From 831a39c241e1254b6ddb8dea3144e77b9bbf44b3 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Tue, 30 Dec 2014 23:52:20 +0100 Subject: Revert "cfg80211: make WEXT compatibility unselectable" This reverts commit 24a0aa212ee2dbe44360288684478d76a8e20a0a. It's causing severe userspace breakage. Namely, all the utilities from wireless-utils which are relying on CONFIG_WEXT (which means tools like 'iwconfig', 'iwlist', etc) are not working anymore. There is a 'iw' utility in newer wireless-tools, which is supposed to be a replacement for all the "deprecated" binaries, but it's far away from being massively adopted. Please see [1] for example of the userspace breakage this is causing. In addition to that, Larry Finger reports [2] that this patch is also causing ipw2200 driver being impossible to build. To me this clearly shows that CONFIG_WEXT is far, far away from being "deprecated enough" to be removed. [1] http://thread.gmane.org/gmane.linux.kernel/1857010 [2] http://thread.gmane.org/gmane.linux.network/343688 Signed-off-by: Jiri Kosina Signed-off-by: Linus Torvalds --- net/wireless/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/Kconfig b/net/wireless/Kconfig index 22ba971741e5..29c8675f9a11 100644 --- a/net/wireless/Kconfig +++ b/net/wireless/Kconfig @@ -175,7 +175,7 @@ config CFG80211_INTERNAL_REGDB Most distributions have a CRDA package. So if unsure, say N. config CFG80211_WEXT - bool + bool "cfg80211 wireless extensions compatibility" depends on CFG80211 select WEXT_CORE help -- cgit v1.2.3 From 843925f33fcc293d80acf2c5c8a78adf3344d49b Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 1 Jan 2015 00:39:23 +1100 Subject: tcp: Do not apply TSO segment limit to non-TSO packets Thomas Jarosch reported IPsec TCP stalls when a PMTU event occurs. In fact the problem was completely unrelated to IPsec. The bug is also reproducible if you just disable TSO/GSO. The problem is that when the MSS goes down, existing queued packet on the TX queue that have not been transmitted yet all look like TSO packets and get treated as such. This then triggers a bug where tcp_mss_split_point tells us to generate a zero-sized packet on the TX queue. Once that happens we're screwed because the zero-sized packet can never be removed by ACKs. Fixes: 1485348d242 ("tcp: Apply device TSO segment limit earlier") Reported-by: Thomas Jarosch Signed-off-by: Herbert Xu Cheers, Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 7f18262e2326..65caf8b95e17 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2019,7 +2019,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now))) break; - if (tso_segs == 1) { + if (tso_segs == 1 || !max_segs) { if (unlikely(!tcp_nagle_test(tp, skb, mss_now, (tcp_skb_is_last(sk, skb) ? nonagle : TCP_NAGLE_PUSH)))) @@ -2032,7 +2032,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, } limit = mss_now; - if (tso_segs > 1 && !tcp_urg_mode(tp)) + if (tso_segs > 1 && max_segs && !tcp_urg_mode(tp)) limit = tcp_mss_split_point(sk, skb, mss_now, min_t(unsigned int, cwnd_quota, -- cgit v1.2.3 From 24cc59d1ebaac54d933dc0b30abcd8bd86193eef Mon Sep 17 00:00:00 2001 From: Ben Pfaff Date: Wed, 31 Dec 2014 08:45:46 -0800 Subject: openvswitch: Consistently include VLAN header in flow and port stats. Until now, when VLAN acceleration was in use, the bytes of the VLAN header were not included in port or flow byte counters. They were however included when VLAN acceleration was not used. This commit corrects the inconsistency, by always including the VLAN header in byte counters. Previous discussion at http://openvswitch.org/pipermail/dev/2014-December/049521.html Reported-by: Motonori Shindo Signed-off-by: Ben Pfaff Reviewed-by: Flavio Leitner Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/flow.c | 5 +++-- net/openvswitch/vport.c | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 70bef2ab7f2b..da2fae0873a5 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -70,6 +70,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, { struct flow_stats *stats; int node = numa_node_id(); + int len = skb->len + (vlan_tx_tag_present(skb) ? VLAN_HLEN : 0); stats = rcu_dereference(flow->stats[node]); @@ -105,7 +106,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, if (likely(new_stats)) { new_stats->used = jiffies; new_stats->packet_count = 1; - new_stats->byte_count = skb->len; + new_stats->byte_count = len; new_stats->tcp_flags = tcp_flags; spin_lock_init(&new_stats->lock); @@ -120,7 +121,7 @@ void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags, stats->used = jiffies; stats->packet_count++; - stats->byte_count += skb->len; + stats->byte_count += len; stats->tcp_flags |= tcp_flags; unlock: spin_unlock(&stats->lock); diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 53f3ebbfceab..2034c6d9cb5a 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -480,7 +480,7 @@ void ovs_vport_receive(struct vport *vport, struct sk_buff *skb, stats = this_cpu_ptr(vport->percpu_stats); u64_stats_update_begin(&stats->syncp); stats->rx_packets++; - stats->rx_bytes += skb->len; + stats->rx_bytes += skb->len + (vlan_tx_tag_present(skb) ? VLAN_HLEN : 0); u64_stats_update_end(&stats->syncp); OVS_CB(skb)->input_vport = vport; -- cgit v1.2.3 From 1e359a5de861a57aa04d92bb620f52a5c1d7f8b1 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 5 Jan 2015 10:28:49 +0100 Subject: Revert "mac80211: Fix accounting of the tailroom-needed counter" This reverts commit ca34e3b5c808385b175650605faa29e71e91991b. It turns out that the p54 and cw2100 drivers assume that there's tailroom even when they don't say they really need it. However, there's currently no way for them to explicitly say they do need it, so for now revert this. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=90331. Cc: stable@vger.kernel.org Fixes: ca34e3b5c808 ("mac80211: Fix accounting of the tailroom-needed counter") Reported-by: Christopher Chavez Bisected-by: Larry Finger Debugged-by: Christian Lamparter Signed-off-by: Johannes Berg --- include/net/mac80211.h | 7 ++----- net/mac80211/key.c | 12 +++++++++--- 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 58d719ddaa60..29c7be8808d5 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1270,8 +1270,7 @@ struct ieee80211_vif *wdev_to_ieee80211_vif(struct wireless_dev *wdev); * * @IEEE80211_KEY_FLAG_GENERATE_IV: This flag should be set by the * driver to indicate that it requires IV generation for this - * particular key. Setting this flag does not necessarily mean that SKBs - * will have sufficient tailroom for ICV or MIC. + * particular key. * @IEEE80211_KEY_FLAG_GENERATE_MMIC: This flag should be set by * the driver for a TKIP key if it requires Michael MIC * generation in software. @@ -1283,9 +1282,7 @@ struct ieee80211_vif *wdev_to_ieee80211_vif(struct wireless_dev *wdev); * @IEEE80211_KEY_FLAG_PUT_IV_SPACE: This flag should be set by the driver * if space should be prepared for the IV, but the IV * itself should not be generated. Do not set together with - * @IEEE80211_KEY_FLAG_GENERATE_IV on the same key. Setting this flag does - * not necessarily mean that SKBs will have sufficient tailroom for ICV or - * MIC. + * @IEEE80211_KEY_FLAG_GENERATE_IV on the same key. * @IEEE80211_KEY_FLAG_RX_MGMT: This key will be used to decrypt received * management frames. The flag can help drivers that have a hardware * crypto implementation that doesn't deal with management frames diff --git a/net/mac80211/key.c b/net/mac80211/key.c index 0bb7038121ac..bd4e46ec32bd 100644 --- a/net/mac80211/key.c +++ b/net/mac80211/key.c @@ -140,7 +140,9 @@ static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key) if (!ret) { key->flags |= KEY_FLAG_UPLOADED_TO_HARDWARE; - if (!(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC)) + if (!((key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC) || + (key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV) || + (key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE))) sdata->crypto_tx_tailroom_needed_cnt--; WARN_ON((key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE) && @@ -188,7 +190,9 @@ static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key) sta = key->sta; sdata = key->sdata; - if (!(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC)) + if (!((key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC) || + (key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV) || + (key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE))) increment_tailroom_need_count(sdata); ret = drv_set_key(key->local, DISABLE_KEY, sdata, @@ -884,7 +888,9 @@ void ieee80211_remove_key(struct ieee80211_key_conf *keyconf) if (key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) { key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE; - if (!(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC)) + if (!((key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIC) || + (key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV) || + (key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE))) increment_tailroom_need_count(key->sdata); } -- cgit v1.2.3 From 329887ad13a3f3d26837ea9fce5a8305a7c983e2 Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Wed, 13 Aug 2014 14:26:56 +0200 Subject: batman-adv: fix and simplify condition when bonding should be used The current condition actually does NOT consider bonding when the interface the packet came in from is the soft interface, which is the opposite of what it should do (and the comment describes). Fix that and slightly simplify the condition. Reported-by: Ray Gibson Signed-off-by: Simon Wunderlich Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/routing.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 35f76f2f7824..6648f321864d 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -443,11 +443,13 @@ batadv_find_router(struct batadv_priv *bat_priv, router = batadv_orig_router_get(orig_node, recv_if); + if (!router) + return router; + /* only consider bonding for recv_if == BATADV_IF_DEFAULT (first hop) * and if activated. */ - if (recv_if == BATADV_IF_DEFAULT || !atomic_read(&bat_priv->bonding) || - !router) + if (!(recv_if == BATADV_IF_DEFAULT && atomic_read(&bat_priv->bonding))) return router; /* bonding: loop through the list of possible routers found -- cgit v1.2.3 From 2c667a339c7a6f23e1b7e23eed06f2564cff0c91 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Thu, 30 Oct 2014 06:23:40 +0100 Subject: batman-adv: fix delayed foreign originator recognition MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently it can happen that the reception of an OGM from a new originator is not being accepted. More precisely it can happen that an originator struct gets allocated and initialized (batadv_orig_node_new()), even the TQ gets calculated and set correctly (batadv_iv_ogm_calc_tq()) but still the periodic orig_node purging thread will decide to delete it if it has a chance to jump between these two function calls. This is because batadv_orig_node_new() initializes the last_seen value to zero and its caller (batadv_iv_ogm_orig_get()) makes it visible to other threads by adding it to the hash table already. batadv_iv_ogm_calc_tq() will set the last_seen variable to the correct, current time a few lines later but if the purging thread jumps in between that it will think that the orig_node timed out and will wrongly schedule it for deletion already. If the purging interval is the same as the originator interval (which is the default: 1 second), then this game can continue for several rounds until the random OGM jitter added enough difference between these two (in tests, two to about four rounds seemed common). Fixing this by initializing the last_seen variable of an orig_node to the current time before adding it to the hash table. Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/originator.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 6a484514cd3e..648bdbacee08 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -678,6 +678,7 @@ struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv, atomic_set(&orig_node->last_ttvn, 0); orig_node->tt_buff = NULL; orig_node->tt_buff_len = 0; + orig_node->last_seen = jiffies; reset_time = jiffies - 1 - msecs_to_jiffies(BATADV_RESET_PROTECTION_MS); orig_node->bcast_seqno_reset = reset_time; #ifdef CONFIG_BATMAN_ADV_MCAST -- cgit v1.2.3 From f44d54077a69b1c990aeef49cf9b813fb274cfee Mon Sep 17 00:00:00 2001 From: Martin Hundebøll Date: Tue, 11 Nov 2014 16:22:23 +0100 Subject: batman-adv: fix lock class for decoding hash in network-coding.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit batadv_has_set_lock_class() is called with the wrong hash table as first argument (probably due to a copy-paste error), which leads to false positives when running with lockdep. Introduced-by: 612d2b4fe0a1ff2f8389462a6f8be34e54124c05 ("batman-adv: network coding - save overheard and tx packets for decoding") Signed-off-by: Martin Hundebøll Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/network-coding.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c index 8d04d174669e..fab47f1f3ef9 100644 --- a/net/batman-adv/network-coding.c +++ b/net/batman-adv/network-coding.c @@ -133,7 +133,7 @@ int batadv_nc_mesh_init(struct batadv_priv *bat_priv) if (!bat_priv->nc.decoding_hash) goto err; - batadv_hash_set_lock_class(bat_priv->nc.coding_hash, + batadv_hash_set_lock_class(bat_priv->nc.decoding_hash, &batadv_nc_decoding_hash_lock_class_key); INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker); -- cgit v1.2.3 From e8829f007e982a9a8fb4023109233d5f344d4657 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Thu, 30 Oct 2014 05:40:46 +0100 Subject: batman-adv: fix counter for multicast supporting nodes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A miscounting of nodes having multicast optimizations enabled can lead to multicast packet loss in the following scenario: If the first OGM a node receives from another one has no multicast optimizations support (no multicast tvlv) then we are missing to increase the counter. This potentially leads to the wrong assumption that we could safely use multicast optimizations. Fixings this by increasing the counter if the initial OGM has the multicast TVLV unset, too. Introduced by 60432d756cf06e597ef9da511402dd059b112447 ("batman-adv: Announce new capability via multicast TVLV") Reported-by: Tobias Hachmer Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/multicast.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index ab6bb2af1d45..d3503fbfec93 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -685,11 +685,13 @@ static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv, if (orig_initialized) atomic_dec(&bat_priv->mcast.num_disabled); orig->capabilities |= BATADV_ORIG_CAPA_HAS_MCAST; - /* If mcast support is being switched off increase the disabled - * mcast node counter. + /* If mcast support is being switched off or if this is an initial + * OGM without mcast support then increase the disabled mcast + * node counter. */ } else if (!orig_mcast_enabled && - orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST) { + (orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST || + !orig_initialized)) { atomic_inc(&bat_priv->mcast.num_disabled); orig->capabilities &= ~BATADV_ORIG_CAPA_HAS_MCAST; } -- cgit v1.2.3 From a5164886b0bdadd662f9715a7541432c4d1a0d99 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Thu, 30 Oct 2014 05:40:47 +0100 Subject: batman-adv: fix multicast counter when purging originators MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When purging an orig_node we should only decrease counter tracking the number of nodes without multicast optimizations support if it was increased through this orig_node before. A not yet quite initialized orig_node (meaning it did not have its turn in the mcast-tvlv handler so far) which gets purged would not adhere to this and will lead to a counter imbalance. Fixing this by adding a check whether the orig_node is mcast-initalized before decreasing the counter in the mcast-orig_node-purging routine. Introduced by 60432d756cf06e597ef9da511402dd059b112447 ("batman-adv: Announce new capability via multicast TVLV") Reported-by: Tobias Hachmer Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/multicast.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index d3503fbfec93..b24e4bb64fb5 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -740,7 +740,8 @@ void batadv_mcast_purge_orig(struct batadv_orig_node *orig) { struct batadv_priv *bat_priv = orig->bat_priv; - if (!(orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST)) + if (!(orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST) && + orig->capa_initialized & BATADV_ORIG_CAPA_HAS_MCAST) atomic_dec(&bat_priv->mcast.num_disabled); batadv_mcast_want_unsnoop_update(bat_priv, orig, BATADV_NO_FLAGS); -- cgit v1.2.3 From 9d31b3ce81683ce3c9fd10afa70892e373b21067 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Sat, 13 Dec 2014 23:32:15 +0100 Subject: batman-adv: fix potential TT client + orig-node memory leak MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes a potential memory leak which can occur once an originator times out. On timeout the according global translation table entry might not get purged correctly. Furthermore, the non purged TT entry will cause its orig-node to leak, too. Which additionally can lead to the new multicast optimization feature not kicking in because of a therefore bogus counter. In detail: The batadv_tt_global_entry->orig_list holds the reference to the orig-node. Usually this reference is released after BATADV_PURGE_TIMEOUT through: _batadv_purge_orig()-> batadv_purge_orig_node()->batadv_update_route()->_batadv_update_route()-> batadv_tt_global_del_orig() which purges this global tt entry and releases the reference to the orig-node. However, if between two batadv_purge_orig_node() calls the orig-node timeout grew to 2*BATADV_PURGE_TIMEOUT then this call path isn't reached. Instead the according orig-node is removed from the originator hash in _batadv_purge_orig(), the batadv_update_route() part is skipped and won't be reached anymore. Fixing the issue by moving batadv_tt_global_del_orig() out of the rcu callback. Signed-off-by: Linus Lüssing Acked-by: Antonio Quartulli Signed-off-by: Marek Lindner Signed-off-by: Antonio Quartulli --- net/batman-adv/originator.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 648bdbacee08..bea8198d0198 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -570,9 +570,6 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu) batadv_frag_purge_orig(orig_node, NULL); - batadv_tt_global_del_orig(orig_node->bat_priv, orig_node, -1, - "originator timed out"); - if (orig_node->bat_priv->bat_algo_ops->bat_orig_free) orig_node->bat_priv->bat_algo_ops->bat_orig_free(orig_node); @@ -978,6 +975,9 @@ static void _batadv_purge_orig(struct batadv_priv *bat_priv) if (batadv_purge_orig_node(bat_priv, orig_node)) { batadv_gw_node_delete(bat_priv, orig_node); hlist_del_rcu(&orig_node->hash_entry); + batadv_tt_global_del_orig(orig_node->bat_priv, + orig_node, -1, + "originator timed out"); batadv_orig_node_free_ref(orig_node); continue; } -- cgit v1.2.3