc *sch) unsigned int i; for (i = 0; i < q->clhash.hashsize; i++) { - hlist_for_each_entry(cl, &q->clhash.hash[i], cl_common.hnode) + hlist_for_each_entry(cl, &q->clhash.hash[i], cl_common.hnode) { tcf_block_put(cl->block); + cl->block = NULL; + } } for (i = 0; i < q->clhash.hashsize; i++) { hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], @@ -1575,7 +1540,12 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) } if (cl->qdisc->q.qlen == 1) { - set_active(cl, qdisc_pkt_len(skb)); + unsigned int len = qdisc_pkt_len(skb); + + if (cl->cl_flags & HFSC_RSC) + init_ed(cl, len); + if (cl->cl_flags & HFSC_FSC) + init_vf(cl, len); /* * If this is the first packet, isolate the head so an eventual * head drop before the first dequeue operation has no chance @@ -1639,18 +1609,18 @@ hfsc_dequeue(struct Qdisc *sch) if (realtime) cl->cl_cumul += qdisc_pkt_len(skb); - if (cl->qdisc->q.qlen != 0) { - if (cl->cl_flags & HFSC_RSC) { + if (cl->cl_flags & HFSC_RSC) { + if (cl->qdisc->q.qlen != 0) { /* update ed */ next_len = qdisc_peek_len(cl->qdisc); if (realtime) update_ed(cl, next_len); else update_d(cl, next_len); + } else { + /* the class becomes passive */ + eltree_remove(cl); } - } else { - /* the class becomes passive */ - set_passive(cl); } qdisc_bstats_update(sch, skb); @@ -1666,8 +1636,7 @@ static const struct Qdisc_class_ops hfsc_class_ops = { .graft = hfsc_graft_class, .leaf = hfsc_class_leaf, .qlen_notify = hfsc_qlen_notify, - .get = hfsc_get_class, - .put = hfsc_put_class, + .find = hfsc_search_class, .bind_tcf = hfsc_bind_tcf, .unbind_tcf = hfsc_unbind_tcf, .tcf_block = hfsc_tcf_block, diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 51d3ba682af9..73a53c08091b 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -477,6 +477,9 @@ static void hhf_destroy(struct Qdisc *sch) kvfree(q->hhf_valid_bits[i]); } + if (!q->hh_flows) + return; + for (i = 0; i < HH_FLOWS_CNT; i++) { struct hh_flow_state *flow, *next; struct list_head *head = &q->hh_flows[i]; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 203286ab4427..7e148376ba52 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -107,7 +107,6 @@ struct htb_class { struct tcf_proto __rcu *filter_list; /* class attached filters */ struct tcf_block *block; int filter_cnt; - int refcnt; /* usage count of this class */ int level; /* our level (see above) */ unsigned int children; @@ -193,6 +192,10 @@ static inline struct htb_class *htb_find(u32 handle, struct Qdisc *sch) return container_of(clc, struct htb_class, common); } +static unsigned long htb_search(struct Qdisc *sch, u32 handle) +{ + return (unsigned long)htb_find(handle, sch); +} /** * htb_classify - classify a packet into class * @@ -1017,6 +1020,9 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) int err; int i; + qdisc_watchdog_init(&q->watchdog, sch); + INIT_WORK(&q->work, htb_work_func); + if (!opt) return -EINVAL; @@ -1041,8 +1047,6 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) for (i = 0; i < TC_HTB_NUMPRIO; i++) INIT_LIST_HEAD(q->drops + i); - qdisc_watchdog_init(&q->watchdog, sch); - INIT_WORK(&q->work, htb_work_func); qdisc_skb_head_init(&q->direct_queue); if (tb[TCA_HTB_DIRECT_QLEN]) @@ -1186,16 +1190,7 @@ static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg) { struct htb_class *cl = (struct htb_class *)arg; - if (cl->un.leaf.q->q.qlen == 0) - htb_deactivate(qdisc_priv(sch), cl); -} - -static unsigned long htb_get(struct Qdisc *sch, u32 classid) -{ - struct htb_class *cl = htb_find(classid, sch); - if (cl) - cl->refcnt++; - return (unsigned long)cl; + htb_deactivate(qdisc_priv(sch), cl); } static inline int htb_parent_last_child(struct htb_class *cl) @@ -1258,8 +1253,10 @@ static void htb_destroy(struct Qdisc *sch) tcf_block_put(q->block); for (i = 0; i < q->clhash.hashsize; i++) { - hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { tcf_block_put(cl->block); + cl->block = NULL; + } } for (i = 0; i < q->clhash.hashsize; i++) { hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], @@ -1315,22 +1312,10 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) if (last_child) htb_parent_to_leaf(q, cl, new_q); - BUG_ON(--cl->refcnt == 0); - /* - * This shouldn't happen: we "hold" one cops->get() when called - * from tc_ctl_tclass; the destroy method is done from cops->put(). - */ - sch_tree_unlock(sch); - return 0; -} - -static void htb_put(struct Qdisc *sch, unsigned long arg) -{ - struct htb_class *cl = (struct htb_class *)arg; - if (--cl->refcnt == 0) - htb_destroy_class(sch, cl); + htb_destroy_class(sch, cl); + return 0; } static int htb_change_class(struct Qdisc *sch, u32 classid, @@ -1421,7 +1406,6 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, } } - cl->refcnt = 1; cl->children = 0; INIT_LIST_HEAD(&cl->un.leaf.drop_list); RB_CLEAR_NODE(&cl->pq_node); @@ -1597,8 +1581,7 @@ static const struct Qdisc_class_ops htb_class_ops = { .graft = htb_graft, .leaf = htb_leaf, .qlen_notify = htb_qlen_notify, - .get = htb_get, - .put = htb_put, + .find = htb_search, .change = htb_change_class, .delete = htb_delete, .walk = htb_walk, diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index d8a9bebcab90..44de4ee51ce9 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -27,23 +27,18 @@ static struct Qdisc *ingress_leaf(struct Qdisc *sch, unsigned long arg) return NULL; } -static unsigned long ingress_get(struct Qdisc *sch, u32 classid) +static unsigned long ingress_find(struct Qdisc *sch, u32 classid) { return TC_H_MIN(classid) + 1; } -static bool ingress_cl_offload(u32 classid) -{ - return true; -} - static unsigned long ingress_bind_filter(struct Qdisc *sch, unsigned long parent, u32 classid) { - return ingress_get(sch, classid); + return ingress_find(sch, classid); } -static void ingress_put(struct Qdisc *sch, unsigned long cl) +static void ingress_unbind_filter(struct Qdisc *sch, unsigned long cl) { } @@ -99,13 +94,11 @@ nla_put_failure: static const struct Qdisc_class_ops ingress_class_ops = { .leaf = ingress_leaf, - .get = ingress_get, - .put = ingress_put, + .find = ingress_find, .walk = ingress_walk, .tcf_block = ingress_tcf_block, - .tcf_cl_offload = ingress_cl_offload, .bind_tcf = ingress_bind_filter, - .unbind_tcf = ingress_put, + .unbind_tcf = ingress_unbind_filter, }; static struct Qdisc_ops ingress_qdisc_ops __read_mostly = { @@ -123,7 +116,7 @@ struct clsact_sched_data { struct tcf_block *egress_block; }; -static unsigned long clsact_get(struct Qdisc *sch, u32 classid) +static unsigned long clsact_find(struct Qdisc *sch, u32 classid) { switch (TC_H_MIN(classid)) { case TC_H_MIN(TC_H_MIN_INGRESS): @@ -134,15 +127,10 @@ static unsigned long clsact_get(struct Qdisc *sch, u32 classid) } } -static bool clsact_cl_offload(u32 classid) -{ - return TC_H_MIN(classid) == TC_H_MIN(TC_H_MIN_INGRESS); -} - static unsigned long clsact_bind_filter(struct Qdisc *sch, unsigned long parent, u32 classid) { - return clsact_get(sch, classid); + return clsact_find(sch, classid); } static struct tcf_block *clsact_tcf_block(struct Qdisc *sch, unsigned long cl) @@ -194,13 +182,11 @@ static void clsact_destroy(struct Qdisc *sch) static const struct Qdisc_class_ops clsact_class_ops = { .leaf = ingress_leaf, - .get = clsact_get, - .put = ingress_put, + .find = clsact_find, .walk = ingress_walk, .tcf_block = clsact_tcf_block, - .tcf_cl_offload = clsact_cl_offload, .bind_tcf = clsact_bind_filter, - .unbind_tcf = ingress_put, + .unbind_tcf = ingress_unbind_filter, }; static struct Qdisc_ops clsact_qdisc_ops __read_mostly = { diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c index cadfdd4f1e52..f3a3e507422b 100644 --- a/net/sched/sch_mq.c +++ b/net/sched/sch_mq.c @@ -165,7 +165,7 @@ static struct Qdisc *mq_leaf(struct Qdisc *sch, unsigned long cl) return dev_queue->qdisc_sleeping; } -static unsigned long mq_get(struct Qdisc *sch, u32 classid) +static unsigned long mq_find(struct Qdisc *sch, u32 classid) { unsigned int ntx = TC_H_MIN(classid); @@ -174,10 +174,6 @@ static unsigned long mq_get(struct Qdisc *sch, u32 classid) return ntx; } -static void mq_put(struct Qdisc *sch, unsigned long cl) -{ -} - static int mq_dump_class(struct Qdisc *sch, unsigned long cl, struct sk_buff *skb, struct tcmsg *tcm) { @@ -223,8 +219,7 @@ static const struct Qdisc_class_ops mq_class_ops = { .select_queue = mq_select_queue, .graft = mq_graft, .leaf = mq_leaf, - .get = mq_get, - .put = mq_put, + .find = mq_find, .walk = mq_walk, .dump = mq_dump_class, .dump_stats = mq_dump_class_stats, diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index e0c02725cd48..6bcdfe6e7b63 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -39,11 +39,9 @@ static void mqprio_destroy(struct Qdisc *sch) } if (priv->hw_offload && dev->netdev_ops->ndo_setup_tc) { - struct tc_mqprio_qopt offload = { 0 }; - struct tc_to_netdev tc = { .type = TC_SETUP_MQPRIO, - { .mqprio = &offload } }; + struct tc_mqprio_qopt mqprio = {}; - dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0, 0, &tc); + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_MQPRIO, &mqprio); } else { netdev_set_num_tc(dev, 0); } @@ -148,16 +146,14 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) * supplied and verified mapping */ if (qopt->hw) { - struct tc_mqprio_qopt offload = *qopt; - struct tc_to_netdev tc = { .type = TC_SETUP_MQPRIO, - { .mqprio = &offload } }; + struct tc_mqprio_qopt mqprio = *qopt; - err = dev->netdev_ops->ndo_setup_tc(dev, sch->handle, - 0, 0, &tc); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_MQPRIO, + &mqprio); if (err) return err; - priv->hw_offload = offload.hw; + priv->hw_offload = mqprio.hw; } else { netdev_set_num_tc(dev, qopt->num_tc); for (i = 0; i < qopt->num_tc; i++) @@ -281,7 +277,7 @@ static struct Qdisc *mqprio_leaf(struct Qdisc *sch, unsigned long cl) return dev_queue->qdisc_sleeping; } -static unsigned long mqprio_get(struct Qdisc *sch, u32 classid) +static unsigned long mqprio_find(struct Qdisc *sch, u32 classid) { struct net_device *dev = qdisc_dev(sch); unsigned int ntx = TC_H_MIN(classid); @@ -291,10 +287,6 @@ static unsigned long mqprio_get(struct Qdisc *sch, u32 classid) return ntx; } -static void mqprio_put(struct Qdisc *sch, unsigned long cl) -{ -} - static int mqprio_dump_class(struct Qdisc *sch, unsigned long cl, struct sk_buff *skb, struct tcmsg *tcm) { @@ -407,8 +399,7 @@ static void mqprio_walk(struct Qdisc *sch, struct qdisc_walker *arg) static const struct Qdisc_class_ops mqprio_class_ops = { .graft = mqprio_graft, .leaf = mqprio_leaf, - .get = mqprio_get, - .put = mqprio_put, + .find = mqprio_find, .walk = mqprio_walk, .dump = mqprio_dump_class, .dump_stats = mqprio_dump_class_stats, diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index f143b7bbaa0d..ff4fc3e0facd 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -257,12 +257,7 @@ static int multiq_init(struct Qdisc *sch, struct nlattr *opt) for (i = 0; i < q->max_bands; i++) q->queues[i] = &noop_qdisc; - err = multiq_tune(sch, opt); - - if (err) - kfree(q->queues); - - return err; + return multiq_tune(sch, opt); } static int multiq_dump(struct Qdisc *sch, struct sk_buff *skb) @@ -306,7 +301,7 @@ multiq_leaf(struct Qdisc *sch, unsigned long arg) return q->queues[band]; } -static unsigned long multiq_get(struct Qdisc *sch, u32 classid) +static unsigned long multiq_find(struct Qdisc *sch, u32 classid) { struct multiq_sched_data *q = qdisc_priv(sch); unsigned long band = TC_H_MIN(classid); @@ -319,11 +314,11 @@ static unsigned long multiq_get(struct Qdisc *sch, u32 classid) static unsigned long multiq_bind(struct Qdisc *sch, unsigned long parent, u32 classid) { - return multiq_get(sch, classid); + return multiq_find(sch, classid); } -static void multiq_put(struct Qdisc *q, unsigned long cl) +static void multiq_unbind(struct Qdisc *q, unsigned long cl) { } @@ -385,12 +380,11 @@ static struct tcf_block *multiq_tcf_block(struct Qdisc *sch, unsigned long cl) static const struct Qdisc_class_ops multiq_class_ops = { .graft = multiq_graft, .leaf = multiq_leaf, - .get = multiq_get, - .put = multiq_put, + .find = multiq_find, .walk = multiq_walk, .tcf_block = multiq_tcf_block, .bind_tcf = multiq_bind, - .unbind_tcf = multiq_put, + .unbind_tcf = multiq_unbind, .dump = multiq_dump_class, .dump_stats = multiq_dump_class_stats, }; diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 1b3dd6190e93..b1266e75ca43 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -933,11 +933,11 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt) struct netem_sched_data *q = qdisc_priv(sch); int ret; + qdisc_watchdog_init(&q->watchdog, sch); + if (!opt) return -EINVAL; - qdisc_watchdog_init(&q->watchdog, sch); - q->loss_model = CLG_RANDOM; ret = netem_change(sch, opt); if (ret) @@ -1096,15 +1096,11 @@ static struct Qdisc *netem_leaf(struct Qdisc *sch, unsigned long arg) return q->qdisc; } -static unsigned long netem_get(struct Qdisc *sch, u32 classid) +static unsigned long netem_find(struct Qdisc *sch, u32 classid) { return 1; } -static void netem_put(struct Qdisc *sch, unsigned long arg) -{ -} - static void netem_walk(struct Qdisc *sch, struct qdisc_walker *walker) { if (!walker->stop) { @@ -1120,8 +1116,7 @@ static void netem_walk(struct Qdisc *sch, struct qdisc_walker *walker) static const struct Qdisc_class_ops netem_class_ops = { .graft = netem_graft, .leaf = netem_leaf, - .get = netem_get, - .put = netem_put, + .find = netem_find, .walk = netem_walk, .dump = netem_dump_class, }; diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index e3e364cc9a70..f31b28f788c0 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -260,7 +260,7 @@ prio_leaf(struct Qdisc *sch, unsigned long arg) return q->queues[band]; } -static unsigned long prio_get(struct Qdisc *sch, u32 classid) +static unsigned long prio_find(struct Qdisc *sch, u32 classid) { struct prio_sched_data *q = qdisc_priv(sch); unsigned long band = TC_H_MIN(classid); @@ -272,11 +272,11 @@ static unsigned long prio_get(struct Qdisc *sch, u32 classid) static unsigned long prio_bind(struct Qdisc *sch, unsigned long parent, u32 classid) { - return prio_get(sch, classid); + return prio_find(sch, classid); } -static void prio_put(struct Qdisc *q, unsigned long cl) +static void prio_unbind(struct Qdisc *q, unsigned long cl) { } @@ -338,12 +338,11 @@ static struct tcf_block *prio_tcf_block(struct Qdisc *sch, unsigned long cl) static const struct Qdisc_class_ops prio_class_ops = { .graft = prio_graft, .leaf = prio_leaf, - .get = prio_get, - .put = prio_put, + .find = prio_find, .walk = prio_walk, .tcf_block = prio_tcf_block, .bind_tcf = prio_bind, - .unbind_tcf = prio_put, + .unbind_tcf = prio_unbind, .dump = prio_dump_class, .dump_stats = prio_dump_class_stats, }; diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 0e16dfda0bd7..cd661a7f81e6 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -132,7 +132,6 @@ struct qfq_aggregate; struct qfq_class { struct Qdisc_class_common common; - unsigned int refcnt; unsigned int filter_cnt; struct gnet_stats_basic_packed bstats; @@ -477,7 +476,6 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (cl == NULL) return -ENOBUFS; - cl->refcnt = 1; cl->common.classid = classid; cl->deficit = lmax; @@ -555,32 +553,15 @@ static int qfq_delete_class(struct Qdisc *sch, unsigned long arg) qfq_purge_queue(cl); qdisc_class_hash_remove(&q->clhash, &cl->common); - BUG_ON(--cl->refcnt == 0); - /* - * This shouldn't happen: we "hold" one cops->get() when called - * from tc_ctl_tclass; the destroy method is done from cops->put(). - */ - sch_tree_unlock(sch); - return 0; -} - -static unsigned long qfq_get_class(struct Qdisc *sch, u32 classid) -{ - struct qfq_class *cl = qfq_find_class(sch, classid); - if (cl != NULL) - cl->refcnt++; - - return (unsigned long)cl; + qfq_destroy_class(sch, cl); + return 0; } -static void qfq_put_class(struct Qdisc *sch, unsigned long arg) +static unsigned long qfq_search_class(struct Qdisc *sch, u32 classid) { - struct qfq_class *cl = (struct qfq_class *)arg; - - if (--cl->refcnt == 0) - qfq_destroy_class(sch, cl); + return (unsigned long)qfq_find_class(sch, classid); } static struct tcf_block *qfq_tcf_block(struct Qdisc *sch, unsigned long cl) @@ -1428,8 +1409,7 @@ static void qfq_qlen_notify(struct Qdisc *sch, unsigned long arg) struct qfq_sched *q = qdisc_priv(sch); struct qfq_class *cl = (struct qfq_class *)arg; - if (cl->qdisc->q.qlen == 0) - qfq_deactivate_class(q, cl); + qfq_deactivate_class(q, cl); } static int qfq_init_qdisc(struct Qdisc *sch, struct nlattr *opt) @@ -1511,8 +1491,7 @@ static void qfq_destroy_qdisc(struct Qdisc *sch) static const struct Qdisc_class_ops qfq_class_ops = { .change = qfq_change_class, .delete = qfq_delete_class, - .get = qfq_get_class, - .put = qfq_put_class, + .find = qfq_search_class, .tcf_block = qfq_tcf_block, .bind_tcf = qfq_bind_tcf, .unbind_tcf = qfq_unbind_tcf, diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index 11292adce412..93b9d70a9b28 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -311,15 +311,11 @@ static struct Qdisc *red_leaf(struct Qdisc *sch, unsigned long arg) return q->qdisc; } -static unsigned long red_get(struct Qdisc *sch, u32 classid) +static unsigned long red_find(struct Qdisc *sch, u32 classid) { return 1; } -static void red_put(struct Qdisc *sch, unsigned long arg) -{ -} - static void red_walk(struct Qdisc *sch, struct qdisc_walker *walker) { if (!walker->stop) { @@ -335,8 +331,7 @@ static void red_walk(struct Qdisc *sch, struct qdisc_walker *walker) static const struct Qdisc_class_ops red_class_ops = { .graft = red_graft, .leaf = red_leaf, - .get = red_get, - .put = red_put, + .find = red_find, .walk = red_walk, .dump = red_dump_class, }; diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c index 11fb6ec878d6..cc39e170b4aa 100644 --- a/net/sched/sch_sfb.c +++ b/net/sched/sch_sfb.c @@ -632,12 +632,12 @@ static struct Qdisc *sfb_leaf(struct Qdisc *sch, unsigned long arg) return q->qdisc; } -static unsigned long sfb_get(struct Qdisc *sch, u32 classid) +static unsigned long sfb_find(struct Qdisc *sch, u32 classid) { return 1; } -static void sfb_put(struct Qdisc *sch, unsigned long arg) +static void sfb_unbind(struct Qdisc *sch, unsigned long arg) { } @@ -683,14 +683,13 @@ static unsigned long sfb_bind(struct Qdisc *sch, unsigned long parent, static const struct Qdisc_class_ops sfb_class_ops = { .graft = sfb_graft, .leaf = sfb_leaf, - .get = sfb_get, - .put = sfb_put, + .find = sfb_find, .change = sfb_change_class, .delete = sfb_delete, .walk = sfb_walk, .tcf_block = sfb_tcf_block, .bind_tcf = sfb_bind, - .unbind_tcf = sfb_put, + .unbind_tcf = sfb_unbind, .dump = sfb_dump_class, }; diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index f80ea2cc5f1f..74ea863b8240 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -292,7 +292,7 @@ static inline void slot_queue_add(struct sfq_slot *slot, struct sk_buff *skb) slot->skblist_prev = skb; } -static unsigned int sfq_drop(struct Qdisc *sch) +static unsigned int sfq_drop(struct Qdisc *sch, struct sk_buff **to_free) { struct sfq_sched_data *q = qdisc_priv(sch); sfq_index x, d = q->cur_depth; @@ -310,9 +310,8 @@ drop: slot->backlog -= len; sfq_dec(q, x); sch->q.qlen--; - qdisc_qstats_drop(sch); qdisc_qstats_backlog_dec(sch, skb); - kfree_skb(skb); + qdisc_drop(skb, sch, to_free); return len; } @@ -360,7 +359,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) if (hash == 0) { if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; } hash--; @@ -437,6 +436,7 @@ congestion_drop: qdisc_drop(head, sch, to_free); slot_queue_add(slot, skb); + qdisc_tree_reduce_backlog(sch, 0, delta); return NET_XMIT_CN; } @@ -464,12 +464,14 @@ enqueue: return NET_XMIT_SUCCESS; qlen = slot->qlen; - dropped = sfq_drop(sch); + dropped = sfq_drop(sch, to_free); /* Return Congestion Notification only if we dropped a packet * from this flow. */ - if (qlen != slot->qlen) + if (qlen != slot->qlen) { + qdisc_tree_reduce_backlog(sch, 0, dropped - qdisc_pkt_len(skb)); return NET_XMIT_CN; + } /* As we dropped a packet, better let upper stack know this */ qdisc_tree_reduce_backlog(sch, 1, dropped); @@ -625,6 +627,8 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt) struct tc_sfq_qopt_v1 *ctl_v1 = NULL; unsigned int qlen, dropped = 0; struct red_parms *p = NULL; + struct sk_buff *to_free = NULL; + struct sk_buff *tail = NULL; if (opt->nla_len < nla_attr_size(sizeof(*ctl))) return -EINVAL; @@ -671,8 +675,13 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt) } qlen = sch->q.qlen; - while (sch->q.qlen > q->limit) - dropped += sfq_drop(sch); + while (sch->q.qlen > q->limit) { + dropped += sfq_drop(sch, &to_free); + if (!tail) + tail = to_free; + } + + rtnl_kfree_skbs(to_free, tail); qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); del_timer(&q->perturb_timer); @@ -713,13 +722,13 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt) int i; int err; + setup_deferrable_timer(&q->perturb_timer, sfq_perturbation, + (unsigned long)sch); + err = tcf_block_get(&q->block, &q->filter_list); if (err) return err; - setup_deferrable_timer(&q->perturb_timer, sfq_perturbation, - (unsigned long)sch); - for (i = 0; i < SFQ_MAX_DEPTH + 1; i++) { q->dep[i].next = i + SFQ_MAX_FLOWS; q->dep[i].prev = i + SFQ_MAX_FLOWS; @@ -805,7 +814,7 @@ static struct Qdisc *sfq_leaf(struct Qdisc *sch, unsigned long arg) return NULL; } -static unsigned long sfq_get(struct Qdisc *sch, u32 classid) +static unsigned long sfq_find(struct Qdisc *sch, u32 classid) { return 0; } @@ -818,7 +827,7 @@ static unsigned long sfq_bind(struct Qdisc *sch, unsigned long parent, return 0; } -static void sfq_put(struct Qdisc *q, unsigned long cl) +static void sfq_unbind(struct Qdisc *q, unsigned long cl) { } @@ -882,11 +891,10 @@ static void sfq_walk(struct Qdisc *sch, struct qdisc_walker *arg) static const struct Qdisc_class_ops sfq_class_ops = { .leaf = sfq_leaf, - .get = sfq_get, - .put = sfq_put, + .find = sfq_find, .tcf_block = sfq_tcf_block, .bind_tcf = sfq_bind, - .unbind_tcf = sfq_put, + .unbind_tcf = sfq_unbind, .dump = sfq_dump_class, .dump_stats = sfq_dump_class_stats, .walk = sfq_walk, diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index b2e4b6ad241a..120f4f365967 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -425,12 +425,13 @@ static int tbf_init(struct Qdisc *sch, struct nlattr *opt) { struct tbf_sched_data *q = qdisc_priv(sch); + qdisc_watchdog_init(&q->watchdog, sch); + q->qdisc = &noop_qdisc; + if (opt == NULL) return -EINVAL; q->t_c = ktime_get_ns(); - qdisc_watchdog_init(&q->watchdog, sch); - q->qdisc = &noop_qdisc; return tbf_change(sch, opt); } @@ -510,15 +511,11 @@ static struct Qdisc *tbf_leaf(struct Qdisc *sch, unsigned long arg) return q->qdisc; } -static unsigned long tbf_get(struct Qdisc *sch, u32 classid) +static unsigned long tbf_find(struct Qdisc *sch, u32 classid) { return 1; } -static void tbf_put(struct Qdisc *sch, unsigned long arg) -{ -} - static void tbf_walk(struct Qdisc *sch, struct qdisc_walker *walker) { if (!walker->stop) { @@ -534,8 +531,7 @@ static void tbf_walk(struct Qdisc *sch, struct qdisc_walker *walker) static const struct Qdisc_class_ops tbf_class_ops = { .graft = tbf_graft, .leaf = tbf_leaf, - .get = tbf_get, - .put = tbf_put, + .find = tbf_find, .walk = tbf_walk, .dump = tbf_dump_class, }; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 40ec83679d6e..dfb9651e818b 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -63,11 +63,11 @@ static void sctp_assoc_free_asconf_queue(struct sctp_association *asoc); /* 1st Level Abstractions. */ /* Initialize a new association from provided memory. */ -static struct sctp_association *sctp_association_init(struct sctp_association *asoc, - const struct sctp_endpoint *ep, - const struct sock *sk, - sctp_scope_t scope, - gfp_t gfp) +static struct sctp_association *sctp_association_init( + struct sctp_association *asoc, + const struct sctp_endpoint *ep, + const struct sock *sk, + enum sctp_scope scope, gfp_t gfp) { struct net *net = sock_net(sk); struct sctp_sock *sp; @@ -301,9 +301,8 @@ fail_init: /* Allocate and initialize a new association */ struct sctp_association *sctp_association_new(const struct sctp_endpoint *ep, - const struct sock *sk, - sctp_scope_t scope, - gfp_t gfp) + const struct sock *sk, + enum sctp_scope scope, gfp_t gfp) { struct sctp_association *asoc; @@ -797,7 +796,7 @@ void sctp_assoc_del_nonprimary_peers(struct sctp_association *asoc, */ void sctp_assoc_control_transport(struct sctp_association *asoc, struct sctp_transport *transport, - sctp_transport_cmd_t command, + enum sctp_transport_cmd command, sctp_sn_error_t error) { struct sctp_ulpevent *event; @@ -1022,11 +1021,11 @@ static void sctp_assoc_bh_rcv(struct work_struct *work) container_of(work, struct sctp_association, base.inqueue.immediate); struct net *net = sock_net(asoc->base.sk); + union sctp_subtype subtype; struct sctp_endpoint *ep; struct sctp_chunk *chunk; struct sctp_inq *inqueue; int state; - sctp_subtype_t subtype; int error = 0; /* The association should be held so we should be safe. */ @@ -1564,7 +1563,7 @@ void sctp_assoc_rwnd_decrease(struct sctp_association *asoc, unsigned int len) * local endpoint and the remote peer. */ int sctp_assoc_set_bind_addr_from_ep(struct sctp_association *asoc, - sctp_scope_t scope, gfp_t gfp) + enum sctp_scope scope, gfp_t gfp) { int flags; diff --git a/net/sctp/auth.c b/net/sctp/auth.c index e001b01b0e68..00667c50efa7 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -185,9 +185,9 @@ static int sctp_auth_compare_vectors(struct sctp_auth_bytes *vector1, * are called the two key vectors. */ static struct sctp_auth_bytes *sctp_auth_make_key_vector( - sctp_random_param_t *random, - sctp_chunks_param_t *chunks, - sctp_hmac_algo_param_t *hmacs, + struct sctp_random_param *random, + struct sctp_chunks_param *chunks, + struct sctp_hmac_algo_param *hmacs, gfp_t gfp) { struct sctp_auth_bytes *new; @@ -226,10 +226,9 @@ static struct sctp_auth_bytes *sctp_auth_make_local_vector( gfp_t gfp) { return sctp_auth_make_key_vector( - (sctp_random_param_t *)asoc->c.auth_random, - (sctp_chunks_param_t *)asoc->c.auth_chunks, - (sctp_hmac_algo_param_t *)asoc->c.auth_hmacs, - gfp); + (struct sctp_random_param *)asoc->c.auth_random, + (struct sctp_chunks_param *)asoc->c.auth_chunks, + (struct sctp_hmac_algo_param *)asoc->c.auth_hmacs, gfp); } /* Make a key vector based on peer's parameters */ diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c index 1ebc184a0e23..7df3704982f5 100644 --- a/net/sctp/bind_addr.c +++ b/net/sctp/bind_addr.c @@ -45,9 +45,9 @@ #include <net/sctp/sm.h> /* Forward declarations for internal helpers. */ -static int sctp_copy_one_addr(struct net *, struct sctp_bind_addr *, - union sctp_addr *, sctp_scope_t scope, gfp_t gfp, - int flags); +static int sctp_copy_one_addr(struct net *net, struct sctp_bind_addr *dest, + union sctp_addr *addr, enum sctp_scope scope, + gfp_t gfp, int flags); static void sctp_bind_addr_clean(struct sctp_bind_addr *); /* First Level Abstractions. */ @@ -57,7 +57,7 @@ static void sctp_bind_addr_clean(struct sctp_bind_addr *); */ int sctp_bind_addr_copy(struct net *net, struct sctp_bind_addr *dest, const struct sctp_bind_addr *src, - sctp_scope_t scope, gfp_t gfp, + enum sctp_scope scope, gfp_t gfp, int flags) { struct sctp_sockaddr_entry *addr; @@ -440,9 +440,8 @@ union sctp_addr *sctp_find_unmatch_addr(struct sctp_bind_addr *bp, /* Copy out addresses from the global local address list. */ static int sctp_copy_one_addr(struct net *net, struct sctp_bind_addr *dest, - union sctp_addr *addr, - sctp_scope_t scope, gfp_t gfp, - int flags) + union sctp_addr *addr, enum sctp_scope scope, + gfp_t gfp, int flags) { int error = 0; @@ -485,9 +484,10 @@ int sctp_is_any(struct sock *sk, const union sctp_addr *addr) } /* Is 'addr' valid for 'scope'? */ -int sctp_in_scope(struct net *net, const union sctp_addr *addr, sctp_scope_t scope) +int sctp_in_scope(struct net *net, const union sctp_addr *addr, + enum sctp_scope scope) { - sctp_scope_t addr_scope = sctp_scope(addr); + enum sctp_scope addr_scope = sctp_scope(addr); /* The unusable SCTP addresses will not be considered with * any defined scopes. @@ -545,7 +545,7 @@ int sctp_is_ep_boundall(struct sock *sk) ********************************************************************/ /* What is the scope of 'addr'? */ -sctp_scope_t sctp_scope(const union sctp_addr *addr) +enum sctp_scope sctp_scope(const union sctp_addr *addr) { struct sctp_af *af; diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index 1323d41e68b8..3afac275ee82 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -201,7 +201,7 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc, struct sctp_hmac *hmac_desc = sctp_auth_asoc_get_hmac(asoc); if (hmac_desc) - max_data -= SCTP_PAD4(sizeof(sctp_auth_chunk_t) + + max_data -= SCTP_PAD4(sizeof(struct sctp_auth_chunk) + hmac_desc->hmac_len); } @@ -221,7 +221,7 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc, asoc->outqueue.out_qlen == 0 && list_empty(&asoc->outqueue.retransmit) && msg_len > max_data) - first_len -= SCTP_PAD4(sizeof(sctp_sack_chunk_t)); + first_len -= SCTP_PAD4(sizeof(struct sctp_sack_chunk)); /* Encourage Cookie-ECHO bundling. */ if (asoc->state < SCTP_STATE_COOKIE_ECHOED) diff --git a/net/sctp/debug.c b/net/sctp/debug.c index 2e47eb2f05cb..3f619fdcbf0a 100644 --- a/net/sctp/debug.c +++ b/net/sctp/debug.c @@ -60,7 +60,7 @@ static const char *const sctp_cid_tbl[SCTP_NUM_BASE_CHUNK_TYPES] = { }; /* Lookup "chunk type" debug name. */ -const char *sctp_cname(const sctp_subtype_t cid) +const char *sctp_cname(const union sctp_subtype cid) { if (cid.chunk <= SCTP_CID_BASE_MAX) return sctp_cid_tbl[cid.chunk]; @@ -130,7 +130,7 @@ static const char *const sctp_primitive_tbl[SCTP_NUM_PRIMITIVE_TYPES] = { }; /* Lookup primitive debug name. */ -const char *sctp_pname(const sctp_subtype_t id) +const char *sctp_pname(const union sctp_subtype id) { if (id.primitive <= SCTP_EVENT_PRIMITIVE_MAX) return sctp_primitive_tbl[id.primitive]; @@ -143,7 +143,7 @@ static const char *const sctp_other_tbl[] = { }; /* Lookup "other" debug name. */ -const char *sctp_oname(const sctp_subtype_t id) +const char *sctp_oname(const union sctp_subtype id) { if (id.other <= SCTP_EVENT_OTHER_MAX) return sctp_other_tbl[id.other]; @@ -165,7 +165,7 @@ static const char *const sctp_timer_tbl[] = { }; /* Lookup timer debug name. */ -const char *sctp_tname(const sctp_subtype_t id) +const char *sctp_tname(const union sctp_subtype id) { BUILD_BUG_ON(SCTP_EVENT_TIMEOUT_MAX + 1 != ARRAY_SIZE(sctp_timer_tbl)); diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 0e86f988f836..ee1e601a0b11 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -73,13 +73,13 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, * variables. There are arrays that we encode directly * into parameters to make the rest of the operations easier. */ - auth_hmacs = kzalloc(sizeof(sctp_hmac_algo_param_t) + - sizeof(__u16) * SCTP_AUTH_NUM_HMACS, gfp); + auth_hmacs = kzalloc(sizeof(*auth_hmacs) + + sizeof(__u16) * SCTP_AUTH_NUM_HMACS, gfp); if (!auth_hmacs) goto nomem; - auth_chunks = kzalloc(sizeof(sctp_chunks_param_t) + - SCTP_NUM_CHUNK_TYPES, gfp); + auth_chunks = kzalloc(sizeof(*auth_chunks) + + SCTP_NUM_CHUNK_TYPES, gfp); if (!auth_chunks) goto nomem; @@ -382,8 +382,8 @@ static void sctp_endpoint_bh_rcv(struct work_struct *work) struct sctp_transport *transport; struct sctp_chunk *chunk; struct sctp_inq *inqueue; - sctp_subtype_t subtype; - sctp_state_t state; + union sctp_subtype subtype; + enum sctp_state state; int error = 0; int first_time = 1; /* is this the first time through the loop */ diff --git a/net/sctp/input.c b/net/sctp/input.c index 41eb2ec10460..92a07141fd07 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -1111,7 +1111,7 @@ static struct sctp_association *__sctp_rcv_asconf_lookup( __be16 peer_port, struct sctp_transport **transportp) { - sctp_addip_chunk_t *asconf = (struct sctp_addip_chunk *)ch; + struct sctp_addip_chunk *asconf = (struct sctp_addip_chunk *)ch; struct sctp_af *af; union sctp_addr_param *param; union sctp_addr paddr; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 2a186b201ad2..51c488769590 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -243,8 +243,8 @@ static void sctp_v6_get_dst(struct sctp_transport *t, union sctp_addr *saddr, union sctp_addr *daddr = &t->ipaddr; union sctp_addr dst_saddr; struct in6_addr *final_p, final; + enum sctp_scope scope; __u8 matchlen = 0; - sctp_scope_t scope; memset(fl6, 0, sizeof(struct flowi6)); fl6->daddr = daddr->v6.sin6_addr; @@ -497,7 +497,7 @@ static void sctp_v6_from_addr_param(union sctp_addr *addr, static int sctp_v6_to_addr_param(const union sctp_addr *addr, union sctp_addr_param *param) { - int length = sizeof(sctp_ipv6addr_param_t); + int length = sizeof(struct sctp_ipv6addr_param); param->v6.param_hdr.type = SCTP_PARAM_IPV6_ADDRESS; param->v6.param_hdr.length = htons(length); @@ -512,7 +512,9 @@ static void sctp_v6_to_addr(union sctp_addr *addr, struct in6_addr *saddr, { addr->sa.sa_family = AF_INET6; addr->v6.sin6_port = port; + addr->v6.sin6_flowinfo = 0; addr->v6.sin6_addr = *saddr; + addr->v6.sin6_scope_id = 0; } /* Compare addresses exactly. @@ -624,10 +626,10 @@ static int sctp_v6_addr_valid(union sctp_addr *addr, } /* What is the scope of 'addr'? */ -static sctp_scope_t sctp_v6_scope(union sctp_addr *addr) +static enum sctp_scope sctp_v6_scope(union sctp_addr *addr) { + enum sctp_scope retval; int v6scope; - sctp_scope_t retval; /* The IPv6 scope is really a set of bit fields. * See IFA_* in <net/if_inet6.h>. Map to a generic SCTP scope. diff --git a/net/sctp/objcnt.c b/net/sctp/objcnt.c index 105ac3327b28..aeea6da81441 100644 --- a/net/sctp/objcnt.c +++ b/net/sctp/objcnt.c @@ -57,7 +57,7 @@ SCTP_DBG_OBJCNT(keys); /* An array to make it easy to pretty print the debug information * to the proc fs. */ -static sctp_dbg_objcnt_entry_t sctp_dbg_objcnt[] = { +static struct sctp_dbg_objcnt_entry sctp_dbg_objcnt[] = { SCTP_DBG_OBJCNT_ENTRY(sock), SCTP_DBG_OBJCNT_ENTRY(ep), SCTP_DBG_OBJCNT_ENTRY(assoc), diff --git a/net/sctp/output.c b/net/sctp/output.c index 9d8504985744..4a865cd06d76 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -57,15 +57,15 @@ #include <net/sctp/checksum.h> /* Forward declarations for private helpers. */ -static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk); -static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk); +static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk); +static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk); static void sctp_packet_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk); -static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, - struct sctp_chunk *chunk, - u16 chunk_len); + struct sctp_chunk *chunk); +static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet, + struct sctp_chunk *chunk, + u16 chunk_len); static void sctp_packet_reset(struct sctp_packet *packet) { @@ -181,11 +181,11 @@ void sctp_packet_free(struct sctp_packet *packet) * as it can fit in the packet, but any more data that does not fit in this * packet can be sent only after receiving the COOKIE_ACK. */ -sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk, - int one_packet, gfp_t gfp) +enum sctp_xmit sctp_packet_transmit_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk, + int one_packet, gfp_t gfp) { - sctp_xmit_t retval; + enum sctp_xmit retval; pr_debug("%s: packet:%p size:%zu chunk:%p size:%d\n", __func__, packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1); @@ -218,12 +218,12 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, } /* Try to bundle an auth chunk into the packet. */ -static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_bundle_auth(struct sctp_packet *pkt, + struct sctp_chunk *chunk) { struct sctp_association *asoc = pkt->transport->asoc; + enum sctp_xmit retval = SCTP_XMIT_OK; struct sctp_chunk *auth; - sctp_xmit_t retval = SCTP_XMIT_OK; /* if we don't have an association, we can't do authentication */ if (!asoc) @@ -254,10 +254,10 @@ static sctp_xmit_t sctp_packet_bundle_auth(struct sctp_packet *pkt, } /* Try to bundle a SACK with the packet. */ -static sctp_xmit_t sctp_packet_bundle_sack(struct sctp_packet *pkt, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_bundle_sack(struct sctp_packet *pkt, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; + enum sctp_xmit retval = SCTP_XMIT_OK; /* If sending DATA and haven't aleady bundled a SACK, try to * bundle one in to the packet. @@ -299,11 +299,11 @@ out: /* Append a chunk to the offered packet reporting back any inability to do * so. */ -static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk) +static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; __u16 chunk_len = SCTP_PAD4(ntohs(chunk->chunk_hdr->length)); + enum sctp_xmit retval = SCTP_XMIT_OK; /* Check to see if this chunk will fit into the packet */ retval = sctp_packet_will_fit(packet, chunk, chunk_len); @@ -353,10 +353,10 @@ finish: /* Append a chunk to the offered packet reporting back any inability to do * so. */ -sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *packet, - struct sctp_chunk *chunk) +enum sctp_xmit sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) { - sctp_xmit_t retval = SCTP_XMIT_OK; + enum sctp_xmit retval = SCTP_XMIT_OK; pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk); @@ -653,8 +653,8 @@ out: ********************************************************************/ /* This private function check to see if a chunk can be added */ -static sctp_xmit_t sctp_packet_can_append_data(struct sctp_packet *packet, - struct sctp_chunk *chunk) +static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk) { size_t datasize, rwnd, inflight, flight_size; struct sctp_transport *transport = packet->transport; @@ -762,12 +762,12 @@ static void sctp_packet_append_data(struct sctp_packet *packet, sctp_chunk_assign_ssn(chunk); } -static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, - struct sctp_chunk *chunk, - u16 chunk_len) +static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet, + struct sctp_chunk *chunk, + u16 chunk_len) { + enum sctp_xmit retval = SCTP_XMIT_OK; size_t psize, pmtu, maxsize; - sctp_xmit_t retval = SCTP_XMIT_OK; psize = packet->size; if (packet->transport->asoc) diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index e8762702a313..2966ff400755 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -534,7 +534,7 @@ void sctp_retransmit_mark(struct sctp_outq *q, * one packet out. */ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport, - sctp_retransmit_reason_t reason) + enum sctp_retransmit_reason reason) { struct net *net = sock_net(q->asoc->base.sk); @@ -594,14 +594,14 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport, static int sctp_outq_flush_rtx(struct sctp_outq *q, struct sctp_packet *pkt, int rtx_timeout, int *start_timer) { - struct list_head *lqueue; struct sctp_transport *transport = pkt->transport; - sctp_xmit_t status; struct sctp_chunk *chunk, *chunk1; - int fast_rtx; + struct list_head *lqueue; + enum sctp_xmit status; int error = 0; int timer = 0; int done = 0; + int fast_rtx; lqueue = &q->retransmit; fast_rtx = q->fast_rtx; @@ -781,7 +781,7 @@ static void sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp) struct sctp_transport *transport = NULL; struct sctp_transport *new_transport; struct sctp_chunk *chunk, *tmp; - sctp_xmit_t status; + enum sctp_xmit status; int error = 0; int start_timer = 0; int one_packet = 0; @@ -1197,7 +1197,7 @@ sctp_flush_out: static void sctp_sack_update_unack_data(struct sctp_association *assoc, struct sctp_sackhdr *sack) { - sctp_sack_variable_t *frags; + union sctp_sack_variable *frags; __u16 unack_data; int i; @@ -1224,7 +1224,7 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk) struct sctp_transport *transport; struct sctp_chunk *tchunk = NULL; struct list_head *lchunk, *transport_list, *temp; - sctp_sack_variable_t *frags = sack->variable; + union sctp_sack_variable *frags = sack->variable; __u32 sack_ctsn, ctsn, tsn; __u32 highest_tsn, highest_new_tsn; __u32 sack_a_rwnd; @@ -1736,10 +1736,10 @@ static void sctp_mark_missing(struct sctp_outq *q, /* Is the given TSN acked by this packet? */ static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn) { - int i; - sctp_sack_variable_t *frags; - __u16 tsn_offset, blocks; __u32 ctsn = ntohl(sack->cum_tsn_ack); + union sctp_sack_variable *frags; + __u16 tsn_offset, blocks; + int i; if (TSN_lte(tsn, ctsn)) goto pass; diff --git a/net/sctp/primitive.c b/net/sctp/primitive.c index f0553a022859..c0817f7a8964 100644 --- a/net/sctp/primitive.c +++ b/net/sctp/primitive.c @@ -53,8 +53,8 @@ int sctp_primitive_ ## name(struct net *net, struct sctp_association *asoc, \ void *arg) { \ int error = 0; \ - sctp_event_t event_type; sctp_subtype_t subtype; \ - sctp_state_t state; \ + enum sctp_event event_type; union sctp_subtype subtype; \ + enum sctp_state state; \ struct sctp_endpoint *ep; \ \ event_type = SCTP_EVENT_T_PRIMITIVE; \ diff --git a/net/sctp/probe.c b/net/sctp/probe.c index 6cc2152e0740..1280f85a598d 100644 --- a/net/sctp/probe.c +++ b/net/sctp/probe.c @@ -127,12 +127,13 @@ static const struct file_operations sctpprobe_fops = { .llseek = noop_llseek, }; -static sctp_disposition_t jsctp_sf_eat_sack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition jsctp_sf_eat_sack( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; struct sk_buff *skb = chunk->skb; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 989a900383b5..fcd80feb293f 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -196,7 +196,7 @@ static void sctp_free_local_addr_list(struct net *net) /* Copy the local addresses which are valid for 'scope' into 'bp'. */ int sctp_copy_local_addr_list(struct net *net, struct sctp_bind_addr *bp, - sctp_scope_t scope, gfp_t gfp, int copy_flags) + enum sctp_scope scope, gfp_t gfp, int copy_flags) { struct sctp_sockaddr_entry *addr; union sctp_addr laddr; @@ -292,7 +292,7 @@ static void sctp_v4_from_addr_param(union sctp_addr *addr, static int sctp_v4_to_addr_param(const union sctp_addr *addr, union sctp_addr_param *param) { - int length = sizeof(sctp_ipv4addr_param_t); + int length = sizeof(struct sctp_ipv4addr_param); param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS; param->v4.param_hdr.length = htons(length); @@ -400,9 +400,9 @@ static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp) * IPv4 scoping can be controlled through sysctl option * net.sctp.addr_scope_policy */ -static sctp_scope_t sctp_v4_scope(union sctp_addr *addr) +static enum sctp_scope sctp_v4_scope(union sctp_addr *addr) { - sctp_scope_t retval; + enum sctp_scope retval; /* Check for unusable SCTP addresses. */ if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) { diff --git a/net/sctp/sctp_diag.c b/net/sctp/sctp_diag.c index 9a647214a91e..e99518e79b52 100644 --- a/net/sctp/sctp_diag.c +++ b/net/sctp/sctp_diag.c @@ -70,7 +70,8 @@ static int inet_diag_msg_sctpladdrs_fill(struct sk_buff *skb, info = nla_data(attr); list_for_each_entry_rcu(laddr, address_list, list) { - memcpy(info, &laddr->a, addrlen); + memcpy(info, &laddr->a, sizeof(laddr->a)); + memset(info + sizeof(laddr->a), 0, addrlen - sizeof(laddr->a)); info += addrlen; } @@ -93,7 +94,9 @@ static int inet_diag_msg_sctpaddrs_fill(struct sk_buff *skb, info = nla_data(attr); list_for_each_entry(from, &asoc->peer.transport_addr_list, transports) { - memcpy(info, &from->ipaddr, addrlen); + memcpy(info, &from->ipaddr, sizeof(from->ipaddr)); + memset(info + sizeof(from->ipaddr), 0, + addrlen - sizeof(from->ipaddr)); info += addrlen; } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 4e16b02ed832..ca8f196b6c6c 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -69,7 +69,8 @@ static struct sctp_chunk *sctp_make_data(const struct sctp_association *asoc, static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc, __u8 type, __u8 flags, int paylen, gfp_t gfp); -static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep, +static struct sctp_cookie_param *sctp_pack_cookie( + const struct sctp_endpoint *ep, const struct sctp_association *asoc, const struct sctp_chunk *init_chunk, int *cookie_len, @@ -131,17 +132,17 @@ static const struct sctp_paramhdr prsctp_param = { * provided chunk, as most cause codes will be embedded inside an * abort chunk. */ -void sctp_init_cause(struct sctp_chunk *chunk, __be16 cause_code, - size_t paylen) +void sctp_init_cause(struct sctp_chunk *chunk, __be16 cause_code, + size_t paylen) { - sctp_errhdr_t err; + struct sctp_errhdr err; __u16 len; /* Cause code constants are now defined in network order. */ err.cause = cause_code; - len = sizeof(sctp_errhdr_t) + paylen; + len = sizeof(err) + paylen; err.length = htons(len); - chunk->subh.err_hdr = sctp_addto_chunk(chunk, sizeof(sctp_errhdr_t), &err); + chunk->subh.err_hdr = sctp_addto_chunk(chunk, sizeof(err), &err); } /* A helper to initialize an op error inside a @@ -150,21 +151,21 @@ void sctp_init_cause(struct sctp_chunk *chunk, __be16 cause_code, * if there isn't enough space in the op error chunk */ static int sctp_init_cause_fixed(struct sctp_chunk *chunk, __be16 cause_code, - size_t paylen) + size_t paylen) { - sctp_errhdr_t err; + struct sctp_errhdr err; __u16 len; /* Cause code constants are now defined in network order. */ err.cause = cause_code; - len = sizeof(sctp_errhdr_t) + paylen; + len = sizeof(err) + paylen; err.length = htons(len); if (skb_tailroom(chunk->skb) < len) return -ENOSPC; - chunk->subh.err_hdr = sctp_addto_chunk_fixed(chunk, - sizeof(sctp_errhdr_t), - &err); + + chunk->subh.err_hdr = sctp_addto_chunk_fixed(chunk, sizeof(err), &err); + return 0; } /* 3.3.2 Initiation (INIT) (1) @@ -212,32 +213,31 @@ static int sctp_init_cause_fixed(struct sctp_chunk *chunk, __be16 cause_code, * Supported Address Types (Note 4) Optional 12 */ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, - const struct sctp_bind_addr *bp, - gfp_t gfp, int vparam_len) + 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; + struct sctp_paramhdr *auth_hmacs = NULL; + struct sctp_supported_addrs_param sat; struct sctp_endpoint *ep = asoc->ep; - struct sctp_inithdr init; - union sctp_params addrs; - size_t chunksize; struct sctp_chunk *retval = NULL; int num_types, addrs_len = 0; + struct sctp_inithdr init; + union sctp_params addrs; struct sctp_sock *sp; - sctp_supported_addrs_param_t sat; + __u8 extensions[4]; + size_t chunksize; __be16 types[2]; - sctp_adaptation_ind_param_t aiparam; - sctp_supported_ext_param_t ext_param; int num_ext = 0; - __u8 extensions[3]; - struct sctp_paramhdr *auth_chunks = NULL, - *auth_hmacs = NULL; /* RFC 2960 3.3.2 Initiation (INIT) (1) * * Note 1: The INIT chunks can contain multiple addresses that * can be IPv4 and/or IPv6 in any combination. */ - retval = NULL; /* Convert the provided bind address list to raw format. */ addrs = sctp_bind_addrs_to_raw(bp, &addrs_len, gfp); @@ -305,8 +305,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, /* If we have any extensions to report, account for that */ if (num_ext) - chunksize += SCTP_PAD4(sizeof(sctp_supported_ext_param_t) + - num_ext); + chunksize += SCTP_PAD4(sizeof(ext_param) + num_ext); /* RFC 2960 3.3.2 Initiation (INIT) (1) * @@ -348,10 +347,8 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, */ if (num_ext) { ext_param.param_hdr.type = SCTP_PARAM_SUPPORTED_EXT; - ext_param.param_hdr.length = - htons(sizeof(sctp_supported_ext_param_t) + num_ext); - sctp_addto_chunk(retval, sizeof(sctp_supported_ext_param_t), - &ext_param); + ext_param.param_hdr.length = htons(sizeof(ext_param) + num_ext); + sctp_addto_chunk(retval, sizeof(ext_param), &ext_param); sctp_addto_param(retval, num_ext, extensions); } @@ -382,26 +379,24 @@ nodata: } struct sctp_chunk *sctp_make_init_ack(const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - gfp_t gfp, int unkparam_len) + const struct sctp_chunk *chunk, + gfp_t gfp, int unkparam_len) { + struct sctp_supported_ext_param ext_param; + struct sctp_adaptation_ind_param aiparam; + struct sctp_paramhdr *auth_chunks = NULL; + struct sctp_paramhdr *auth_random = NULL; + struct sctp_paramhdr *auth_hmacs = NULL; + struct sctp_chunk *retval = NULL; + struct sctp_cookie_param *cookie; struct sctp_inithdr initack; - struct sctp_chunk *retval; union sctp_params addrs; struct sctp_sock *sp; - int addrs_len; - sctp_cookie_param_t *cookie; - int cookie_len; + __u8 extensions[4]; size_t chunksize; - sctp_adaptation_ind_param_t aiparam; - sctp_supported_ext_param_t ext_param; int num_ext = 0; - __u8 extensions[3]; - struct sctp_paramhdr *auth_chunks = NULL, - *auth_hmacs = NULL, - *auth_random = NULL; - - retval = NULL; + int cookie_len; + int addrs_len; /* Note: there may be no addresses to embed. */ addrs = sctp_bind_addrs_to_raw(&asoc->base.bind_addr, &addrs_len, gfp); @@ -468,8 +463,7 @@ struct sctp_chunk *sctp_make_init_ack(const struct sctp_association *asoc, } if (num_ext) - chunksize += SCTP_PAD4(sizeof(sctp_supported_ext_param_t) + - num_ext); + chunksize += SCTP_PAD4(sizeof(ext_param) + num_ext); /* Now allocate and fill out the chunk. */ retval = sctp_make_control(asoc, SCTP_CID_INIT_ACK, 0, chunksize, gfp); @@ -495,10 +489,8 @@ struct sctp_chunk *sctp_make_init_ack(const struct sctp_association *asoc, sctp_addto_chunk(retval, sizeof(ecap_param), &ecap_param); if (num_ext) { ext_param.param_hdr.type = SCTP_PARAM_SUPPORTED_EXT; - ext_param.param_hdr.length = - htons(sizeof(sctp_supported_ext_param_t) + num_ext); - sctp_addto_chunk(retval, sizeof(sctp_supported_ext_param_t), - &ext_param); + ext_param.param_hdr.length = htons(sizeof(ext_param) + num_ext); + sctp_addto_chunk(retval, sizeof(ext_param), &ext_param); sctp_addto_param(retval, num_ext, extensions); } if (asoc->peer.prsctp_capable) @@ -567,11 +559,11 @@ nomem_cookie: * to insure interoperability. */ struct sctp_chunk *sctp_make_cookie_echo(const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_chunk *chunk) { struct sctp_chunk *retval; - void *cookie; int cookie_len; + void *cookie; cookie = asoc->peer.cookie; cookie_len = asoc->peer.cookie_len; @@ -619,7 +611,7 @@ nodata: * Set to zero on transmit and ignored on receipt. */ struct sctp_chunk *sctp_make_cookie_ack(const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_chunk *chunk) { struct sctp_chunk *retval; @@ -664,15 +656,15 @@ struct sctp_chunk *sctp_make_cookie_ack(const struct sctp_association *asoc, * Note: The CWR is considered a Control chunk. */ struct sctp_chunk *sctp_make_cwr(const struct sctp_association *asoc, - const __u32 lowest_tsn, - const struct sctp_chunk *chunk) + const __u32 lowest_tsn, + const struct sctp_chunk *chunk) { struct sctp_chunk *retval; - sctp_cwrhdr_t cwr; + struct sctp_cwrhdr cwr; cwr.lowest_tsn = htonl(lowest_tsn); retval = sctp_make_control(asoc, SCTP_CID_ECN_CWR, 0, - sizeof(sctp_cwrhdr_t), GFP_ATOMIC); + sizeof(cwr), GFP_ATOMIC); if (!retval) goto nodata; @@ -699,14 +691,14 @@ nodata: /* Make an ECNE chunk. This is a congestion experienced report. */ struct sctp_chunk *sctp_make_ecne(const struct sctp_association *asoc, - const __u32 lowest_tsn) + const __u32 lowest_tsn) { struct sctp_chunk *retval; - sctp_ecnehdr_t ecne; + struct sctp_ecnehdr ecne; ecne.lowest_tsn = htonl(lowest_tsn); retval = sctp_make_control(asoc, SCTP_CID_ECN_ECNE, 0, - sizeof(sctp_ecnehdr_t), GFP_ATOMIC); + sizeof(ecne), GFP_ATOMIC); if (!retval) goto nodata; retval->subh.ecne_hdr = @@ -720,9 +712,9 @@ nodata: * parameters. However, do not populate the data payload. */ struct sctp_chunk *sctp_make_datafrag_empty(struct sctp_association *asoc, - const struct sctp_sndrcvinfo *sinfo, - int data_len, __u8 flags, __u16 ssn, - gfp_t gfp) + const struct sctp_sndrcvinfo *sinfo, + int data_len, __u8 flags, __u16 ssn, + gfp_t gfp) { struct sctp_chunk *retval; struct sctp_datahdr dp; @@ -760,15 +752,15 @@ nodata: */ struct sctp_chunk *sctp_make_sack(const struct sctp_association *asoc) { - struct sctp_chunk *retval; - struct sctp_sackhdr sack; - int len; - __u32 ctsn; - __u16 num_gabs, num_dup_tsns; - struct sctp_association *aptr = (struct sctp_association *)asoc; struct sctp_tsnmap *map = (struct sctp_tsnmap *)&asoc->peer.tsn_map; + struct sctp_association *aptr = (struct sctp_association *)asoc; struct sctp_gap_ack_block gabs[SCTP_MAX_GABS]; + __u16 num_gabs, num_dup_tsns; struct sctp_transport *trans; + struct sctp_chunk *retval; + struct sctp_sackhdr sack; + __u32 ctsn; + int len; memset(gabs, 0, sizeof(gabs)); ctsn = sctp_tsnmap_get_ctsn(map); @@ -862,15 +854,15 @@ nodata: struct sctp_chunk *sctp_make_shutdown(const struct sctp_association *asoc, const struct sctp_chunk *chunk) { + struct sctp_shutdownhdr shut; struct sctp_chunk *retval; - sctp_shutdownhdr_t shut; __u32 ctsn; ctsn = sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map); shut.cum_tsn_ack = htonl(ctsn); retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN, 0, - sizeof(sctp_shutdownhdr_t), GFP_ATOMIC); + sizeof(shut), GFP_ATOMIC); if (!retval) goto nodata; @@ -884,7 +876,7 @@ nodata: } struct sctp_chunk *sctp_make_shutdown_ack(const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_chunk *chunk) { struct sctp_chunk *retval; @@ -907,8 +899,8 @@ struct sctp_chunk *sctp_make_shutdown_ack(const struct sctp_association *asoc, } struct sctp_chunk *sctp_make_shutdown_complete( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk) { struct sctp_chunk *retval; __u8 flags = 0; @@ -941,8 +933,8 @@ struct sctp_chunk *sctp_make_shutdown_complete( * association, except when responding to an INIT (sctpimpguide 2.41). */ struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - const size_t hint) + const struct sctp_chunk *chunk, + const size_t hint) { struct sctp_chunk *retval; __u8 flags = 0; @@ -978,14 +970,15 @@ struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc, /* Helper to create ABORT with a NO_USER_DATA error. */ struct sctp_chunk *sctp_make_abort_no_data( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk, __u32 tsn) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + __u32 tsn) { struct sctp_chunk *retval; __be32 payload; - retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) - + sizeof(tsn)); + retval = sctp_make_abort(asoc, chunk, + sizeof(struct sctp_errhdr) + sizeof(tsn)); if (!retval) goto no_mem; @@ -1020,7 +1013,8 @@ struct sctp_chunk *sctp_make_abort_user(const struct sctp_association *asoc, void *payload = NULL; int err; - retval = sctp_make_abort(asoc, NULL, sizeof(sctp_errhdr_t) + paylen); + retval = sctp_make_abort(asoc, NULL, + sizeof(struct sctp_errhdr) + paylen); if (!retval) goto err_chunk; @@ -1058,8 +1052,8 @@ err_chunk: static void *sctp_addto_param(struct sctp_chunk *chunk, int len, const void *data) { - void *target; int chunklen = ntohs(chunk->chunk_hdr->length); + void *target; target = skb_put(chunk->skb, len); @@ -1077,16 +1071,16 @@ static void *sctp_addto_param(struct sctp_chunk *chunk, int len, /* Make an ABORT chunk with a PROTOCOL VIOLATION cause code. */ struct sctp_chunk *sctp_make_abort_violation( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - const __u8 *payload, - const size_t paylen) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + const __u8 *payload, + const size_t paylen) { struct sctp_chunk *retval; struct sctp_paramhdr phdr; - retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) + paylen + - sizeof(phdr)); + retval = sctp_make_abort(asoc, chunk, sizeof(struct sctp_errhdr) + + paylen + sizeof(phdr)); if (!retval) goto end; @@ -1103,14 +1097,14 @@ end: } struct sctp_chunk *sctp_make_violation_paramlen( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - struct sctp_paramhdr *param) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + struct sctp_paramhdr *param) { - struct sctp_chunk *retval; static const char error[] = "The following parameter had invalid length:"; - size_t payload_len = sizeof(error) + sizeof(sctp_errhdr_t) + + size_t payload_len = sizeof(error) + sizeof(struct sctp_errhdr) + sizeof(*param); + struct sctp_chunk *retval; retval = sctp_make_abort(asoc, chunk, payload_len); if (!retval) @@ -1126,12 +1120,12 @@ nodata: } struct sctp_chunk *sctp_make_violation_max_retrans( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk) { - struct sctp_chunk *retval; static const char error[] = "Association exceeded its max_retans count"; - size_t payload_len = sizeof(error) + sizeof(sctp_errhdr_t); + size_t payload_len = sizeof(error) + sizeof(struct sctp_errhdr); + struct sctp_chunk *retval; retval = sctp_make_abort(asoc, chunk, payload_len); if (!retval) @@ -1146,10 +1140,10 @@ nodata: /* Make a HEARTBEAT chunk. */ struct sctp_chunk *sctp_make_heartbeat(const struct sctp_association *asoc, - const struct sctp_transport *transport) + const struct sctp_transport *transport) { + struct sctp_sender_hb_info hbinfo; struct sctp_chunk *retval; - sctp_sender_hb_info_t hbinfo; retval = sctp_make_control(asoc, SCTP_CID_HEARTBEAT, 0, sizeof(hbinfo), GFP_ATOMIC); @@ -1158,7 +1152,7 @@ struct sctp_chunk *sctp_make_heartbeat(const struct sctp_association *asoc, goto nodata; hbinfo.param_hdr.type = SCTP_PARAM_HEARTBEAT_INFO; - hbinfo.param_hdr.length = htons(sizeof(sctp_sender_hb_info_t)); + hbinfo.param_hdr.length = htons(sizeof(hbinfo)); hbinfo.daddr = transport->ipaddr; hbinfo.sent_at = jiffies; hbinfo.hb_nonce = transport->hb_nonce; @@ -1175,8 +1169,9 @@ nodata: } struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - const void *payload, const size_t paylen) + const struct sctp_chunk *chunk, + const void *payload, + const size_t paylen) { struct sctp_chunk *retval; @@ -1207,14 +1202,15 @@ nodata: * This routine can be used for containing multiple causes in the chunk. */ static struct sctp_chunk *sctp_make_op_error_space( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - size_t size) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + size_t size) { struct sctp_chunk *retval; retval = sctp_make_control(asoc, SCTP_CID_ERROR, 0, - sizeof(sctp_errhdr_t) + size, GFP_ATOMIC); + sizeof(struct sctp_errhdr) + size, + GFP_ATOMIC); if (!retval) goto nodata; @@ -1240,8 +1236,8 @@ nodata: * to report all the errors, if the incoming chunk is large */ static inline struct sctp_chunk *sctp_make_op_error_fixed( - const struct sctp_association *asoc, - const struct sctp_chunk *chunk) + const struct sctp_association *asoc, + const struct sctp_chunk *chunk) { size_t size = asoc ? asoc->pathmtu : 0; @@ -1253,9 +1249,9 @@ static inline struct sctp_chunk *sctp_make_op_error_fixed( /* Create an Operation Error chunk. */ struct sctp_chunk *sctp_make_op_error(const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - __be16 cause_code, const void *payload, - size_t paylen, size_t reserve_tail) + const struct sctp_chunk *chunk, + __be16 cause_code, const void *payload, + size_t paylen, size_t reserve_tail) { struct sctp_chunk *retval; @@ -1274,9 +1270,9 @@ nodata: struct sctp_chunk *sctp_make_auth(const struct sctp_association *asoc) { - struct sctp_chunk *retval; - struct sctp_hmac *hmac_desc; struct sctp_authhdr auth_hdr; + struct sctp_hmac *hmac_desc; + struct sctp_chunk *retval; __u8 *hmac; /* Get the first hmac that the peer told us to use */ @@ -1285,16 +1281,16 @@ struct sctp_chunk *sctp_make_auth(const struct sctp_association *asoc) return NULL; retval = sctp_make_control(asoc, SCTP_CID_AUTH, 0, - hmac_desc->hmac_len + sizeof(sctp_authhdr_t), - GFP_ATOMIC); + hmac_desc->hmac_len + sizeof(auth_hdr), + GFP_ATOMIC); if (!retval) return NULL; auth_hdr.hmac_id = htons(hmac_desc->hmac_id); auth_hdr.shkey_id = htons(asoc->active_key_id); - retval->subh.auth_hdr = sctp_addto_chunk(retval, sizeof(sctp_authhdr_t), - &auth_hdr); + retval->subh.auth_hdr = sctp_addto_chunk(retval, sizeof(auth_hdr), + &auth_hdr); hmac = skb_put_zero(retval->skb, hmac_desc->hmac_len); @@ -1322,8 +1318,8 @@ struct sctp_chunk *sctp_make_auth(const struct sctp_association *asoc) * */ struct sctp_chunk *sctp_chunkify(struct sk_buff *skb, - const struct sctp_association *asoc, - struct sock *sk, gfp_t gfp) + const struct sctp_association *asoc, + struct sock *sk, gfp_t gfp) { struct sctp_chunk *retval; @@ -1375,11 +1371,11 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk) * arguments, reserving enough space for a 'paylen' byte payload. */ static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc, - __u8 type, __u8 flags, int paylen, - gfp_t gfp) + __u8 type, __u8 flags, int paylen, + gfp_t gfp) { - struct sctp_chunk *retval; struct sctp_chunkhdr *chunk_hdr; + struct sctp_chunk *retval; struct sk_buff *skb; struct sock *sk; @@ -1473,9 +1469,9 @@ void sctp_chunk_put(struct sctp_chunk *ch) */ void *sctp_addto_chunk(struct sctp_chunk *chunk, int len, const void *data) { - void *target; int chunklen = ntohs(chunk->chunk_hdr->length); int padlen = SCTP_PAD4(chunklen) - chunklen; + void *target; skb_put_zero(chunk->skb, padlen); target = skb_put_data(chunk->skb, data, len); @@ -1528,11 +1524,10 @@ int sctp_user_addto_chunk(struct sctp_chunk *chunk, int len, */ void sctp_chunk_assign_ssn(struct sctp_chunk *chunk) { - struct sctp_datamsg *msg; - struct sctp_chunk *lchunk; struct sctp_stream *stream; - __u16 ssn; - __u16 sid; + struct sctp_chunk *lchunk; + struct sctp_datamsg *msg; + __u16 ssn, sid; if (chunk->has_ssn) return; @@ -1577,12 +1572,12 @@ void sctp_chunk_assign_tsn(struct sctp_chunk *chunk) /* Create a CLOSED association to use with an incoming packet. */ struct sctp_association *sctp_make_temp_asoc(const struct sctp_endpoint *ep, - struct sctp_chunk *chunk, - gfp_t gfp) + struct sctp_chunk *chunk, + gfp_t gfp) { struct sctp_association *asoc; + enum sctp_scope scope; struct sk_buff *skb; - sctp_scope_t scope; /* Create the bare association. */ scope = sctp_scope(sctp_source(chunk)); @@ -1601,14 +1596,15 @@ nodata: /* Build a cookie representing asoc. * This INCLUDES the param header needed to put the cookie in the INIT ACK. */ -static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const struct sctp_chunk *init_chunk, - int *cookie_len, - const __u8 *raw_addrs, int addrs_len) +static struct sctp_cookie_param *sctp_pack_cookie( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const struct sctp_chunk *init_chunk, + int *cookie_len, const __u8 *raw_addrs, + int addrs_len) { - sctp_cookie_param_t *retval; struct sctp_signed_cookie *cookie; + struct sctp_cookie_param *retval; int headersize, bodysize; /* Header size is static data prior to the actual cookie, including @@ -1692,19 +1688,19 @@ nodata: /* Unpack the cookie from COOKIE ECHO chunk, recreating the association. */ struct sctp_association *sctp_unpack_cookie( - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - struct sctp_chunk *chunk, gfp_t gfp, - int *error, struct sctp_chunk **errp) + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, gfp_t gfp, + int *error, struct sctp_chunk **errp) { struct sctp_association *retval = NULL; + int headersize, bodysize, fixed_size; struct sctp_signed_cookie *cookie; + struct sk_buff *skb = chunk->skb; struct sctp_cookie *bear_cookie; - int headersize, bodysize, fixed_size; __u8 *digest = ep->digest; + enum sctp_scope scope; unsigned int len; - sctp_scope_t scope; - struct sk_buff *skb = chunk->skb; ktime_t kt; /* Header size is static data prior to the actual cookie, including @@ -1976,8 +1972,8 @@ static int sctp_process_hn_param(const struct sctp_association *asoc, static int sctp_verify_ext_param(struct net *net, union sctp_params param) { __u16 num_ext = ntohs(param.p->length) - sizeof(struct sctp_paramhdr); - int have_auth = 0; int have_asconf = 0; + int have_auth = 0; int i; for (i = 0; i < num_ext; i++) { @@ -2007,10 +2003,10 @@ static int sctp_verify_ext_param(struct net *net, union sctp_params param) } static void sctp_process_ext_param(struct sctp_association *asoc, - union sctp_params param) + union sctp_params param) { - struct net *net = sock_net(asoc->base.sk); __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++) { @@ -2067,10 +2063,11 @@ static void sctp_process_ext_param(struct sctp_association *asoc, * SCTP_IERROR_ERROR - stop and report an error. * SCTP_IERROR_NOMEME - out of memory. */ -static sctp_ierror_t sctp_process_unk_param(const struct sctp_association *asoc, - union sctp_params param, - struct sctp_chunk *chunk, - struct sctp_chunk **errp) +static enum sctp_ierror sctp_process_unk_param( + const struct sctp_association *asoc, + union sctp_params param, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) { int retval = SCTP_IERROR_NO_ERROR; @@ -2119,13 +2116,13 @@ static sctp_ierror_t sctp_process_unk_param(const struct sctp_association *asoc, * SCTP_IERROR_ERROR - stop processing, trigger an ERROR * SCTP_IERROR_NO_ERROR - continue with the chunk */ -static sctp_ierror_t sctp_verify_param(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - union sctp_params param, - enum sctp_cid cid, - struct sctp_chunk *chunk, - struct sctp_chunk **err_chunk) +static enum sctp_ierror sctp_verify_param(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + union sctp_params param, + enum sctp_cid cid, + struct sctp_chunk *chunk, + struct sctp_chunk **err_chunk) { struct sctp_hmac_algo_param *hmacs; int retval = SCTP_IERROR_NO_ERROR; @@ -2310,13 +2307,13 @@ int sctp_process_init(struct sctp_association *asoc, struct sctp_chunk *chunk, struct sctp_init_chunk *peer_init, gfp_t gfp) { struct net *net = sock_net(asoc->base.sk); - union sctp_params param; struct sctp_transport *transport; struct list_head *pos, *temp; - struct sctp_af *af; + union sctp_params param; union sctp_addr addr; - char *cookie; + struct sctp_af *af; int src_match = 0; + char *cookie; /* We must include the address that the INIT packet came from. * This is the only address that matters for an INIT packet. @@ -2500,16 +2497,15 @@ static int sctp_process_param(struct sctp_association *asoc, gfp_t gfp) { struct net *net = sock_net(asoc->base.sk); - union sctp_addr addr; - int i; - __u16 sat; - int retval = 1; - sctp_scope_t scope; - u32 stale; - struct sctp_af *af; + struct sctp_endpoint *ep = asoc->ep; union sctp_addr_param *addr_param; struct sctp_transport *t; - struct sctp_endpoint *ep = asoc->ep; + enum sctp_scope scope; + union sctp_addr addr; + struct sctp_af *af; + int retval = 1, i; + u32 stale; + __u16 sat; /* We maintain all INIT parameters in network byte order all the * time. This allows us to not worry about whether the parameters @@ -2617,7 +2613,7 @@ do_addr_param: if (!net->sctp.addip_enable) goto fall_through; - addr_param = param.v + sizeof(sctp_addip_param_t); + addr_param = param.v + sizeof(struct sctp_addip_param); af = sctp_get_af_specific(param_type2af(addr_param->p.type)); if (af == NULL) @@ -2754,7 +2750,7 @@ static struct sctp_chunk *sctp_make_asconf(struct sctp_association *asoc, union sctp_addr *addr, int vparam_len) { - sctp_addiphdr_t asconf; + struct sctp_addiphdr asconf; struct sctp_chunk *retval; int length = sizeof(asconf) + vparam_len; union sctp_addr_param addrparam; @@ -2807,22 +2803,20 @@ static struct sctp_chunk *sctp_make_asconf(struct sctp_association *asoc, * */ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc, - union sctp_addr *laddr, - struct sockaddr *addrs, - int addrcnt, - __be16 flags) -{ - sctp_addip_param_t param; - struct sctp_chunk *retval; - union sctp_addr_param addr_param; - union sctp_addr *addr; - void *addr_buf; - struct sctp_af *af; - int paramlen = sizeof(param); - int addr_param_len = 0; - int totallen = 0; - int i; - int del_pickup = 0; + union sctp_addr *laddr, + struct sockaddr *addrs, + int addrcnt, __be16 flags) +{ + union sctp_addr_param addr_param; + struct sctp_addip_param param; + int paramlen = sizeof(param); + struct sctp_chunk *retval; + int addr_param_len = 0; + union sctp_addr *addr; + int totallen = 0, i; + int del_pickup = 0; + struct sctp_af *af; + void *addr_buf; /* Get total length of all the address parameters. */ addr_buf = addrs; @@ -2898,12 +2892,12 @@ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc, struct sctp_chunk *sctp_make_asconf_set_prim(struct sctp_association *asoc, union sctp_addr *addr) { - sctp_addip_param_t param; - struct sctp_chunk *retval; - int len = sizeof(param); - union sctp_addr_param addrparam; - int addrlen; - struct sctp_af *af = sctp_get_af_specific(addr->v4.sin_family); + struct sctp_af *af = sctp_get_af_specific(addr->v4.sin_family); + union sctp_addr_param addrparam; + struct sctp_addip_param param; + struct sctp_chunk *retval; + int len = sizeof(param); + int addrlen; addrlen = af->to_addr_param(addr, &addrparam); if (!addrlen) @@ -2947,9 +2941,9 @@ struct sctp_chunk *sctp_make_asconf_set_prim(struct sctp_association *asoc, static struct sctp_chunk *sctp_make_asconf_ack(const struct sctp_association *asoc, __u32 serial, int vparam_len) { - sctp_addiphdr_t asconf; - struct sctp_chunk *retval; - int length = sizeof(asconf) + vparam_len; + struct sctp_addiphdr asconf; + struct sctp_chunk *retval; + int length = sizeof(asconf) + vparam_len; /* Create the chunk. */ retval = sctp_make_control(asoc, SCTP_CID_ASCONF_ACK, 0, length, @@ -2967,13 +2961,14 @@ static struct sctp_chunk *sctp_make_asconf_ack(const struct sctp_association *as /* Add response parameters to an ASCONF_ACK chunk. */ static void sctp_add_asconf_response(struct sctp_chunk *chunk, __be32 crr_id, - __be16 err_code, sctp_addip_param_t *asconf_param) + __be16 err_code, + struct sctp_addip_param *asconf_param) { - sctp_addip_param_t ack_param; - sctp_errhdr_t err_param; - int asconf_param_len = 0; - int err_param_len = 0; - __be16 response_type; + struct sctp_addip_param ack_param; + struct sctp_errhdr err_param; + int asconf_param_len = 0; + int err_param_len = 0; + __be16 response_type; if (SCTP_ERROR_NO_ERROR == err_code) { response_type = SCTP_PARAM_SUCCESS_REPORT; @@ -3008,15 +3003,15 @@ static void sctp_add_asconf_response(struct sctp_chunk *chunk, __be32 crr_id, /* Process a asconf parameter. */ static __be16 sctp_process_asconf_param(struct sctp_association *asoc, - struct sctp_chunk *asconf, - sctp_addip_param_t *asconf_param) + struct sctp_chunk *asconf, + struct sctp_addip_param *asconf_param) { + union sctp_addr_param *addr_param; struct sctp_transport *peer; - struct sctp_af *af; union sctp_addr addr; - union sctp_addr_param *addr_param; + struct sctp_af *af; - addr_param = (void *)asconf_param + sizeof(sctp_addip_param_t); + addr_param = (void *)asconf_param + sizeof(*asconf_param); if (asconf_param->param_hdr.type != SCTP_PARAM_ADD_IP && asconf_param->param_hdr.type != SCTP_PARAM_DEL_IP && @@ -3141,10 +3136,11 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, struct sctp_chunk *chunk, bool addr_param_needed, struct sctp_paramhdr **errp) { - sctp_addip_chunk_t *addip = (sctp_addip_chunk_t *) chunk->chunk_hdr; - union sctp_params param; + struct sctp_addip_chunk *addip; bool addr_param_seen = false; + union sctp_params param; + addip = (struct sctp_addip_chunk *)chunk->chunk_hdr; sctp_walk_params(param, addip, addip_hdr.params) { size_t length = ntohs(param.p->length); @@ -3153,7 +3149,7 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, case SCTP_PARAM_ERR_CAUSE: break; case SCTP_PARAM_IPV4_ADDRESS: - if (length != sizeof(sctp_ipv4addr_param_t)) + if (length != sizeof(struct sctp_ipv4addr_param)) return false; /* ensure there is only one addr param and it's in the * beginning of addip_hdr params, or we reject it. @@ -3163,7 +3159,7 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, addr_param_seen = true; break; case SCTP_PARAM_IPV6_ADDRESS: - if (length != sizeof(sctp_ipv6addr_param_t)) + if (length != sizeof(struct sctp_ipv6addr_param)) return false; if (param.v != addip->addip_hdr.params) return false; @@ -3176,13 +3172,13 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, if (addr_param_needed && !addr_param_seen) return false; length = ntohs(param.addip->param_hdr.length); - if (length < sizeof(sctp_addip_param_t) + + if (length < sizeof(struct sctp_addip_param) + sizeof(**errp)) return false; break; case SCTP_PARAM_SUCCESS_REPORT: case SCTP_PARAM_ADAPTATION_LAYER_IND: - if (length != sizeof(sctp_addip_param_t)) + if (length != sizeof(struct sctp_addip_param)) return false; break; default: @@ -3208,24 +3204,24 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc, struct sctp_chunk *asconf) { - sctp_addip_chunk_t *addip = (sctp_addip_chunk_t *) asconf->chunk_hdr; + union sctp_addr_param *addr_param; + struct sctp_addip_chunk *addip; + struct sctp_chunk *asconf_ack; bool all_param_pass = true; + struct sctp_addiphdr *hdr; + int length = 0, chunk_len; union sctp_params param; - sctp_addiphdr_t *hdr; - union sctp_addr_param *addr_param; - struct sctp_chunk *asconf_ack; - __be16 err_code; - int length = 0; - int chunk_len; - __u32 serial; + __be16 err_code; + __u32 serial; + addip = (struct sctp_addip_chunk *)asconf->chunk_hdr; chunk_len = ntohs(asconf->chunk_hdr->length) - sizeof(struct sctp_chunkhdr); - hdr = (sctp_addiphdr_t *)asconf->skb->data; + hdr = (struct sctp_addiphdr *)asconf->skb->data; serial = ntohl(hdr->serial); /* Skip the addiphdr and store a pointer to address parameter. */ - length = sizeof(sctp_addiphdr_t); + length = sizeof(*hdr); addr_param = (union sctp_addr_param *)(asconf->skb->data + length); chunk_len -= length; @@ -3291,16 +3287,16 @@ done: /* Process a asconf parameter that is successfully acked. */ static void sctp_asconf_param_success(struct sctp_association *asoc, - sctp_addip_param_t *asconf_param) + struct sctp_addip_param *asconf_param) { - struct sctp_af *af; - union sctp_addr addr; struct sctp_bind_addr *bp = &asoc->base.bind_addr; union sctp_addr_param *addr_param; - struct sctp_transport *transport; struct sctp_sockaddr_entry *saddr; + struct sctp_transport *transport; + union sctp_addr addr; + struct sctp_af *af; - addr_param = (void *)asconf_param + sizeof(sctp_addip_param_t); + addr_param = (void *)asconf_param + sizeof(*asconf_param); /* We have checked the packet before, so we do not check again. */ af = sctp_get_af_specific(param_type2af(addr_param->p.type)); @@ -3351,14 +3347,14 @@ static void sctp_asconf_param_success(struct sctp_association *asoc, * specific success indication is present for the parameter. */ static __be16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, - sctp_addip_param_t *asconf_param, - int no_err) + struct sctp_addip_param *asconf_param, + int no_err) { - sctp_addip_param_t *asconf_ack_param; - sctp_errhdr_t *err_param; - int length; - int asconf_ack_len; - __be16 err_code; + struct sctp_addip_param *asconf_ack_param; + struct sctp_errhdr *err_param; + int asconf_ack_len; + __be16 err_code; + int length; if (no_err) err_code = SCTP_ERROR_NO_ERROR; @@ -3371,9 +3367,9 @@ static __be16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, /* Skip the addiphdr from the asconf_ack chunk and store a pointer to * the first asconf_ack parameter. */ - length = sizeof(sctp_addiphdr_t); - asconf_ack_param = (sctp_addip_param_t *)(asconf_ack->skb->data + - length); + length = sizeof(struct sctp_addiphdr); + asconf_ack_param = (struct sctp_addip_param *)(asconf_ack->skb->data + + length); asconf_ack_len -= length; while (asconf_ack_len > 0) { @@ -3382,7 +3378,7 @@ static __be16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, case SCTP_PARAM_SUCCESS_REPORT: return SCTP_ERROR_NO_ERROR; case SCTP_PARAM_ERR_CAUSE: - length = sizeof(sctp_addip_param_t); + length = sizeof(*asconf_ack_param); err_param = (void *)asconf_ack_param + length; asconf_ack_len -= length; if (asconf_ack_len > 0) @@ -3407,20 +3403,20 @@ static __be16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, int sctp_process_asconf_ack(struct sctp_association *asoc, struct sctp_chunk *asconf_ack) { - struct sctp_chunk *asconf = asoc->addip_last_asconf; - union sctp_addr_param *addr_param; - sctp_addip_param_t *asconf_param; - int length = 0; - int asconf_len = asconf->skb->len; - int all_param_pass = 0; - int no_err = 1; - int retval = 0; - __be16 err_code = SCTP_ERROR_NO_ERROR; + struct sctp_chunk *asconf = asoc->addip_last_asconf; + struct sctp_addip_param *asconf_param; + __be16 err_code = SCTP_ERROR_NO_ERROR; + union sctp_addr_param *addr_param; + int asconf_len = asconf->skb->len; + int all_param_pass = 0; + int length = 0; + int no_err = 1; + int retval = 0; /* Skip the chunkhdr and addiphdr from the last asconf sent and store * a pointer to address parameter. */ - length = sizeof(sctp_addip_chunk_t); + length = sizeof(struct sctp_addip_chunk); addr_param = (union sctp_addr_param *)(asconf->skb->data + length); asconf_len -= length; @@ -3436,7 +3432,7 @@ int sctp_process_asconf_ack(struct sctp_association *asoc, * failures are indicated, then all request(s) are considered * successful. */ - if (asconf_ack->skb->len == sizeof(sctp_addiphdr_t)) + if (asconf_ack->skb->len == sizeof(struct sctp_addiphdr)) all_param_pass = 1; /* Process the TLVs contained in the last sent ASCONF chunk. */ @@ -3542,9 +3538,8 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc, * \ \ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ -static struct sctp_chunk *sctp_make_reconf( - const struct sctp_association *asoc, - int length) +static struct sctp_chunk *sctp_make_reconf(const struct sctp_association *asoc, + int length) { struct sctp_reconf_chunk *reconf; struct sctp_chunk *retval; @@ -3595,9 +3590,9 @@ static struct sctp_chunk *sctp_make_reconf( * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ struct sctp_chunk *sctp_make_strreset_req( - const struct sctp_association *asoc, - __u16 stream_num, __u16 *stream_list, - bool out, bool in) + const struct sctp_association *asoc, + __u16 stream_num, __u16 *stream_list, + bool out, bool in) { struct sctp_strreset_outreq outreq; __u16 stream_len = stream_num * 2; @@ -3649,7 +3644,7 @@ struct sctp_chunk *sctp_make_strreset_req( * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ struct sctp_chunk *sctp_make_strreset_tsnreq( - const struct sctp_association *asoc) + const struct sctp_association *asoc) { struct sctp_strreset_tsnreq tsnreq; __u16 length = sizeof(tsnreq); @@ -3680,8 +3675,8 @@ struct sctp_chunk *sctp_make_strreset_tsnreq( * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ struct sctp_chunk *sctp_make_strreset_addstrm( - const struct sctp_association *asoc, - __u16 out, __u16 in) + const struct sctp_association *asoc, + __u16 out, __u16 in) { struct sctp_strreset_addstrm addstrm; __u16 size = sizeof(addstrm); @@ -3725,9 +3720,8 @@ struct sctp_chunk *sctp_make_strreset_addstrm( * | Result | * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ -struct sctp_chunk *sctp_make_strreset_resp( - const struct sctp_association *asoc, - __u32 result, __u32 sn) +struct sctp_chunk *sctp_make_strreset_resp(const struct sctp_association *asoc, + __u32 result, __u32 sn) { struct sctp_strreset_resp resp; __u16 length = sizeof(resp); @@ -3762,10 +3756,10 @@ struct sctp_chunk *sctp_make_strreset_resp( * | Receiver's Next TSN (optional) | * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ */ -struct sctp_chunk *sctp_make_strreset_tsnresp( - struct sctp_association *asoc, - __u32 result, __u32 sn, - __u32 sender_tsn, __u32 receiver_tsn) +struct sctp_chunk *sctp_make_strreset_tsnresp(struct sctp_association *asoc, + __u32 result, __u32 sn, + __u32 sender_tsn, + __u32 receiver_tsn) { struct sctp_strreset_resptsn tsnresp; __u16 length = sizeof(tsnresp); diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index d6e5e9e0fd6d..e6a2974e020e 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -51,22 +51,23 @@ #include <net/sctp/sctp.h> #include <net/sctp/sm.h> -static int sctp_cmd_interpreter(sctp_event_t event_type, - sctp_subtype_t subtype, - sctp_state_t state, +static int sctp_cmd_interpreter(enum sctp_event event_type, + union sctp_subtype subtype, + enum sctp_state state, struct sctp_endpoint *ep, struct sctp_association *asoc, void *event_arg, - sctp_disposition_t status, - sctp_cmd_seq_t *commands, + enum sctp_disposition status, + struct sctp_cmd_seq *commands, gfp_t gfp); -static int sctp_side_effects(sctp_event_t event_type, sctp_subtype_t subtype, - sctp_state_t state, +static int sctp_side_effects(enum sctp_event event_type, + union sctp_subtype subtype, + enum sctp_state state, struct sctp_endpoint *ep, struct sctp_association **asoc, void *event_arg, - sctp_disposition_t status, - sctp_cmd_seq_t *commands, + enum sctp_disposition status, + struct sctp_cmd_seq *commands, gfp_t gfp); /******************************************************************** @@ -96,8 +97,8 @@ static void sctp_do_ecn_ce_work(struct sctp_association *asoc, * that was originally marked with the CE bit. */ static struct sctp_chunk *sctp_do_ecn_ecne_work(struct sctp_association *asoc, - __u32 lowest_tsn, - struct sctp_chunk *chunk) + __u32 lowest_tsn, + struct sctp_chunk *chunk) { struct sctp_chunk *repl; @@ -149,11 +150,11 @@ static void sctp_do_ecn_cwr_work(struct sctp_association *asoc, /* Generate SACK if necessary. We call this at the end of a packet. */ static int sctp_gen_sack(struct sctp_association *asoc, int force, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { + struct sctp_transport *trans = asoc->peer.last_data_from; __u32 ctsn, max_tsn_seen; struct sctp_chunk *sack; - struct sctp_transport *trans = asoc->peer.last_data_from; int error = 0; if (force || @@ -243,11 +244,11 @@ nomem: */ void sctp_generate_t3_rtx_event(unsigned long peer) { - int error; struct sctp_transport *transport = (struct sctp_transport *) peer; struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); + int error; /* Check whether a task is in the sock. */ @@ -280,7 +281,7 @@ out_unlock: * for timeouts which use the association as their parameter. */ static void sctp_generate_timeout_event(struct sctp_association *asoc, - sctp_event_timeout_t timeout_type) + enum sctp_event_timeout timeout_type) { struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -360,12 +361,12 @@ static void sctp_generate_autoclose_event(unsigned long data) */ void sctp_generate_heartbeat_event(unsigned long data) { - int error = 0; struct sctp_transport *transport = (struct sctp_transport *) data; struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); u32 elapsed, timeout; + int error = 0; bh_lock_sock(sk); if (sock_owned_by_user(sk)) { @@ -405,7 +406,7 @@ out_unlock: */ void sctp_generate_proto_unreach_event(unsigned long data) { - struct sctp_transport *transport = (struct sctp_transport *) data; + struct sctp_transport *transport = (struct sctp_transport *)data; struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -471,7 +472,7 @@ out_unlock: /* Inject a SACK Timeout event into the state machine. */ static void sctp_generate_sack_event(unsigned long data) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = (struct sctp_association *)data; sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_SACK); } @@ -505,7 +506,7 @@ sctp_timer_event_t *sctp_timer_events[SCTP_NUM_TIMEOUT_TYPES] = { * notification SHOULD be sent to the upper layer. * */ -static void sctp_do_8_2_transport_strike(sctp_cmd_seq_t *commands, +static void sctp_do_8_2_transport_strike(struct sctp_cmd_seq *commands, struct sctp_association *asoc, struct sctp_transport *transport, int is_hb) @@ -577,7 +578,7 @@ static void sctp_do_8_2_transport_strike(sctp_cmd_seq_t *commands, } /* Worker routine to handle INIT command failure. */ -static void sctp_cmd_init_failed(sctp_cmd_seq_t *commands, +static void sctp_cmd_init_failed(struct sctp_cmd_seq *commands, struct sctp_association *asoc, unsigned int error) { @@ -600,15 +601,16 @@ static void sctp_cmd_init_failed(sctp_cmd_seq_t *commands, } /* Worker routine to handle SCTP_CMD_ASSOC_FAILED. */ -static void sctp_cmd_assoc_failed(sctp_cmd_seq_t *commands, +static void sctp_cmd_assoc_failed(struct sctp_cmd_seq *commands, struct sctp_association *asoc, - sctp_event_t event_type, - sctp_subtype_t subtype, + enum sctp_event event_type, + union sctp_subtype subtype, struct sctp_chunk *chunk, unsigned int error) { struct sctp_ulpevent *event; struct sctp_chunk *abort; + /* Cancel any partial delivery in progress. */ sctp_ulpq_abort_pd(&asoc->ulpq, GFP_ATOMIC); @@ -644,7 +646,7 @@ static void sctp_cmd_assoc_failed(sctp_cmd_seq_t *commands, * since all other cases use "temporary" associations and can do all * their work in statefuns directly. */ -static int sctp_cmd_process_init(sctp_cmd_seq_t *commands, +static int sctp_cmd_process_init(struct sctp_cmd_seq *commands, struct sctp_association *asoc, struct sctp_chunk *chunk, struct sctp_init_chunk *peer_init, @@ -666,7 +668,7 @@ static int sctp_cmd_process_init(sctp_cmd_seq_t *commands, } /* Helper function to break out starting up of heartbeat timers. */ -static void sctp_cmd_hb_timers_start(sctp_cmd_seq_t *cmds, +static void sctp_cmd_hb_timers_start(struct sctp_cmd_seq *cmds, struct sctp_association *asoc) { struct sctp_transport *t; @@ -679,7 +681,7 @@ static void sctp_cmd_hb_timers_start(sctp_cmd_seq_t *cmds, sctp_transport_reset_hb_timer(t); } -static void sctp_cmd_hb_timers_stop(sctp_cmd_seq_t *cmds, +static void sctp_cmd_hb_timers_stop(struct sctp_cmd_seq *cmds, struct sctp_association *asoc) { struct sctp_transport *t; @@ -694,7 +696,7 @@ static void sctp_cmd_hb_timers_stop(sctp_cmd_seq_t *cmds, } /* Helper function to stop any pending T3-RTX timers */ -static void sctp_cmd_t3_rtx_timers_stop(sctp_cmd_seq_t *cmds, +static void sctp_cmd_t3_rtx_timers_stop(struct sctp_cmd_seq *cmds, struct sctp_association *asoc) { struct sctp_transport *t; @@ -708,12 +710,12 @@ static void sctp_cmd_t3_rtx_timers_stop(sctp_cmd_seq_t *cmds, /* Helper function to handle the reception of an HEARTBEAT ACK. */ -static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, +static void sctp_cmd_transport_on(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, struct sctp_transport *t, struct sctp_chunk *chunk) { - sctp_sender_hb_info_t *hbinfo; + struct sctp_sender_hb_info *hbinfo; int was_unconfirmed = 0; /* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of the @@ -767,7 +769,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, if (t->rto_pending == 0) t->rto_pending = 1; - hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data; + hbinfo = (struct sctp_sender_hb_info *)chunk->skb->data; sctp_transport_update_rto(t, (jiffies - hbinfo->sent_at)); /* Update the heartbeat timer. */ @@ -779,7 +781,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* Helper function to process the process SACK command. */ -static int sctp_cmd_process_sack(sctp_cmd_seq_t *cmds, +static int sctp_cmd_process_sack(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, struct sctp_chunk *chunk) { @@ -801,7 +803,7 @@ static int sctp_cmd_process_sack(sctp_cmd_seq_t *cmds, /* Helper function to set the timeout value for T2-SHUTDOWN timer and to set * the transport for a shutdown chunk. */ -static void sctp_cmd_setup_t2(sctp_cmd_seq_t *cmds, +static void sctp_cmd_setup_t2(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, struct sctp_chunk *chunk) { @@ -818,7 +820,7 @@ static void sctp_cmd_setup_t2(sctp_cmd_seq_t *cmds, asoc->timeouts[SCTP_EVENT_TIMEOUT_T2_SHUTDOWN] = t->rto; } -static void sctp_cmd_assoc_update(sctp_cmd_seq_t *cmds, +static void sctp_cmd_assoc_update(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, struct sctp_association *new) { @@ -828,7 +830,7 @@ static void sctp_cmd_assoc_update(sctp_cmd_seq_t *cmds, if (!sctp_assoc_update(asoc, new)) return; - abort = sctp_make_abort(asoc, NULL, sizeof(sctp_errhdr_t)); + abort = sctp_make_abort(asoc, NULL, sizeof(struct sctp_errhdr)); if (abort) { sctp_init_cause(abort, SCTP_ERROR_RSRC_LOW, 0); sctp_add_cmd_sf(cmds, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); @@ -841,9 +843,9 @@ static void sctp_cmd_assoc_update(sctp_cmd_seq_t *cmds, } /* Helper function to change the state of an association. */ -static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, +static void sctp_cmd_new_state(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, - sctp_state_t state) + enum sctp_state state) { struct sock *sk = asoc->base.sk; @@ -901,7 +903,7 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, } /* Helper function to delete an association. */ -static void sctp_cmd_delete_tcb(sctp_cmd_seq_t *cmds, +static void sctp_cmd_delete_tcb(struct sctp_cmd_seq *cmds, struct sctp_association *asoc) { struct sock *sk = asoc->base.sk; @@ -923,9 +925,9 @@ static void sctp_cmd_delete_tcb(sctp_cmd_seq_t *cmds, * destination address (we use active path instead of primary path just * because primary path may be inactive. */ -static void sctp_cmd_setup_t4(sctp_cmd_seq_t *cmds, - struct sctp_association *asoc, - struct sctp_chunk *chunk) +static void sctp_cmd_setup_t4(struct sctp_cmd_seq *cmds, + struct sctp_association *asoc, + struct sctp_chunk *chunk) { struct sctp_transport *t; @@ -935,7 +937,7 @@ static void sctp_cmd_setup_t4(sctp_cmd_seq_t *cmds, } /* Process an incoming Operation Error Chunk. */ -static void sctp_cmd_process_operr(sctp_cmd_seq_t *cmds, +static void sctp_cmd_process_operr(struct sctp_cmd_seq *cmds, struct sctp_association *asoc, struct sctp_chunk *chunk) { @@ -990,6 +992,7 @@ static void sctp_cmd_process_fwdtsn(struct sctp_ulpq *ulpq, struct sctp_chunk *chunk) { struct sctp_fwdtsn_skip *skip; + /* Walk through all the skipped SSNs */ sctp_walk_fwdtsn(skip, chunk) { sctp_ulpq_skip(ulpq, ntohs(skip->stream), ntohs(skip->ssn)); @@ -1002,8 +1005,8 @@ static void sctp_cmd_process_fwdtsn(struct sctp_ulpq *ulpq, static void sctp_cmd_del_non_primary(struct sctp_association *asoc) { struct sctp_transport *t; - struct list_head *pos; struct list_head *temp; + struct list_head *pos; list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { t = list_entry(pos, struct sctp_transport, transports); @@ -1024,9 +1027,9 @@ static void sctp_cmd_set_sk_err(struct sctp_association *asoc, int error) } /* Helper function to generate an association change event */ -static void sctp_cmd_assoc_change(sctp_cmd_seq_t *commands, - struct sctp_association *asoc, - u8 state) +static void sctp_cmd_assoc_change(struct sctp_cmd_seq *commands, + struct sctp_association *asoc, + u8 state) { struct sctp_ulpevent *ev; @@ -1039,7 +1042,7 @@ static void sctp_cmd_assoc_change(sctp_cmd_seq_t *commands, } /* Helper function to generate an adaptation indication event */ -static void sctp_cmd_adaptation_ind(sctp_cmd_seq_t *commands, +static void sctp_cmd_adaptation_ind(struct sctp_cmd_seq *commands, struct sctp_association *asoc) { struct sctp_ulpevent *ev; @@ -1052,8 +1055,8 @@ static void sctp_cmd_adaptation_ind(sctp_cmd_seq_t *commands, static void sctp_cmd_t1_timer_update(struct sctp_association *asoc, - sctp_event_timeout_t timer, - char *name) + enum sctp_event_timeout timer, + char *name) { struct sctp_transport *t; @@ -1139,22 +1142,20 @@ static void sctp_cmd_send_asconf(struct sctp_association *asoc) * If you want to understand all of lksctp, this is a * good place to start. */ -int sctp_do_sm(struct net *net, sctp_event_t event_type, sctp_subtype_t subtype, - sctp_state_t state, - struct sctp_endpoint *ep, - struct sctp_association *asoc, - void *event_arg, - gfp_t gfp) +int sctp_do_sm(struct net *net, enum sctp_event event_type, + union sctp_subtype subtype, enum sctp_state state, + struct sctp_endpoint *ep, struct sctp_association *asoc, + void *event_arg, gfp_t gfp) { - sctp_cmd_seq_t commands; - const sctp_sm_table_entry_t *state_fn; - sctp_disposition_t status; - int error = 0; - typedef const char *(printfn_t)(sctp_subtype_t); + typedef const char *(printfn_t)(union sctp_subtype); static printfn_t *table[] = { NULL, sctp_cname, sctp_tname, sctp_oname, sctp_pname, }; printfn_t *debug_fn __attribute__ ((unused)) = table[event_type]; + const struct sctp_sm_table_entry *state_fn; + struct sctp_cmd_seq commands; + enum sctp_disposition status; + int error = 0; /* Look up the state function, run it, and then process the * side effects. These three steps are the heart of lksctp. @@ -1178,13 +1179,14 @@ int sctp_do_sm(struct net *net, sctp_event_t event_type, sctp_subtype_t subtype, /***************************************************************** * This the master state function side effect processing function. *****************************************************************/ -static int sctp_side_effects(sctp_event_t event_type, sctp_subtype_t subtype, - sctp_state_t state, +static int sctp_side_effects(enum sctp_event event_type, + union sctp_subtype subtype, + enum sctp_state state, struct sctp_endpoint *ep, struct sctp_association **asoc, void *event_arg, - sctp_disposition_t status, - sctp_cmd_seq_t *commands, + enum sctp_disposition status, + struct sctp_cmd_seq *commands, gfp_t gfp) { int error; @@ -1263,29 +1265,27 @@ bail: ********************************************************************/ /* This is the side-effect interpreter. */ -static int sctp_cmd_interpreter(sctp_event_t event_type, - sctp_subtype_t subtype, - sctp_state_t state, +static int sctp_cmd_interpreter(enum sctp_event event_type, + union sctp_subtype subtype, + enum sctp_state state, struct sctp_endpoint *ep, struct sctp_association *asoc, void *event_arg, - sctp_disposition_t status, - sctp_cmd_seq_t *commands, + enum sctp_disposition status, + struct sctp_cmd_seq *commands, gfp_t gfp) { - struct sock *sk = ep->base.sk; - struct sctp_sock *sp = sctp_sk(sk); - int error = 0; - int force; - sctp_cmd_t *cmd; - struct sctp_chunk *new_obj; - struct sctp_chunk *chunk = NULL; + struct sctp_sock *sp = sctp_sk(ep->base.sk); + struct sctp_chunk *chunk = NULL, *new_obj; struct sctp_packet *packet; + struct sctp_sackhdr sackh; struct timer_list *timer; - unsigned long timeout; struct sctp_transport *t; - struct sctp_sackhdr sackh; + unsigned long timeout; + struct sctp_cmd *cmd; int local_cork = 0; + int error = 0; + int force; if (SCTP_EVENT_T_TIMEOUT != event_type) chunk = event_arg; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index b2a74c3823ee..8f8ccded13e4 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -59,103 +59,110 @@ #include <net/sctp/sm.h> #include <net/sctp/structs.h> -static struct sctp_packet *sctp_abort_pkt_new(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - struct sctp_chunk *chunk, - const void *payload, - size_t paylen); +static struct sctp_packet *sctp_abort_pkt_new( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + const void *payload, size_t paylen); static int sctp_eat_data(const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands); -static struct sctp_packet *sctp_ootb_pkt_new(struct net *net, - const struct sctp_association *asoc, - const struct sctp_chunk *chunk); + struct sctp_cmd_seq *commands); +static struct sctp_packet *sctp_ootb_pkt_new( + struct net *net, + const struct sctp_association *asoc, + const struct sctp_chunk *chunk); static void sctp_send_stale_cookie_err(struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, const struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_chunk *err_chunk); -static sctp_disposition_t sctp_sf_do_5_2_6_stale(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands); -static sctp_disposition_t sctp_sf_shut_8_4_5(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands); -static sctp_disposition_t sctp_sf_tabort_8_4_8(struct net *net, +static enum sctp_disposition sctp_sf_do_5_2_6_stale( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands); +static enum sctp_disposition sctp_sf_shut_8_4_5( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands); +static enum sctp_disposition sctp_sf_tabort_8_4_8( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands); + struct sctp_cmd_seq *commands); static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk); -static sctp_disposition_t sctp_stop_t1_and_abort(struct net *net, - sctp_cmd_seq_t *commands, - __be16 error, int sk_err, - const struct sctp_association *asoc, - struct sctp_transport *transport); +static enum sctp_disposition sctp_stop_t1_and_abort( + struct net *net, + struct sctp_cmd_seq *commands, + __be16 error, int sk_err, + const struct sctp_association *asoc, + struct sctp_transport *transport); -static sctp_disposition_t sctp_sf_abort_violation( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - void *arg, - sctp_cmd_seq_t *commands, - const __u8 *payload, - const size_t paylen); +static enum sctp_disposition sctp_sf_abort_violation( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + void *arg, + struct sctp_cmd_seq *commands, + const __u8 *payload, + const size_t paylen); -static sctp_disposition_t sctp_sf_violation_chunklen( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands); +static enum sctp_disposition sctp_sf_violation_chunklen( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands); -static sctp_disposition_t sctp_sf_violation_paramlen( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, void *ext, - sctp_cmd_seq_t *commands); +static enum sctp_disposition sctp_sf_violation_paramlen( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, void *ext, + struct sctp_cmd_seq *commands); -static sctp_disposition_t sctp_sf_violation_ctsn( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands); +static enum sctp_disposition sctp_sf_violation_ctsn( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands); -static sctp_disposition_t sctp_sf_violation_chunk( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands); +static enum sctp_disposition sctp_sf_violation_chunk( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands); -static sctp_ierror_t sctp_sf_authenticate(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - struct sctp_chunk *chunk); +static enum sctp_ierror sctp_sf_authenticate( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + struct sctp_chunk *chunk); -static sctp_disposition_t __sctp_sf_do_9_1_abort(struct net *net, +static enum sctp_disposition __sctp_sf_do_9_1_abort( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands); + struct sctp_cmd_seq *commands); /* Small helper function that checks if the chunk length * is of the appropriate length. The 'required_length' argument @@ -164,8 +171,8 @@ static sctp_disposition_t __sctp_sf_do_9_1_abort(struct net *net, * false = Invalid length * */ -static inline bool -sctp_chunk_length_valid(struct sctp_chunk *chunk, __u16 required_length) +static inline bool sctp_chunk_length_valid(struct sctp_chunk *chunk, + __u16 required_length) { __u16 chunk_length = ntohs(chunk->chunk_hdr->length); @@ -213,12 +220,11 @@ sctp_chunk_length_valid(struct sctp_chunk *chunk, __u16 required_length) * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_4_C(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_4_C(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; struct sctp_ulpevent *ev; @@ -299,19 +305,17 @@ sctp_disposition_t sctp_sf_do_4_C(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_1B_init(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_1B_init(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; - struct sctp_chunk *repl; + struct sctp_chunk *chunk = arg, *repl, *err_chunk; + struct sctp_unrecognized_param *unk_param; struct sctp_association *new_asoc; - struct sctp_chunk *err_chunk; struct sctp_packet *packet; - sctp_unrecognized_param_t *unk_param; int len; /* 6.10 Bundling @@ -435,7 +439,7 @@ sctp_disposition_t sctp_sf_do_5_1B_init(struct net *net, * construct the parameters in INIT ACK by copying the * ERROR causes over. */ - unk_param = (sctp_unrecognized_param_t *) + unk_param = (struct sctp_unrecognized_param *) ((__u8 *)(err_chunk->chunk_hdr) + sizeof(struct sctp_chunkhdr)); /* Replace the cause code with the "Unrecognized parameter" @@ -495,15 +499,15 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_1C_ack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_1C_ack(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; struct sctp_init_chunk *initchunk; + struct sctp_chunk *chunk = arg; struct sctp_chunk *err_chunk; struct sctp_packet *packet; @@ -518,7 +522,7 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(struct net *net, return sctp_sf_violation_chunk(net, ep, asoc, type, arg, commands); /* Make sure that the INIT-ACK chunk has a valid length */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_initack_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_initack_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); /* Grab the INIT header. */ @@ -530,7 +534,7 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(struct net *net, (struct sctp_init_chunk *)chunk->chunk_hdr, chunk, &err_chunk)) { - sctp_error_t error = SCTP_ERROR_NO_RESOURCE; + enum sctp_error error = SCTP_ERROR_NO_RESOURCE; /* This chunk contains fatal error. It is to be discarded. * Send an ABORT, with causes. If there are no causes, @@ -645,20 +649,21 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_1D_ce(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_1D_ce(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; + struct sctp_ulpevent *ev, *ai_ev = NULL; struct sctp_association *new_asoc; struct sctp_init_chunk *peer_init; - struct sctp_chunk *repl; - struct sctp_ulpevent *ev, *ai_ev = NULL; - int error = 0; + struct sctp_chunk *chunk = arg; struct sctp_chunk *err_chk_p; + struct sctp_chunk *repl; struct sock *sk; + int error = 0; /* If the packet is an OOTB packet which is temporarily on the * control endpoint, respond with an ABORT. @@ -758,7 +763,7 @@ sctp_disposition_t sctp_sf_do_5_1D_ce(struct net *net, */ if (chunk->auth_chunk) { struct sctp_chunk auth; - sctp_ierror_t ret; + enum sctp_ierror ret; /* Make sure that we and the peer are AUTH capable */ if (!net->sctp.auth_enable || !new_asoc->peer.auth_capable) { @@ -872,11 +877,12 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_1E_ca(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_1E_ca(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; struct sctp_ulpevent *ev; @@ -950,11 +956,12 @@ nomem: } /* Generate and sendout a heartbeat packet. */ -static sctp_disposition_t sctp_sf_heartbeat(const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_heartbeat( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_transport *transport = (struct sctp_transport *) arg; struct sctp_chunk *reply; @@ -975,12 +982,12 @@ static sctp_disposition_t sctp_sf_heartbeat(const struct sctp_endpoint *ep, } /* Generate a HEARTBEAT packet on the given transport. */ -sctp_disposition_t sctp_sf_sendbeat_8_3(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_sendbeat_8_3(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_transport *transport = (struct sctp_transport *) arg; @@ -1023,11 +1030,12 @@ sctp_disposition_t sctp_sf_sendbeat_8_3(struct net *net, } /* resend asoc strreset_chunk. */ -sctp_disposition_t sctp_sf_send_reconf(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_send_reconf(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_transport *transport = arg; @@ -1074,12 +1082,11 @@ sctp_disposition_t sctp_sf_send_reconf(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_beat_8_3(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_beat_8_3(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { struct sctp_paramhdr *param_hdr; struct sctp_chunk *chunk = arg; @@ -1090,7 +1097,8 @@ sctp_disposition_t sctp_sf_beat_8_3(struct net *net, return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the HEARTBEAT chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_heartbeat_chunk_t))) + if (!sctp_chunk_length_valid(chunk, + sizeof(struct sctp_heartbeat_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); @@ -1098,7 +1106,7 @@ sctp_disposition_t sctp_sf_beat_8_3(struct net *net, * respond with a HEARTBEAT ACK that contains the Heartbeat * Information field copied from the received HEARTBEAT chunk. */ - chunk->subh.hb_hdr = (sctp_heartbeathdr_t *)chunk->skb->data; + chunk->subh.hb_hdr = (struct sctp_heartbeathdr *)chunk->skb->data; param_hdr = (struct sctp_paramhdr *)chunk->subh.hb_hdr; paylen = ntohs(chunk->chunk_hdr->length) - sizeof(struct sctp_chunkhdr); @@ -1148,34 +1156,32 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_backbeat_8_3(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_backbeat_8_3(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { + struct sctp_sender_hb_info *hbinfo; struct sctp_chunk *chunk = arg; - union sctp_addr from_addr; struct sctp_transport *link; - sctp_sender_hb_info_t *hbinfo; unsigned long max_interval; + union sctp_addr from_addr; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the HEARTBEAT-ACK chunk has a valid length. */ if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_chunkhdr) + - sizeof(sctp_sender_hb_info_t))) + sizeof(*hbinfo))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data; + hbinfo = (struct sctp_sender_hb_info *)chunk->skb->data; /* Make sure that the length of the parameter is what we expect */ - if (ntohs(hbinfo->param_hdr.length) != - sizeof(sctp_sender_hb_info_t)) { + if (ntohs(hbinfo->param_hdr.length) != sizeof(*hbinfo)) return SCTP_DISPOSITION_DISCARD; - } from_addr = hbinfo->daddr; link = sctp_assoc_lookup_paddr(asoc, &from_addr); @@ -1227,15 +1233,15 @@ sctp_disposition_t sctp_sf_backbeat_8_3(struct net *net, */ static int sctp_sf_send_restart_abort(struct net *net, union sctp_addr *ssa, struct sctp_chunk *init, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { - int len; - struct sctp_packet *pkt; + struct sctp_af *af = sctp_get_af_specific(ssa->v4.sin_family); union sctp_addr_param *addrparm; struct sctp_errhdr *errhdr; + char buffer[sizeof(*errhdr) + sizeof(*addrparm)]; struct sctp_endpoint *ep; - char buffer[sizeof(struct sctp_errhdr)+sizeof(union sctp_addr_param)]; - struct sctp_af *af = sctp_get_af_specific(ssa->v4.sin_family); + struct sctp_packet *pkt; + int len; /* Build the error on the stack. We are way to malloc crazy * throughout the code today. @@ -1245,7 +1251,7 @@ static int sctp_sf_send_restart_abort(struct net *net, union sctp_addr *ssa, /* Copy into a parm format. */ len = af->to_addr_param(ssa, addrparm); - len += sizeof(sctp_errhdr_t); + len += sizeof(*errhdr); errhdr->cause = SCTP_ERROR_RESTART; errhdr->length = htons(len); @@ -1292,7 +1298,7 @@ static bool list_has_sctp_addr(const struct list_head *list, static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, const struct sctp_association *asoc, struct sctp_chunk *init, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { struct net *net = sock_net(new_asoc->base.sk); struct sctp_transport *new_addr; @@ -1412,20 +1418,19 @@ static char sctp_tietags_compare(struct sctp_association *new_asoc, /* Common helper routine for both duplicate and simulataneous INIT * chunk handling. */ -static sctp_disposition_t sctp_sf_do_unexpected_init( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_do_unexpected_init( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - sctp_disposition_t retval; - struct sctp_chunk *chunk = arg; - struct sctp_chunk *repl; + struct sctp_chunk *chunk = arg, *repl, *err_chunk; + struct sctp_unrecognized_param *unk_param; struct sctp_association *new_asoc; - struct sctp_chunk *err_chunk; + enum sctp_disposition retval; struct sctp_packet *packet; - sctp_unrecognized_param_t *unk_param; int len; /* 6.10 Bundling @@ -1555,7 +1560,7 @@ static sctp_disposition_t sctp_sf_do_unexpected_init( * construct the parameters in INIT ACK by copying the * ERROR causes over. */ - unk_param = (sctp_unrecognized_param_t *) + unk_param = (struct sctp_unrecognized_param *) ((__u8 *)(err_chunk->chunk_hdr) + sizeof(struct sctp_chunkhdr)); /* Replace the cause code with the "Unrecognized parameter" @@ -1626,12 +1631,13 @@ cleanup: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_2_1_siminit(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_2_1_siminit( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* Call helper to do the real work for both simulataneous and * duplicate INIT chunk handling. @@ -1680,12 +1686,13 @@ sctp_disposition_t sctp_sf_do_5_2_1_siminit(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_2_2_dupinit(struct net *net, +enum sctp_disposition sctp_sf_do_5_2_2_dupinit( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { /* Call helper to do the real work for both simulataneous and * duplicate INIT chunk handling. @@ -1703,11 +1710,13 @@ sctp_disposition_t sctp_sf_do_5_2_2_dupinit(struct net *net, * An unexpected INIT ACK usually indicates the processing of an old or * duplicated INIT chunk. */ -sctp_disposition_t sctp_sf_do_5_2_3_initack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_5_2_3_initack( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* Per the above section, we'll discard the chunk if we have an * endpoint. If this is an OOTB INIT-ACK, treat it as such. @@ -1723,18 +1732,19 @@ sctp_disposition_t sctp_sf_do_5_2_3_initack(struct net *net, * Section 5.2.4 * A) In this case, the peer may have restarted. */ -static sctp_disposition_t sctp_sf_do_dupcook_a(struct net *net, +static enum sctp_disposition sctp_sf_do_dupcook_a( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_association *new_asoc) { struct sctp_init_chunk *peer_init; + enum sctp_disposition disposition; struct sctp_ulpevent *ev; struct sctp_chunk *repl; struct sctp_chunk *err; - sctp_disposition_t disposition; /* new_asoc is a brand-new association, so these are not yet * side effects--it is safe to run them here. @@ -1838,11 +1848,12 @@ nomem: * after responding to the local endpoint's INIT */ /* This case represents an initialization collision. */ -static sctp_disposition_t sctp_sf_do_dupcook_b(struct net *net, +static enum sctp_disposition sctp_sf_do_dupcook_b( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_association *new_asoc) { struct sctp_init_chunk *peer_init; @@ -1909,11 +1920,12 @@ nomem: * but a new tag of its own. */ /* This case represents an initialization collision. */ -static sctp_disposition_t sctp_sf_do_dupcook_c(struct net *net, +static enum sctp_disposition sctp_sf_do_dupcook_c( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_association *new_asoc) { /* The cookie should be silently discarded. @@ -1931,11 +1943,12 @@ static sctp_disposition_t sctp_sf_do_dupcook_c(struct net *net, * enter the ESTABLISHED state, if it has not already done so. */ /* This case represents an initialization collision. */ -static sctp_disposition_t sctp_sf_do_dupcook_d(struct net *net, +static enum sctp_disposition sctp_sf_do_dupcook_d( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_association *new_asoc) { struct sctp_ulpevent *ev = NULL, *ai_ev = NULL; @@ -2026,19 +2039,20 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_5_2_4_dupcook(struct net *net, +enum sctp_disposition sctp_sf_do_5_2_4_dupcook( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { - sctp_disposition_t retval; - struct sctp_chunk *chunk = arg; struct sctp_association *new_asoc; + struct sctp_chunk *chunk = arg; + enum sctp_disposition retval; + struct sctp_chunk *err_chk_p; int error = 0; char action; - struct sctp_chunk *err_chk_p; /* Make sure that the chunk has a valid length from the protocol * perspective. In this case check to make sure we have at least @@ -2144,13 +2158,13 @@ nomem: * * See sctp_sf_do_9_1_abort(). */ -sctp_disposition_t sctp_sf_shutdown_pending_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_shutdown_pending_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -2167,7 +2181,7 @@ sctp_disposition_t sctp_sf_shutdown_pending_abort( * as we do not know its true length. So, to be safe, discard the * packet. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_abort_chunk))) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* ADD-IP: Special case for ABORT chunks @@ -2187,12 +2201,13 @@ sctp_disposition_t sctp_sf_shutdown_pending_abort( * * See sctp_sf_do_9_1_abort(). */ -sctp_disposition_t sctp_sf_shutdown_sent_abort(struct net *net, +enum sctp_disposition sctp_sf_shutdown_sent_abort( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -2209,7 +2224,7 @@ sctp_disposition_t sctp_sf_shutdown_sent_abort(struct net *net, * as we do not know its true length. So, to be safe, discard the * packet. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_abort_chunk))) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* ADD-IP: Special case for ABORT chunks @@ -2237,13 +2252,13 @@ sctp_disposition_t sctp_sf_shutdown_sent_abort(struct net *net, * * See sctp_sf_do_9_1_abort(). */ -sctp_disposition_t sctp_sf_shutdown_ack_sent_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_shutdown_ack_sent_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* The same T2 timer, so we should be able to use * common function with the SHUTDOWN-SENT state. @@ -2265,15 +2280,16 @@ sctp_disposition_t sctp_sf_shutdown_ack_sent_abort( * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_cookie_echoed_err(struct net *net, +enum sctp_disposition sctp_sf_cookie_echoed_err( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; - sctp_errhdr_t *err; + struct sctp_errhdr *err; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); @@ -2281,7 +2297,7 @@ sctp_disposition_t sctp_sf_cookie_echoed_err(struct net *net, /* Make sure that the ERROR chunk has a valid length. * The parameter walking depends on this as well. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_operr_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_operr_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); @@ -2329,20 +2345,20 @@ sctp_disposition_t sctp_sf_cookie_echoed_err(struct net *net, * * The return value is the disposition of the chunk. */ -static sctp_disposition_t sctp_sf_do_5_2_6_stale(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_do_5_2_6_stale( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; - u32 stale; - sctp_cookie_preserve_param_t bht; - sctp_errhdr_t *err; - struct sctp_chunk *reply; - struct sctp_bind_addr *bp; int attempts = asoc->init_err_counter + 1; + struct sctp_chunk *chunk = arg, *reply; + struct sctp_cookie_preserve_param bht; + struct sctp_bind_addr *bp; + struct sctp_errhdr *err; + u32 stale; if (attempts > asoc->max_init_attempts) { sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, @@ -2352,7 +2368,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(struct net *net, return SCTP_DISPOSITION_DELETE_TCB; } - err = (sctp_errhdr_t *)(chunk->skb->data); + err = (struct sctp_errhdr *)(chunk->skb->data); /* When calculating the time extension, an implementation * SHOULD use the RTT information measured based on the @@ -2368,7 +2384,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(struct net *net, * to give ample time to retransmit the new cookie and thus * yield a higher probability of success on the reattempt. */ - stale = ntohl(*(__be32 *)((u8 *)err + sizeof(sctp_errhdr_t))); + stale = ntohl(*(__be32 *)((u8 *)err + sizeof(*err))); stale = (stale * 2) / 1000; bht.param_hdr.type = SCTP_PARAM_COOKIE_PRESERVATIVE; @@ -2452,12 +2468,13 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_9_1_abort(struct net *net, +enum sctp_disposition sctp_sf_do_9_1_abort( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -2474,7 +2491,7 @@ sctp_disposition_t sctp_sf_do_9_1_abort(struct net *net, * as we do not know its true length. So, to be safe, discard the * packet. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_abort_chunk))) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* ADD-IP: Special case for ABORT chunks @@ -2489,27 +2506,29 @@ sctp_disposition_t sctp_sf_do_9_1_abort(struct net *net, return __sctp_sf_do_9_1_abort(net, ep, asoc, type, arg, commands); } -static sctp_disposition_t __sctp_sf_do_9_1_abort(struct net *net, +static enum sctp_disposition __sctp_sf_do_9_1_abort( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { + __be16 error = SCTP_ERROR_NO_ERROR; struct sctp_chunk *chunk = arg; unsigned int len; - __be16 error = SCTP_ERROR_NO_ERROR; /* See if we have an error cause code in the chunk. */ len = ntohs(chunk->chunk_hdr->length); if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr)) { + struct sctp_errhdr *err; - sctp_errhdr_t *err; sctp_walk_errors(err, chunk->chunk_hdr); if ((void *)err != (void *)chunk->chunk_end) - return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); + return sctp_sf_pdiscard(net, ep, asoc, type, arg, + commands); - error = ((sctp_errhdr_t *)chunk->skb->data)->cause; + error = ((struct sctp_errhdr *)chunk->skb->data)->cause; } sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ECONNRESET)); @@ -2526,16 +2545,17 @@ static sctp_disposition_t __sctp_sf_do_9_1_abort(struct net *net, * * See sctp_sf_do_9_1_abort() above. */ -sctp_disposition_t sctp_sf_cookie_wait_abort(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_wait_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { + __be16 error = SCTP_ERROR_NO_ERROR; struct sctp_chunk *chunk = arg; unsigned int len; - __be16 error = SCTP_ERROR_NO_ERROR; if (!sctp_vtag_verify_either(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); @@ -2550,13 +2570,13 @@ sctp_disposition_t sctp_sf_cookie_wait_abort(struct net *net, * as we do not know its true length. So, to be safe, discard the * packet. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_abort_chunk))) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* See if we have an error cause code in the chunk. */ len = ntohs(chunk->chunk_hdr->length); if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr)) - error = ((sctp_errhdr_t *)chunk->skb->data)->cause; + error = ((struct sctp_errhdr *)chunk->skb->data)->cause; return sctp_stop_t1_and_abort(net, commands, error, ECONNREFUSED, asoc, chunk->transport); @@ -2565,12 +2585,13 @@ sctp_disposition_t sctp_sf_cookie_wait_abort(struct net *net, /* * Process an incoming ICMP as an ABORT. (COOKIE-WAIT state) */ -sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(struct net *net, +enum sctp_disposition sctp_sf_cookie_wait_icmp_abort( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { return sctp_stop_t1_and_abort(net, commands, SCTP_ERROR_NO_ERROR, ENOPROTOOPT, asoc, @@ -2580,12 +2601,13 @@ sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(struct net *net, /* * Process an ABORT. (COOKIE-ECHOED state) */ -sctp_disposition_t sctp_sf_cookie_echoed_abort(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_echoed_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* There is a single T1 timer, so we should be able to use * common function with the COOKIE-WAIT state. @@ -2598,11 +2620,12 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(struct net *net, * * This is common code called by several sctp_sf_*_abort() functions above. */ -static sctp_disposition_t sctp_stop_t1_and_abort(struct net *net, - sctp_cmd_seq_t *commands, - __be16 error, int sk_err, - const struct sctp_association *asoc, - struct sctp_transport *transport) +static enum sctp_disposition sctp_stop_t1_and_abort( + struct net *net, + struct sctp_cmd_seq *commands, + __be16 error, int sk_err, + const struct sctp_association *asoc, + struct sctp_transport *transport) { pr_debug("%s: ABORT received (INIT)\n", __func__); @@ -2652,16 +2675,17 @@ static sctp_disposition_t sctp_stop_t1_and_abort(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_9_2_shutdown(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { + enum sctp_disposition disposition; struct sctp_chunk *chunk = arg; - sctp_shutdownhdr_t *sdh; - sctp_disposition_t disposition; + struct sctp_shutdownhdr *sdh; struct sctp_ulpevent *ev; __u32 ctsn; @@ -2669,14 +2693,13 @@ sctp_disposition_t sctp_sf_do_9_2_shutdown(struct net *net, return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the SHUTDOWN chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, - sizeof(struct sctp_shutdown_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_shutdown_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); /* Convert the elaborate header. */ - sdh = (sctp_shutdownhdr_t *)chunk->skb->data; - skb_pull(chunk->skb, sizeof(sctp_shutdownhdr_t)); + sdh = (struct sctp_shutdownhdr *)chunk->skb->data; + skb_pull(chunk->skb, sizeof(*sdh)); chunk->subh.shutdown_hdr = sdh; ctsn = ntohl(sdh->cum_tsn_ack); @@ -2742,27 +2765,27 @@ out: * The Cumulative TSN Ack of the received SHUTDOWN chunk * MUST be processed. */ -sctp_disposition_t sctp_sf_do_9_2_shut_ctsn(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_shut_ctsn( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; - sctp_shutdownhdr_t *sdh; + struct sctp_shutdownhdr *sdh; __u32 ctsn; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the SHUTDOWN chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, - sizeof(struct sctp_shutdown_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_shutdown_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - sdh = (sctp_shutdownhdr_t *)chunk->skb->data; + sdh = (struct sctp_shutdownhdr *)chunk->skb->data; ctsn = ntohl(sdh->cum_tsn_ack); if (TSN_lt(ctsn, asoc->ctsn_ack_point)) { @@ -2796,14 +2819,15 @@ sctp_disposition_t sctp_sf_do_9_2_shut_ctsn(struct net *net, * that belong to this association, it should discard the INIT chunk and * retransmit the SHUTDOWN ACK chunk. */ -sctp_disposition_t sctp_sf_do_9_2_reshutack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_reshutack( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = (struct sctp_chunk *) arg; + struct sctp_chunk *chunk = arg; struct sctp_chunk *reply; /* Make sure that the chunk has a valid length */ @@ -2860,26 +2884,26 @@ nomem: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_ecn_cwr(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_ecn_cwr(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - sctp_cwrhdr_t *cwr; struct sctp_chunk *chunk = arg; + struct sctp_cwrhdr *cwr; u32 lowest_tsn; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_ecne_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_ecne_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - cwr = (sctp_cwrhdr_t *) chunk->skb->data; - skb_pull(chunk->skb, sizeof(sctp_cwrhdr_t)); + cwr = (struct sctp_cwrhdr *)chunk->skb->data; + skb_pull(chunk->skb, sizeof(*cwr)); lowest_tsn = ntohl(cwr->lowest_tsn); @@ -2916,25 +2940,24 @@ sctp_disposition_t sctp_sf_do_ecn_cwr(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_ecne(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_ecne(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { - sctp_ecnehdr_t *ecne; struct sctp_chunk *chunk = arg; + struct sctp_ecnehdr *ecne; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_ecne_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_ecne_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - ecne = (sctp_ecnehdr_t *) chunk->skb->data; - skb_pull(chunk->skb, sizeof(sctp_ecnehdr_t)); + ecne = (struct sctp_ecnehdr *)chunk->skb->data; + skb_pull(chunk->skb, sizeof(*ecne)); /* If this is a newer ECNE than the last CWR packet we sent out */ sctp_add_cmd_sf(commands, SCTP_CMD_ECN_ECNE, @@ -2973,15 +2996,15 @@ sctp_disposition_t sctp_sf_do_ecne(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_eat_data_6_2(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_data_6_2(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { + union sctp_arg force = SCTP_NOFORCE(); struct sctp_chunk *chunk = arg; - sctp_arg_t force = SCTP_NOFORCE(); int error; if (!sctp_vtag_verify(chunk, asoc)) { @@ -3093,12 +3116,13 @@ discard_noforce: * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_eat_data_fast_4_4(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_data_fast_4_4( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; int error; @@ -3184,22 +3208,22 @@ sctp_disposition_t sctp_sf_eat_data_fast_4_4(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_eat_sack_6_2(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_sack_6_2(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; - sctp_sackhdr_t *sackh; + struct sctp_sackhdr *sackh; __u32 ctsn; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the SACK chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_sack_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_sack_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); @@ -3258,12 +3282,13 @@ sctp_disposition_t sctp_sf_eat_sack_6_2(struct net *net, * * The return value is the disposition of the chunk. */ -static sctp_disposition_t sctp_sf_tabort_8_4_8(struct net *net, +static enum sctp_disposition sctp_sf_tabort_8_4_8( + struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { struct sctp_packet *packet = NULL; struct sctp_chunk *chunk = arg; @@ -3308,21 +3333,21 @@ static sctp_disposition_t sctp_sf_tabort_8_4_8(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_operr_notify(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_operr_notify(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; - sctp_errhdr_t *err; + struct sctp_errhdr *err; if (!sctp_vtag_verify(chunk, asoc)) return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); /* Make sure that the ERROR chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_operr_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_operr_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); sctp_walk_errors(err, chunk->chunk_hdr); @@ -3346,12 +3371,12 @@ sctp_disposition_t sctp_sf_operr_notify(struct net *net, * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_9_2_final(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_final(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; struct sctp_chunk *reply; @@ -3429,20 +3454,19 @@ nomem: * receiver of the OOTB packet shall discard the OOTB packet and take * no further action. */ -sctp_disposition_t sctp_sf_ootb(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_ootb(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; struct sk_buff *skb = chunk->skb; struct sctp_chunkhdr *ch; - sctp_errhdr_t *err; - __u8 *ch_end; - int ootb_shut_ack = 0; + struct sctp_errhdr *err; int ootb_cookie_ack = 0; + int ootb_shut_ack = 0; + __u8 *ch_end; SCTP_INC_STATS(net, SCTP_MIB_OUTOFBLUES); @@ -3518,16 +3542,17 @@ sctp_disposition_t sctp_sf_ootb(struct net *net, * (endpoint, asoc, type, arg, commands) * * Outputs - * (sctp_disposition_t) + * (enum sctp_disposition) * * The return value is the disposition of the chunk. */ -static sctp_disposition_t sctp_sf_shut_8_4_5(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_shut_8_4_5( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_packet *packet = NULL; struct sctp_chunk *chunk = arg; @@ -3584,12 +3609,12 @@ static sctp_disposition_t sctp_sf_shut_8_4_5(struct net *net, * chunks. --piggy ] * */ -sctp_disposition_t sctp_sf_do_8_5_1_E_sa(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_8_5_1_E_sa(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -3609,17 +3634,18 @@ sctp_disposition_t sctp_sf_do_8_5_1_E_sa(struct net *net, } /* ADDIP Section 4.2 Upon reception of an ASCONF Chunk. */ -sctp_disposition_t sctp_sf_do_asconf(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_asconf(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; - struct sctp_chunk *asconf_ack = NULL; - struct sctp_paramhdr *err_param = NULL; - sctp_addiphdr_t *hdr; - __u32 serial; + struct sctp_paramhdr *err_param = NULL; + struct sctp_chunk *asconf_ack = NULL; + struct sctp_chunk *chunk = arg; + struct sctp_addiphdr *hdr; + __u32 serial; if (!sctp_vtag_verify(chunk, asoc)) { sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, @@ -3634,14 +3660,15 @@ sctp_disposition_t sctp_sf_do_asconf(struct net *net, * described in [I-D.ietf-tsvwg-sctp-auth]. */ if (!net->sctp.addip_noauth && !chunk->auth) - return sctp_sf_discard_chunk(net, ep, asoc, type, arg, commands); + return sctp_sf_discard_chunk(net, ep, asoc, type, arg, + commands); /* Make sure that the ASCONF ADDIP chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, sizeof(sctp_addip_chunk_t))) + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_addip_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - hdr = (sctp_addiphdr_t *)chunk->skb->data; + hdr = (struct sctp_addiphdr *)chunk->skb->data; serial = ntohl(hdr->serial); /* Verify the ASCONF chunk before processing it. */ @@ -3725,18 +3752,19 @@ sctp_disposition_t sctp_sf_do_asconf(struct net *net, * When building TLV parameters for the ASCONF Chunk that will add or * delete IP addresses the D0 to D13 rules should be applied: */ -sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_asconf_ack(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *asconf_ack = arg; - struct sctp_chunk *last_asconf = asoc->addip_last_asconf; - struct sctp_chunk *abort; - struct sctp_paramhdr *err_param = NULL; - sctp_addiphdr_t *addip_hdr; - __u32 sent_serial, rcvd_serial; + struct sctp_chunk *last_asconf = asoc->addip_last_asconf; + struct sctp_paramhdr *err_param = NULL; + struct sctp_chunk *asconf_ack = arg; + struct sctp_addiphdr *addip_hdr; + __u32 sent_serial, rcvd_serial; + struct sctp_chunk *abort; if (!sctp_vtag_verify(asconf_ack, asoc)) { sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, @@ -3751,14 +3779,16 @@ sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, * described in [I-D.ietf-tsvwg-sctp-auth]. */ if (!net->sctp.addip_noauth && !asconf_ack->auth) - return sctp_sf_discard_chunk(net, ep, asoc, type, arg, commands); + return sctp_sf_discard_chunk(net, ep, asoc, type, arg, + commands); /* Make sure that the ADDIP chunk has a valid length. */ - if (!sctp_chunk_length_valid(asconf_ack, sizeof(sctp_addip_chunk_t))) + if (!sctp_chunk_length_valid(asconf_ack, + sizeof(struct sctp_addip_chunk))) return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, commands); - addip_hdr = (sctp_addiphdr_t *)asconf_ack->skb->data; + addip_hdr = (struct sctp_addiphdr *)asconf_ack->skb->data; rcvd_serial = ntohl(addip_hdr->serial); /* Verify the ASCONF-ACK chunk before processing it. */ @@ -3767,7 +3797,7 @@ sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, (void *)err_param, commands); if (last_asconf) { - addip_hdr = (sctp_addiphdr_t *)last_asconf->subh.addip_hdr; + addip_hdr = (struct sctp_addiphdr *)last_asconf->subh.addip_hdr; sent_serial = ntohl(addip_hdr->serial); } else { sent_serial = asoc->addip_serial - 1; @@ -3782,7 +3812,7 @@ sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, if (ADDIP_SERIAL_gte(rcvd_serial, sent_serial + 1) && !(asoc->addip_last_asconf)) { abort = sctp_make_abort(asoc, asconf_ack, - sizeof(sctp_errhdr_t)); + sizeof(struct sctp_errhdr)); if (abort) { sctp_init_cause(abort, SCTP_ERROR_ASCONF_ACK, 0); sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, @@ -3818,7 +3848,7 @@ sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, } abort = sctp_make_abort(asoc, asconf_ack, - sizeof(sctp_errhdr_t)); + sizeof(struct sctp_errhdr)); if (abort) { sctp_init_cause(abort, SCTP_ERROR_RSRC_LOW, 0); sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, @@ -3841,11 +3871,12 @@ sctp_disposition_t sctp_sf_do_asconf_ack(struct net *net, } /* RE-CONFIG Section 5.2 Upon reception of an RECONF Chunk. */ -sctp_disposition_t sctp_sf_do_reconf(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_reconf(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_paramhdr *err_param = NULL; struct sctp_chunk *chunk = arg; @@ -3917,15 +3948,15 @@ sctp_disposition_t sctp_sf_do_reconf(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_eat_fwd_tsn(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_fwd_tsn(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; struct sctp_fwdtsn_hdr *fwdtsn_hdr; + struct sctp_chunk *chunk = arg; struct sctp_fwdtsn_skip *skip; __u16 len; __u32 tsn; @@ -3987,16 +4018,16 @@ discard_noforce: return SCTP_DISPOSITION_DISCARD; } -sctp_disposition_t sctp_sf_eat_fwd_tsn_fast( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_fwd_tsn_fast( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; struct sctp_fwdtsn_hdr *fwdtsn_hdr; + struct sctp_chunk *chunk = arg; struct sctp_fwdtsn_skip *skip; __u16 len; __u32 tsn; @@ -4079,23 +4110,23 @@ gen_shutdown: * * The return value is the disposition of the chunk. */ -static sctp_ierror_t sctp_sf_authenticate(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - struct sctp_chunk *chunk) +static enum sctp_ierror sctp_sf_authenticate( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + struct sctp_chunk *chunk) { struct sctp_authhdr *auth_hdr; + __u8 *save_digest, *digest; struct sctp_hmac *hmac; unsigned int sig_len; __u16 key_id; - __u8 *save_digest; - __u8 *digest; /* Pull in the auth header, so we can do some more verification */ auth_hdr = (struct sctp_authhdr *)chunk->skb->data; chunk->subh.auth_hdr = auth_hdr; - skb_pull(chunk->skb, sizeof(struct sctp_authhdr)); + skb_pull(chunk->skb, sizeof(*auth_hdr)); /* Make sure that we support the HMAC algorithm from the auth * chunk. @@ -4114,7 +4145,8 @@ static sctp_ierror_t sctp_sf_authenticate(struct net *net, /* Make sure that the length of the signature matches what * we expect. */ - sig_len = ntohs(chunk->chunk_hdr->length) - sizeof(sctp_auth_chunk_t); + sig_len = ntohs(chunk->chunk_hdr->length) - + sizeof(struct sctp_auth_chunk); hmac = sctp_auth_get_hmac(ntohs(auth_hdr->hmac_id)); if (sig_len != hmac->hmac_len) return SCTP_IERROR_PROTO_VIOLATION; @@ -4136,8 +4168,8 @@ static sctp_ierror_t sctp_sf_authenticate(struct net *net, memset(digest, 0, sig_len); sctp_auth_calculate_hmac(asoc, chunk->skb, - (struct sctp_auth_chunk *)chunk->chunk_hdr, - GFP_ATOMIC); + (struct sctp_auth_chunk *)chunk->chunk_hdr, + GFP_ATOMIC); /* Discard the packet if the digests do not match */ if (memcmp(save_digest, digest, sig_len)) { @@ -4153,17 +4185,16 @@ nomem: return SCTP_IERROR_NOMEM; } -sctp_disposition_t sctp_sf_eat_auth(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_eat_auth(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { - struct sctp_authhdr *auth_hdr; struct sctp_chunk *chunk = arg; + struct sctp_authhdr *auth_hdr; struct sctp_chunk *err_chunk; - sctp_ierror_t error; + enum sctp_ierror error; /* Make sure that the peer has AUTH capable */ if (!asoc->peer.auth_capable) @@ -4250,12 +4281,12 @@ sctp_disposition_t sctp_sf_eat_auth(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_unk_chunk(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_unk_chunk(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *unk_chunk = arg; struct sctp_chunk *err_chunk; @@ -4330,12 +4361,12 @@ sctp_disposition_t sctp_sf_unk_chunk(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_discard_chunk(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_discard_chunk(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -4370,12 +4401,11 @@ sctp_disposition_t sctp_sf_discard_chunk(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_pdiscard(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_pdiscard(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { SCTP_INC_STATS(net, SCTP_MIB_IN_PKT_DISCARDS); sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET, SCTP_NULL()); @@ -4398,12 +4428,12 @@ sctp_disposition_t sctp_sf_pdiscard(struct net *net, * We simply tag the chunk as a violation. The state machine will log * the violation and continue. */ -sctp_disposition_t sctp_sf_violation(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_violation(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -4418,14 +4448,14 @@ sctp_disposition_t sctp_sf_violation(struct net *net, /* * Common function to handle a protocol violation. */ -static sctp_disposition_t sctp_sf_abort_violation( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - void *arg, - sctp_cmd_seq_t *commands, - const __u8 *payload, - const size_t paylen) +static enum sctp_disposition sctp_sf_abort_violation( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + void *arg, + struct sctp_cmd_seq *commands, + const __u8 *payload, + const size_t paylen) { struct sctp_packet *packet = NULL; struct sctp_chunk *chunk = arg; @@ -4454,11 +4484,10 @@ static sctp_disposition_t sctp_sf_abort_violation( /* Treat INIT-ACK as a special case during COOKIE-WAIT. */ if (chunk->chunk_hdr->type == SCTP_CID_INIT_ACK && !asoc->peer.i.init_tag) { - sctp_initack_chunk_t *initack; + struct sctp_initack_chunk *initack; - initack = (sctp_initack_chunk_t *)chunk->chunk_hdr; - if (!sctp_chunk_length_valid(chunk, - sizeof(sctp_initack_chunk_t))) + initack = (struct sctp_initack_chunk *)chunk->chunk_hdr; + if (!sctp_chunk_length_valid(chunk, sizeof(*initack))) abort->chunk_hdr->flags |= SCTP_CHUNK_FLAG_T; else { unsigned int inittag; @@ -4521,7 +4550,7 @@ nomem: * Handle a protocol violation when the chunk length is invalid. * "Invalid" length is identified as smaller than the minimal length a * given chunk can be. For example, a SACK chunk has invalid length - * if its length is set to be smaller than the size of sctp_sack_chunk_t. + * if its length is set to be smaller than the size of struct sctp_sack_chunk. * * We inform the other end by sending an ABORT with a Protocol Violation * error code. @@ -4536,18 +4565,18 @@ nomem: * * Generate an ABORT chunk and terminate the association. */ -static sctp_disposition_t sctp_sf_violation_chunklen( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_violation_chunklen( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { static const char err_str[] = "The following chunk had invalid length:"; return sctp_sf_abort_violation(net, ep, asoc, arg, commands, err_str, - sizeof(err_str)); + sizeof(err_str)); } /* @@ -4556,17 +4585,17 @@ static sctp_disposition_t sctp_sf_violation_chunklen( * or accumulated length in multi parameters exceeds the end of the chunk, * the length is considered as invalid. */ -static sctp_disposition_t sctp_sf_violation_paramlen( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, void *ext, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_violation_paramlen( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, void *ext, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = arg; struct sctp_paramhdr *param = ext; struct sctp_chunk *abort = NULL; + struct sctp_chunk *chunk = arg; if (sctp_auth_recv_cid(SCTP_CID_ABORT, asoc)) goto discard; @@ -4599,18 +4628,18 @@ nomem: * We inform the other end by sending an ABORT with a Protocol Violation * error code. */ -static sctp_disposition_t sctp_sf_violation_ctsn( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_violation_ctsn( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { static const char err_str[] = "The cumulative tsn ack beyond the max tsn currently sent:"; return sctp_sf_abort_violation(net, ep, asoc, arg, commands, err_str, - sizeof(err_str)); + sizeof(err_str)); } /* Handle protocol violation of an invalid chunk bundling. For example, @@ -4619,13 +4648,13 @@ static sctp_disposition_t sctp_sf_violation_ctsn( * statement from the specs. Additionally, there might be an attacker * on the path and we may not want to continue this communication. */ -static sctp_disposition_t sctp_sf_violation_chunk( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +static enum sctp_disposition sctp_sf_violation_chunk( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { static const char err_str[] = "The following chunk violates protocol:"; @@ -4633,7 +4662,7 @@ static sctp_disposition_t sctp_sf_violation_chunk( return sctp_sf_violation(net, ep, asoc, type, arg, commands); return sctp_sf_abort_violation(net, ep, asoc, arg, commands, err_str, - sizeof(err_str)); + sizeof(err_str)); } /*************************************************************************** * These are the state functions for handling primitive (Section 10) events. @@ -4695,15 +4724,15 @@ static sctp_disposition_t sctp_sf_violation_chunk( * * The return value is a disposition. */ -sctp_disposition_t sctp_sf_do_prm_asoc(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_prm_asoc(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *repl; struct sctp_association *my_asoc; + struct sctp_chunk *repl; /* The comment below says that we enter COOKIE-WAIT AFTER * sending the INIT, but that doesn't actually work in our @@ -4807,12 +4836,12 @@ nomem: * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_prm_send(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_prm_send(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_datamsg *msg = arg; @@ -4846,15 +4875,15 @@ sctp_disposition_t sctp_sf_do_prm_send(struct net *net, * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_9_2_prm_shutdown( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_prm_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - int disposition; + enum sctp_disposition disposition; /* From 9.2 Shutdown of an Association * Upon receipt of the SHUTDOWN primitive from its upper @@ -4872,6 +4901,7 @@ sctp_disposition_t sctp_sf_do_9_2_prm_shutdown( disposition = sctp_sf_do_9_2_start_shutdown(net, ep, asoc, type, arg, commands); } + return disposition; } @@ -4902,13 +4932,13 @@ sctp_disposition_t sctp_sf_do_9_2_prm_shutdown( * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_9_1_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_1_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* From 9.1 Abort of an Association * Upon receipt of the ABORT primitive from its upper @@ -4940,12 +4970,12 @@ sctp_disposition_t sctp_sf_do_9_1_prm_abort( } /* We tried an illegal operation on an association which is closed. */ -sctp_disposition_t sctp_sf_error_closed(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_error_closed(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_ERROR, SCTP_ERROR(-EINVAL)); return SCTP_DISPOSITION_CONSUME; @@ -4954,12 +4984,13 @@ sctp_disposition_t sctp_sf_error_closed(struct net *net, /* We tried an illegal operation on an association which is shutting * down. */ -sctp_disposition_t sctp_sf_error_shutdown(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_error_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_ERROR, SCTP_ERROR(-ESHUTDOWN)); @@ -4980,13 +5011,13 @@ sctp_disposition_t sctp_sf_error_shutdown(struct net *net, * Outputs * (timers) */ -sctp_disposition_t sctp_sf_cookie_wait_prm_shutdown( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_wait_prm_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); @@ -5015,12 +5046,13 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_shutdown( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_cookie_echoed_prm_shutdown( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_echoed_prm_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* There is a single T1 timer, so we should be able to use * common function with the COOKIE-WAIT state. @@ -5042,13 +5074,13 @@ sctp_disposition_t sctp_sf_cookie_echoed_prm_shutdown( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_cookie_wait_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_wait_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *abort = arg; @@ -5091,13 +5123,13 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_abort( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_cookie_echoed_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_cookie_echoed_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* There is a single T1 timer, so we should be able to use * common function with the COOKIE-WAIT state. @@ -5117,13 +5149,13 @@ sctp_disposition_t sctp_sf_cookie_echoed_prm_abort( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_shutdown_pending_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_shutdown_pending_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* Stop the T5-shutdown guard timer. */ sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, @@ -5144,13 +5176,13 @@ sctp_disposition_t sctp_sf_shutdown_pending_prm_abort( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_shutdown_sent_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_shutdown_sent_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* Stop the T2-shutdown timer. */ sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, @@ -5175,13 +5207,13 @@ sctp_disposition_t sctp_sf_shutdown_sent_prm_abort( * Outputs * (timers) */ -sctp_disposition_t sctp_sf_shutdown_ack_sent_prm_abort( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_shutdown_ack_sent_prm_abort( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { /* The same T2 timer, so we should be able to use * common function with the SHUTDOWN-SENT state. @@ -5211,13 +5243,13 @@ sctp_disposition_t sctp_sf_shutdown_ack_sent_prm_abort( * o destination transport address - the transport address of the * association on which a heartbeat should be issued. */ -sctp_disposition_t sctp_sf_do_prm_requestheartbeat( +enum sctp_disposition sctp_sf_do_prm_requestheartbeat( struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, - const sctp_subtype_t type, + const union sctp_subtype type, void *arg, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { if (SCTP_DISPOSITION_NOMEM == sctp_sf_heartbeat(ep, asoc, type, (struct sctp_transport *)arg, commands)) @@ -5244,12 +5276,12 @@ sctp_disposition_t sctp_sf_do_prm_requestheartbeat( * When an endpoint has an ASCONF signaled change to be sent to the * remote endpoint it should do A1 to A9 */ -sctp_disposition_t sctp_sf_do_prm_asconf(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_prm_asconf(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -5261,11 +5293,12 @@ sctp_disposition_t sctp_sf_do_prm_asconf(struct net *net, } /* RE-CONFIG Section 5.1 RECONF Chunk Procedures */ -sctp_disposition_t sctp_sf_do_prm_reconf(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_prm_reconf(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = arg; @@ -5278,13 +5311,13 @@ sctp_disposition_t sctp_sf_do_prm_reconf(struct net *net, * * The return value is the disposition of the primitive. */ -sctp_disposition_t sctp_sf_ignore_primitive( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_ignore_primitive( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { pr_debug("%s: primitive type:%d is ignored\n", __func__, type.primitive); @@ -5302,13 +5335,13 @@ sctp_disposition_t sctp_sf_ignore_primitive( * subscribes to this event, if there is no data to be sent or * retransmit, the stack will immediately send up this notification. */ -sctp_disposition_t sctp_sf_do_no_pending_tsn( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_no_pending_tsn( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_ulpevent *event; @@ -5334,13 +5367,13 @@ sctp_disposition_t sctp_sf_do_no_pending_tsn( * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_9_2_start_shutdown( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_start_shutdown( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *reply; @@ -5404,15 +5437,15 @@ nomem: * * The return value is the disposition. */ -sctp_disposition_t sctp_sf_do_9_2_shutdown_ack( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_9_2_shutdown_ack( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *chunk = (struct sctp_chunk *) arg; + struct sctp_chunk *chunk = arg; struct sctp_chunk *reply; /* There are 2 ways of getting here: @@ -5424,12 +5457,14 @@ sctp_disposition_t sctp_sf_do_9_2_shutdown_ack( */ if (chunk) { if (!sctp_vtag_verify(chunk, asoc)) - return sctp_sf_pdiscard(net, ep, asoc, type, arg, commands); + return sctp_sf_pdiscard(net, ep, asoc, type, arg, + commands); /* Make sure that the SHUTDOWN chunk has a valid length. */ - if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_shutdown_chunk_t))) - return sctp_sf_violation_chunklen(net, ep, asoc, type, arg, - commands); + if (!sctp_chunk_length_valid( + chunk, sizeof(struct sctp_shutdown_chunk))) + return sctp_sf_violation_chunklen(net, ep, asoc, type, + arg, commands); } /* If it has no more outstanding DATA chunks, the SHUTDOWN receiver @@ -5476,12 +5511,12 @@ nomem: * * The return value is the disposition of the event. */ -sctp_disposition_t sctp_sf_ignore_other(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_ignore_other(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { pr_debug("%s: the event other type:%d is ignored\n", __func__, type.other); @@ -5504,12 +5539,12 @@ sctp_disposition_t sctp_sf_ignore_other(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_do_6_3_3_rtx(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_6_3_3_rtx(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_transport *transport = arg; @@ -5592,12 +5627,12 @@ sctp_disposition_t sctp_sf_do_6_3_3_rtx(struct net *net, * allow. However, an SCTP transmitter MUST NOT be more aggressive than * the following algorithms allow. */ -sctp_disposition_t sctp_sf_do_6_2_sack(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_do_6_2_sack(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { SCTP_INC_STATS(net, SCTP_MIB_DELAY_SACK_EXPIREDS); sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE()); @@ -5623,16 +5658,17 @@ sctp_disposition_t sctp_sf_do_6_2_sack(struct net *net, * (timers, events) * */ -sctp_disposition_t sctp_sf_t1_init_timer_expire(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_t1_init_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { + int attempts = asoc->init_err_counter + 1; struct sctp_chunk *repl = NULL; struct sctp_bind_addr *bp; - int attempts = asoc->init_err_counter + 1; pr_debug("%s: timer T1 expired (INIT)\n", __func__); @@ -5687,15 +5723,16 @@ sctp_disposition_t sctp_sf_t1_init_timer_expire(struct net *net, * (timers, events) * */ -sctp_disposition_t sctp_sf_t1_cookie_timer_expire(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_t1_cookie_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - struct sctp_chunk *repl = NULL; int attempts = asoc->init_err_counter + 1; + struct sctp_chunk *repl = NULL; pr_debug("%s: timer T1 expired (COOKIE-ECHO)\n", __func__); @@ -5737,12 +5774,13 @@ sctp_disposition_t sctp_sf_t1_cookie_timer_expire(struct net *net, * the T2-Shutdown timer, giving its peer ample opportunity to transmit * all of its queued DATA chunks that have not yet been sent. */ -sctp_disposition_t sctp_sf_t2_timer_expire(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_t2_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *reply = NULL; @@ -5807,13 +5845,13 @@ nomem: * ADDIP Section 4.1 ASCONF CHunk Procedures * If the T4 RTO timer expires the endpoint should do B1 to B5 */ -sctp_disposition_t sctp_sf_t4_timer_expire( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_t4_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *chunk = asoc->addip_last_asconf; struct sctp_transport *transport = chunk->transport; @@ -5879,12 +5917,13 @@ sctp_disposition_t sctp_sf_t4_timer_expire( * At the expiration of this timer the sender SHOULD abort the association * by sending an ABORT chunk. */ -sctp_disposition_t sctp_sf_t5_timer_expire(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_t5_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { struct sctp_chunk *reply = NULL; @@ -5915,15 +5954,15 @@ nomem: * The work that needs to be done is same as when SHUTDOWN is initiated by * the user. So this routine looks same as sctp_sf_do_9_2_prm_shutdown(). */ -sctp_disposition_t sctp_sf_autoclose_timer_expire( - struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_autoclose_timer_expire( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { - int disposition; + enum sctp_disposition disposition; SCTP_INC_STATS(net, SCTP_MIB_AUTOCLOSE_EXPIREDS); @@ -5943,6 +5982,7 @@ sctp_disposition_t sctp_sf_autoclose_timer_expire( disposition = sctp_sf_do_9_2_start_shutdown(net, ep, asoc, type, arg, commands); } + return disposition; } @@ -5958,12 +5998,11 @@ sctp_disposition_t sctp_sf_autoclose_timer_expire( * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_not_impl(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_not_impl(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { return SCTP_DISPOSITION_NOT_IMPL; } @@ -5976,12 +6015,11 @@ sctp_disposition_t sctp_sf_not_impl(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_bug(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_bug(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, struct sctp_cmd_seq *commands) { return SCTP_DISPOSITION_BUG; } @@ -5997,12 +6035,12 @@ sctp_disposition_t sctp_sf_bug(struct net *net, * * The return value is the disposition of the chunk. */ -sctp_disposition_t sctp_sf_timer_ignore(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - const sctp_subtype_t type, - void *arg, - sctp_cmd_seq_t *commands) +enum sctp_disposition sctp_sf_timer_ignore(struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const union sctp_subtype type, + void *arg, + struct sctp_cmd_seq *commands) { pr_debug("%s: timer %d ignored\n", __func__, type.chunk); @@ -6017,9 +6055,9 @@ sctp_disposition_t sctp_sf_timer_ignore(struct net *net, static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk) { struct sctp_sackhdr *sack; + __u16 num_dup_tsns; unsigned int len; __u16 num_blocks; - __u16 num_dup_tsns; /* Protect ourselves from reading too far into * the skb from a bogus sender. @@ -6041,12 +6079,12 @@ static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk) /* Create an ABORT packet to be sent as a response, with the specified * error causes. */ -static struct sctp_packet *sctp_abort_pkt_new(struct net *net, - const struct sctp_endpoint *ep, - const struct sctp_association *asoc, - struct sctp_chunk *chunk, - const void *payload, - size_t paylen) +static struct sctp_packet *sctp_abort_pkt_new( + struct net *net, + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + const void *payload, size_t paylen) { struct sctp_packet *packet; struct sctp_chunk *abort; @@ -6083,14 +6121,14 @@ static struct sctp_packet *sctp_abort_pkt_new(struct net *net, } /* Allocate a packet for responding in the OOTB conditions. */ -static struct sctp_packet *sctp_ootb_pkt_new(struct net *net, - const struct sctp_association *asoc, - const struct sctp_chunk *chunk) +static struct sctp_packet *sctp_ootb_pkt_new( + struct net *net, + const struct sctp_association *asoc, + const struct sctp_chunk *chunk) { - struct sctp_packet *packet; struct sctp_transport *transport; - __u16 sport; - __u16 dport; + struct sctp_packet *packet; + __u16 sport, dport; __u32 vtag; /* Get the source and destination port from the inbound packet. */ @@ -6107,9 +6145,9 @@ static struct sctp_packet *sctp_ootb_pkt_new(struct net *net, switch (chunk->chunk_hdr->type) { case SCTP_CID_INIT_ACK: { - sctp_initack_chunk_t *initack; + struct sctp_initack_chunk *initack; - initack = (sctp_initack_chunk_t *)chunk->chunk_hdr; + initack = (struct sctp_initack_chunk *)chunk->chunk_hdr; vtag = ntohl(initack->init_hdr.init_tag); break; } @@ -6168,7 +6206,7 @@ static void sctp_send_stale_cookie_err(struct net *net, const struct sctp_endpoint *ep, const struct sctp_association *asoc, const struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands, + struct sctp_cmd_seq *commands, struct sctp_chunk *err_chunk) { struct sctp_packet *packet; @@ -6197,20 +6235,19 @@ static void sctp_send_stale_cookie_err(struct net *net, /* Process a data chunk */ static int sctp_eat_data(const struct sctp_association *asoc, struct sctp_chunk *chunk, - sctp_cmd_seq_t *commands) + struct sctp_cmd_seq *commands) { - struct sctp_datahdr *data_hdr; - struct sctp_chunk *err; - size_t datalen; - sctp_verb_t deliver; - int tmp; - __u32 tsn; struct sctp_tsnmap *map = (struct sctp_tsnmap *)&asoc->peer.tsn_map; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); - u16 ssn; - u16 sid; + struct sctp_datahdr *data_hdr; + struct sctp_chunk *err; + enum sctp_verb deliver; + size_t datalen; u8 ordered = 0; + u16 ssn, sid; + __u32 tsn; + int tmp; data_hdr = (struct sctp_datahdr *)chunk->skb->data; chunk->subh.data_hdr = data_hdr; diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 3e958c1c4b95..79b6bee5b768 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -45,26 +45,27 @@ #include <net/sctp/sctp.h> #include <net/sctp/sm.h> -static const sctp_sm_table_entry_t +static const struct sctp_sm_table_entry primitive_event_table[SCTP_NUM_PRIMITIVE_TYPES][SCTP_STATE_NUM_STATES]; -static const sctp_sm_table_entry_t +static const struct sctp_sm_table_entry other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_STATE_NUM_STATES]; -static const sctp_sm_table_entry_t +static const struct sctp_sm_table_entry timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][SCTP_STATE_NUM_STATES]; -static const sctp_sm_table_entry_t *sctp_chunk_event_lookup(struct net *net, - enum sctp_cid cid, - sctp_state_t state); +static const struct sctp_sm_table_entry *sctp_chunk_event_lookup( + struct net *net, + enum sctp_cid cid, + enum sctp_state state); -static const sctp_sm_table_entry_t bug = { +static const struct sctp_sm_table_entry bug = { .fn = sctp_sf_bug, .name = "sctp_sf_bug" }; #define DO_LOOKUP(_max, _type, _table) \ ({ \ - const sctp_sm_table_entry_t *rtn; \ + const struct sctp_sm_table_entry *rtn; \ \ if ((event_subtype._type > (_max))) { \ pr_warn("table %p possible attack: event %d exceeds max %d\n", \ @@ -76,10 +77,11 @@ static const sctp_sm_table_entry_t bug = { rtn; \ }) -const sctp_sm_table_entry_t *sctp_sm_lookup_event(struct net *net, - sctp_event_t event_type, - sctp_state_t state, - sctp_subtype_t event_subtype) +const struct sctp_sm_table_entry *sctp_sm_lookup_event( + struct net *net, + enum sctp_event event_type, + enum sctp_state state, + union sctp_subtype event_subtype) { switch (event_type) { case SCTP_EVENT_T_CHUNK: @@ -392,7 +394,8 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(struct net *net, * * For base protocol (RFC 2960). */ -static const sctp_sm_table_entry_t chunk_event_table[SCTP_NUM_BASE_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +chunk_event_table[SCTP_NUM_BASE_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_DATA, TYPE_SCTP_INIT, TYPE_SCTP_INIT_ACK, @@ -451,7 +454,8 @@ static const sctp_sm_table_entry_t chunk_event_table[SCTP_NUM_BASE_CHUNK_TYPES][ /* The primary index for this table is the chunk type. * The secondary index for this table is the state. */ -static const sctp_sm_table_entry_t addip_chunk_event_table[SCTP_NUM_ADDIP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +addip_chunk_event_table[SCTP_NUM_ADDIP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_ASCONF, TYPE_SCTP_ASCONF_ACK, }; /*state_fn_t addip_chunk_event_table[][] */ @@ -478,7 +482,8 @@ static const sctp_sm_table_entry_t addip_chunk_event_table[SCTP_NUM_ADDIP_CHUNK_ /* The primary index for this table is the chunk type. * The secondary index for this table is the state. */ -static const sctp_sm_table_entry_t prsctp_chunk_event_table[SCTP_NUM_PRSCTP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +prsctp_chunk_event_table[SCTP_NUM_PRSCTP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_FWD_TSN, }; /*state_fn_t prsctp_chunk_event_table[][] */ @@ -504,7 +509,8 @@ static const sctp_sm_table_entry_t prsctp_chunk_event_table[SCTP_NUM_PRSCTP_CHUN /* The primary index for this table is the chunk type. * The secondary index for this table is the state. */ -static const sctp_sm_table_entry_t reconf_chunk_event_table[SCTP_NUM_RECONF_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +reconf_chunk_event_table[SCTP_NUM_RECONF_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_RECONF, }; /*state_fn_t reconf_chunk_event_table[][] */ @@ -530,11 +536,12 @@ static const sctp_sm_table_entry_t reconf_chunk_event_table[SCTP_NUM_RECONF_CHUN /* The primary index for this table is the chunk type. * The secondary index for this table is the state. */ -static const sctp_sm_table_entry_t auth_chunk_event_table[SCTP_NUM_AUTH_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +auth_chunk_event_table[SCTP_NUM_AUTH_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_AUTH, }; /*state_fn_t auth_chunk_event_table[][] */ -static const sctp_sm_table_entry_t +static const struct sctp_sm_table_entry chunk_event_table_unknown[SCTP_STATE_NUM_STATES] = { /* SCTP_STATE_CLOSED */ TYPE_SCTP_FUNC(sctp_sf_ootb), @@ -691,7 +698,8 @@ chunk_event_table_unknown[SCTP_STATE_NUM_STATES] = { /* The primary index for this table is the primitive type. * The secondary index for this table is the state. */ -static const sctp_sm_table_entry_t primitive_event_table[SCTP_NUM_PRIMITIVE_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +primitive_event_table[SCTP_NUM_PRIMITIVE_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_PRIMITIVE_ASSOCIATE, TYPE_SCTP_PRIMITIVE_SHUTDOWN, TYPE_SCTP_PRIMITIVE_ABORT, @@ -739,7 +747,8 @@ static const sctp_sm_table_entry_t primitive_event_table[SCTP_NUM_PRIMITIVE_TYPE TYPE_SCTP_FUNC(sctp_sf_ignore_other), \ } -static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_OTHER_NO_PENDING_TSN, TYPE_SCTP_OTHER_ICMP_PROTO_UNREACH, }; @@ -953,7 +962,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_ TYPE_SCTP_FUNC(sctp_sf_timer_ignore), \ } -static const sctp_sm_table_entry_t timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][SCTP_STATE_NUM_STATES] = { +static const struct sctp_sm_table_entry +timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][SCTP_STATE_NUM_STATES] = { TYPE_SCTP_EVENT_TIMEOUT_NONE, TYPE_SCTP_EVENT_TIMEOUT_T1_COOKIE, TYPE_SCTP_EVENT_TIMEOUT_T1_INIT, @@ -967,9 +977,10 @@ static const sctp_sm_table_entry_t timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][S TYPE_SCTP_EVENT_TIMEOUT_AUTOCLOSE, }; -static const sctp_sm_table_entry_t *sctp_chunk_event_lookup(struct net *net, - enum sctp_cid cid, - sctp_state_t state) +static const struct sctp_sm_table_entry *sctp_chunk_event_lookup( + struct net *net, + enum sctp_cid cid, + enum sctp_state state) { if (state > SCTP_STATE_MAX) return &bug; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 1db478e34520..1b00a1e09b93 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -100,8 +100,9 @@ static int sctp_send_asconf(struct sctp_association *asoc, struct sctp_chunk *chunk); static int sctp_do_bind(struct sock *, union sctp_addr *, int); static int sctp_autobind(struct sock *sk); -static void sctp_sock_migrate(struct sock *, struct sock *, - struct sctp_association *, sctp_socket_type_t); +static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, + struct sctp_association *assoc, + enum sctp_socket_type type); static unsigned long sctp_memory_pressure; static atomic_long_t sctp_memory_allocated; @@ -1055,7 +1056,7 @@ static int __sctp_connect(struct sock *sk, struct sctp_association *asoc2; struct sctp_transport *transport; union sctp_addr to; - sctp_scope_t scope; + enum sctp_scope scope; long timeo; int err = 0; int addrcnt = 0; @@ -1593,7 +1594,8 @@ static int sctp_error(struct sock *sk, int flags, int err) */ /* BUG: We do not implement the equivalent of sk_stream_wait_memory(). */ -static int sctp_msghdr_parse(const struct msghdr *, sctp_cmsgs_t *); +static int sctp_msghdr_parse(const struct msghdr *msg, + struct sctp_cmsgs *cmsgs); static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) { @@ -1609,8 +1611,8 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) struct sctp_sndrcvinfo *sinfo; struct sctp_initmsg *sinit; sctp_assoc_t associd = 0; - sctp_cmsgs_t cmsgs = { NULL }; - sctp_scope_t scope; + struct sctp_cmsgs cmsgs = { NULL }; + enum sctp_scope scope; bool fill_sinfo_ttl = false, wait_connect = false; struct sctp_datamsg *datamsg; int msg_flags = msg->msg_flags; @@ -4538,8 +4540,7 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc, info->sctpi_ictrlchunks = asoc->stats.ictrlchunks; prim = asoc->peer.primary_path; - memcpy(&info->sctpi_p_address, &prim->ipaddr, - sizeof(struct sockaddr_storage)); + memcpy(&info->sctpi_p_address, &prim->ipaddr, sizeof(prim->ipaddr)); info->sctpi_p_state = prim->state; info->sctpi_p_cwnd = prim->cwnd; info->sctpi_p_srtt = prim->srtt; @@ -7445,10 +7446,10 @@ static int sctp_autobind(struct sock *sk) * msg_control * points here */ -static int sctp_msghdr_parse(const struct msghdr *msg, sctp_cmsgs_t *cmsgs) +static int sctp_msghdr_parse(const struct msghdr *msg, struct sctp_cmsgs *cmsgs) { - struct cmsghdr *cmsg; struct msghdr *my_msg = (struct msghdr *)msg; + struct cmsghdr *cmsg; for_each_cmsghdr(cmsg, my_msg) { if (!CMSG_OK(my_msg, cmsg)) @@ -8085,7 +8086,7 @@ static inline void sctp_copy_descendant(struct sock *sk_to, */ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, struct sctp_association *assoc, - sctp_socket_type_t type) + enum sctp_socket_type type) { struct sctp_sock *oldsp = sctp_sk(oldsk); struct sctp_sock *newsp = sctp_sk(newsk); diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 0e732f68c2bf..ef7ca44d6e6a 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -46,7 +46,7 @@ static int timer_max = 86400000; /* ms in one day */ static int int_max = INT_MAX; static int sack_timer_min = 1; static int sack_timer_max = 500; -static int addr_scope_max = 3; /* check sctp_scope_policy_t in include/net/sctp/constants.h for max entries */ +static int addr_scope_max = SCTP_SCOPE_POLICY_MAX; static int rwnd_scale_max = 16; static int rto_alpha_min = 0; static int rto_beta_min = 0; diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 80a97c8501a7..2d9bd3776bc8 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -490,7 +490,7 @@ void sctp_transport_raise_cwnd(struct sctp_transport *transport, * detected. */ void sctp_transport_lower_cwnd(struct sctp_transport *transport, - sctp_lower_cwnd_t reason) + enum sctp_lower_cwnd reason) { struct sctp_association *asoc = transport->asoc; diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index 5f86c5062a98..67abc0194f30 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -371,19 +371,19 @@ sctp_ulpevent_make_remote_error(const struct sctp_association *asoc, struct sctp_chunk *chunk, __u16 flags, gfp_t gfp) { - struct sctp_ulpevent *event; struct sctp_remote_error *sre; + struct sctp_ulpevent *event; + struct sctp_errhdr *ch; struct sk_buff *skb; - sctp_errhdr_t *ch; __be16 cause; int elen; - ch = (sctp_errhdr_t *)(chunk->skb->data); + ch = (struct sctp_errhdr *)(chunk->skb->data); cause = ch->cause; - elen = SCTP_PAD4(ntohs(ch->length)) - sizeof(sctp_errhdr_t); + elen = SCTP_PAD4(ntohs(ch->length)) - sizeof(*ch); /* Pull off the ERROR header. */ - skb_pull(chunk->skb, sizeof(sctp_errhdr_t)); + skb_pull(chunk->skb, sizeof(*ch)); /* Copy the skb to a new skb with room for us to prepend * notification with. diff --git a/net/smc/Kconfig b/net/smc/Kconfig index 33954852f3f8..c717ef0896aa 100644 --- a/net/smc/Kconfig +++ b/net/smc/Kconfig @@ -8,10 +8,6 @@ config SMC The Linux implementation of the SMC-R solution is designed as a separate socket family SMC. - Warning: SMC will expose all memory for remote reads and writes - once a connection is established. Don't enable this option except - for tightly controlled lab environment. - Select this option if you want to run SMC socket applications config SMC_DIAG diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 6793d7348cc8..8c6d24b2995d 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -338,6 +338,12 @@ static int smc_clnt_conf_first_link(struct smc_sock *smc, union ib_gid *gid) return SMC_CLC_DECL_INTERR; smc_wr_remember_qp_attr(link); + + rc = smc_wr_reg_send(link, + smc->conn.rmb_desc->mr_rx[SMC_SINGLE_LINK]); + if (rc) + return SMC_CLC_DECL_INTERR; + /* send CONFIRM LINK response over RoCE fabric */ rc = smc_llc_send_confirm_link(link, link->smcibdev->mac[link->ibport - 1], @@ -430,12 +436,8 @@ static int smc_connect_rdma(struct smc_sock *smc) smc_conn_save_peer_info(smc, &aclc); - rc = smc_sndbuf_create(smc); - if (rc) { - reason_code = SMC_CLC_DECL_MEM; - goto decline_rdma_unlock; - } - rc = smc_rmb_create(smc); + /* create send buffer and rmb */ + rc = smc_buf_create(smc); if (rc) { reason_code = SMC_CLC_DECL_MEM; goto decline_rdma_unlock; @@ -459,7 +461,20 @@ static int smc_connect_rdma(struct smc_sock *smc) reason_code = SMC_CLC_DECL_INTERR; goto decline_rdma_unlock; } + } else { + struct smc_buf_desc *buf_desc = smc->conn.rmb_desc; + + if (!buf_desc->reused) { + /* register memory region for new rmb */ + rc = smc_wr_reg_send(link, + buf_desc->mr_rx[SMC_SINGLE_LINK]); + if (rc) { + reason_code = SMC_CLC_DECL_INTERR; + goto decline_rdma_unlock; + } + } } + smc_rmb_sync_sg_for_device(&smc->conn); rc = smc_clc_send_confirm(smc); if (rc) @@ -692,6 +707,12 @@ static int smc_serv_conf_first_link(struct smc_sock *smc) int rc; link = &lgr->lnk[SMC_SINGLE_LINK]; + + rc = smc_wr_reg_send(link, + smc->conn.rmb_desc->mr_rx[SMC_SINGLE_LINK]); + if (rc) + return SMC_CLC_DECL_INTERR; + /* send CONFIRM LINK request to client over the RoCE fabric */ rc = smc_llc_send_confirm_link(link, link->smcibdev->mac[link->ibport - 1], @@ -779,11 +800,6 @@ static void smc_listen_work(struct work_struct *work) mutex_lock(&smc_create_lgr_pending); local_contact = smc_conn_create(new_smc, peeraddr.sin_addr.s_addr, smcibdev, ibport, &pclc.lcl, 0); - if (local_contact == SMC_REUSE_CONTACT) - /* lock no longer needed, free it due to following - * smc_clc_wait_msg() call - */ - mutex_unlock(&smc_create_lgr_pending); if (local_contact < 0) { rc = local_contact; if (rc == -ENOMEM) @@ -794,12 +810,8 @@ static void smc_listen_work(struct work_struct *work) } link = &new_smc->conn.lgr->lnk[SMC_SINGLE_LINK]; - rc = smc_sndbuf_create(new_smc); - if (rc) { - reason_code = SMC_CLC_DECL_MEM; - goto decline_rdma; - } - rc = smc_rmb_create(new_smc); + /* create send buffer and rmb */ + rc = smc_buf_create(new_smc); if (rc) { reason_code = SMC_CLC_DECL_MEM; goto decline_rdma; @@ -808,6 +820,21 @@ static void smc_listen_work(struct work_struct *work) smc_close_init(new_smc); smc_rx_init(new_smc); + if (local_contact != SMC_FIRST_CONTACT) { + struct smc_buf_desc *buf_desc = new_smc->conn.rmb_desc; + + if (!buf_desc->reused) { + /* register memory region for new rmb */ + rc = smc_wr_reg_send(link, + buf_desc->mr_rx[SMC_SINGLE_LINK]); + if (rc) { + reason_code = SMC_CLC_DECL_INTERR; + goto decline_rdma; + } + } + } + smc_rmb_sync_sg_for_device(&new_smc->conn); + rc = smc_clc_send_accept(new_smc, local_contact); if (rc) goto out_err; @@ -853,8 +880,7 @@ out_connected: if (newsmcsk->sk_state == SMC_INIT) newsmcsk->sk_state = SMC_ACTIVE; enqueue: - if (local_contact == SMC_FIRST_CONTACT) - mutex_unlock(&smc_create_lgr_pending); + mutex_unlock(&smc_create_lgr_pending); lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING); if (lsmc->sk.sk_state == SMC_LISTEN) { smc_accept_enqueue(&lsmc->sk, newsmcsk); diff --git a/net/smc/smc_clc.c b/net/smc/smc_clc.c index 03ec058d18df..3934913ab835 100644 --- a/net/smc/smc_clc.c +++ b/net/smc/smc_clc.c @@ -204,13 +204,13 @@ int smc_clc_send_confirm(struct smc_sock *smc) memcpy(&cclc.lcl.mac, &link->smcibdev->mac[link->ibport - 1], ETH_ALEN); hton24(cclc.qpn, link->roce_qp->qp_num); cclc.rmb_rkey = - htonl(conn->rmb_desc->rkey[SMC_SINGLE_LINK]); + htonl(conn->rmb_desc->mr_rx[SMC_SINGLE_LINK]->rkey); cclc.conn_idx = 1; /* for now: 1 RMB = 1 RMBE */ cclc.rmbe_alert_token = htonl(conn->alert_token_local); cclc.qp_mtu = min(link->path_mtu, link->peer_mtu); cclc.rmbe_size = conn->rmbe_size_short; - cclc.rmb_dma_addr = - cpu_to_be64((u64)conn->rmb_desc->dma_addr[SMC_SINGLE_LINK]); + cclc.rmb_dma_addr = cpu_to_be64( + (u64)sg_dma_address(conn->rmb_desc->sgt[SMC_SINGLE_LINK].sgl)); hton24(cclc.psn, link->psn_initial); memcpy(cclc.trl.eyecatcher, SMC_EYECATCHER, sizeof(SMC_EYECATCHER)); @@ -256,13 +256,13 @@ int smc_clc_send_accept(struct smc_sock *new_smc, int srv_first_contact) memcpy(&aclc.lcl.mac, link->smcibdev->mac[link->ibport - 1], ETH_ALEN); hton24(aclc.qpn, link->roce_qp->qp_num); aclc.rmb_rkey = - htonl(conn->rmb_desc->rkey[SMC_SINGLE_LINK]); + htonl(conn->rmb_desc->mr_rx[SMC_SINGLE_LINK]->rkey); aclc.conn_idx = 1; /* as long as 1 RMB = 1 RMBE */ aclc.rmbe_alert_token = htonl(conn->alert_token_local); aclc.qp_mtu = link->path_mtu; aclc.rmbe_size = conn->rmbe_size_short, - aclc.rmb_dma_addr = - cpu_to_be64((u64)conn->rmb_desc->dma_addr[SMC_SINGLE_LINK]); + aclc.rmb_dma_addr = cpu_to_be64( + (u64)sg_dma_address(conn->rmb_desc->sgt[SMC_SINGLE_LINK].sgl)); hton24(aclc.psn, link->psn_initial); memcpy(aclc.trl.eyecatcher, SMC_EYECATCHER, sizeof(SMC_EYECATCHER)); diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 3ac09a629ea1..1a16d51e2330 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -175,7 +175,6 @@ static int smc_lgr_create(struct smc_sock *smc, __be32 peer_in_addr, rc = smc_wr_alloc_link_mem(lnk); if (rc) goto free_lgr; - init_waitqueue_head(&lnk->wr_tx_wait); rc = smc_ib_create_protection_domain(lnk); if (rc) goto free_link_mem; @@ -207,17 +206,14 @@ out: return rc; } -static void smc_sndbuf_unuse(struct smc_connection *conn) +static void smc_buf_unuse(struct smc_connection *conn) { if (conn->sndbuf_desc) { conn->sndbuf_desc->used = 0; conn->sndbuf_size = 0; } -} - -static void smc_rmb_unuse(struct smc_connection *conn) -{ if (conn->rmb_desc) { + conn->rmb_desc->reused = true; conn->rmb_desc->used = 0; conn->rmbe_size = 0; } @@ -232,8 +228,7 @@ void smc_conn_free(struct smc_connection *conn) return; smc_cdc_tx_dismiss_slots(conn); smc_lgr_unregister_conn(conn); - smc_rmb_unuse(conn); - smc_sndbuf_unuse(conn); + smc_buf_unuse(conn); } static void smc_link_clear(struct smc_link *lnk) @@ -246,48 +241,57 @@ static void smc_link_clear(struct smc_link *lnk) smc_wr_free_link_mem(lnk); } -static void smc_lgr_free_sndbufs(struct smc_link_group *lgr) +static void smc_buf_free(struct smc_buf_desc *buf_desc, struct smc_link *lnk, + bool is_rmb) { - struct smc_buf_desc *sndbuf_desc, *bf_desc; - int i; - - for (i = 0; i < SMC_RMBE_SIZES; i++) { - list_for_each_entry_safe(sndbuf_desc, bf_desc, &lgr->sndbufs[i], - list) { - list_del(&sndbuf_desc->list); - smc_ib_buf_unmap(lgr->lnk[SMC_SINGLE_LINK].smcibdev, - smc_uncompress_bufsize(i), - sndbuf_desc, DMA_TO_DEVICE); - kfree(sndbuf_desc->cpu_addr); - kfree(sndbuf_desc); - } + if (is_rmb) { + if (buf_desc->mr_rx[SMC_SINGLE_LINK]) + smc_ib_put_memory_region( + buf_desc->mr_rx[SMC_SINGLE_LINK]); + smc_ib_buf_unmap_sg(lnk->smcibdev, buf_desc, + DMA_FROM_DEVICE); + } else { + smc_ib_buf_unmap_sg(lnk->smcibdev, buf_desc, + DMA_TO_DEVICE); } + sg_free_table(&buf_desc->sgt[SMC_SINGLE_LINK]); + if (buf_desc->cpu_addr) + free_pages((unsigned long)buf_desc->cpu_addr, buf_desc->order); + kfree(buf_desc); } -static void smc_lgr_free_rmbs(struct smc_link_group *lgr) +static void __smc_lgr_free_bufs(struct smc_link_group *lgr, bool is_rmb) { - struct smc_buf_desc *rmb_desc, *bf_desc; struct smc_link *lnk = &lgr->lnk[SMC_SINGLE_LINK]; + struct smc_buf_desc *buf_desc, *bf_desc; + struct list_head *buf_list; int i; for (i = 0; i < SMC_RMBE_SIZES; i++) { - list_for_each_entry_safe(rmb_desc, bf_desc, &lgr->rmbs[i], + if (is_rmb) + buf_list = &lgr->rmbs[i]; + else + buf_list = &lgr->sndbufs[i]; + list_for_each_entry_safe(buf_desc, bf_desc, buf_list, list) { - list_del(&rmb_desc->list); - smc_ib_buf_unmap(lnk->smcibdev, - smc_uncompress_bufsize(i), - rmb_desc, DMA_FROM_DEVICE); - kfree(rmb_desc->cpu_addr); - kfree(rmb_desc); + list_del(&buf_desc->list); + smc_buf_free(buf_desc, lnk, is_rmb); } } } +static void smc_lgr_free_bufs(struct smc_link_group *lgr) +{ + /* free send buffers */ + __smc_lgr_free_bufs(lgr, false); + /* free rmbs */ + __smc_lgr_free_bufs(lgr, true); +} + /* remove a link group */ void smc_lgr_free(struct smc_link_group *lgr) { - smc_lgr_free_rmbs(lgr); - smc_lgr_free_sndbufs(lgr); + smc_lgr_free_bufs(lgr); smc_link_clear(&lgr->lnk[SMC_SINGLE_LINK]); kfree(lgr); } @@ -452,45 +456,25 @@ out: return rc ? rc : local_contact; } -/* try to reuse a sndbuf description slot of the sndbufs list for a certain - * buf_size; if not available, return NULL +/* try to reuse a sndbuf or rmb description slot for a certain + * buffer size; if not available, return NULL */ static inline -struct smc_buf_desc *smc_sndbuf_get_slot(struct smc_link_group *lgr, - int compressed_bufsize) +struct smc_buf_desc *smc_buf_get_slot(struct smc_link_group *lgr, + int compressed_bufsize, + rwlock_t *lock, + struct list_head *buf_list) { - struct smc_buf_desc *sndbuf_slot; - - read_lock_bh(&lgr->sndbufs_lock); - list_for_each_entry(sndbuf_slot, &lgr->sndbufs[compressed_bufsize], - list) { - if (cmpxchg(&sndbuf_slot->used, 0, 1) == 0) { - read_unlock_bh(&lgr->sndbufs_lock); - return sndbuf_slot; - } - } - read_unlock_bh(&lgr->sndbufs_lock); - return NULL; -} + struct smc_buf_desc *buf_slot; -/* try to reuse an rmb description slot of the rmbs list for a certain - * rmbe_size; if not available, return NULL - */ -static inline -struct smc_buf_desc *smc_rmb_get_slot(struct smc_link_group *lgr, - int compressed_bufsize) -{ - struct smc_buf_desc *rmb_slot; - - read_lock_bh(&lgr->rmbs_lock); - list_for_each_entry(rmb_slot, &lgr->rmbs[compressed_bufsize], - list) { - if (cmpxchg(&rmb_slot->used, 0, 1) == 0) { - read_unlock_bh(&lgr->rmbs_lock); - return rmb_slot; + read_lock_bh(lock); + list_for_each_entry(buf_slot, buf_list, list) { + if (cmpxchg(&buf_slot->used, 0, 1) == 0) { + read_unlock_bh(lock); + return buf_slot; } } - read_unlock_bh(&lgr->rmbs_lock); + read_unlock_bh(lock); return NULL; } @@ -503,136 +487,186 @@ static inline int smc_rmb_wnd_update_limit(int rmbe_size) return min_t(int, rmbe_size / 10, SOCK_MIN_SNDBUF / 2); } -/* create the tx buffer for an SMC socket */ -int smc_sndbuf_create(struct smc_sock *smc) +static struct smc_buf_desc *smc_new_buf_create(struct smc_link_group *lgr, + bool is_rmb, int bufsize) { - struct smc_connection *conn = &smc->conn; - struct smc_link_group *lgr = conn->lgr; - int tmp_bufsize, tmp_bufsize_short; - struct smc_buf_desc *sndbuf_desc; + struct smc_buf_desc *buf_desc; + struct smc_link *lnk; int rc; - /* use socket send buffer size (w/o overhead) as start value */ - for (tmp_bufsize_short = smc_compress_bufsize(smc->sk.sk_sndbuf / 2); - tmp_bufsize_short >= 0; tmp_bufsize_short--) { - tmp_bufsize = smc_uncompress_bufsize(tmp_bufsize_short); - /* check for reusable sndbuf_slot in the link group */ - sndbuf_desc = smc_sndbuf_get_slot(lgr, tmp_bufsize_short); - if (sndbuf_desc) { - memset(sndbuf_desc->cpu_addr, 0, tmp_bufsize); - break; /* found reusable slot */ - } - /* try to alloc a new send buffer */ - sndbuf_desc = kzalloc(sizeof(*sndbuf_desc), GFP_KERNEL); - if (!sndbuf_desc) - break; /* give up with -ENOMEM */ - sndbuf_desc->cpu_addr = kzalloc(tmp_bufsize, - GFP_KERNEL | __GFP_NOWARN | - __GFP_NOMEMALLOC | - __GFP_NORETRY); - if (!sndbuf_desc->cpu_addr) { - kfree(sndbuf_desc); - sndbuf_desc = NULL; - /* if send buffer allocation has failed, - * try a smaller one - */ - continue; - } - rc = smc_ib_buf_map(lgr->lnk[SMC_SINGLE_LINK].smcibdev, - tmp_bufsize, sndbuf_desc, - DMA_TO_DEVICE); + /* try to alloc a new buffer */ + buf_desc = kzalloc(sizeof(*buf_desc), GFP_KERNEL); + if (!buf_desc) + return ERR_PTR(-ENOMEM); + + buf_desc->cpu_addr = + (void *)__get_free_pages(GFP_KERNEL | __GFP_NOWARN | + __GFP_NOMEMALLOC | + __GFP_NORETRY | __GFP_ZERO, + get_order(bufsize)); + if (!buf_desc->cpu_addr) { + kfree(buf_desc); + return ERR_PTR(-EAGAIN); + } + buf_desc->order = get_order(bufsize); + + /* build the sg table from the pages */ + lnk = &lgr->lnk[SMC_SINGLE_LINK]; + rc = sg_alloc_table(&buf_desc->sgt[SMC_SINGLE_LINK], 1, + GFP_KERNEL); + if (rc) { + smc_buf_free(buf_desc, lnk, is_rmb); + return ERR_PTR(rc); + } + sg_set_buf(buf_desc->sgt[SMC_SINGLE_LINK].sgl, + buf_desc->cpu_addr, bufsize); + + /* map sg table to DMA address */ + rc = smc_ib_buf_map_sg(lnk->smcibdev, buf_desc, + is_rmb ? DMA_FROM_DEVICE : DMA_TO_DEVICE); + /* SMC protocol depends on mapping to one DMA address only */ + if (rc != 1) { + smc_buf_free(buf_desc, lnk, is_rmb); + return ERR_PTR(-EAGAIN); + } + + /* create a new memory region for the RMB */ + if (is_rmb) { + rc = smc_ib_get_memory_region(lnk->roce_pd, + IB_ACCESS_REMOTE_WRITE | + IB_ACCESS_LOCAL_WRITE, + buf_desc); if (rc) { - kfree(sndbuf_desc->cpu_addr); - kfree(sndbuf_desc); - sndbuf_desc = NULL; - continue; /* if mapping failed, try smaller one */ + smc_buf_free(buf_desc, lnk, is_rmb); + return ERR_PTR(rc); } - sndbuf_desc->used = 1; - write_lock_bh(&lgr->sndbufs_lock); - list_add(&sndbuf_desc->list, - &lgr->sndbufs[tmp_bufsize_short]); - write_unlock_bh(&lgr->sndbufs_lock); - break; - } - if (sndbuf_desc && sndbuf_desc->cpu_addr) { - conn->sndbuf_desc = sndbuf_desc; - conn->sndbuf_size = tmp_bufsize; - smc->sk.sk_sndbuf = tmp_bufsize * 2; - atomic_set(&conn->sndbuf_space, tmp_bufsize); - return 0; - } else { - return -ENOMEM; } + + return buf_desc; } -/* create the RMB for an SMC socket (even though the SMC protocol - * allows more than one RMB-element per RMB, the Linux implementation - * uses just one RMB-element per RMB, i.e. uses an extra RMB for every - * connection in a link group - */ -int smc_rmb_create(struct smc_sock *smc) +static int __smc_buf_create(struct smc_sock *smc, bool is_rmb) { struct smc_connection *conn = &smc->conn; struct smc_link_group *lgr = conn->lgr; - int tmp_bufsize, tmp_bufsize_short; - struct smc_buf_desc *rmb_desc; - int rc; + struct smc_buf_desc *buf_desc = NULL; + struct list_head *buf_list; + int bufsize, bufsize_short; + int sk_buf_size; + rwlock_t *lock; + + if (is_rmb) + /* use socket recv buffer size (w/o overhead) as start value */ + sk_buf_size = smc->sk.sk_rcvbuf / 2; + else + /* use socket send buffer size (w/o overhead) as start value */ + sk_buf_size = smc->sk.sk_sndbuf / 2; + + for (bufsize_short = smc_compress_bufsize(smc->sk.sk_sndbuf / 2); + bufsize_short >= 0; bufsize_short--) { + + if (is_rmb) { + lock = &lgr->rmbs_lock; + buf_list = &lgr->rmbs[bufsize_short]; + } else { + lock = &lgr->sndbufs_lock; + buf_list = &lgr->sndbufs[bufsize_short]; + } + bufsize = smc_uncompress_bufsize(bufsize_short); + if ((1 << get_order(bufsize)) > SG_MAX_SINGLE_ALLOC) + continue; - /* use socket recv buffer size (w/o overhead) as start value */ - for (tmp_bufsize_short = smc_compress_bufsize(smc->sk.sk_rcvbuf / 2); - tmp_bufsize_short >= 0; tmp_bufsize_short--) { - tmp_bufsize = smc_uncompress_bufsize(tmp_bufsize_short); - /* check for reusable rmb_slot in the link group */ - rmb_desc = smc_rmb_get_slot(lgr, tmp_bufsize_short); - if (rmb_desc) { - memset(rmb_desc->cpu_addr, 0, tmp_bufsize); + /* check for reusable slot in the link group */ + buf_desc = smc_buf_get_slot(lgr, bufsize_short, lock, buf_list); + if (buf_desc) { + memset(buf_desc->cpu_addr, 0, bufsize); break; /* found reusable slot */ } - /* try to alloc a new RMB */ - rmb_desc = kzalloc(sizeof(*rmb_desc), GFP_KERNEL); - if (!rmb_desc) - break; /* give up with -ENOMEM */ - rmb_desc->cpu_addr = kzalloc(tmp_bufsize, - GFP_KERNEL | __GFP_NOWARN | - __GFP_NOMEMALLOC | - __GFP_NORETRY); - if (!rmb_desc->cpu_addr) { - kfree(rmb_desc); - rmb_desc = NULL; - /* if RMB allocation has failed, - * try a smaller one - */ + + buf_desc = smc_new_buf_create(lgr, is_rmb, bufsize); + if (PTR_ERR(buf_desc) == -ENOMEM) + break; + if (IS_ERR(buf_desc)) continue; - } - rc = smc_ib_buf_map(lgr->lnk[SMC_SINGLE_LINK].smcibdev, - tmp_bufsize, rmb_desc, - DMA_FROM_DEVICE); - if (rc) { - kfree(rmb_desc->cpu_addr); - kfree(rmb_desc); - rmb_desc = NULL; - continue; /* if mapping failed, try smaller one */ - } - rmb_desc->rkey[SMC_SINGLE_LINK] = - lgr->lnk[SMC_SINGLE_LINK].roce_pd->unsafe_global_rkey; - rmb_desc->used = 1; - write_lock_bh(&lgr->rmbs_lock); - list_add(&rmb_desc->list, - &lgr->rmbs[tmp_bufsize_short]); - write_unlock_bh(&lgr->rmbs_lock); - break; + + buf_desc->used = 1; + write_lock_bh(lock); + list_add(&buf_desc->list, buf_list); + write_unlock_bh(lock); + break; /* found */ } - if (rmb_desc && rmb_desc->cpu_addr) { - conn->rmb_desc = rmb_desc; - conn->rmbe_size = tmp_bufsize; - conn->rmbe_size_short = tmp_bufsize_short; - smc->sk.sk_rcvbuf = tmp_bufsize * 2; + + if (IS_ERR(buf_desc)) + return -ENOMEM; + + if (is_rmb) { + conn->rmb_desc = buf_desc; + conn->rmbe_size = bufsize; + conn->rmbe_size_short = bufsize_short; + smc->sk.sk_rcvbuf = bufsize * 2; atomic_set(&conn->bytes_to_rcv, 0); - conn->rmbe_update_limit = smc_rmb_wnd_update_limit(tmp_bufsize); - return 0; + conn->rmbe_update_limit = smc_rmb_wnd_update_limit(bufsize); } else { - return -ENOMEM; + conn->sndbuf_desc = buf_desc; + conn->sndbuf_size = bufsize; + smc->sk.sk_sndbuf = bufsize * 2; + atomic_set(&conn->sndbuf_space, bufsize); } + return 0; +} + +void smc_sndbuf_sync_sg_for_cpu(struct smc_connection *conn) +{ + struct smc_link_group *lgr = conn->lgr; + + smc_ib_sync_sg_for_cpu(lgr->lnk[SMC_SINGLE_LINK].smcibdev, + conn->sndbuf_desc, DMA_TO_DEVICE); +} + +void smc_sndbuf_sync_sg_for_device(struct smc_connection *conn) +{ + struct smc_link_group *lgr = conn->lgr; + + smc_ib_sync_sg_for_device(lgr->lnk[SMC_SINGLE_LINK].smcibdev, + conn->sndbuf_desc, DMA_TO_DEVICE); +} + +void smc_rmb_sync_sg_for_cpu(struct smc_connection *conn) +{ + struct smc_link_group *lgr = conn->lgr; + + smc_ib_sync_sg_for_cpu(lgr->lnk[SMC_SINGLE_LINK].smcibdev, + conn->rmb_desc, DMA_FROM_DEVICE); +} + +void smc_rmb_sync_sg_for_device(struct smc_connection *conn) +{ + struct smc_link_group *lgr = conn->lgr; + + smc_ib_sync_sg_for_device(lgr->lnk[SMC_SINGLE_LINK].smcibdev, + conn->rmb_desc, DMA_FROM_DEVICE); +} + +/* create the send and receive buffer for an SMC socket; + * receive buffers are called RMBs; + * (even though the SMC protocol allows more than one RMB-element per RMB, + * the Linux implementation uses just one RMB-element per RMB, i.e. uses an + * extra RMB for every connection in a link group + */ +int smc_buf_create(struct smc_sock *smc) +{ + int rc; + + /* create send buffer */ + rc = __smc_buf_create(smc, false); + if (rc) + return rc; + /* create rmb */ + rc = __smc_buf_create(smc, true); + if (rc) + smc_buf_free(smc->conn.sndbuf_desc, + &smc->conn.lgr->lnk[SMC_SINGLE_LINK], false); + return rc; } static inline int smc_rmb_reserve_rtoken_idx(struct smc_link_group *lgr) diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index b013cb43a327..19c44bf4e391 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -37,6 +37,14 @@ struct smc_wr_buf { u8 raw[SMC_WR_BUF_SIZE]; }; +#define SMC_WR_REG_MR_WAIT_TIME (5 * HZ)/* wait time for ib_wr_reg_mr result */ + +enum smc_wr_reg_state { + POSTED, /* ib_wr_reg_mr request posted */ + CONFIRMED, /* ib_wr_reg_mr response: successful */ + FAILED /* ib_wr_reg_mr response: failure */ +}; + struct smc_link { struct smc_ib_device *smcibdev; /* ib-device */ u8 ibport; /* port - values 1 | 2 */ @@ -65,6 +73,10 @@ struct smc_link { u64 wr_rx_id; /* seq # of last recv WR */ u32 wr_rx_cnt; /* number of WR recv buffers */ + struct ib_reg_wr wr_reg; /* WR register memory region */ + wait_queue_head_t wr_reg_wait; /* wait for wr_reg result */ + enum smc_wr_reg_state wr_reg_state; /* state of wr_reg request */ + union ib_gid gid; /* gid matching used vlan id */ u32 peer_qpn; /* QP number of peer */ enum ib_mtu path_mtu; /* used mtu */ @@ -90,14 +102,15 @@ struct smc_link { /* tx/rx buffer list element for sndbufs list and rmbs list of a lgr */ struct smc_buf_desc { struct list_head list; - u64 dma_addr[SMC_LINKS_PER_LGR_MAX]; - /* mapped address of buffer */ void *cpu_addr; /* virtual address of buffer */ - u32 rkey[SMC_LINKS_PER_LGR_MAX]; - /* for rmb only: - * rkey provided to peer + struct sg_table sgt[SMC_LINKS_PER_LGR_MAX];/* virtual buffer */ + struct ib_mr *mr_rx[SMC_LINKS_PER_LGR_MAX]; + /* for rmb only: memory region + * incl. rkey provided to peer */ + u32 order; /* allocation order */ u32 used; /* currently used / unused */ + bool reused; /* new created / reused */ }; struct smc_rtoken { /* address/key of remote RMB */ @@ -173,9 +186,11 @@ struct smc_clc_msg_accept_confirm; void smc_lgr_free(struct smc_link_group *lgr); void smc_lgr_terminate(struct smc_link_group *lgr); -int smc_sndbuf_create(struct smc_sock *smc); -int smc_rmb_create(struct smc_sock *smc); +int smc_buf_create(struct smc_sock *smc); int smc_rmb_rtoken_handling(struct smc_connection *conn, struct smc_clc_msg_accept_confirm *clc); - +void smc_sndbuf_sync_sg_for_cpu(struct smc_connection *conn); +void smc_sndbuf_sync_sg_for_device(struct smc_connection *conn); +void smc_rmb_sync_sg_for_cpu(struct smc_connection *conn); +void smc_rmb_sync_sg_for_device(struct smc_connection *conn); #endif diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index b31715505a35..547e0e113b17 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -13,6 +13,7 @@ #include <linux/random.h> #include <linux/workqueue.h> +#include <linux/scatterlist.h> #include <rdma/ib_verbs.h> #include "smc_pnet.h" @@ -192,8 +193,7 @@ int smc_ib_create_protection_domain(struct smc_link *lnk) { int rc; - lnk->roce_pd = ib_alloc_pd(lnk->smcibdev->ibdev, - IB_PD_UNSAFE_GLOBAL_RKEY); + lnk->roce_pd = ib_alloc_pd(lnk->smcibdev->ibdev, 0); rc = PTR_ERR_OR_ZERO(lnk->roce_pd); if (IS_ERR(lnk->roce_pd)) lnk->roce_pd = NULL; @@ -232,10 +232,10 @@ int smc_ib_create_queue_pair(struct smc_link *lnk) .recv_cq = lnk->smcibdev->roce_cq_recv, .srq = NULL, .cap = { - .max_send_wr = SMC_WR_BUF_CNT, /* include unsolicited rdma_writes as well, * there are max. 2 RDMA_WRITE per 1 WR_SEND */ + .max_send_wr = SMC_WR_BUF_CNT * 3, .max_recv_wr = SMC_WR_BUF_CNT * 3, .max_send_sge = SMC_IB_MAX_SEND_SGE, .max_recv_sge = 1, @@ -254,33 +254,117 @@ int smc_ib_create_queue_pair(struct smc_link *lnk) return rc; } -/* map a new TX or RX buffer to DMA */ -int smc_ib_buf_map(struct smc_ib_device *smcibdev, int buf_size, - struct smc_buf_desc *buf_slot, - enum dma_data_direction data_direction) +void smc_ib_put_memory_region(struct ib_mr *mr) { - int rc = 0; + ib_dereg_mr(mr); +} - if (buf_slot->dma_addr[SMC_SINGLE_LINK]) - return rc; /* already mapped */ - buf_slot->dma_addr[SMC_SINGLE_LINK] = - ib_dma_map_single(smcibdev->ibdev, buf_slot->cpu_addr, - buf_size, data_direction); - if (ib_dma_mapping_error(smcibdev->ibdev, - buf_slot->dma_addr[SMC_SINGLE_LINK])) - rc = -EIO; - return rc; +static int smc_ib_map_mr_sg(struct smc_buf_desc *buf_slot) +{ + unsigned int offset = 0; + int sg_num; + + /* map the largest prefix of a dma mapped SG list */ + sg_num = ib_map_mr_sg(buf_slot->mr_rx[SMC_SINGLE_LINK], + buf_slot->sgt[SMC_SINGLE_LINK].sgl, + buf_slot->sgt[SMC_SINGLE_LINK].orig_nents, + &offset, PAGE_SIZE); + + return sg_num; +} + +/* Allocate a memory region and map the dma mapped SG list of buf_slot */ +int smc_ib_get_memory_region(struct ib_pd *pd, int access_flags, + struct smc_buf_desc *buf_slot) +{ + if (buf_slot->mr_rx[SMC_SINGLE_LINK]) + return 0; /* already done */ + + buf_slot->mr_rx[SMC_SINGLE_LINK] = + ib_alloc_mr(pd, IB_MR_TYPE_MEM_REG, 1 << buf_slot->order); + if (IS_ERR(buf_slot->mr_rx[SMC_SINGLE_LINK])) { + int rc; + + rc = PTR_ERR(buf_slot->mr_rx[SMC_SINGLE_LINK]); + buf_slot->mr_rx[SMC_SINGLE_LINK] = NULL; + return rc; + } + + if (smc_ib_map_mr_sg(buf_slot) != 1) + return -EINVAL; + + return 0; } -void smc_ib_buf_unmap(struct smc_ib_device *smcibdev, int buf_size, +/* synchronize buffer usage for cpu access */ +void smc_ib_sync_sg_for_cpu(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction) +{ + struct scatterlist *sg; + unsigned int i; + + /* for now there is just one DMA address */ + for_each_sg(buf_slot->sgt[SMC_SINGLE_LINK].sgl, sg, + buf_slot->sgt[SMC_SINGLE_LINK].nents, i) { + if (!sg_dma_len(sg)) + break; + ib_dma_sync_single_for_cpu(smcibdev->ibdev, + sg_dma_address(sg), + sg_dma_len(sg), + data_direction); + } +} + +/* synchronize buffer usage for device access */ +void smc_ib_sync_sg_for_device(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction) +{ + struct scatterlist *sg; + unsigned int i; + + /* for now there is just one DMA address */ + for_each_sg(buf_slot->sgt[SMC_SINGLE_LINK].sgl, sg, + buf_slot->sgt[SMC_SINGLE_LINK].nents, i) { + if (!sg_dma_len(sg)) + break; + ib_dma_sync_single_for_device(smcibdev->ibdev, + sg_dma_address(sg), + sg_dma_len(sg), + data_direction); + } +} + +/* Map a new TX or RX buffer SG-table to DMA */ +int smc_ib_buf_map_sg(struct smc_ib_device *smcibdev, struct smc_buf_desc *buf_slot, enum dma_data_direction data_direction) { - if (!buf_slot->dma_addr[SMC_SINGLE_LINK]) + int mapped_nents; + + mapped_nents = ib_dma_map_sg(smcibdev->ibdev, + buf_slot->sgt[SMC_SINGLE_LINK].sgl, + buf_slot->sgt[SMC_SINGLE_LINK].orig_nents, + data_direction); + if (!mapped_nents) + return -ENOMEM; + + return mapped_nents; +} + +void smc_ib_buf_unmap_sg(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction) +{ + if (!buf_slot->sgt[SMC_SINGLE_LINK].sgl->dma_address) return; /* already unmapped */ - ib_dma_unmap_single(smcibdev->ibdev, *buf_slot->dma_addr, buf_size, - data_direction); - buf_slot->dma_addr[SMC_SINGLE_LINK] = 0; + + ib_dma_unmap_sg(smcibdev->ibdev, + buf_slot->sgt[SMC_SINGLE_LINK].sgl, + buf_slot->sgt[SMC_SINGLE_LINK].orig_nents, + data_direction); + buf_slot->sgt[SMC_SINGLE_LINK].sgl->dma_address = 0; } static int smc_ib_fill_gid_and_mac(struct smc_ib_device *smcibdev, u8 ibport) diff --git a/net/smc/smc_ib.h b/net/smc/smc_ib.h index b567152a526d..9b927a33d5e6 100644 --- a/net/smc/smc_ib.h +++ b/net/smc/smc_ib.h @@ -51,12 +51,12 @@ int smc_ib_register_client(void) __init; void smc_ib_unregister_client(void); bool smc_ib_port_active(struct smc_ib_device *smcibdev, u8 ibport); int smc_ib_remember_port_attr(struct smc_ib_device *smcibdev, u8 ibport); -int smc_ib_buf_map(struct smc_ib_device *smcibdev, int buf_size, - struct smc_buf_desc *buf_slot, - enum dma_data_direction data_direction); -void smc_ib_buf_unmap(struct smc_ib_device *smcibdev, int bufsize, +int smc_ib_buf_map_sg(struct smc_ib_device *smcibdev, struct smc_buf_desc *buf_slot, enum dma_data_direction data_direction); +void smc_ib_buf_unmap_sg(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction); void smc_ib_dealloc_protection_domain(struct smc_link *lnk); int smc_ib_create_protection_domain(struct smc_link *lnk); void smc_ib_destroy_queue_pair(struct smc_link *lnk); @@ -65,6 +65,13 @@ int smc_ib_ready_link(struct smc_link *lnk); int smc_ib_modify_qp_rts(struct smc_link *lnk); int smc_ib_modify_qp_reset(struct smc_link *lnk); long smc_ib_setup_per_ibdev(struct smc_ib_device *smcibdev); - - +int smc_ib_get_memory_region(struct ib_pd *pd, int access_flags, + struct smc_buf_desc *buf_slot); +void smc_ib_put_memory_region(struct ib_mr *mr); +void smc_ib_sync_sg_for_cpu(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction); +void smc_ib_sync_sg_for_device(struct smc_ib_device *smcibdev, + struct smc_buf_desc *buf_slot, + enum dma_data_direction data_direction); #endif diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index f0c8b089f770..b17a333e9bb0 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -170,6 +170,7 @@ copy: copylen, conn->rmbe_size - cons.count); chunk_len_sum = chunk_len; chunk_off = cons.count; + smc_rmb_sync_sg_for_cpu(conn); for (chunk = 0; chunk < 2; chunk++) { if (!(flags & MSG_TRUNC)) { rc = memcpy_to_msg(msg, rcvbuf_base + chunk_off, @@ -177,6 +178,7 @@ copy: if (rc) { if (!read_done) read_done = -EFAULT; + smc_rmb_sync_sg_for_device(conn); goto out; } } @@ -190,6 +192,7 @@ copy: chunk_len_sum += chunk_len; chunk_off = 0; /* modulo offset in recv ring buffer */ } + smc_rmb_sync_sg_for_device(conn); /* update cursors */ if (!(flags & MSG_PEEK)) { diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index 21ec1832ab51..3c656beb8820 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -174,10 +174,12 @@ int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len) copylen, conn->sndbuf_size - tx_cnt_prep); chunk_len_sum = chunk_len; chunk_off = tx_cnt_prep; + smc_sndbuf_sync_sg_for_cpu(conn); for (chunk = 0; chunk < 2; chunk++) { rc = memcpy_from_msg(sndbuf_base + chunk_off, msg, chunk_len); if (rc) { + smc_sndbuf_sync_sg_for_device(conn); if (send_done) return send_done; goto out_err; @@ -192,6 +194,7 @@ int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len) chunk_len_sum += chunk_len; chunk_off = 0; /* modulo offset in send ring buffer */ } + smc_sndbuf_sync_sg_for_device(conn); /* update cursors */ smc_curs_add(conn->sndbuf_size, &prep, copylen); smc_curs_write(&conn->tx_curs_prep, @@ -277,6 +280,7 @@ static int smc_tx_rdma_writes(struct smc_connection *conn) struct smc_link_group *lgr = conn->lgr; int to_send, rmbespace; struct smc_link *link; + dma_addr_t dma_addr; int num_sges; int rc; @@ -334,12 +338,11 @@ static int smc_tx_rdma_writes(struct smc_connection *conn) src_len = conn->sndbuf_size - sent.count; } src_len_sum = src_len; + dma_addr = sg_dma_address(conn->sndbuf_desc->sgt[SMC_SINGLE_LINK].sgl); for (dstchunk = 0; dstchunk < 2; dstchunk++) { num_sges = 0; for (srcchunk = 0; srcchunk < 2; srcchunk++) { - sges[srcchunk].addr = - conn->sndbuf_desc->dma_addr[SMC_SINGLE_LINK] + - src_off; + sges[srcchunk].addr = dma_addr + src_off; sges[srcchunk].length = src_len; sges[srcchunk].lkey = link->roce_pd->local_dma_lkey; num_sges++; diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index 874ee9f9d796..ab56bda66783 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -68,6 +68,16 @@ static inline void smc_wr_tx_process_cqe(struct ib_wc *wc) int i; link = wc->qp->qp_context; + + if (wc->opcode == IB_WC_REG_MR) { + if (wc->status) + link->wr_reg_state = FAILED; + else + link->wr_reg_state = CONFIRMED; + wake_up(&link->wr_reg_wait); + return; + } + pnd_snd_idx = smc_wr_tx_find_pending_index(link, wc->wr_id); if (pnd_snd_idx == link->wr_tx_cnt) return; @@ -243,6 +253,52 @@ int smc_wr_tx_send(struct smc_link *link, struct smc_wr_tx_pend_priv *priv) return rc; } +/* Register a memory region and wait for result. */ +int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr) +{ + struct ib_send_wr *failed_wr = NULL; + int rc; + + ib_req_notify_cq(link->smcibdev->roce_cq_send, + IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS); + link->wr_reg_state = POSTED; + link->wr_reg.wr.wr_id = (u64)(uintptr_t)mr; + link->wr_reg.mr = mr; + link->wr_reg.key = mr->rkey; + failed_wr = &link->wr_reg.wr; + rc = ib_post_send(link->roce_qp, &link->wr_reg.wr, &failed_wr); + WARN_ON(failed_wr != &link->wr_reg.wr); + if (rc) + return rc; + + rc = wait_event_interruptible_timeout(link->wr_reg_wait, + (link->wr_reg_state != POSTED), + SMC_WR_REG_MR_WAIT_TIME); + if (!rc) { + /* timeout - terminate connections */ + struct smc_link_group *lgr; + + lgr = container_of(link, struct smc_link_group, + lnk[SMC_SINGLE_LINK]); + smc_lgr_terminate(lgr); + return -EPIPE; + } + if (rc == -ERESTARTSYS) + return -EINTR; + switch (link->wr_reg_state) { + case CONFIRMED: + rc = 0; + break; + case FAILED: + rc = -EIO; + break; + case POSTED: + rc = -EPIPE; + break; + } + return rc; +} + void smc_wr_tx_dismiss_slots(struct smc_link *link, u8 wr_rx_hdr_type, smc_wr_tx_filter filter, smc_wr_tx_dismisser dismisser, @@ -458,6 +514,11 @@ static void smc_wr_init_sge(struct smc_link *lnk) lnk->wr_rx_ibs[i].sg_list = &lnk->wr_rx_sges[i]; lnk->wr_rx_ibs[i].num_sge = 1; } + lnk->wr_reg.wr.next = NULL; + lnk->wr_reg.wr.num_sge = 0; + lnk->wr_reg.wr.send_flags = IB_SEND_SIGNALED; + lnk->wr_reg.wr.opcode = IB_WR_REG_MR; + lnk->wr_reg.access = IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE; } void smc_wr_free_link(struct smc_link *lnk) @@ -602,6 +663,8 @@ int smc_wr_create_link(struct smc_link *lnk) smc_wr_init_sge(lnk); memset(lnk->wr_tx_mask, 0, BITS_TO_LONGS(SMC_WR_BUF_CNT) * sizeof(*lnk->wr_tx_mask)); + init_waitqueue_head(&lnk->wr_tx_wait); + init_waitqueue_head(&lnk->wr_reg_wait); return rc; dma_unmap: diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h index 0b9beeda6053..45eb53833052 100644 --- a/net/smc/smc_wr.h +++ b/net/smc/smc_wr.h @@ -102,5 +102,6 @@ void smc_wr_tx_dismiss_slots(struct smc_link *lnk, u8 wr_rx_hdr_type, int smc_wr_rx_register_handler(struct smc_wr_rx_handler *handler); int smc_wr_rx_post_init(struct smc_link *link); void smc_wr_rx_cq_handler(struct ib_cq *ib_cq, void *cq_context); +int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr); #endif /* SMC_WR_H */ diff --git a/net/socket.c b/net/socket.c index bf2122691fba..c729625eb5d3 100644 --- a/net/socket.c +++ b/net/socket.c @@ -652,6 +652,20 @@ int kernel_sendmsg(struct socket *sock, struct msghdr *msg, } EXPORT_SYMBOL(kernel_sendmsg); +int kernel_sendmsg_locked(struct sock *sk, struct msghdr *msg, + struct kvec *vec, size_t num, size_t size) +{ + struct socket *sock = sk->sk_socket; + + if (!sock->ops->sendmsg_locked) + return sock_no_sendmsg_locked(sk, msg, size); + + iov_iter_kvec(&msg->msg_iter, WRITE | ITER_KVEC, vec, num, size); + + return sock->ops->sendmsg_locked(sk, msg, msg_data_left(msg)); +} +EXPORT_SYMBOL(kernel_sendmsg_locked); + static bool skb_is_err_queue(const struct sk_buff *skb) { /* pkt_type of skbs enqueued on the error queue are set to @@ -1916,7 +1930,7 @@ static int copy_msghdr_from_user(struct msghdr *kmsg, if (copy_from_user(&msg, umsg, sizeof(*umsg))) return -EFAULT; - kmsg->msg_control = msg.msg_control; + kmsg->msg_control = (void __force *)msg.msg_control; kmsg->msg_controllen = msg.msg_controllen; kmsg->msg_flags = msg.msg_flags; @@ -1935,7 +1949,8 @@ static int copy_msghdr_from_user(struct msghdr *kmsg, if (msg.msg_name && kmsg->msg_namelen) { if (!save_addr) { - err = move_addr_to_kernel(msg.msg_name, kmsg->msg_namelen, + err = move_addr_to_kernel(msg.msg_name, + kmsg->msg_namelen, kmsg->msg_name); if (err < 0) return err; @@ -3375,6 +3390,19 @@ int kernel_sendpage(struct socket *sock, struct page *page, int offset, } EXPORT_SYMBOL(kernel_sendpage); +int kernel_sendpage_locked(struct sock *sk, struct page *page, int offset, + size_t size, int flags) +{ + struct socket *sock = sk->sk_socket; + + if (sock->ops->sendpage_locked) + return sock->ops->sendpage_locked(sk, page, offset, size, + flags); + + return sock_no_sendpage_locked(sk, page, offset, size, flags); +} +EXPORT_SYMBOL(kernel_sendpage_locked); + int kernel_sock_ioctl(struct socket *sock, int cmd, unsigned long arg) { mm_segment_t oldfs = get_fs(); @@ -3404,7 +3432,6 @@ u32 kernel_sock_ip_overhead(struct sock *sk) struct inet_sock *inet; struct ip_options_rcu *opt; u32 overhead = 0; - bool owned_by_user; #if IS_ENABLED(CONFIG_IPV6) struct ipv6_pinfo *np; struct ipv6_txoptions *optv6 = NULL; @@ -3413,13 +3440,12 @@ u32 kernel_sock_ip_overhead(struct sock *sk) if (!sk) return overhead; - owned_by_user = sock_owned_by_user(sk); switch (sk->sk_family) { case AF_INET: inet = inet_sk(sk); overhead += sizeof(struct iphdr); opt = rcu_dereference_protected(inet->inet_opt, - owned_by_user); + sock_owned_by_user(sk)); if (opt) overhead += opt->opt.optlen; return overhead; @@ -3429,7 +3455,7 @@ u32 kernel_sock_ip_overhead(struct sock *sk) overhead += sizeof(struct ipv6hdr); if (np) optv6 = rcu_dereference_protected(np->opt, - owned_by_user); + sock_owned_by_user(sk)); if (optv6) overhead += (optv6->opt_flen + optv6->opt_nflen); return overhead; diff --git a/net/strparser/strparser.c b/net/strparser/strparser.c index b5c279b22680..d4ea46a5f233 100644 --- a/net/strparser/strparser.c +++ b/net/strparser/strparser.c @@ -29,44 +29,46 @@ static struct workqueue_struct *strp_wq; -struct _strp_rx_msg { - /* Internal cb structure. struct strp_rx_msg must be first for passing +struct _strp_msg { + /* Internal cb structure. struct strp_msg must be first for passing * to upper layer. */ - struct strp_rx_msg strp; + struct strp_msg strp; int accum_len; int early_eaten; }; -static inline struct _strp_rx_msg *_strp_rx_msg(struct sk_buff *skb) +static inline struct _strp_msg *_strp_msg(struct sk_buff *skb) { - return (struct _strp_rx_msg *)((void *)skb->cb + + return (struct _strp_msg *)((void *)skb->cb + offsetof(struct qdisc_skb_cb, data)); } /* Lower lock held */ -static void strp_abort_rx_strp(struct strparser *strp, int err) +static void strp_abort_strp(struct strparser *strp, int err) { - struct sock *csk = strp->sk; - /* Unrecoverable error in receive */ - del_timer(&strp->rx_msg_timer); + del_timer(&strp->msg_timer); - if (strp->rx_stopped) + if (strp->stopped) return; - strp->rx_stopped = 1; + strp->stopped = 1; + + if (strp->sk) { + struct sock *sk = strp->sk; - /* Report an error on the lower socket */ - csk->sk_err = err; - csk->sk_error_report(csk); + /* Report an error on the lower socket */ + sk->sk_err = err; + sk->sk_error_report(sk); + } } -static void strp_start_rx_timer(struct strparser *strp) +static void strp_start_timer(struct strparser *strp, long timeo) { - if (strp->sk->sk_rcvtimeo) - mod_timer(&strp->rx_msg_timer, strp->sk->sk_rcvtimeo); + if (timeo) + mod_timer(&strp->msg_timer, timeo); } /* Lower lock held */ @@ -74,46 +76,55 @@ static void strp_parser_err(struct strparser *strp, int err, read_descriptor_t *desc) { desc->error = err; - kfree_skb(strp->rx_skb_head); - strp->rx_skb_head = NULL; + kfree_skb(strp->skb_head); + strp->skb_head = NULL; strp->cb.abort_parser(strp, err); } static inline int strp_peek_len(struct strparser *strp) { - struct socket *sock = strp->sk->sk_socket; + if (strp->sk) { + struct socket *sock = strp->sk->sk_socket; + + return sock->ops->peek_len(sock); + } - return sock->ops->peek_len(sock); + /* If we don't have an associated socket there's nothing to peek. + * Return int max to avoid stopping the strparser. + */ + + return INT_MAX; } /* Lower socket lock held */ -static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, - unsigned int orig_offset, size_t orig_len) +static int __strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, + unsigned int orig_offset, size_t orig_len, + size_t max_msg_size, long timeo) { struct strparser *strp = (struct strparser *)desc->arg.data; - struct _strp_rx_msg *rxm; + struct _strp_msg *stm; struct sk_buff *head, *skb; size_t eaten = 0, cand_len; ssize_t extra; int err; bool cloned_orig = false; - if (strp->rx_paused) + if (strp->paused) return 0; - head = strp->rx_skb_head; + head = strp->skb_head; if (head) { /* Message already in progress */ - rxm = _strp_rx_msg(head); - if (unlikely(rxm->early_eaten)) { + stm = _strp_msg(head); + if (unlikely(stm->early_eaten)) { /* Already some number of bytes on the receive sock - * data saved in rx_skb_head, just indicate they + * data saved in skb_head, just indicate they * are consumed. */ - eaten = orig_len <= rxm->early_eaten ? - orig_len : rxm->early_eaten; - rxm->early_eaten -= eaten; + eaten = orig_len <= stm->early_eaten ? + orig_len : stm->early_eaten; + stm->early_eaten -= eaten; return eaten; } @@ -126,12 +137,12 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, */ orig_skb = skb_clone(orig_skb, GFP_ATOMIC); if (!orig_skb) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); desc->error = -ENOMEM; return 0; } if (!pskb_pull(orig_skb, orig_offset)) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); kfree_skb(orig_skb); desc->error = -ENOMEM; return 0; @@ -140,13 +151,13 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, orig_offset = 0; } - if (!strp->rx_skb_nextp) { + if (!strp->skb_nextp) { /* We are going to append to the frags_list of head. * Need to unshare the frag_list. */ err = skb_unclone(head, GFP_ATOMIC); if (err) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); desc->error = err; return 0; } @@ -165,20 +176,20 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, skb = alloc_skb(0, GFP_ATOMIC); if (!skb) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); desc->error = -ENOMEM; return 0; } skb->len = head->len; skb->data_len = head->len; skb->truesize = head->truesize; - *_strp_rx_msg(skb) = *_strp_rx_msg(head); - strp->rx_skb_nextp = &head->next; + *_strp_msg(skb) = *_strp_msg(head); + strp->skb_nextp = &head->next; skb_shinfo(skb)->frag_list = head; - strp->rx_skb_head = skb; + strp->skb_head = skb; head = skb; } else { - strp->rx_skb_nextp = + strp->skb_nextp = &skb_shinfo(head)->frag_list; } } @@ -188,112 +199,112 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, /* Always clone since we will consume something */ skb = skb_clone(orig_skb, GFP_ATOMIC); if (!skb) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); desc->error = -ENOMEM; break; } cand_len = orig_len - eaten; - head = strp->rx_skb_head; + head = strp->skb_head; if (!head) { head = skb; - strp->rx_skb_head = head; - /* Will set rx_skb_nextp on next packet if needed */ - strp->rx_skb_nextp = NULL; - rxm = _strp_rx_msg(head); - memset(rxm, 0, sizeof(*rxm)); - rxm->strp.offset = orig_offset + eaten; + strp->skb_head = head; + /* Will set skb_nextp on next packet if needed */ + strp->skb_nextp = NULL; + stm = _strp_msg(head); + memset(stm, 0, sizeof(*stm)); + stm->strp.offset = orig_offset + eaten; } else { /* Unclone since we may be appending to an skb that we * already share a frag_list with. */ err = skb_unclone(skb, GFP_ATOMIC); if (err) { - STRP_STATS_INCR(strp->stats.rx_mem_fail); + STRP_STATS_INCR(strp->stats.mem_fail); desc->error = err; break; } - rxm = _strp_rx_msg(head); - *strp->rx_skb_nextp = skb; - strp->rx_skb_nextp = &skb->next; + stm = _strp_msg(head); + *strp->skb_nextp = skb; + strp->skb_nextp = &skb->next; head->data_len += skb->len; head->len += skb->len; head->truesize += skb->truesize; } - if (!rxm->strp.full_len) { + if (!stm->strp.full_len) { ssize_t len; len = (*strp->cb.parse_msg)(strp, head); if (!len) { /* Need more header to determine length */ - if (!rxm->accum_len) { + if (!stm->accum_len) { /* Start RX timer for new message */ - strp_start_rx_timer(strp); + strp_start_timer(strp, timeo); } - rxm->accum_len += cand_len; + stm->accum_len += cand_len; eaten += cand_len; - STRP_STATS_INCR(strp->stats.rx_need_more_hdr); + STRP_STATS_INCR(strp->stats.need_more_hdr); WARN_ON(eaten != orig_len); break; } else if (len < 0) { - if (len == -ESTRPIPE && rxm->accum_len) { + if (len == -ESTRPIPE && stm->accum_len) { len = -ENODATA; - strp->rx_unrecov_intr = 1; + strp->unrecov_intr = 1; } else { - strp->rx_interrupted = 1; + strp->interrupted = 1; } strp_parser_err(strp, len, desc); break; - } else if (len > strp->sk->sk_rcvbuf) { + } else if (len > max_msg_size) { /* Message length exceeds maximum allowed */ - STRP_STATS_INCR(strp->stats.rx_msg_too_big); + STRP_STATS_INCR(strp->stats.msg_too_big); strp_parser_err(strp, -EMSGSIZE, desc); break; } else if (len <= (ssize_t)head->len - - skb->len - rxm->strp.offset) { + skb->len - stm->strp.offset) { /* Length must be into new skb (and also * greater than zero) */ - STRP_STATS_INCR(strp->stats.rx_bad_hdr_len); + STRP_STATS_INCR(strp->stats.bad_hdr_len); strp_parser_err(strp, -EPROTO, desc); break; } - rxm->strp.full_len = len; + stm->strp.full_len = len; } - extra = (ssize_t)(rxm->accum_len + cand_len) - - rxm->strp.full_len; + extra = (ssize_t)(stm->accum_len + cand_len) - + stm->strp.full_len; if (extra < 0) { /* Message not complete yet. */ - if (rxm->strp.full_len - rxm->accum_len > + if (stm->strp.full_len - stm->accum_len > strp_peek_len(strp)) { - /* Don't have the whole messages in the socket - * buffer. Set strp->rx_need_bytes to wait for + /* Don't have the whole message in the socket + * buffer. Set strp->need_bytes to wait for * the rest of the message. Also, set "early * eaten" since we've already buffered the skb * but don't consume yet per strp_read_sock. */ - if (!rxm->accum_len) { + if (!stm->accum_len) { /* Start RX timer for new message */ - strp_start_rx_timer(strp); + strp_start_timer(strp, timeo); } - strp->rx_need_bytes = rxm->strp.full_len - - rxm->accum_len; - rxm->accum_len += cand_len; - rxm->early_eaten = cand_len; - STRP_STATS_ADD(strp->stats.rx_bytes, cand_len); + strp->need_bytes = stm->strp.full_len - + stm->accum_len; + stm->accum_len += cand_len; + stm->early_eaten = cand_len; + STRP_STATS_ADD(strp->stats.bytes, cand_len); desc->count = 0; /* Stop reading socket */ break; } - rxm->accum_len += cand_len; + stm->accum_len += cand_len; eaten += cand_len; WARN_ON(eaten != orig_len); break; @@ -308,14 +319,14 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, eaten += (cand_len - extra); /* Hurray, we have a new message! */ - del_timer(&strp->rx_msg_timer); - strp->rx_skb_head = NULL; - STRP_STATS_INCR(strp->stats.rx_msgs); + del_timer(&strp->msg_timer); + strp->skb_head = NULL; + STRP_STATS_INCR(strp->stats.msgs); /* Give skb to upper layer */ strp->cb.rcv_msg(strp, head); - if (unlikely(strp->rx_paused)) { + if (unlikely(strp->paused)) { /* Upper layer paused strp */ break; } @@ -324,11 +335,33 @@ static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, if (cloned_orig) kfree_skb(orig_skb); - STRP_STATS_ADD(strp->stats.rx_bytes, eaten); + STRP_STATS_ADD(strp->stats.bytes, eaten); return eaten; } +int strp_process(struct strparser *strp, struct sk_buff *orig_skb, + unsigned int orig_offset, size_t orig_len, + size_t max_msg_size, long timeo) +{ + read_descriptor_t desc; /* Dummy arg to strp_recv */ + + desc.arg.data = strp; + + return __strp_recv(&desc, orig_skb, orig_offset, orig_len, + max_msg_size, timeo); +} +EXPORT_SYMBOL_GPL(strp_process); + +static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, + unsigned int orig_offset, size_t orig_len) +{ + struct strparser *strp = (struct strparser *)desc->arg.data; + + return __strp_recv(desc, orig_skb, orig_offset, orig_len, + strp->sk->sk_rcvbuf, strp->sk->sk_rcvtimeo); +} + static int default_read_sock_done(struct strparser *strp, int err) { return err; @@ -340,6 +373,9 @@ static int strp_read_sock(struct strparser *strp) struct socket *sock = strp->sk->sk_socket; read_descriptor_t desc; + if (unlikely(!sock || !sock->ops || !sock->ops->read_sock)) + return -EBUSY; + desc.arg.data = strp; desc.error = 0; desc.count = 1; /* give more than one skb per call */ @@ -355,101 +391,124 @@ static int strp_read_sock(struct strparser *strp) /* Lower sock lock held */ void strp_data_ready(struct strparser *strp) { - if (unlikely(strp->rx_stopped)) + if (unlikely(strp->stopped)) return; - /* This check is needed to synchronize with do_strp_rx_work. - * do_strp_rx_work acquires a process lock (lock_sock) whereas + /* This check is needed to synchronize with do_strp_work. + * do_strp_work acquires a process lock (lock_sock) whereas * the lock held here is bh_lock_sock. The two locks can be * held by different threads at the same time, but bh_lock_sock * allows a thread in BH context to safely check if the process * lock is held. In this case, if the lock is held, queue work. */ if (sock_owned_by_user(strp->sk)) { - queue_work(strp_wq, &strp->rx_work); + queue_work(strp_wq, &strp->work); return; } - if (strp->rx_paused) + if (strp->paused) return; - if (strp->rx_need_bytes) { - if (strp_peek_len(strp) >= strp->rx_need_bytes) - strp->rx_need_bytes = 0; + if (strp->need_bytes) { + if (strp_peek_len(strp) >= strp->need_bytes) + strp->need_bytes = 0; else return; } if (strp_read_sock(strp) == -ENOMEM) - queue_work(strp_wq, &strp->rx_work); + queue_work(strp_wq, &strp->work); } EXPORT_SYMBOL_GPL(strp_data_ready); -static void do_strp_rx_work(struct strparser *strp) +static void do_strp_work(struct strparser *strp) { read_descriptor_t rd_desc; - struct sock *csk = strp->sk; /* We need the read lock to synchronize with strp_data_ready. We * need the socket lock for calling strp_read_sock. */ - lock_sock(csk); + strp->cb.lock(strp); - if (unlikely(strp->rx_stopped)) + if (unlikely(strp->stopped)) goto out; - if (strp->rx_paused) + if (strp->paused) goto out; rd_desc.arg.data = strp; if (strp_read_sock(strp) == -ENOMEM) - queue_work(strp_wq, &strp->rx_work); + queue_work(strp_wq, &strp->work); out: - release_sock(csk); + strp->cb.unlock(strp); } -static void strp_rx_work(struct work_struct *w) +static void strp_work(struct work_struct *w) { - do_strp_rx_work(container_of(w, struct strparser, rx_work)); + do_strp_work(container_of(w, struct strparser, work)); } -static void strp_rx_msg_timeout(unsigned long arg) +static void strp_msg_timeout(unsigned long arg) { struct strparser *strp = (struct strparser *)arg; /* Message assembly timed out */ - STRP_STATS_INCR(strp->stats.rx_msg_timeouts); - lock_sock(strp->sk); + STRP_STATS_INCR(strp->stats.msg_timeouts); + strp->cb.lock(strp); strp->cb.abort_parser(strp, ETIMEDOUT); + strp->cb.unlock(strp); +} + +static void strp_sock_lock(struct strparser *strp) +{ + lock_sock(strp->sk); +} + +static void strp_sock_unlock(struct strparser *strp) +{ release_sock(strp->sk); } -int strp_init(struct strparser *strp, struct sock *csk, - struct strp_callbacks *cb) +int strp_init(struct strparser *strp, struct sock *sk, + const struct strp_callbacks *cb) { - struct socket *sock = csk->sk_socket; if (!cb || !cb->rcv_msg || !cb->parse_msg) return -EINVAL; - if (!sock->ops->read_sock || !sock->ops->peek_len) - return -EAFNOSUPPORT; - - memset(strp, 0, sizeof(*strp)); + /* The sk (sock) arg determines the mode of the stream parser. + * + * If the sock is set then the strparser is in receive callback mode. + * The upper layer calls strp_data_ready to kick receive processing + * and strparser calls the read_sock function on the socket to + * get packets. + * + * If the sock is not set then the strparser is in general mode. + * The upper layer calls strp_process for each skb to be parsed. + */ - strp->sk = csk; + if (!sk) { + if (!cb->lock || !cb->unlock) + return -EINVAL; + } - setup_timer(&strp->rx_msg_timer, strp_rx_msg_timeout, - (unsigned long)strp); + memset(strp, 0, sizeof(*strp)); - INIT_WORK(&strp->rx_work, strp_rx_work); + strp->sk = sk; + strp->cb.lock = cb->lock ? : strp_sock_lock; + strp->cb.unlock = cb->unlock ? : strp_sock_unlock; strp->cb.rcv_msg = cb->rcv_msg; strp->cb.parse_msg = cb->parse_msg; strp->cb.read_sock_done = cb->read_sock_done ? : default_read_sock_done; - strp->cb.abort_parser = cb->abort_parser ? : strp_abort_rx_strp; + strp->cb.abort_parser = cb->abort_parser ? : strp_abort_strp; + + setup_timer(&strp->msg_timer, strp_msg_timeout, + (unsigned long)strp); + + INIT_WORK(&strp->work, strp_work); return 0; } @@ -457,12 +516,12 @@ EXPORT_SYMBOL_GPL(strp_init); void strp_unpause(struct strparser *strp) { - strp->rx_paused = 0; + strp->paused = 0; - /* Sync setting rx_paused with RX work */ + /* Sync setting paused with RX work */ smp_mb(); - queue_work(strp_wq, &strp->rx_work); + queue_work(strp_wq, &strp->work); } EXPORT_SYMBOL_GPL(strp_unpause); @@ -471,27 +530,27 @@ EXPORT_SYMBOL_GPL(strp_unpause); */ void strp_done(struct strparser *strp) { - WARN_ON(!strp->rx_stopped); + WARN_ON(!strp->stopped); - del_timer_sync(&strp->rx_msg_timer); - cancel_work_sync(&strp->rx_work); + del_timer_sync(&strp->msg_timer); + cancel_work_sync(&strp->work); - if (strp->rx_skb_head) { - kfree_skb(strp->rx_skb_head); - strp->rx_skb_head = NULL; + if (strp->skb_head) { + kfree_skb(strp->skb_head); + strp->skb_head = NULL; } } EXPORT_SYMBOL_GPL(strp_done); void strp_stop(struct strparser *strp) { - strp->rx_stopped = 1; + strp->stopped = 1; } EXPORT_SYMBOL_GPL(strp_stop); void strp_check_rcv(struct strparser *strp) { - queue_work(strp_wq, &strp->rx_work); + queue_work(strp_wq, &strp->work); } EXPORT_SYMBOL_GPL(strp_check_rcv); diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 2b720fa35c4f..e18500151236 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -421,6 +421,9 @@ static void svc_data_ready(struct sock *sk) dprintk("svc: socket %p(inet %p), busy=%d\n", svsk, sk, test_bit(XPT_BUSY, &svsk->sk_xprt.xpt_flags)); + + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_odata(sk); if (!test_and_set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags)) svc_xprt_enqueue(&svsk->sk_xprt); @@ -437,6 +440,9 @@ static void svc_write_space(struct sock *sk) if (svsk) { dprintk("svc: socket %p(inet %p), write_space busy=%d\n", svsk, sk, test_bit(XPT_BUSY, &svsk->sk_xprt.xpt_flags)); + + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_owspace(sk); svc_xprt_enqueue(&svsk->sk_xprt); } @@ -760,8 +766,12 @@ static void svc_tcp_listen_data_ready(struct sock *sk) dprintk("svc: socket %p TCP (listen) state change %d\n", sk, sk->sk_state); - if (svsk) + if (svsk) { + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_odata(sk); + } + /* * This callback may called twice when a new connection * is established as a child socket inherits everything @@ -794,6 +804,8 @@ static void svc_tcp_state_change(struct sock *sk) if (!svsk) printk("svc: socket %p: no user data\n", sk); else { + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_ostate(sk); if (sk->sk_state != TCP_ESTABLISHED) { set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags); @@ -1381,12 +1393,18 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, return ERR_PTR(err); } - inet->sk_user_data = svsk; svsk->sk_sock = sock; svsk->sk_sk = inet; svsk->sk_ostate = inet->sk_state_change; svsk->sk_odata = inet->sk_data_ready; svsk->sk_owspace = inet->sk_write_space; + /* + * This barrier is necessary in order to prevent race condition + * with svc_data_ready(), svc_listen_data_ready() and others + * when calling callbacks above. + */ + wmb(); + inet->sk_user_data = svsk; /* Initialize the socket */ if (sock->type == SOCK_DGRAM) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index d5b54c020dec..4f154d388748 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1624,6 +1624,8 @@ static void xs_tcp_state_change(struct sock *sk) if (test_and_clear_bit(XPRT_SOCK_CONNECTING, &transport->sock_state)) xprt_clear_connecting(xprt); + if (sk->sk_err) + xprt_wake_pending_tasks(xprt, -sk->sk_err); xs_sock_mark_closed(xprt); } out: diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index 25dc67ef9d37..0531b41d1f2d 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -343,8 +343,6 @@ static size_t switchdev_obj_size(const struct switchdev_obj *obj) switch (obj->id) { case SWITCHDEV_OBJ_ID_PORT_VLAN: return sizeof(struct switchdev_obj_port_vlan); - case SWITCHDEV_OBJ_ID_PORT_FDB: - return sizeof(struct switchdev_obj_port_fdb); case SWITCHDEV_OBJ_ID_PORT_MDB: return sizeof(struct switchdev_obj_port_mdb); default: @@ -534,43 +532,6 @@ int switchdev_port_obj_del(struct net_device *dev, } EXPORT_SYMBOL_GPL(switchdev_port_obj_del); -/** - * switchdev_port_obj_dump - Dump port objects - * - * @dev: port device - * @id: object ID - * @obj: object to dump - * @cb: function to call with a filled object - * - * rtnl_lock must be held. - */ -int switchdev_port_obj_dump(struct net_device *dev, struct switchdev_obj *obj, - switchdev_obj_dump_cb_t *cb) -{ - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - int err = -EOPNOTSUPP; - - ASSERT_RTNL(); - - if (ops && ops->switchdev_port_obj_dump) - return ops->switchdev_port_obj_dump(dev, obj, cb); - - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to dump objects on - * first port at bottom of stack. - */ - - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = switchdev_port_obj_dump(lower_dev, obj, cb); - break; - } - - return err; -} -EXPORT_SYMBOL_GPL(switchdev_port_obj_dump); - static ATOMIC_NOTIFIER_HEAD(switchdev_notif_chain); /** @@ -613,486 +574,6 @@ int call_switchdev_notifiers(unsigned long val, struct net_device *dev, } EXPORT_SYMBOL_GPL(call_switchdev_notifiers); -struct switchdev_vlan_dump { - struct switchdev_obj_port_vlan vlan; - struct sk_buff *skb; - u32 filter_mask; - u16 flags; - u16 begin; - u16 end; -}; - -static int switchdev_port_vlan_dump_put(struct switchdev_vlan_dump *dump) -{ - struct bridge_vlan_info vinfo; - - vinfo.flags = dump->flags; - - if (dump->begin == 0 && dump->end == 0) { - return 0; - } else if (dump->begin == dump->end) { - vinfo.vid = dump->begin; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - } else { - vinfo.vid = dump->begin; - vinfo.flags |= BRIDGE_VLAN_INFO_RANGE_BEGIN; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - vinfo.vid = dump->end; - vinfo.flags &= ~BRIDGE_VLAN_INFO_RANGE_BEGIN; - vinfo.flags |= BRIDGE_VLAN_INFO_RANGE_END; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - } - - return 0; -} - -static int switchdev_port_vlan_dump_cb(struct switchdev_obj *obj) -{ - struct switchdev_obj_port_vlan *vlan = SWITCHDEV_OBJ_PORT_VLAN(obj); - struct switchdev_vlan_dump *dump = - container_of(vlan, struct switchdev_vlan_dump, vlan); - int err = 0; - - if (vlan->vid_begin > vlan->vid_end) - return -EINVAL; - - if (dump->filter_mask & RTEXT_FILTER_BRVLAN) { - dump->flags = vlan->flags; - for (dump->begin = dump->end = vlan->vid_begin; - dump->begin <= vlan->vid_end; - dump->begin++, dump->end++) { - err = switchdev_port_vlan_dump_put(dump); - if (err) - return err; - } - } else if (dump->filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED) { - if (dump->begin > vlan->vid_begin && - dump->begin >= vlan->vid_end) { - if ((dump->begin - 1) == vlan->vid_end && - dump->flags == vlan->flags) { - /* prepend */ - dump->begin = vlan->vid_begin; - } else { - err = switchdev_port_vlan_dump_put(dump); - dump->flags = vlan->flags; - dump->begin = vlan->vid_begin; - dump->end = vlan->vid_end; - } - } else if (dump->end <= vlan->vid_begin && - dump->end < vlan->vid_end) { - if ((dump->end + 1) == vlan->vid_begin && - dump->flags == vlan->flags) { - /* append */ - dump->end = vlan->vid_end; - } else { - err = switchdev_port_vlan_dump_put(dump); - dump->flags = vlan->flags; - dump->begin = vlan->vid_begin; - dump->end = vlan->vid_end; - } - } else { - err = -EINVAL; - } - } - - return err; -} - -static int switchdev_port_vlan_fill(struct sk_buff *skb, struct net_device *dev, - u32 filter_mask) -{ - struct switchdev_vlan_dump dump = { - .vlan.obj.orig_dev = dev, - .vlan.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, - .skb = skb, - .filter_mask = filter_mask, - }; - int err = 0; - - if ((filter_mask & RTEXT_FILTER_BRVLAN) || - (filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED)) { - err = switchdev_port_obj_dump(dev, &dump.vlan.obj, - switchdev_port_vlan_dump_cb); - if (err) - goto err_out; - if (filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED) - /* last one */ - err = switchdev_port_vlan_dump_put(&dump); - } - -err_out: - return err == -EOPNOTSUPP ? 0 : err; -} - -/** - * switchdev_port_bridge_getlink - Get bridge port attributes - * - * @dev: port device - * - * Called for SELF on rtnl_bridge_getlink to get bridge port - * attributes. - */ -int switchdev_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, u32 filter_mask, - int nlflags) -{ - struct switchdev_attr attr = { - .orig_dev = dev, - .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, - }; - u16 mode = BRIDGE_MODE_UNDEF; - u32 mask = BR_LEARNING | BR_LEARNING_SYNC | BR_FLOOD; - int err; - - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; - - err = switchdev_port_attr_get(dev, &attr); - if (err && err != -EOPNOTSUPP) - return err; - - return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, - attr.u.brport_flags, mask, nlflags, - filter_mask, switchdev_port_vlan_fill); -} -EXPORT_SYMBOL_GPL(switchdev_port_bridge_getlink); - -static int switchdev_port_br_setflag(struct net_device *dev, - struct nlattr *nlattr, - unsigned long brport_flag) -{ - struct switchdev_attr attr = { - .orig_dev = dev, - .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, - }; - u8 flag = nla_get_u8(nlattr); - int err; - - err = switchdev_port_attr_get(dev, &attr); - if (err) - return err; - - if (flag) - attr.u.brport_flags |= brport_flag; - else - attr.u.brport_flags &= ~brport_flag; - - return switchdev_port_attr_set(dev, &attr); -} - -static const struct nla_policy -switchdev_port_bridge_policy[IFLA_BRPORT_MAX + 1] = { - [IFLA_BRPORT_STATE] = { .type = NLA_U8 }, - [IFLA_BRPORT_COST] = { .type = NLA_U32 }, - [IFLA_BRPORT_PRIORITY] = { .type = NLA_U16 }, - [IFLA_BRPORT_MODE] = { .type = NLA_U8 }, - [IFLA_BRPORT_GUARD] = { .type = NLA_U8 }, - [IFLA_BRPORT_PROTECT] = { .type = NLA_U8 }, - [IFLA_BRPORT_FAST_LEAVE] = { .type = NLA_U8 }, - [IFLA_BRPORT_LEARNING] = { .type = NLA_U8 }, - [IFLA_BRPORT_LEARNING_SYNC] = { .type = NLA_U8 }, - [IFLA_BRPORT_UNICAST_FLOOD] = { .type = NLA_U8 }, -}; - -static int switchdev_port_br_setlink_protinfo(struct net_device *dev, - struct nlattr *protinfo) -{ - struct nlattr *attr; - int rem; - int err; - - err = nla_validate_nested(protinfo, IFLA_BRPORT_MAX, - switchdev_port_bridge_policy, NULL); - if (err) - return err; - - nla_for_each_nested(attr, protinfo, rem) { - switch (nla_type(attr)) { - case IFLA_BRPORT_LEARNING: - err = switchdev_port_br_setflag(dev, attr, - BR_LEARNING); - break; - case IFLA_BRPORT_LEARNING_SYNC: - err = switchdev_port_br_setflag(dev, attr, - BR_LEARNING_SYNC); - break; - case IFLA_BRPORT_UNICAST_FLOOD: - err = switchdev_port_br_setflag(dev, attr, BR_FLOOD); - break; - default: - err = -EOPNOTSUPP; - break; - } - if (err) - return err; - } - - return 0; -} - -static int switchdev_port_br_afspec(struct net_device *dev, - struct nlattr *afspec, - int (*f)(struct net_device *dev, - const struct switchdev_obj *obj)) -{ - struct nlattr *attr; - struct bridge_vlan_info *vinfo; - struct switchdev_obj_port_vlan vlan = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, - }; - int rem; - int err; - - nla_for_each_nested(attr, afspec, rem) { - if (nla_type(attr) != IFLA_BRIDGE_VLAN_INFO) - continue; - if (nla_len(attr) != sizeof(struct bridge_vlan_info)) - return -EINVAL; - vinfo = nla_data(attr); - if (!vinfo->vid || vinfo->vid >= VLAN_VID_MASK) - return -EINVAL; - vlan.flags = vinfo->flags; - if (vinfo->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) { - if (vlan.vid_begin) - return -EINVAL; - vlan.vid_begin = vinfo->vid; - /* don't allow range of pvids */ - if (vlan.flags & BRIDGE_VLAN_INFO_PVID) - return -EINVAL; - } else if (vinfo->flags & BRIDGE_VLAN_INFO_RANGE_END) { - if (!vlan.vid_begin) - return -EINVAL; - vlan.vid_end = vinfo->vid; - if (vlan.vid_end <= vlan.vid_begin) - return -EINVAL; - err = f(dev, &vlan.obj); - if (err) - return err; - vlan.vid_begin = 0; - } else { - if (vlan.vid_begin) - return -EINVAL; - vlan.vid_begin = vinfo->vid; - vlan.vid_end = vinfo->vid; - err = f(dev, &vlan.obj); - if (err) - return err; - vlan.vid_begin = 0; - } - } - - return 0; -} - -/** - * switchdev_port_bridge_setlink - Set bridge port attributes - * - * @dev: port device - * @nlh: netlink header - * @flags: netlink flags - * - * Called for SELF on rtnl_bridge_setlink to set bridge port - * attributes. - */ -int switchdev_port_bridge_setlink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) -{ - struct nlattr *protinfo; - struct nlattr *afspec; - int err = 0; - - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; - - protinfo = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_PROTINFO); - if (protinfo) { - err = switchdev_port_br_setlink_protinfo(dev, protinfo); - if (err) - return err; - } - - afspec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_AF_SPEC); - if (afspec) - err = switchdev_port_br_afspec(dev, afspec, - switchdev_port_obj_add); - - return err; -} -EXPORT_SYMBOL_GPL(switchdev_port_bridge_setlink); - -/** - * switchdev_port_bridge_dellink - Set bridge port attributes - * - * @dev: port device - * @nlh: netlink header - * @flags: netlink flags - * - * Called for SELF on rtnl_bridge_dellink to set bridge port - * attributes. - */ -int switchdev_port_bridge_dellink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) -{ - struct nlattr *afspec; - - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; - - afspec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_AF_SPEC); - if (afspec) - return switchdev_port_br_afspec(dev, afspec, - switchdev_port_obj_del); - - return 0; -} -EXPORT_SYMBOL_GPL(switchdev_port_bridge_dellink); - -/** - * switchdev_port_fdb_add - Add FDB (MAC/VLAN) entry to port - * - * @ndmsg: netlink hdr - * @nlattr: netlink attributes - * @dev: port device - * @addr: MAC address to add - * @vid: VLAN to add - * - * Add FDB entry to switch device. - */ -int switchdev_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, const unsigned char *addr, - u16 vid, u16 nlm_flags) -{ - struct switchdev_obj_port_fdb fdb = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .vid = vid, - }; - - ether_addr_copy(fdb.addr, addr); - return switchdev_port_obj_add(dev, &fdb.obj); -} -EXPORT_SYMBOL_GPL(switchdev_port_fdb_add); - -/** - * switchdev_port_fdb_del - Delete FDB (MAC/VLAN) entry from port - * - * @ndmsg: netlink hdr - * @nlattr: netlink attributes - * @dev: port device - * @addr: MAC address to delete - * @vid: VLAN to delete - * - * Delete FDB entry from switch device. - */ -int switchdev_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, const unsigned char *addr, - u16 vid) -{ - struct switchdev_obj_port_fdb fdb = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .vid = vid, - }; - - ether_addr_copy(fdb.addr, addr); - return switchdev_port_obj_del(dev, &fdb.obj); -} -EXPORT_SYMBOL_GPL(switchdev_port_fdb_del); - -struct switchdev_fdb_dump { - struct switchdev_obj_port_fdb fdb; - struct net_device *dev; - struct sk_buff *skb; - struct netlink_callback *cb; - int idx; -}; - -static int switchdev_port_fdb_dump_cb(struct switchdev_obj *obj) -{ - struct switchdev_obj_port_fdb *fdb = SWITCHDEV_OBJ_PORT_FDB(obj); - struct switchdev_fdb_dump *dump = - container_of(fdb, struct switchdev_fdb_dump, fdb); - u32 portid = NETLINK_CB(dump->cb->skb).portid; - u32 seq = dump->cb->nlh->nlmsg_seq; - struct nlmsghdr *nlh; - struct ndmsg *ndm; - - if (dump->idx < dump->cb->args[2]) - goto skip; - - nlh = nlmsg_put(dump->skb, portid, seq, RTM_NEWNEIGH, - sizeof(*ndm), NLM_F_MULTI); - if (!nlh) - return -EMSGSIZE; - - ndm = nlmsg_data(nlh); - ndm->ndm_family = AF_BRIDGE; - ndm->ndm_pad1 = 0; - ndm->ndm_pad2 = 0; - ndm->ndm_flags = NTF_SELF; - ndm->ndm_type = 0; - ndm->ndm_ifindex = dump->dev->ifindex; - ndm->ndm_state = fdb->ndm_state; - - if (nla_put(dump->skb, NDA_LLADDR, ETH_ALEN, fdb->addr)) - goto nla_put_failure; - - if (fdb->vid && nla_put_u16(dump->skb, NDA_VLAN, fdb->vid)) - goto nla_put_failure; - - nlmsg_end(dump->skb, nlh); - -skip: - dump->idx++; - return 0; - -nla_put_failure: - nlmsg_cancel(dump->skb, nlh); - return -EMSGSIZE; -} - -/** - * switchdev_port_fdb_dump - Dump port FDB (MAC/VLAN) entries - * - * @skb: netlink skb - * @cb: netlink callback - * @dev: port device - * @filter_dev: filter device - * @idx: - * - * Dump FDB entries from switch device. - */ -int switchdev_port_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb, - struct net_device *dev, - struct net_device *filter_dev, int *idx) -{ - struct switchdev_fdb_dump dump = { - .fdb.obj.orig_dev = dev, - .fdb.obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .dev = dev, - .skb = skb, - .cb = cb, - .idx = *idx, - }; - int err; - - err = switchdev_port_obj_dump(dev, &dump.fdb.obj, - switchdev_port_fdb_dump_cb); - *idx = dump.idx; - return err; -} -EXPORT_SYMBOL_GPL(switchdev_port_fdb_dump); - bool switchdev_port_same_parent_id(struct net_device *a, struct net_device *b) { diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index d174ee3254ee..ac1d66d7e1fd 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -65,6 +65,8 @@ static struct tipc_bearer *bearer_get(struct net *net, int bearer_id) } static void bearer_disable(struct net *net, struct tipc_bearer *b); +static int tipc_l2_rcv_msg(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev); /** * tipc_media_find - locates specified media object by name @@ -365,30 +367,6 @@ static int tipc_reset_bearer(struct net *net, struct tipc_bearer *b) return 0; } -/* tipc_bearer_reset_all - reset all links on all bearers - */ -void tipc_bearer_reset_all(struct net *net) -{ - struct tipc_bearer *b; - int i; - - for (i = 0; i < MAX_BEARERS; i++) { - b = bearer_get(net, i); - if (b) - clear_bit_unlock(0, &b->up); - } - for (i = 0; i < MAX_BEARERS; i++) { - b = bearer_get(net, i); - if (b) - tipc_reset_bearer(net, b); - } - for (i = 0; i < MAX_BEARERS; i++) { - b = bearer_get(net, i); - if (b) - test_and_set_bit_lock(0, &b->up); - } -} - /** * bearer_disable * @@ -428,6 +406,10 @@ int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, /* Associate TIPC bearer with L2 bearer */ rcu_assign_pointer(b->media_ptr, dev); + b->pt.dev = dev; + b->pt.type = htons(ETH_P_TIPC); + b->pt.func = tipc_l2_rcv_msg; + dev_add_pack(&b->pt); memset(&b->bcast_addr, 0, sizeof(b->bcast_addr)); memcpy(b->bcast_addr.value, dev->broadcast, b->media->hwaddr_len); b->bcast_addr.media_id = b->media->type_id; @@ -447,6 +429,7 @@ void tipc_disable_l2_media(struct tipc_bearer *b) struct net_device *dev; dev = (struct net_device *)rtnl_dereference(b->media_ptr); + dev_remove_pack(&b->pt); RCU_INIT_POINTER(dev->tipc_ptr, NULL); synchronize_net(); dev_put(dev); @@ -594,11 +577,12 @@ static int tipc_l2_rcv_msg(struct sk_buff *skb, struct net_device *dev, struct tipc_bearer *b; rcu_read_lock(); - b = rcu_dereference_rtnl(dev->tipc_ptr); + b = rcu_dereference_rtnl(dev->tipc_ptr) ?: + rcu_dereference_rtnl(orig_dev->tipc_ptr); if (likely(b && test_bit(0, &b->up) && - (skb->pkt_type <= PACKET_BROADCAST))) { + (skb->pkt_type <= PACKET_MULTICAST))) { skb->next = NULL; - tipc_rcv(dev_net(dev), skb, b); + tipc_rcv(dev_net(b->pt.dev), skb, b); rcu_read_unlock(); return NET_RX_SUCCESS; } @@ -659,11 +643,6 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, return NOTIFY_OK; } -static struct packet_type tipc_packet_type __read_mostly = { - .type = htons(ETH_P_TIPC), - .func = tipc_l2_rcv_msg, -}; - static struct notifier_block notifier = { .notifier_call = tipc_l2_device_event, .priority = 0, @@ -671,19 +650,12 @@ static struct notifier_block notifier = { int tipc_bearer_setup(void) { - int err; - - err = register_netdevice_notifier(¬ifier); - if (err) - return err; - dev_add_pack(&tipc_packet_type); - return 0; + return register_netdevice_notifier(¬ifier); } void tipc_bearer_cleanup(void) { unregister_netdevice_notifier(¬ifier); - dev_remove_pack(&tipc_packet_type); } void tipc_bearer_stop(struct net *net) diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 635c9086e19a..42d6eeeb646d 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -131,6 +131,7 @@ struct tipc_media { * @name: bearer name (format = media:interface) * @media: ptr to media structure associated with bearer * @bcast_addr: media address used in broadcasting + * @pt: packet type for bearer * @rcu: rcu struct for tipc_bearer * @priority: default link priority for bearer * @window: default window size for bearer @@ -151,6 +152,7 @@ struct tipc_bearer { char name[TIPC_MAX_BEARER_NAME]; struct tipc_media *media; struct tipc_media_addr bcast_addr; + struct packet_type pt; struct rcu_head rcu; u32 priority; u32 window; @@ -210,7 +212,6 @@ void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest); struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name); int tipc_bearer_get_name(struct net *net, char *name, u32 bearer_id); struct tipc_media *tipc_media_find(const char *name); -void tipc_bearer_reset_all(struct net *net); int tipc_bearer_setup(void); void tipc_bearer_cleanup(void); void tipc_bearer_stop(struct net *net); diff --git a/net/tipc/link.c b/net/tipc/link.c index 60820dc35a08..ac0144f532aa 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -978,15 +978,15 @@ static void link_retransmit_failure(struct tipc_link *l, struct sk_buff *skb) struct tipc_msg *hdr = buf_msg(skb); pr_warn("Retransmission failure on link <%s>\n", l->name); - link_print(l, "Resetting link "); + 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)); } -int tipc_link_retrans(struct tipc_link *l, u16 from, u16 to, - struct sk_buff_head *xmitq) +int tipc_link_retrans(struct tipc_link *l, struct tipc_link *nacker, + u16 from, u16 to, struct sk_buff_head *xmitq) { struct sk_buff *_skb, *skb = skb_peek(&l->transmq); struct tipc_msg *hdr; @@ -997,11 +997,14 @@ int tipc_link_retrans(struct tipc_link *l, u16 from, u16 to, return 0; /* Detect repeated retransmit failures on same packet */ - if (likely(l->last_retransm != buf_seqno(skb))) { - l->last_retransm = buf_seqno(skb); - l->stale_count = 1; - } else if (++l->stale_count > 100) { + if (nacker->last_retransm != buf_seqno(skb)) { + nacker->last_retransm = buf_seqno(skb); + nacker->stale_count = 1; + } else if (++nacker->stale_count > 100) { link_retransmit_failure(l, skb); + nacker->stale_count = 0; + if (link_is_bc_sndlink(l)) + return TIPC_LINK_DOWN_EVT; return tipc_link_fsm_evt(l, LINK_FAILURE_EVT); } @@ -1528,7 +1531,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, /* If NACK, retransmit will now start at right position */ if (gap) { - rc = tipc_link_retrans(l, ack + 1, ack + gap, xmitq); + rc = tipc_link_retrans(l, l, ack + 1, ack + gap, xmitq); l->stats.recv_nacks++; } @@ -1680,7 +1683,7 @@ int tipc_link_bc_sync_rcv(struct tipc_link *l, struct tipc_msg *hdr, return rc; if (link_bc_retr_eval(snd_l, &from, &to)) - rc = tipc_link_retrans(snd_l, from, to, xmitq); + rc = tipc_link_retrans(snd_l, l, from, to, xmitq); l->snd_nxt = peers_snd_nxt; if (link_bc_rcv_gap(l)) @@ -1775,7 +1778,7 @@ int tipc_link_bc_nack_rcv(struct tipc_link *l, struct sk_buff *skb, if (dnode == tipc_own_addr(l->net)) { tipc_link_bc_ack_rcv(l, acked, xmitq); - rc = tipc_link_retrans(l->bc_sndlink, from, to, xmitq); + rc = tipc_link_retrans(l->bc_sndlink, l, from, to, xmitq); l->stats.recv_nacks++; return rc; } diff --git a/net/tipc/msg.c b/net/tipc/msg.c index ab3087687a32..6ef379f004ac 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -479,13 +479,14 @@ bool tipc_msg_make_bundle(struct sk_buff **skb, struct tipc_msg *msg, bool tipc_msg_reverse(u32 own_node, struct sk_buff **skb, int err) { struct sk_buff *_skb = *skb; - struct tipc_msg *hdr = buf_msg(_skb); + struct tipc_msg *hdr; struct tipc_msg ohdr; - int dlen = min_t(uint, msg_data_sz(hdr), MAX_FORWARD_SIZE); + int dlen; if (skb_linearize(_skb)) goto exit; hdr = buf_msg(_skb); + dlen = min_t(uint, msg_data_sz(hdr), MAX_FORWARD_SIZE); if (msg_dest_droppable(hdr)) goto exit; if (msg_errcode(hdr)) @@ -511,8 +512,11 @@ bool tipc_msg_reverse(u32 own_node, struct sk_buff **skb, int err) pskb_expand_head(_skb, BUF_HEADROOM, BUF_TAILROOM, GFP_ATOMIC)) goto exit; + /* reassign after skb header modifications */ + hdr = buf_msg(_skb); /* Now reverse the concerned fields */ msg_set_errcode(hdr, err); + msg_set_non_seq(hdr, 0); msg_set_origport(hdr, msg_destport(&ohdr)); msg_set_destport(hdr, msg_origport(&ohdr)); msg_set_destnode(hdr, msg_prevnode(&ohdr)); diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c index 9bfe886ab330..e48f0b2c01b9 100644 --- a/net/tipc/netlink_compat.c +++ b/net/tipc/netlink_compat.c @@ -258,13 +258,15 @@ static int tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd, arg = nlmsg_new(0, GFP_KERNEL); if (!arg) { kfree_skb(msg->rep); + msg->rep = NULL; return -ENOMEM; } err = __tipc_nl_compat_dumpit(cmd, msg, arg); - if (err) + if (err) { kfree_skb(msg->rep); - + msg->rep = NULL; + } kfree_skb(arg); return err; @@ -1215,7 +1217,7 @@ send: return err; } -static struct genl_ops tipc_genl_compat_ops[] = { +static const struct genl_ops tipc_genl_compat_ops[] = { { .cmd = TIPC_GENL_CMD, .doit = tipc_nl_compat_recv, diff --git a/net/tipc/node.c b/net/tipc/node.c index aeef8011ac7d..198dbc7adbe1 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1126,8 +1126,8 @@ int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 addr, strncpy(linkname, tipc_link_name(link), len); err = 0; } -exit: tipc_node_read_unlock(node); +exit: tipc_node_put(node); return err; } @@ -1284,7 +1284,7 @@ static void tipc_node_bc_sync_rcv(struct tipc_node *n, struct tipc_msg *hdr, rc = tipc_bcast_sync_rcv(n->net, n->bc_entry.link, hdr); if (rc & TIPC_LINK_DOWN_EVT) { - tipc_bearer_reset_all(n->net); + tipc_node_reset_links(n); return; } @@ -1351,15 +1351,9 @@ static void tipc_node_bc_rcv(struct net *net, struct sk_buff *skb, int bearer_id if (!skb_queue_empty(&be->inputq1)) tipc_node_mcast_rcv(n); - if (rc & TIPC_LINK_DOWN_EVT) { - /* Reception reassembly failure => reset all links to peer */ - if (!tipc_link_is_up(be->link)) - tipc_node_reset_links(n); - - /* Retransmission failure => reset all links to all peers */ - if (!tipc_link_is_up(tipc_bc_sndlink(net))) - tipc_bearer_reset_all(net); - } + /* If reassembly or retransmission failure => reset all links to peer */ + if (rc & TIPC_LINK_DOWN_EVT) + tipc_node_reset_links(n); tipc_node_put(n); } @@ -1455,10 +1449,8 @@ static bool tipc_node_check_state(struct tipc_node *n, struct sk_buff *skb, /* Initiate synch mode if applicable */ if ((usr == TUNNEL_PROTOCOL) && (mtyp == SYNCH_MSG) && (oseqno == 1)) { syncpt = iseqno + exp_pkts - 1; - if (!tipc_link_is_up(l)) { - tipc_link_fsm_evt(l, LINK_ESTABLISH_EVT); + if (!tipc_link_is_up(l)) __tipc_node_link_up(n, bearer_id, xmitq); - } if (n->state == SELF_UP_PEER_UP) { n->sync_point = syncpt; tipc_link_fsm_evt(l, LINK_SYNCH_BEGIN_EVT); @@ -1559,6 +1551,8 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b) /* Check/update node state before receiving */ if (unlikely(skb)) { + if (unlikely(skb_linearize(skb))) + goto discard; tipc_node_write_lock(n); if (tipc_node_check_state(n, skb, bearer_id, &xmitq)) { if (le->link) { diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 101e3597338f..d50edd6e0019 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2255,8 +2255,8 @@ void tipc_sk_reinit(struct net *net) do { tsk = ERR_PTR(rhashtable_walk_start(&iter)); - if (tsk) - continue; + if (IS_ERR(tsk)) + goto walk_stop; while ((tsk = rhashtable_walk_next(&iter)) && !IS_ERR(tsk)) { spin_lock_bh(&tsk->sk.sk_lock.slock); @@ -2265,7 +2265,7 @@ void tipc_sk_reinit(struct net *net) msg_set_orignode(msg, tn->own_addr); spin_unlock_bh(&tsk->sk.sk_lock.slock); } - +walk_stop: rhashtable_walk_stop(&iter); } while (tsk == ERR_PTR(-EAGAIN)); } diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 0bf91cd3733c..be3d9e3183dc 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -52,7 +52,6 @@ struct tipc_subscriber { struct list_head subscrp_list; }; -static void tipc_subscrp_delete(struct tipc_subscription *sub); static void tipc_subscrb_put(struct tipc_subscriber *subscriber); /** @@ -197,15 +196,19 @@ static void tipc_subscrb_subscrp_delete(struct tipc_subscriber *subscriber, { struct list_head *subscription_list = &subscriber->subscrp_list; struct tipc_subscription *sub, *temp; + u32 timeout; spin_lock_bh(&subscriber->lock); list_for_each_entry_safe(sub, temp, subscription_list, subscrp_list) { if (s && memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) continue; - tipc_nametbl_unsubscribe(sub); - list_del(&sub->subscrp_list); - tipc_subscrp_delete(sub); + timeout = htohl(sub->evt.s.timeout, sub->swap); + if (timeout == TIPC_WAIT_FOREVER || del_timer(&sub->timer)) { + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscrp_list); + tipc_subscrp_put(sub); + } if (s) break; @@ -236,18 +239,12 @@ static void tipc_subscrb_delete(struct tipc_subscriber *subscriber) tipc_subscrb_put(subscriber); } -static void tipc_subscrp_delete(struct tipc_subscription *sub) -{ - u32 timeout = htohl(sub->evt.s.timeout, sub->swap); - - if (timeout == TIPC_WAIT_FOREVER || del_timer(&sub->timer)) - tipc_subscrp_put(sub); -} - static void tipc_subscrp_cancel(struct tipc_subscr *s, struct tipc_subscriber *subscriber) { + tipc_subscrb_get(subscriber); tipc_subscrb_subscrp_delete(subscriber, s); + tipc_subscrb_put(subscriber); } static struct tipc_subscription *tipc_subscrp_create(struct net *net, diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 7b52a380d710..7f46bab4ce5c 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1528,26 +1528,13 @@ static inline bool too_many_unix_fds(struct task_struct *p) return false; } -#define MAX_RECURSION_LEVEL 4 - static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) { int i; - unsigned char max_level = 0; if (too_many_unix_fds(current)) return -ETOOMANYREFS; - for (i = scm->fp->count - 1; i >= 0; i--) { - struct sock *sk = unix_get_socket(scm->fp->fp[i]); - - if (sk) - max_level = max(max_level, - unix_sk(sk)->recursion_level); - } - if (unlikely(max_level > MAX_RECURSION_LEVEL)) - return -ETOOMANYREFS; - /* * Need to duplicate file references for the sake of garbage * collection. Otherwise a socket in the fps might become a @@ -1559,7 +1546,7 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) for (i = scm->fp->count - 1; i >= 0; i--) unix_inflight(scm->fp->user, scm->fp->fp[i]); - return max_level; + return 0; } static int unix_scm_to_skb(struct scm_cookie *scm, struct sk_buff *skb, bool send_fds) @@ -1649,7 +1636,6 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, struct sk_buff *skb; long timeo; struct scm_cookie scm; - int max_level; int data_len = 0; int sk_locked; @@ -1701,7 +1687,6 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, err = unix_scm_to_skb(&scm, skb, true); if (err < 0) goto out_free; - max_level = err + 1; skb_put(skb, len - data_len); skb->data_len = data_len; @@ -1819,8 +1804,6 @@ restart_locked: __net_timestamp(skb); maybe_add_creds(skb, sock, other); skb_queue_tail(&other->sk_receive_queue, skb); - if (max_level > unix_sk(other)->recursion_level) - unix_sk(other)->recursion_level = max_level; unix_state_unlock(other); other->sk_data_ready(other); sock_put(other); @@ -1855,7 +1838,6 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, int sent = 0; struct scm_cookie scm; bool fds_sent = false; - int max_level; int data_len; wait_for_unix_gc(); @@ -1905,7 +1887,6 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, kfree_skb(skb); goto out_err; } - max_level = err + 1; fds_sent = true; skb_put(skb, size - data_len); @@ -1925,8 +1906,6 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, maybe_add_creds(skb, sock, other); skb_queue_tail(&other->sk_receive_queue, skb); - if (max_level > unix_sk(other)->recursion_level) - unix_sk(other)->recursion_level = max_level; unix_state_unlock(other); other->sk_data_ready(other); sent += size; @@ -2304,10 +2283,7 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state, */ mutex_lock(&u->iolock); - if (flags & MSG_PEEK) - skip = sk_peek_offset(sk, flags); - else - skip = 0; + skip = max(sk_peek_offset(sk, flags), 0); do { int chunk; @@ -2324,7 +2300,6 @@ redo: last_len = last ? last->len : 0; again: if (skb == NULL) { - unix_sk(sk)->recursion_level = 0; if (copied >= target) goto unlock; diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig index 8831e7c42167..a24369d175fd 100644 --- a/net/vmw_vsock/Kconfig +++ b/net/vmw_vsock/Kconfig @@ -46,3 +46,15 @@ config VIRTIO_VSOCKETS_COMMON This option is selected by any driver which needs to access the virtio_vsock. The module will be called vmw_vsock_virtio_transport_common. + +config HYPERV_VSOCKETS + tristate "Hyper-V transport for Virtual Sockets" + depends on VSOCKETS && HYPERV + help + This module implements a Hyper-V transport for Virtual Sockets. + + Enable this transport if your Virtual Machine host supports Virtual + Sockets over Hyper-V VMBus. + + To compile this driver as a module, choose M here: the module will be + called hv_sock. If unsure, say N. diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile index 09fc2eb29dc8..e63d574234a9 100644 --- a/net/vmw_vsock/Makefile +++ b/net/vmw_vsock/Makefile @@ -2,6 +2,7 @@ obj-$(CONFIG_VSOCKETS) += vsock.o obj-$(CONFIG_VMWARE_VMCI_VSOCKETS) += vmw_vsock_vmci_transport.o obj-$(CONFIG_VIRTIO_VSOCKETS) += vmw_vsock_virtio_transport.o obj-$(CONFIG_VIRTIO_VSOCKETS_COMMON) += vmw_vsock_virtio_transport_common.o +obj-$(CONFIG_HYPERV_VSOCKETS) += hv_sock.o vsock-y += af_vsock.o af_vsock_tap.o vsock_addr.o @@ -11,3 +12,5 @@ vmw_vsock_vmci_transport-y += vmci_transport.o vmci_transport_notify.o \ vmw_vsock_virtio_transport-y += virtio_transport.o vmw_vsock_virtio_transport_common-y += virtio_transport_common.o + +hv_sock-y += hyperv_transport.o diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c new file mode 100644 index 000000000000..14ed5a344cdf --- /dev/null +++ b/net/vmw_vsock/hyperv_transport.c @@ -0,0 +1,904 @@ +/* + * Hyper-V transport for vsock + * + * Hyper-V Sockets supplies a byte-stream based communication mechanism + * between the host and the VM. This driver implements the necessary + * support in the VM by introducing the new vsock transport. + * + * Copyright (c) 2017, Microsoft Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/hyperv.h> +#include <net/sock.h> +#include <net/af_vsock.h> + +/* The host side's design of the feature requires 6 exact 4KB pages for + * recv/send rings respectively -- this is suboptimal considering memory + * consumption, however unluckily we have to live with it, before the + * host comes up with a better design in the future. + */ +#define PAGE_SIZE_4K 4096 +#define RINGBUFFER_HVS_RCV_SIZE (PAGE_SIZE_4K * 6) +#define RINGBUFFER_HVS_SND_SIZE (PAGE_SIZE_4K * 6) + +/* The MTU is 16KB per the host side's design */ +#define HVS_MTU_SIZE (1024 * 16) + +struct vmpipe_proto_header { + u32 pkt_type; + u32 data_size; +}; + +/* For recv, we use the VMBus in-place packet iterator APIs to directly copy + * data from the ringbuffer into the userspace buffer. + */ +struct hvs_recv_buf { + /* The header before the payload data */ + struct vmpipe_proto_header hdr; + + /* The payload */ + u8 data[HVS_MTU_SIZE]; +}; + +/* We can send up to HVS_MTU_SIZE bytes of payload to the host, but let's use + * a small size, i.e. HVS_SEND_BUF_SIZE, to minimize the dynamically-allocated + * buffer, because tests show there is no significant performance difference. + * + * Note: the buffer can be eliminated in the future when we add new VMBus + * ringbuffer APIs that allow us to directly copy data from userspace buffer + * to VMBus ringbuffer. + */ +#define HVS_SEND_BUF_SIZE (PAGE_SIZE_4K - sizeof(struct vmpipe_proto_header)) + +struct hvs_send_buf { + /* The header before the payload data */ + struct vmpipe_proto_header hdr; + + /* The payload */ + u8 data[HVS_SEND_BUF_SIZE]; +}; + +#define HVS_HEADER_LEN (sizeof(struct vmpacket_descriptor) + \ + sizeof(struct vmpipe_proto_header)) + +/* See 'prev_indices' in hv_ringbuffer_read(), hv_ringbuffer_write(), and + * __hv_pkt_iter_next(). + */ +#define VMBUS_PKT_TRAILER_SIZE (sizeof(u64)) + +#define HVS_PKT_LEN(payload_len) (HVS_HEADER_LEN + \ + ALIGN((payload_len), 8) + \ + VMBUS_PKT_TRAILER_SIZE) + +union hvs_service_id { + uuid_le srv_id; + + struct { + unsigned int svm_port; + unsigned char b[sizeof(uuid_le) - sizeof(unsigned int)]; + }; +}; + +/* Per-socket state (accessed via vsk->trans) */ +struct hvsock { + struct vsock_sock *vsk; + + uuid_le vm_srv_id; + uuid_le host_srv_id; + + struct vmbus_channel *chan; + struct vmpacket_descriptor *recv_desc; + + /* The length of the payload not delivered to userland yet */ + u32 recv_data_len; + /* The offset of the payload */ + u32 recv_data_off; + + /* Have we sent the zero-length packet (FIN)? */ + bool fin_sent; +}; + +/* In the VM, we support Hyper-V Sockets with AF_VSOCK, and the endpoint is + * <cid, port> (see struct sockaddr_vm). Note: cid is not really used here: + * when we write apps to connect to the host, we can only use VMADDR_CID_ANY + * or VMADDR_CID_HOST (both are equivalent) as the remote cid, and when we + * write apps to bind() & listen() in the VM, we can only use VMADDR_CID_ANY + * as the local cid. + * + * On the host, Hyper-V Sockets are supported by Winsock AF_HYPERV: + * https://docs.microsoft.com/en-us/virtualization/hyper-v-on-windows/user- + * guide/make-integration-service, and the endpoint is <VmID, ServiceId> with + * the below sockaddr: + * + * struct SOCKADDR_HV + * { + * ADDRESS_FAMILY Family; + * USHORT Reserved; + * GUID VmId; + * GUID ServiceId; + * }; + * Note: VmID is not used by Linux VM and actually it isn't transmitted via + * VMBus, because here it's obvious the host and the VM can easily identify + * each other. Though the VmID is useful on the host, especially in the case + * of Windows container, Linux VM doesn't need it at all. + * + * To make use of the AF_VSOCK infrastructure in Linux VM, we have to limit + * the available GUID space of SOCKADDR_HV so that we can create a mapping + * between AF_VSOCK port and SOCKADDR_HV Service GUID. The rule of writing + * Hyper-V Sockets apps on the host and in Linux VM is: + * + **************************************************************************** + * The only valid Service GUIDs, from the perspectives of both the host and * + * Linux VM, that can be connected by the other end, must conform to this * + * format: <port>-facb-11e6-bd58-64006a7986d3, and the "port" must be in * + * this range [0, 0x7FFFFFFF]. * + **************************************************************************** + * + * When we write apps on the host to connect(), the GUID ServiceID is used. + * When we write apps in Linux VM to connect(), we only need to specify the + * port and the driver will form the GUID and use that to request the host. + * + * From the perspective of Linux VM: + * 1. the local ephemeral port (i.e. the local auto-bound port when we call + * connect() without explicit bind()) is generated by __vsock_bind_stream(), + * and the range is [1024, 0xFFFFFFFF). + * 2. the remote ephemeral port (i.e. the auto-generated remote port for + * a connect request initiated by the host's connect()) is generated by + * hvs_remote_addr_init() and the range is [0x80000000, 0xFFFFFFFF). + */ + +#define MAX_LISTEN_PORT ((u32)0x7FFFFFFF) +#define MAX_VM_LISTEN_PORT MAX_LISTEN_PORT +#define MAX_HOST_LISTEN_PORT MAX_LISTEN_PORT +#define MIN_HOST_EPHEMERAL_PORT (MAX_HOST_LISTEN_PORT + 1) + +/* 00000000-facb-11e6-bd58-64006a7986d3 */ +static const uuid_le srv_id_template = + UUID_LE(0x00000000, 0xfacb, 0x11e6, 0xbd, 0x58, + 0x64, 0x00, 0x6a, 0x79, 0x86, 0xd3); + +static bool is_valid_srv_id(const uuid_le *id) +{ + return !memcmp(&id->b[4], &srv_id_template.b[4], sizeof(uuid_le) - 4); +} + +static unsigned int get_port_by_srv_id(const uuid_le *svr_id) +{ + return *((unsigned int *)svr_id); +} + +static void hvs_addr_init(struct sockaddr_vm *addr, const uuid_le *svr_id) +{ + unsigned int port = get_port_by_srv_id(svr_id); + + vsock_addr_init(addr, VMADDR_CID_ANY, port); +} + +static void hvs_remote_addr_init(struct sockaddr_vm *remote, + struct sockaddr_vm *local) +{ + static u32 host_ephemeral_port = MIN_HOST_EPHEMERAL_PORT; + struct sock *sk; + + vsock_addr_init(remote, VMADDR_CID_ANY, VMADDR_PORT_ANY); + + while (1) { + /* Wrap around ? */ + if (host_ephemeral_port < MIN_HOST_EPHEMERAL_PORT || + host_ephemeral_port == VMADDR_PORT_ANY) + host_ephemeral_port = MIN_HOST_EPHEMERAL_PORT; + + remote->svm_port = host_ephemeral_port++; + + sk = vsock_find_connected_socket(remote, local); + if (!sk) { + /* Found an available ephemeral port */ + return; + } + + /* Release refcnt got in vsock_find_connected_socket */ + sock_put(sk); + } +} + +static void hvs_set_channel_pending_send_size(struct vmbus_channel *chan) +{ + set_channel_pending_send_size(chan, + HVS_PKT_LEN(HVS_SEND_BUF_SIZE)); + + /* See hvs_stream_has_space(): we must make sure the host has seen + * the new pending send size, before we can re-check the writable + * bytes. + */ + virt_mb(); +} + +static void hvs_clear_channel_pending_send_size(struct vmbus_channel *chan) +{ + set_channel_pending_send_size(chan, 0); + + /* Ditto */ + virt_mb(); +} + +static bool hvs_channel_readable(struct vmbus_channel *chan) +{ + u32 readable = hv_get_bytes_to_read(&chan->inbound); + + /* 0-size payload means FIN */ + return readable >= HVS_PKT_LEN(0); +} + +static int hvs_channel_readable_payload(struct vmbus_channel *chan) +{ + u32 readable = hv_get_bytes_to_read(&chan->inbound); + + if (readable > HVS_PKT_LEN(0)) { + /* At least we have 1 byte to read. We don't need to return + * the exact readable bytes: see vsock_stream_recvmsg() -> + * vsock_stream_has_data(). + */ + return 1; + } + + if (readable == HVS_PKT_LEN(0)) { + /* 0-size payload means FIN */ + return 0; + } + + /* No payload or FIN */ + return -1; +} + +static size_t hvs_channel_writable_bytes(struct vmbus_channel *chan) +{ + u32 writeable = hv_get_bytes_to_write(&chan->outbound); + size_t ret; + + /* The ringbuffer mustn't be 100% full, and we should reserve a + * zero-length-payload packet for the FIN: see hv_ringbuffer_write() + * and hvs_shutdown(). + */ + if (writeable <= HVS_PKT_LEN(1) + HVS_PKT_LEN(0)) + return 0; + + ret = writeable - HVS_PKT_LEN(1) - HVS_PKT_LEN(0); + + return round_down(ret, 8); +} + +static int hvs_send_data(struct vmbus_channel *chan, + struct hvs_send_buf *send_buf, size_t to_write) +{ + send_buf->hdr.pkt_type = 1; + send_buf->hdr.data_size = to_write; + return vmbus_sendpacket(chan, &send_buf->hdr, + sizeof(send_buf->hdr) + to_write, + 0, VM_PKT_DATA_INBAND, 0); +} + +static void hvs_channel_cb(void *ctx) +{ + struct sock *sk = (struct sock *)ctx; + struct vsock_sock *vsk = vsock_sk(sk); + struct hvsock *hvs = vsk->trans; + struct vmbus_channel *chan = hvs->chan; + + if (hvs_channel_readable(chan)) + sk->sk_data_ready(sk); + + /* See hvs_stream_has_space(): when we reach here, the writable bytes + * may be already less than HVS_PKT_LEN(HVS_SEND_BUF_SIZE). + */ + if (hv_get_bytes_to_write(&chan->outbound) > 0) + sk->sk_write_space(sk); +} + +static void hvs_close_connection(struct vmbus_channel *chan) +{ + struct sock *sk = get_per_channel_state(chan); + struct vsock_sock *vsk = vsock_sk(sk); + + sk->sk_state = SS_UNCONNECTED; + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown |= SEND_SHUTDOWN | RCV_SHUTDOWN; + + sk->sk_state_change(sk); +} + +static void hvs_open_connection(struct vmbus_channel *chan) +{ + uuid_le *if_instance, *if_type; + unsigned char conn_from_host; + + struct sockaddr_vm addr; + struct sock *sk, *new = NULL; + struct vsock_sock *vnew; + struct hvsock *hvs, *hvs_new; + int ret; + + if_type = &chan->offermsg.offer.if_type; + if_instance = &chan->offermsg.offer.if_instance; + conn_from_host = chan->offermsg.offer.u.pipe.user_def[0]; + + /* The host or the VM should only listen on a port in + * [0, MAX_LISTEN_PORT] + */ + if (!is_valid_srv_id(if_type) || + get_port_by_srv_id(if_type) > MAX_LISTEN_PORT) + return; + + hvs_addr_init(&addr, conn_from_host ? if_type : if_instance); + sk = vsock_find_bound_socket(&addr); + if (!sk) + return; + + if ((conn_from_host && sk->sk_state != VSOCK_SS_LISTEN) || + (!conn_from_host && sk->sk_state != SS_CONNECTING)) + goto out; + + if (conn_from_host) { + if (sk->sk_ack_backlog >= sk->sk_max_ack_backlog) + goto out; + + new = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL, + sk->sk_type, 0); + if (!new) + goto out; + + new->sk_state = SS_CONNECTING; + vnew = vsock_sk(new); + hvs_new = vnew->trans; + hvs_new->chan = chan; + } else { + hvs = vsock_sk(sk)->trans; + hvs->chan = chan; + } + + set_channel_read_mode(chan, HV_CALL_DIRECT); + ret = vmbus_open(chan, RINGBUFFER_HVS_SND_SIZE, + RINGBUFFER_HVS_RCV_SIZE, NULL, 0, + hvs_channel_cb, conn_from_host ? new : sk); + if (ret != 0) { + if (conn_from_host) { + hvs_new->chan = NULL; + sock_put(new); + } else { + hvs->chan = NULL; + } + goto out; + } + + set_per_channel_state(chan, conn_from_host ? new : sk); + vmbus_set_chn_rescind_callback(chan, hvs_close_connection); + + if (conn_from_host) { + new->sk_state = SS_CONNECTED; + sk->sk_ack_backlog++; + + hvs_addr_init(&vnew->local_addr, if_type); + hvs_remote_addr_init(&vnew->remote_addr, &vnew->local_addr); + + hvs_new->vm_srv_id = *if_type; + hvs_new->host_srv_id = *if_instance; + + vsock_insert_connected(vnew); + + lock_sock(sk); + vsock_enqueue_accept(sk, new); + release_sock(sk); + } else { + sk->sk_state = SS_CONNECTED; + sk->sk_socket->state = SS_CONNECTED; + + vsock_insert_connected(vsock_sk(sk)); + } + + sk->sk_state_change(sk); + +out: + /* Release refcnt obtained when we called vsock_find_bound_socket() */ + sock_put(sk); +} + +static u32 hvs_get_local_cid(void) +{ + return VMADDR_CID_ANY; +} + +static int hvs_sock_init(struct vsock_sock *vsk, struct vsock_sock *psk) +{ + struct hvsock *hvs; + + hvs = kzalloc(sizeof(*hvs), GFP_KERNEL); + if (!hvs) + return -ENOMEM; + + vsk->trans = hvs; + hvs->vsk = vsk; + + return 0; +} + +static int hvs_connect(struct vsock_sock *vsk) +{ + union hvs_service_id vm, host; + struct hvsock *h = vsk->trans; + + vm.srv_id = srv_id_template; + vm.svm_port = vsk->local_addr.svm_port; + h->vm_srv_id = vm.srv_id; + + host.srv_id = srv_id_template; + host.svm_port = vsk->remote_addr.svm_port; + h->host_srv_id = host.srv_id; + + return vmbus_send_tl_connect_request(&h->vm_srv_id, &h->host_srv_id); +} + +static int hvs_shutdown(struct vsock_sock *vsk, int mode) +{ + struct sock *sk = sk_vsock(vsk); + struct vmpipe_proto_header hdr; + struct hvs_send_buf *send_buf; + struct hvsock *hvs; + + if (!(mode & SEND_SHUTDOWN)) + return 0; + + lock_sock(sk); + + hvs = vsk->trans; + if (hvs->fin_sent) + goto out; + + send_buf = (struct hvs_send_buf *)&hdr; + + /* It can't fail: see hvs_channel_writable_bytes(). */ + (void)hvs_send_data(hvs->chan, send_buf, 0); + + hvs->fin_sent = true; +out: + release_sock(sk); + return 0; +} + +static void hvs_release(struct vsock_sock *vsk) +{ + struct hvsock *hvs = vsk->trans; + struct vmbus_channel *chan = hvs->chan; + + if (chan) + hvs_shutdown(vsk, RCV_SHUTDOWN | SEND_SHUTDOWN); + + vsock_remove_sock(vsk); +} + +static void hvs_destruct(struct vsock_sock *vsk) +{ + struct hvsock *hvs = vsk->trans; + struct vmbus_channel *chan = hvs->chan; + + if (chan) + vmbus_hvsock_device_unregister(chan); + + kfree(hvs); +} + +static int hvs_dgram_bind(struct vsock_sock *vsk, struct sockaddr_vm *addr) +{ + return -EOPNOTSUPP; +} + +static int hvs_dgram_dequeue(struct vsock_sock *vsk, struct msghdr *msg, + size_t len, int flags) +{ + return -EOPNOTSUPP; +} + +static int hvs_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote, struct msghdr *msg, + size_t dgram_len) +{ + return -EOPNOTSUPP; +} + +static bool hvs_dgram_allow(u32 cid, u32 port) +{ + return false; +} + +static int hvs_update_recv_data(struct hvsock *hvs) +{ + struct hvs_recv_buf *recv_buf; + u32 payload_len; + + recv_buf = (struct hvs_recv_buf *)(hvs->recv_desc + 1); + payload_len = recv_buf->hdr.data_size; + + if (payload_len > HVS_MTU_SIZE) + return -EIO; + + if (payload_len == 0) + hvs->vsk->peer_shutdown |= SEND_SHUTDOWN; + + hvs->recv_data_len = payload_len; + hvs->recv_data_off = 0; + + return 0; +} + +static ssize_t hvs_stream_dequeue(struct vsock_sock *vsk, struct msghdr *msg, + size_t len, int flags) +{ + struct hvsock *hvs = vsk->trans; + bool need_refill = !hvs->recv_desc; + struct hvs_recv_buf *recv_buf; + u32 to_read; + int ret; + + if (flags & MSG_PEEK) + return -EOPNOTSUPP; + + if (need_refill) { + hvs->recv_desc = hv_pkt_iter_first(hvs->chan); + ret = hvs_update_recv_data(hvs); + if (ret) + return ret; + } + + recv_buf = (struct hvs_recv_buf *)(hvs->recv_desc + 1); + to_read = min_t(u32, len, hvs->recv_data_len); + ret = memcpy_to_msg(msg, recv_buf->data + hvs->recv_data_off, to_read); + if (ret != 0) + return ret; + + hvs->recv_data_len -= to_read; + if (hvs->recv_data_len == 0) { + hvs->recv_desc = hv_pkt_iter_next(hvs->chan, hvs->recv_desc); + if (hvs->recv_desc) { + ret = hvs_update_recv_data(hvs); + if (ret) + return ret; + } + } else { + hvs->recv_data_off += to_read; + } + + return to_read; +} + +static ssize_t hvs_stream_enqueue(struct vsock_sock *vsk, struct msghdr *msg, + size_t len) +{ + struct hvsock *hvs = vsk->trans; + struct vmbus_channel *chan = hvs->chan; + struct hvs_send_buf *send_buf; + ssize_t to_write, max_writable, ret; + + BUILD_BUG_ON(sizeof(*send_buf) != PAGE_SIZE_4K); + + send_buf = kmalloc(sizeof(*send_buf), GFP_KERNEL); + if (!send_buf) + return -ENOMEM; + + max_writable = hvs_channel_writable_bytes(chan); + to_write = min_t(ssize_t, len, max_writable); + to_write = min_t(ssize_t, to_write, HVS_SEND_BUF_SIZE); + + ret = memcpy_from_msg(send_buf->data, msg, to_write); + if (ret < 0) + goto out; + + ret = hvs_send_data(hvs->chan, send_buf, to_write); + if (ret < 0) + goto out; + + ret = to_write; +out: + kfree(send_buf); + return ret; +} + +static s64 hvs_stream_has_data(struct vsock_sock *vsk) +{ + struct hvsock *hvs = vsk->trans; + s64 ret; + + if (hvs->recv_data_len > 0) + return 1; + + switch (hvs_channel_readable_payload(hvs->chan)) { + case 1: + ret = 1; + break; + case 0: + vsk->peer_shutdown |= SEND_SHUTDOWN; + ret = 0; + break; + default: /* -1 */ + ret = 0; + break; + } + + return ret; +} + +static s64 hvs_stream_has_space(struct vsock_sock *vsk) +{ + struct hvsock *hvs = vsk->trans; + struct vmbus_channel *chan = hvs->chan; + s64 ret; + + ret = hvs_channel_writable_bytes(chan); + if (ret > 0) { + hvs_clear_channel_pending_send_size(chan); + } else { + /* See hvs_channel_cb() */ + hvs_set_channel_pending_send_size(chan); + + /* Re-check the writable bytes to avoid race */ + ret = hvs_channel_writable_bytes(chan); + if (ret > 0) + hvs_clear_channel_pending_send_size(chan); + } + + return ret; +} + +static u64 hvs_stream_rcvhiwat(struct vsock_sock *vsk) +{ + return HVS_MTU_SIZE + 1; +} + +static bool hvs_stream_is_active(struct vsock_sock *vsk) +{ + struct hvsock *hvs = vsk->trans; + + return hvs->chan != NULL; +} + +static bool hvs_stream_allow(u32 cid, u32 port) +{ + /* The host's port range [MIN_HOST_EPHEMERAL_PORT, 0xFFFFFFFF) is + * reserved as ephemeral ports, which are used as the host's ports + * when the host initiates connections. + * + * Perform this check in the guest so an immediate error is produced + * instead of a timeout. + */ + if (port > MAX_HOST_LISTEN_PORT) + return false; + + if (cid == VMADDR_CID_HOST) + return true; + + return false; +} + +static +int hvs_notify_poll_in(struct vsock_sock *vsk, size_t target, bool *readable) +{ + struct hvsock *hvs = vsk->trans; + + *readable = hvs_channel_readable(hvs->chan); + return 0; +} + +static +int hvs_notify_poll_out(struct vsock_sock *vsk, size_t target, bool *writable) +{ + *writable = hvs_stream_has_space(vsk) > 0; + + return 0; +} + +static +int hvs_notify_recv_init(struct vsock_sock *vsk, size_t target, + struct vsock_transport_recv_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_recv_pre_block(struct vsock_sock *vsk, size_t target, + struct vsock_transport_recv_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_recv_pre_dequeue(struct vsock_sock *vsk, size_t target, + struct vsock_transport_recv_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_recv_post_dequeue(struct vsock_sock *vsk, size_t target, + ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *d) +{ + return 0; +} + +static +int hvs_notify_send_post_enqueue(struct vsock_sock *vsk, ssize_t written, + struct vsock_transport_send_notify_data *d) +{ + return 0; +} + +static void hvs_set_buffer_size(struct vsock_sock *vsk, u64 val) +{ + /* Ignored. */ +} + +static void hvs_set_min_buffer_size(struct vsock_sock *vsk, u64 val) +{ + /* Ignored. */ +} + +static void hvs_set_max_buffer_size(struct vsock_sock *vsk, u64 val) +{ + /* Ignored. */ +} + +static u64 hvs_get_buffer_size(struct vsock_sock *vsk) +{ + return -ENOPROTOOPT; +} + +static u64 hvs_get_min_buffer_size(struct vsock_sock *vsk) +{ + return -ENOPROTOOPT; +} + +static u64 hvs_get_max_buffer_size(struct vsock_sock *vsk) +{ + return -ENOPROTOOPT; +} + +static struct vsock_transport hvs_transport = { + .get_local_cid = hvs_get_local_cid, + + .init = hvs_sock_init, + .destruct = hvs_destruct, + .release = hvs_release, + .connect = hvs_connect, + .shutdown = hvs_shutdown, + + .dgram_bind = hvs_dgram_bind, + .dgram_dequeue = hvs_dgram_dequeue, + .dgram_enqueue = hvs_dgram_enqueue, + .dgram_allow = hvs_dgram_allow, + + .stream_dequeue = hvs_stream_dequeue, + .stream_enqueue = hvs_stream_enqueue, + .stream_has_data = hvs_stream_has_data, + .stream_has_space = hvs_stream_has_space, + .stream_rcvhiwat = hvs_stream_rcvhiwat, + .stream_is_active = hvs_stream_is_active, + .stream_allow = hvs_stream_allow, + + .notify_poll_in = hvs_notify_poll_in, + .notify_poll_out = hvs_notify_poll_out, + .notify_recv_init = hvs_notify_recv_init, + .notify_recv_pre_block = hvs_notify_recv_pre_block, + .notify_recv_pre_dequeue = hvs_notify_recv_pre_dequeue, + .notify_recv_post_dequeue = hvs_notify_recv_post_dequeue, + .notify_send_init = hvs_notify_send_init, + .notify_send_pre_block = hvs_notify_send_pre_block, + .notify_send_pre_enqueue = hvs_notify_send_pre_enqueue, + .notify_send_post_enqueue = hvs_notify_send_post_enqueue, + + .set_buffer_size = hvs_set_buffer_size, + .set_min_buffer_size = hvs_set_min_buffer_size, + .set_max_buffer_size = hvs_set_max_buffer_size, + .get_buffer_size = hvs_get_buffer_size, + .get_min_buffer_size = hvs_get_min_buffer_size, + .get_max_buffer_size = hvs_get_max_buffer_size, +}; + +static int hvs_probe(struct hv_device *hdev, + const struct hv_vmbus_device_id *dev_id) +{ + struct vmbus_channel *chan = hdev->channel; + + hvs_open_connection(chan); + + /* Always return success to suppress the unnecessary error message + * in vmbus_probe(): on error the host will rescind the device in + * 30 seconds and we can do cleanup at that time in + * vmbus_onoffer_rescind(). + */ + return 0; +} + +static int hvs_remove(struct hv_device *hdev) +{ + struct vmbus_channel *chan = hdev->channel; + + vmbus_close(chan); + + return 0; +} + +/* This isn't really used. See vmbus_match() and vmbus_probe() */ +static const struct hv_vmbus_device_id id_table[] = { + {}, +}; + +static struct hv_driver hvs_drv = { + .name = "hv_sock", + .hvsock = true, + .id_table = id_table, + .probe = hvs_probe, + .remove = hvs_remove, +}; + +static int __init hvs_init(void) +{ + int ret; + + if (vmbus_proto_version < VERSION_WIN10) + return -ENODEV; + + ret = vmbus_driver_register(&hvs_drv); + if (ret != 0) + return ret; + + ret = vsock_core_init(&hvs_transport); + if (ret) { + vmbus_driver_unregister(&hvs_drv); + return ret; + } + + return 0; +} + +static void __exit hvs_exit(void) +{ + vsock_core_exit(); + vmbus_driver_unregister(&hvs_drv); +} + +module_init(hvs_init); +module_exit(hvs_exit); + +MODULE_DESCRIPTION("Hyper-V Sockets"); +MODULE_VERSION("1.0.0"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_NETPROTO(PF_VSOCK); diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 5a1a98df3499..ac095936552d 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -74,7 +74,7 @@ DEFINE_RWLOCK(x25_list_lock); static const struct proto_ops x25_proto_ops; -static struct x25_address null_x25_address = {" "}; +static const struct x25_address null_x25_address = {" "}; #ifdef CONFIG_COMPAT struct compat_x25_subscrip_struct { diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 5f7e8bfa0c2d..acf00104ef31 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -63,7 +63,7 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, xfrm_address_t *daddr; if (!x->type_offload) - return 0; + return -EINVAL; /* We don't yet support UDP encapsulation, TFC padding and ESN. */ if (x->encap || x->tfcpad || (x->props.flags & XFRM_STATE_ESN)) @@ -79,7 +79,8 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, daddr = &x->props.saddr; } - dst = __xfrm_dst_lookup(net, 0, 0, saddr, daddr, x->props.family); + dst = __xfrm_dst_lookup(net, 0, 0, saddr, daddr, + x->props.family, x->props.output_mark); if (IS_ERR(dst)) return 0; @@ -153,6 +154,7 @@ static int xfrm_dev_register(struct net_device *dev) static int xfrm_dev_unregister(struct net_device *dev) { + xfrm_policy_cache_flush(); return NOTIFY_DONE; } @@ -175,8 +177,7 @@ static int xfrm_dev_down(struct net_device *dev) if (dev->features & NETIF_F_HW_ESP) xfrm_dev_state_flush(dev_net(dev), dev, true); - xfrm_garbage_collect(dev_net(dev)); - + xfrm_policy_cache_flush(); return NOTIFY_DONE; } diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 923205e279f7..2515cd2bc5db 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -247,6 +247,11 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop; } + if (xo->status & CRYPTO_INVALID_PROTOCOL) { + XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEPROTOERROR); + goto drop; + } + XFRM_INC_STATS(net, LINUX_MIB_XFRMINBUFFERERROR); goto drop; } @@ -424,6 +429,7 @@ resume: nf_reset(skb); if (decaps) { + skb->sp->olen = 0; skb_dst_drop(skb); gro_cells_receive(&gro_cells, skb); return 0; @@ -434,6 +440,7 @@ resume: err = x->inner_mode->afinfo->transport_finish(skb, xfrm_gro || async); if (xfrm_gro) { + skb->sp->olen = 0; skb_dst_drop(skb); gro_cells_receive(&gro_cells, skb); return err; diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index 8c0b6722aaa8..31a2e6d34dba 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -66,6 +66,9 @@ static int xfrm_output_one(struct sk_buff *skb, int err) goto error_nolock; } + if (x->props.output_mark) + skb->mark = x->props.output_mark; + err = x->outer_mode->output(x, skb); if (err) { XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTSTATEMODEERROR); diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index ff61d8557929..f06253969972 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -24,6 +24,7 @@ #include <linux/netfilter.h> #include <linux/module.h> #include <linux/cache.h> +#include <linux/cpu.h> #include <linux/audit.h> #include <net/dst.h> #include <net/flow.h> @@ -44,6 +45,8 @@ struct xfrm_flo { u8 flags; }; +static DEFINE_PER_CPU(struct xfrm_dst *, xfrm_last_dst); +static struct work_struct *xfrm_pcpu_work __read_mostly; static DEFINE_SPINLOCK(xfrm_policy_afinfo_lock); static struct xfrm_policy_afinfo const __rcu *xfrm_policy_afinfo[AF_INET6 + 1] __read_mostly; @@ -119,7 +122,7 @@ static const struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short fa struct dst_entry *__xfrm_dst_lookup(struct net *net, int tos, int oif, const xfrm_address_t *saddr, const xfrm_address_t *daddr, - int family) + int family, u32 mark) { const struct xfrm_policy_afinfo *afinfo; struct dst_entry *dst; @@ -128,7 +131,7 @@ struct dst_entry *__xfrm_dst_lookup(struct net *net, int tos, int oif, if (unlikely(afinfo == NULL)) return ERR_PTR(-EAFNOSUPPORT); - dst = afinfo->dst_lookup(net, tos, oif, saddr, daddr); + dst = afinfo->dst_lookup(net, tos, oif, saddr, daddr, mark); rcu_read_unlock(); @@ -140,7 +143,7 @@ static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos, int oif, xfrm_address_t *prev_saddr, xfrm_address_t *prev_daddr, - int family) + int family, u32 mark) { struct net *net = xs_net(x); xfrm_address_t *saddr = &x->props.saddr; @@ -156,7 +159,7 @@ static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, daddr = x->coaddr; } - dst = __xfrm_dst_lookup(net, tos, oif, saddr, daddr, family); + dst = __xfrm_dst_lookup(net, tos, oif, saddr, daddr, family, mark); if (!IS_ERR(dst)) { if (prev_saddr != saddr) @@ -246,36 +249,6 @@ expired: xfrm_pol_put(xp); } -static struct flow_cache_object *xfrm_policy_flo_get(struct flow_cache_object *flo) -{ - struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo); - - if (unlikely(pol->walk.dead)) - flo = NULL; - else - xfrm_pol_hold(pol); - - return flo; -} - -static int xfrm_policy_flo_check(struct flow_cache_object *flo) -{ - struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo); - - return !pol->walk.dead; -} - -static void xfrm_policy_flo_delete(struct flow_cache_object *flo) -{ - xfrm_pol_put(container_of(flo, struct xfrm_policy, flo)); -} - -static const struct flow_cache_ops xfrm_policy_fc_ops = { - .get = xfrm_policy_flo_get, - .check = xfrm_policy_flo_check, - .delete = xfrm_policy_flo_delete, -}; - /* Allocate xfrm_policy. Not used here, it is supposed to be used by pfkeyv2 * SPD calls. */ @@ -298,7 +271,6 @@ struct xfrm_policy *xfrm_policy_alloc(struct net *net, gfp_t gfp) (unsigned long)policy); setup_timer(&policy->polq.hold_timer, xfrm_policy_queue_process, (unsigned long)policy); - policy->flo.ops = &xfrm_policy_fc_ops; } return policy; } @@ -798,7 +770,6 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl) else hlist_add_head(&policy->bydst, chain); __xfrm_policy_link(policy, dir); - atomic_inc(&net->xfrm.flow_cache_genid); /* After previous checking, family can either be AF_INET or AF_INET6 */ if (policy->family == AF_INET) @@ -1004,6 +975,8 @@ int xfrm_policy_flush(struct net *net, u8 type, bool task_valid) } if (!cnt) err = -ESRCH; + else + xfrm_policy_cache_flush(); out: spin_unlock_bh(&net->xfrm.xfrm_policy_lock); return err; @@ -1175,7 +1148,7 @@ fail: } static struct xfrm_policy * -__xfrm_policy_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir) +xfrm_policy_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir) { #ifdef CONFIG_XFRM_SUB_POLICY struct xfrm_policy *pol; @@ -1187,61 +1160,6 @@ __xfrm_policy_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir return xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family, dir); } -static int flow_to_policy_dir(int dir) -{ - if (XFRM_POLICY_IN == FLOW_DIR_IN && - XFRM_POLICY_OUT == FLOW_DIR_OUT && - XFRM_POLICY_FWD == FLOW_DIR_FWD) - return dir; - - switch (dir) { - default: - case FLOW_DIR_IN: - return XFRM_POLICY_IN; - case FLOW_DIR_OUT: - return XFRM_POLICY_OUT; - case FLOW_DIR_FWD: - return XFRM_POLICY_FWD; - } -} - -static struct flow_cache_object * -xfrm_policy_lookup(struct net *net, const struct flowi *fl, u16 family, - u8 dir, struct flow_cache_object *old_obj, void *ctx) -{ - struct xfrm_policy *pol; - - if (old_obj) - xfrm_pol_put(container_of(old_obj, struct xfrm_policy, flo)); - - pol = __xfrm_policy_lookup(net, fl, family, flow_to_policy_dir(dir)); - if (IS_ERR_OR_NULL(pol)) - return ERR_CAST(pol); - - /* Resolver returns two references: - * one for cache and one for caller of flow_cache_lookup() */ - xfrm_pol_hold(pol); - - return &pol->flo; -} - -static inline int policy_to_flow_dir(int dir) -{ - if (XFRM_POLICY_IN == FLOW_DIR_IN && - XFRM_POLICY_OUT == FLOW_DIR_OUT && - XFRM_POLICY_FWD == FLOW_DIR_FWD) - return dir; - switch (dir) { - default: - case XFRM_POLICY_IN: - return FLOW_DIR_IN; - case XFRM_POLICY_OUT: - return FLOW_DIR_OUT; - case XFRM_POLICY_FWD: - return FLOW_DIR_FWD; - } -} - static struct xfrm_policy *xfrm_sk_policy_lookup(const struct sock *sk, int dir, const struct flowi *fl, u16 family) { @@ -1261,7 +1179,7 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(const struct sock *sk, int dir, } err = security_xfrm_policy_lookup(pol->security, fl->flowi_secid, - policy_to_flow_dir(dir)); + dir); if (!err) { if (!xfrm_pol_hold_rcu(pol)) goto again; @@ -1422,14 +1340,14 @@ int __xfrm_sk_clone_policy(struct sock *sk, const struct sock *osk) static int xfrm_get_saddr(struct net *net, int oif, xfrm_address_t *local, - xfrm_address_t *remote, unsigned short family) + xfrm_address_t *remote, unsigned short family, u32 mark) { int err; const struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family); if (unlikely(afinfo == NULL)) return -EINVAL; - err = afinfo->get_saddr(net, oif, local, remote); + err = afinfo->get_saddr(net, oif, local, remote, mark); rcu_read_unlock(); return err; } @@ -1460,7 +1378,7 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl, if (xfrm_addr_any(local, tmpl->encap_family)) { error = xfrm_get_saddr(net, fl->flowi_oif, &tmp, remote, - tmpl->encap_family); + tmpl->encap_family, 0); if (error) goto fail; local = &tmp; @@ -1545,58 +1463,6 @@ static int xfrm_get_tos(const struct flowi *fl, int family) return tos; } -static struct flow_cache_object *xfrm_bundle_flo_get(struct flow_cache_object *flo) -{ - struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo); - struct dst_entry *dst = &xdst->u.dst; - - if (xdst->route == NULL) { - /* Dummy bundle - if it has xfrms we were not - * able to build bundle as template resolution failed. - * It means we need to try again resolving. */ - if (xdst->num_xfrms > 0) - return NULL; - } else if (dst->flags & DST_XFRM_QUEUE) { - return NULL; - } else { - /* Real bundle */ - if (stale_bundle(dst)) - return NULL; - } - - dst_hold(dst); - return flo; -} - -static int xfrm_bundle_flo_check(struct flow_cache_object *flo) -{ - struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo); - struct dst_entry *dst = &xdst->u.dst; - - if (!xdst->route) - return 0; - if (stale_bundle(dst)) - return 0; - - return 1; -} - -static void xfrm_bundle_flo_delete(struct flow_cache_object *flo) -{ - struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo); - struct dst_entry *dst = &xdst->u.dst; - - /* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */ - dst->obsolete = DST_OBSOLETE_DEAD; - dst_release_immediate(dst); -} - -static const struct flow_cache_ops xfrm_bundle_fc_ops = { - .get = xfrm_bundle_flo_get, - .check = xfrm_bundle_flo_check, - .delete = xfrm_bundle_flo_delete, -}; - static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family) { const struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family); @@ -1624,7 +1490,6 @@ static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family) struct dst_entry *dst = &xdst->u.dst; memset(dst + 1, 0, sizeof(*xdst) - sizeof(*dst)); - xdst->flo.ops = &xfrm_bundle_fc_ops; } else xdst = ERR_PTR(-ENOBUFS); @@ -1733,7 +1598,8 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy, if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) { family = xfrm[i]->props.family; dst = xfrm_dst_lookup(xfrm[i], tos, fl->flowi_oif, - &saddr, &daddr, family); + &saddr, &daddr, family, + xfrm[i]->props.output_mark); err = PTR_ERR(dst); if (IS_ERR(dst)) goto put_states; @@ -1840,6 +1706,102 @@ static int xfrm_expand_policies(const struct flowi *fl, u16 family, } +static void xfrm_last_dst_update(struct xfrm_dst *xdst, struct xfrm_dst *old) +{ + this_cpu_write(xfrm_last_dst, xdst); + if (old) + dst_release(&old->u.dst); +} + +static void __xfrm_pcpu_work_fn(void) +{ + struct xfrm_dst *old; + + old = this_cpu_read(xfrm_last_dst); + if (old && !xfrm_bundle_ok(old)) + xfrm_last_dst_update(NULL, old); +} + +static void xfrm_pcpu_work_fn(struct work_struct *work) +{ + local_bh_disable(); + rcu_read_lock(); + __xfrm_pcpu_work_fn(); + rcu_read_unlock(); + local_bh_enable(); +} + +void xfrm_policy_cache_flush(void) +{ + struct xfrm_dst *old; + bool found = 0; + int cpu; + + local_bh_disable(); + rcu_read_lock(); + for_each_possible_cpu(cpu) { + old = per_cpu(xfrm_last_dst, cpu); + if (old && !xfrm_bundle_ok(old)) { + if (smp_processor_id() == cpu) { + __xfrm_pcpu_work_fn(); + continue; + } + found = true; + break; + } + } + + rcu_read_unlock(); + local_bh_enable(); + + if (!found) + return; + + get_online_cpus(); + + for_each_possible_cpu(cpu) { + bool bundle_release; + + rcu_read_lock(); + old = per_cpu(xfrm_last_dst, cpu); + bundle_release = old && !xfrm_bundle_ok(old); + rcu_read_unlock(); + + if (!bundle_release) + continue; + + if (cpu_online(cpu)) { + schedule_work_on(cpu, &xfrm_pcpu_work[cpu]); + continue; + } + + rcu_read_lock(); + old = per_cpu(xfrm_last_dst, cpu); + if (old && !xfrm_bundle_ok(old)) { + per_cpu(xfrm_last_dst, cpu) = NULL; + dst_release(&old->u.dst); + } + rcu_read_unlock(); + } + + put_online_cpus(); +} + +static bool xfrm_pol_dead(struct xfrm_dst *xdst) +{ + unsigned int num_pols = xdst->num_pols; + unsigned int pol_dead = 0, i; + + for (i = 0; i < num_pols; i++) + pol_dead |= xdst->pols[i]->walk.dead; + + /* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */ + if (pol_dead) + xdst->u.dst.obsolete = DST_OBSOLETE_DEAD; + + return pol_dead; +} + static struct xfrm_dst * xfrm_resolve_and_create_bundle(struct xfrm_policy **pols, int num_pols, const struct flowi *fl, u16 family, @@ -1847,10 +1809,23 @@ xfrm_resolve_and_create_bundle(struct xfrm_policy **pols, int num_pols, { struct net *net = xp_net(pols[0]); struct xfrm_state *xfrm[XFRM_MAX_DEPTH]; + struct xfrm_dst *xdst, *old; struct dst_entry *dst; - struct xfrm_dst *xdst; int err; + xdst = this_cpu_read(xfrm_last_dst); + if (xdst && + xdst->u.dst.dev == dst_orig->dev && + xdst->num_pols == num_pols && + !xfrm_pol_dead(xdst) && + memcmp(xdst->pols, pols, + sizeof(struct xfrm_policy *) * num_pols) == 0 && + xfrm_bundle_ok(xdst)) { + dst_hold(&xdst->u.dst); + return xdst; + } + + old = xdst; /* Try to instantiate a bundle */ err = xfrm_tmpl_resolve(pols, num_pols, fl, xfrm, family); if (err <= 0) { @@ -1871,6 +1846,9 @@ xfrm_resolve_and_create_bundle(struct xfrm_policy **pols, int num_pols, memcpy(xdst->pols, pols, sizeof(struct xfrm_policy *) * num_pols); xdst->policy_genid = atomic_read(&pols[0]->genid); + atomic_set(&xdst->u.dst.__refcnt, 2); + xfrm_last_dst_update(xdst, old); + return xdst; } @@ -2051,86 +2029,39 @@ free_dst: goto out; } -static struct flow_cache_object * -xfrm_bundle_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir, - struct flow_cache_object *oldflo, void *ctx) +static struct xfrm_dst * +xfrm_bundle_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir, struct xfrm_flo *xflo) { - struct xfrm_flo *xflo = (struct xfrm_flo *)ctx; struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX]; - struct xfrm_dst *xdst, *new_xdst; - int num_pols = 0, num_xfrms = 0, i, err, pol_dead; - - /* Check if the policies from old bundle are usable */ - xdst = NULL; - if (oldflo) { - xdst = container_of(oldflo, struct xfrm_dst, flo); - num_pols = xdst->num_pols; - num_xfrms = xdst->num_xfrms; - pol_dead = 0; - for (i = 0; i < num_pols; i++) { - pols[i] = xdst->pols[i]; - pol_dead |= pols[i]->walk.dead; - } - if (pol_dead) { - /* Mark DST_OBSOLETE_DEAD to fail the next - * xfrm_dst_check() - */ - xdst->u.dst.obsolete = DST_OBSOLETE_DEAD; - dst_release_immediate(&xdst->u.dst); - xdst = NULL; - num_pols = 0; - num_xfrms = 0; - oldflo = NULL; - } - } + int num_pols = 0, num_xfrms = 0, err; + struct xfrm_dst *xdst; /* Resolve policies to use if we couldn't get them from * previous cache entry */ - if (xdst == NULL) { - num_pols = 1; - pols[0] = __xfrm_policy_lookup(net, fl, family, - flow_to_policy_dir(dir)); - err = xfrm_expand_policies(fl, family, pols, + num_pols = 1; + pols[0] = xfrm_policy_lookup(net, fl, family, dir); + err = xfrm_expand_policies(fl, family, pols, &num_pols, &num_xfrms); - if (err < 0) - goto inc_error; - if (num_pols == 0) - return NULL; - if (num_xfrms <= 0) - goto make_dummy_bundle; - } + if (err < 0) + goto inc_error; + if (num_pols == 0) + return NULL; + if (num_xfrms <= 0) + goto make_dummy_bundle; - new_xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family, + xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family, xflo->dst_orig); - if (IS_ERR(new_xdst)) { - err = PTR_ERR(new_xdst); + if (IS_ERR(xdst)) { + err = PTR_ERR(xdst); if (err != -EAGAIN) goto error; - if (oldflo == NULL) - goto make_dummy_bundle; - dst_hold(&xdst->u.dst); - return oldflo; - } else if (new_xdst == NULL) { + goto make_dummy_bundle; + } else if (xdst == NULL) { num_xfrms = 0; - if (oldflo == NULL) - goto make_dummy_bundle; - xdst->num_xfrms = 0; - dst_hold(&xdst->u.dst); - return oldflo; + goto make_dummy_bundle; } - /* Kill the previous bundle */ - if (xdst) { - /* The policies were stolen for newly generated bundle */ - xdst->num_pols = 0; - /* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */ - xdst->u.dst.obsolete = DST_OBSOLETE_DEAD; - dst_release_immediate(&xdst->u.dst); - } - - /* We do need to return one reference for original caller */ - dst_hold(&new_xdst->u.dst); - return &new_xdst->flo; + return xdst; make_dummy_bundle: /* We found policies, but there's no bundles to instantiate: @@ -2146,17 +2077,12 @@ make_dummy_bundle: memcpy(xdst->pols, pols, sizeof(struct xfrm_policy *) * num_pols); dst_hold(&xdst->u.dst); - return &xdst->flo; + return xdst; inc_error: XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR); error: - if (xdst != NULL) { - /* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */ - xdst->u.dst.obsolete = DST_OBSOLETE_DEAD; - dst_release_immediate(&xdst->u.dst); - } else - xfrm_pols_put(pols, num_pols); + xfrm_pols_put(pols, num_pols); return ERR_PTR(err); } @@ -2187,11 +2113,10 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig, const struct sock *sk, int flags) { struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX]; - struct flow_cache_object *flo; struct xfrm_dst *xdst; struct dst_entry *dst, *route; u16 family = dst_orig->ops->family; - u8 dir = policy_to_flow_dir(XFRM_POLICY_OUT); + u8 dir = XFRM_POLICY_OUT; int i, err, num_pols, num_xfrms = 0, drop_pols = 0; dst = NULL; @@ -2226,7 +2151,6 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig, goto no_transform; } - dst_hold(&xdst->u.dst); route = xdst->route; } } @@ -2242,15 +2166,13 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig, !net->xfrm.policy_count[XFRM_POLICY_OUT]) goto nopol; - flo = flow_cache_lookup(net, fl, family, dir, - xfrm_bundle_lookup, &xflo); - if (flo == NULL) + xdst = xfrm_bundle_lookup(net, fl, family, dir, &xflo); + if (xdst == NULL) goto nopol; - if (IS_ERR(flo)) { - err = PTR_ERR(flo); + if (IS_ERR(xdst)) { + err = PTR_ERR(xdst); goto dropdst; } - xdst = container_of(flo, struct xfrm_dst, flo); num_pols = xdst->num_pols; num_xfrms = xdst->num_xfrms; @@ -2449,12 +2371,10 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, int pi; int reverse; struct flowi fl; - u8 fl_dir; int xerr_idx = -1; reverse = dir & ~XFRM_POLICY_MASK; dir &= XFRM_POLICY_MASK; - fl_dir = policy_to_flow_dir(dir); if (__xfrm_decode_session(skb, &fl, family, reverse) < 0) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR); @@ -2486,16 +2406,8 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, } } - if (!pol) { - struct flow_cache_object *flo; - - flo = flow_cache_lookup(net, &fl, family, fl_dir, - xfrm_policy_lookup, NULL); - if (IS_ERR_OR_NULL(flo)) - pol = ERR_CAST(flo); - else - pol = container_of(flo, struct xfrm_policy, flo); - } + if (!pol) + pol = xfrm_policy_lookup(net, &fl, family, dir); if (IS_ERR(pol)) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR); @@ -2641,11 +2553,9 @@ static struct dst_entry *xfrm_dst_check(struct dst_entry *dst, u32 cookie) * notice. That's what we are validating here via the * stale_bundle() check. * - * When an xdst is removed from flow cache, DST_OBSOLETE_DEAD will - * be marked on it. * When a dst is removed from the fib tree, DST_OBSOLETE_DEAD will * be marked on it. - * Both will force stable_bundle() to fail on any xdst bundle with + * This will force stale_bundle() to fail on any xdst bundle with * this dst linked in it. */ if (dst->obsolete < 0 && !stale_bundle(dst)) @@ -2685,18 +2595,6 @@ static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst) return dst; } -void xfrm_garbage_collect(struct net *net) -{ - flow_cache_flush(net); -} -EXPORT_SYMBOL(xfrm_garbage_collect); - -void xfrm_garbage_collect_deferred(struct net *net) -{ - flow_cache_flush_deferred(net); -} -EXPORT_SYMBOL(xfrm_garbage_collect_deferred); - static void xfrm_init_pmtu(struct dst_entry *dst) { do { @@ -3034,14 +2932,9 @@ static int __net_init xfrm_net_init(struct net *net) rv = xfrm_sysctl_init(net); if (rv < 0) goto out_sysctl; - rv = flow_cache_init(net); - if (rv < 0) - goto out; return 0; -out: - xfrm_sysctl_fini(net); out_sysctl: xfrm_policy_fini(net); out_policy: @@ -3054,7 +2947,6 @@ out_statistics: static void __net_exit xfrm_net_exit(struct net *net) { - flow_cache_fini(net); xfrm_sysctl_fini(net); xfrm_policy_fini(net); xfrm_state_fini(net); @@ -3068,7 +2960,15 @@ static struct pernet_operations __net_initdata xfrm_net_ops = { void __init xfrm_init(void) { - flow_cache_hp_init(); + int i; + + xfrm_pcpu_work = kmalloc_array(NR_CPUS, sizeof(*xfrm_pcpu_work), + GFP_KERNEL); + BUG_ON(!xfrm_pcpu_work); + + for (i = 0; i < NR_CPUS; i++) + INIT_WORK(&xfrm_pcpu_work[i], xfrm_pcpu_work_fn); + register_pernet_subsys(&xfrm_net_ops); seqcount_init(&xfrm_policy_hash_generation); xfrm_input_init(); @@ -3308,9 +3208,15 @@ int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, struct xfrm_state *x_new[XFRM_MAX_DEPTH]; struct xfrm_migrate *mp; + /* Stage 0 - sanity checks */ if ((err = xfrm_migrate_check(m, num_migrate)) < 0) goto out; + if (dir >= XFRM_POLICY_MAX) { + err = -EINVAL; + goto out; + } + /* Stage 1 - find policy */ if ((pol = xfrm_migrate_policy_find(sel, dir, type, net)) == NULL) { err = -ENOENT; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 6c0956d10db6..0dab1cd79ce4 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -296,12 +296,14 @@ int xfrm_unregister_type_offload(const struct xfrm_type_offload *type, } EXPORT_SYMBOL(xfrm_unregister_type_offload); -static const struct xfrm_type_offload *xfrm_get_type_offload(u8 proto, unsigned short family) +static const struct xfrm_type_offload * +xfrm_get_type_offload(u8 proto, unsigned short family, bool try_load) { struct xfrm_state_afinfo *afinfo; const struct xfrm_type_offload **typemap; const struct xfrm_type_offload *type; +retry: afinfo = xfrm_state_get_afinfo(family); if (unlikely(afinfo == NULL)) return NULL; @@ -311,6 +313,12 @@ static const struct xfrm_type_offload *xfrm_get_type_offload(u8 proto, unsigned if ((type && !try_module_get(type->owner))) type = NULL; + if (!type && try_load) { + request_module("xfrm-offload-%d-%d", family, proto); + try_load = 0; + goto retry; + } + rcu_read_unlock(); return type; } @@ -724,9 +732,10 @@ restart: } } } - if (cnt) + if (cnt) { err = 0; - + xfrm_policy_cache_flush(); + } out: spin_unlock_bh(&net->xfrm.xfrm_state_lock); return err; @@ -1620,6 +1629,7 @@ int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n, unsigned short family, struct net *net) { + int i; int err = 0; struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family); if (!afinfo) @@ -1628,6 +1638,9 @@ xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n, spin_lock_bh(&net->xfrm.xfrm_state_lock); /*FIXME*/ if (afinfo->tmpl_sort) err = afinfo->tmpl_sort(dst, src, n); + else + for (i = 0; i < n; i++) + dst[i] = src[i]; spin_unlock_bh(&net->xfrm.xfrm_state_lock); rcu_read_unlock(); return err; @@ -1638,6 +1651,7 @@ int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n, unsigned short family) { + int i; int err = 0; struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family); struct net *net = xs_net(*src); @@ -1648,6 +1662,9 @@ xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n, spin_lock_bh(&net->xfrm.xfrm_state_lock); if (afinfo->state_sort) err = afinfo->state_sort(dst, src, n); + else + for (i = 0; i < n; i++) + dst[i] = src[i]; spin_unlock_bh(&net->xfrm.xfrm_state_lock); rcu_read_unlock(); return err; @@ -2164,7 +2181,7 @@ int xfrm_state_mtu(struct xfrm_state *x, int mtu) return mtu - x->props.header_len; } -int __xfrm_init_state(struct xfrm_state *x, bool init_replay) +int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload) { struct xfrm_state_afinfo *afinfo; struct xfrm_mode *inner_mode; @@ -2229,7 +2246,7 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay) if (x->type == NULL) goto error; - x->type_offload = xfrm_get_type_offload(x->id.proto, family); + x->type_offload = xfrm_get_type_offload(x->id.proto, family, offload); err = x->type->init_state(x); if (err) @@ -2257,7 +2274,7 @@ EXPORT_SYMBOL(__xfrm_init_state); int xfrm_init_state(struct xfrm_state *x) { - return __xfrm_init_state(x, true); + return __xfrm_init_state(x, true, false); } EXPORT_SYMBOL(xfrm_init_state); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2be4c6af008a..2bfbd9121e3b 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -584,7 +584,10 @@ static struct xfrm_state *xfrm_state_construct(struct net *net, xfrm_mark_get(attrs, &x->mark); - err = __xfrm_init_state(x, false); + if (attrs[XFRMA_OUTPUT_MARK]) + x->props.output_mark = nla_get_u32(attrs[XFRMA_OUTPUT_MARK]); + + err = __xfrm_init_state(x, false, attrs[XFRMA_OFFLOAD_DEV]); if (err) goto error; @@ -796,7 +799,7 @@ static int copy_user_offload(struct xfrm_state_offload *xso, struct sk_buff *skb return -EMSGSIZE; xuo = nla_data(attr); - + memset(xuo, 0, sizeof(*xuo)); xuo->ifindex = xso->dev->ifindex; xuo->flags = xso->flags; @@ -897,6 +900,11 @@ static int copy_to_user_state_extra(struct xfrm_state *x, ret = copy_user_offload(&x->xso, skb); if (ret) goto out; + if (x->props.output_mark) { + ret = nla_put_u32(skb, XFRMA_OUTPUT_MARK, x->props.output_mark); + if (ret) + goto out; + } if (x->security) ret = copy_sec_ctx(x->security, skb); out: @@ -1815,8 +1823,6 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, out: xfrm_pol_put(xp); - if (delete && err == 0) - xfrm_garbage_collect(net); return err; } @@ -1869,6 +1875,7 @@ static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, const struct return -EMSGSIZE; id = nlmsg_data(nlh); + memset(&id->sa_id, 0, sizeof(id->sa_id)); memcpy(&id->sa_id.daddr, &x->id.daddr, sizeof(x->id.daddr)); id->sa_id.spi = x->id.spi; id->sa_id.family = x->props.family; @@ -2027,7 +2034,6 @@ static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, return 0; return err; } - xfrm_garbage_collect(net); c.data.type = type; c.event = nlh->nlmsg_type; @@ -2457,6 +2463,7 @@ static const struct nla_policy xfrma_policy[XFRMA_MAX+1] = { [XFRMA_PROTO] = { .type = NLA_U8 }, [XFRMA_ADDRESS_FILTER] = { .len = sizeof(struct xfrm_address_filter) }, [XFRMA_OFFLOAD_DEV] = { .len = sizeof(struct xfrm_user_offload) }, + [XFRMA_OUTPUT_MARK] = { .len = NLA_U32 }, }; static const struct nla_policy xfrma_spd_policy[XFRMA_SPD_MAX+1] = { @@ -2578,6 +2585,8 @@ static int build_expire(struct sk_buff *skb, struct xfrm_state *x, const struct ue = nlmsg_data(nlh); copy_to_user_state(x, &ue->state); ue->hard = (c->data.hard != 0) ? 1 : 0; + /* clear the padding bytes */ + memset(&ue->hard + 1, 0, sizeof(*ue) - offsetofend(typeof(*ue), hard)); err = xfrm_mark_put(skb, &x->mark); if (err) @@ -2676,6 +2685,8 @@ static inline size_t xfrm_sa_len(struct xfrm_state *x) l += nla_total_size(sizeof(x->props.extra_flags)); if (x->xso.dev) l += nla_total_size(sizeof(x->xso)); + if (x->props.output_mark) + l += nla_total_size(sizeof(x->props.output_mark)); /* Must count x->lastused as it may become non-zero behind our back. */ l += nla_total_size_64bit(sizeof(u64)); @@ -2715,6 +2726,7 @@ static int xfrm_notify_sa(struct xfrm_state *x, const struct km_event *c) struct nlattr *attr; id = nlmsg_data(nlh); + memset(id, 0, sizeof(*id)); memcpy(&id->daddr, &x->id.daddr, sizeof(id->daddr)); id->spi = x->id.spi; id->family = x->props.family; |