From 52bd2d62ce6758d811edcbd2256eb9ea7f6a56cb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:30:50 -0800 Subject: net: better skb->sender_cpu and skb->napi_id cohabitation skb->sender_cpu and skb->napi_id share a common storage, and we had various bugs about this. We had to call skb_sender_cpu_clear() in some places to not leave a prior skb->napi_id and fool netdev_pick_tx() As suggested by Alexei, we could split the space so that these errors can not happen. 0 value being reserved as the common (not initialized) value, let's reserve [1 .. NR_CPUS] range for valid sender_cpu, and [NR_CPUS+1 .. ~0U] for valid napi_id. This will allow proper busy polling support over tunnels. Signed-off-by: Eric Dumazet Suggested-by: Alexei Starovoitov Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 4355129fff91..c9c394bf0771 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1082,9 +1082,6 @@ static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from) static inline void skb_sender_cpu_clear(struct sk_buff *skb) { -#ifdef CONFIG_XPS - skb->sender_cpu = 0; -#endif } #ifdef NET_SKBUFF_DATA_USES_OFFSET -- cgit v1.2.3-59-g8ed1b From 02d62e86fe892c59a1259d089d4d16ac76977a37 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:30:52 -0800 Subject: net: un-inline sk_busy_loop() There is really little gain from inlining this big function. We'll soon make it even bigger in following patches. This means we no longer need to export napi_by_id() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 --------- include/net/busy_poll.h | 45 +----------------------------------------- net/core/dev.c | 50 +++++++++++++++++++++++++++++++++++++++++++++-- 3 files changed, 49 insertions(+), 55 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 67bfac1abfc1..2020a89df12b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -460,15 +460,6 @@ static inline void napi_complete(struct napi_struct *n) return napi_complete_done(n, 0); } -/** - * napi_by_id - lookup a NAPI by napi_id - * @napi_id: hashed napi_id - * - * lookup @napi_id in napi_hash table - * must be called under rcu_read_lock() - */ -struct napi_struct *napi_by_id(unsigned int napi_id); - /** * napi_hash_add - add a NAPI to global hashtable * @napi: napi context diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h index 1d67fb6b23a0..2fbeb1313c0f 100644 --- a/include/net/busy_poll.h +++ b/include/net/busy_poll.h @@ -72,50 +72,7 @@ static inline bool busy_loop_timeout(unsigned long end_time) return time_after(now, end_time); } -/* when used in sock_poll() nonblock is known at compile time to be true - * so the loop and end_time will be optimized out - */ -static inline bool sk_busy_loop(struct sock *sk, int nonblock) -{ - unsigned long end_time = !nonblock ? sk_busy_loop_end_time(sk) : 0; - const struct net_device_ops *ops; - struct napi_struct *napi; - int rc = false; - - /* - * rcu read lock for napi hash - * bh so we don't race with net_rx_action - */ - rcu_read_lock_bh(); - - napi = napi_by_id(sk->sk_napi_id); - if (!napi) - goto out; - - ops = napi->dev->netdev_ops; - if (!ops->ndo_busy_poll) - goto out; - - do { - rc = ops->ndo_busy_poll(napi); - - if (rc == LL_FLUSH_FAILED) - break; /* permanent failure */ - - if (rc > 0) - /* local bh are disabled so it is ok to use _BH */ - NET_ADD_STATS_BH(sock_net(sk), - LINUX_MIB_BUSYPOLLRXPACKETS, rc); - cpu_relax(); - - } while (!nonblock && skb_queue_empty(&sk->sk_receive_queue) && - !need_resched() && !busy_loop_timeout(end_time)); - - rc = !skb_queue_empty(&sk->sk_receive_queue); -out: - rcu_read_unlock_bh(); - return rc; -} +bool sk_busy_loop(struct sock *sk, int nonblock); /* used in the NIC receive handler to mark the skb */ static inline void skb_mark_napi_id(struct sk_buff *skb, diff --git a/net/core/dev.c b/net/core/dev.c index 2582c24a75c6..74a816b299df 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -96,6 +96,7 @@ #include #include #include +#include #include #include #include @@ -4663,7 +4664,7 @@ void napi_complete_done(struct napi_struct *n, int work_done) EXPORT_SYMBOL(napi_complete_done); /* must be called under rcu_read_lock(), as we dont take a reference */ -struct napi_struct *napi_by_id(unsigned int napi_id) +static struct napi_struct *napi_by_id(unsigned int napi_id) { unsigned int hash = napi_id % HASH_SIZE(napi_hash); struct napi_struct *napi; @@ -4674,7 +4675,52 @@ struct napi_struct *napi_by_id(unsigned int napi_id) return NULL; } -EXPORT_SYMBOL_GPL(napi_by_id); + +#if defined(CONFIG_NET_RX_BUSY_POLL) +bool sk_busy_loop(struct sock *sk, int nonblock) +{ + unsigned long end_time = !nonblock ? sk_busy_loop_end_time(sk) : 0; + const struct net_device_ops *ops; + struct napi_struct *napi; + int rc = false; + + /* + * rcu read lock for napi hash + * bh so we don't race with net_rx_action + */ + rcu_read_lock_bh(); + + napi = napi_by_id(sk->sk_napi_id); + if (!napi) + goto out; + + ops = napi->dev->netdev_ops; + if (!ops->ndo_busy_poll) + goto out; + + do { + rc = ops->ndo_busy_poll(napi); + + if (rc == LL_FLUSH_FAILED) + break; /* permanent failure */ + + if (rc > 0) + /* local bh are disabled so it is ok to use _BH */ + NET_ADD_STATS_BH(sock_net(sk), + LINUX_MIB_BUSYPOLLRXPACKETS, rc); + cpu_relax(); + + } while (!nonblock && skb_queue_empty(&sk->sk_receive_queue) && + !need_resched() && !busy_loop_timeout(end_time)); + + rc = !skb_queue_empty(&sk->sk_receive_queue); +out: + rcu_read_unlock_bh(); + return rc; +} +EXPORT_SYMBOL(sk_busy_loop); + +#endif /* CONFIG_NET_RX_BUSY_POLL */ void napi_hash_add(struct napi_struct *napi) { -- cgit v1.2.3-59-g8ed1b From d64b5e85bfe2fe4c790abcbd16d9ae32391ddd7e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:31:00 -0800 Subject: net: add netif_tx_napi_add() netif_tx_napi_add() is a variant of netif_napi_add() It should be used by drivers that use a napi structure to exclusively poll TX. We do not want to add this kind of napi in napi_hash[] in following patches, adding generic busy polling to all NAPI drivers. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 2 +- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 4 ++-- .../net/ethernet/freescale/fs_enet/fs_enet-main.c | 2 +- drivers/net/ethernet/freescale/gianfar.c | 4 ++-- drivers/net/ethernet/mellanox/mlx4/en_cq.c | 4 ++-- drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c | 4 ++-- drivers/net/ethernet/rocker/rocker.c | 2 +- drivers/net/ethernet/ti/cpsw.c | 2 +- drivers/net/ethernet/ti/netcp_core.c | 2 +- drivers/net/wireless/ath/wil6210/netdev.c | 2 +- include/linux/netdevice.h | 23 +++++++++++++++++++++- net/core/dev.c | 3 ++- 12 files changed, 38 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 858106352ce9..993c780bdfab 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -1216,7 +1216,7 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv, /* Initialize SW view of the ring */ spin_lock_init(&ring->lock); ring->priv = priv; - netif_napi_add(priv->netdev, &ring->napi, bcm_sysport_tx_poll, 64); + netif_tx_napi_add(priv->netdev, &ring->napi, bcm_sysport_tx_poll, 64); ring->index = index; ring->size = size; ring->alloc_size = ring->size; diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 17f017ab4dac..b15a60d787c7 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2041,11 +2041,11 @@ static void bcmgenet_init_tx_napi(struct bcmgenet_priv *priv) for (i = 0; i < priv->hw_params->tx_queues; ++i) { ring = &priv->tx_rings[i]; - netif_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64); + netif_tx_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64); } ring = &priv->tx_rings[DESC_INDEX]; - netif_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64); + netif_tx_napi_add(priv->dev, &ring->napi, bcmgenet_tx_poll, 64); } static void bcmgenet_enable_tx_napi(struct bcmgenet_priv *priv) diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c index cf8e54652df9..48a9c176e0d1 100644 --- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c +++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c @@ -1050,7 +1050,7 @@ static int fs_enet_probe(struct platform_device *ofdev) ndev->netdev_ops = &fs_enet_netdev_ops; ndev->watchdog_timeo = 2 * HZ; netif_napi_add(ndev, &fep->napi, fs_enet_rx_napi, fpi->napi_weight); - netif_napi_add(ndev, &fep->napi_tx, fs_enet_tx_napi, 2); + netif_tx_napi_add(ndev, &fep->napi_tx, fs_enet_tx_napi, 2); ndev->ethtool_ops = &fs_ethtool_ops; diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 3e6b9b437497..c8bc43e99a35 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1347,12 +1347,12 @@ static int gfar_probe(struct platform_device *ofdev) if (priv->poll_mode == GFAR_SQ_POLLING) { netif_napi_add(dev, &priv->gfargrp[i].napi_rx, gfar_poll_rx_sq, GFAR_DEV_WEIGHT); - netif_napi_add(dev, &priv->gfargrp[i].napi_tx, + netif_tx_napi_add(dev, &priv->gfargrp[i].napi_tx, gfar_poll_tx_sq, 2); } else { netif_napi_add(dev, &priv->gfargrp[i].napi_rx, gfar_poll_rx, GFAR_DEV_WEIGHT); - netif_napi_add(dev, &priv->gfargrp[i].napi_tx, + netif_tx_napi_add(dev, &priv->gfargrp[i].napi_tx, gfar_poll_tx, 2); } } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c index eb8a4988de63..3a6176fea78d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c @@ -156,8 +156,8 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, cq->mcq.event = mlx4_en_cq_event; if (cq->is_tx) { - netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq, - NAPI_POLL_WEIGHT); + netif_tx_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq, + NAPI_POLL_WEIGHT); } else { netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64); napi_hash_add(&cq->napi); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c index d4b5085a21fa..7bd6f25b4625 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c @@ -1604,7 +1604,7 @@ int qlcnic_82xx_napi_add(struct qlcnic_adapter *adapter, if (qlcnic_check_multi_tx(adapter) && !adapter->ahw->diag_test) { for (ring = 0; ring < adapter->drv_tx_rings; ring++) { tx_ring = &adapter->tx_ring[ring]; - netif_napi_add(netdev, &tx_ring->napi, qlcnic_tx_poll, + netif_tx_napi_add(netdev, &tx_ring->napi, qlcnic_tx_poll, NAPI_POLL_WEIGHT); } } @@ -2135,7 +2135,7 @@ int qlcnic_83xx_napi_add(struct qlcnic_adapter *adapter, !(adapter->flags & QLCNIC_TX_INTR_SHARED)) { for (ring = 0; ring < adapter->drv_tx_rings; ring++) { tx_ring = &adapter->tx_ring[ring]; - netif_napi_add(netdev, &tx_ring->napi, + netif_tx_napi_add(netdev, &tx_ring->napi, qlcnic_83xx_msix_tx_poll, NAPI_POLL_WEIGHT); } diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c index e9f2349e98bc..a4ab71d43e4e 100644 --- a/drivers/net/ethernet/rocker/rocker.c +++ b/drivers/net/ethernet/rocker/rocker.c @@ -4998,7 +4998,7 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number) dev->netdev_ops = &rocker_port_netdev_ops; dev->ethtool_ops = &rocker_port_ethtool_ops; dev->switchdev_ops = &rocker_port_switchdev_ops; - netif_napi_add(dev, &rocker_port->napi_tx, rocker_port_poll_tx, + netif_tx_napi_add(dev, &rocker_port->napi_tx, rocker_port_poll_tx, NAPI_POLL_WEIGHT); netif_napi_add(dev, &rocker_port->napi_rx, rocker_port_poll_rx, NAPI_POLL_WEIGHT); diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 48b92c9de12a..15322c08de80 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -2469,7 +2469,7 @@ static int cpsw_probe(struct platform_device *pdev) ndev->netdev_ops = &cpsw_netdev_ops; ndev->ethtool_ops = &cpsw_ethtool_ops; netif_napi_add(ndev, &priv->napi_rx, cpsw_rx_poll, CPSW_POLL_WEIGHT); - netif_napi_add(ndev, &priv->napi_tx, cpsw_tx_poll, CPSW_POLL_WEIGHT); + netif_tx_napi_add(ndev, &priv->napi_tx, cpsw_tx_poll, CPSW_POLL_WEIGHT); /* register the network device */ SET_NETDEV_DEV(ndev, &pdev->dev); diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 37b9b39192ec..e5e20e734f21 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -1990,7 +1990,7 @@ static int netcp_create_interface(struct netcp_device *netcp_device, /* NAPI register */ netif_napi_add(ndev, &netcp->rx_napi, netcp_rx_poll, NETCP_NAPI_WEIGHT); - netif_napi_add(ndev, &netcp->tx_napi, netcp_tx_poll, NETCP_NAPI_WEIGHT); + netif_tx_napi_add(ndev, &netcp->tx_napi, netcp_tx_poll, NETCP_NAPI_WEIGHT); /* Register the network device */ ndev->dev_id = 0; diff --git a/drivers/net/wireless/ath/wil6210/netdev.c b/drivers/net/wireless/ath/wil6210/netdev.c index e3b3c8fb4605..56aaa2d4fb0e 100644 --- a/drivers/net/wireless/ath/wil6210/netdev.c +++ b/drivers/net/wireless/ath/wil6210/netdev.c @@ -183,7 +183,7 @@ void *wil_if_alloc(struct device *dev) netif_napi_add(ndev, &wil->napi_rx, wil6210_netdev_poll_rx, WIL6210_NAPI_BUDGET); - netif_napi_add(ndev, &wil->napi_tx, wil6210_netdev_poll_tx, + netif_tx_napi_add(ndev, &wil->napi_tx, wil6210_netdev_poll_tx, WIL6210_NAPI_BUDGET); netif_tx_stop_all_queues(ndev); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2020a89df12b..838935d1cdbb 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -326,7 +326,8 @@ enum { NAPI_STATE_SCHED, /* Poll is scheduled */ NAPI_STATE_DISABLE, /* Disable pending */ NAPI_STATE_NPSVC, /* Netpoll - don't dequeue from poll_list */ - NAPI_STATE_HASHED, /* In NAPI hash */ + NAPI_STATE_HASHED, /* In NAPI hash (busy polling possible) */ + NAPI_STATE_NO_BUSY_POLL,/* Do not add in napi_hash, no busy polling */ }; enum gro_result { @@ -1938,6 +1939,26 @@ static inline void *netdev_priv(const struct net_device *dev) void netif_napi_add(struct net_device *dev, struct napi_struct *napi, int (*poll)(struct napi_struct *, int), int weight); +/** + * netif_tx_napi_add - initialize a napi context + * @dev: network device + * @napi: napi context + * @poll: polling function + * @weight: default weight + * + * This variant of netif_napi_add() should be used from drivers using NAPI + * to exclusively poll a TX queue. + * This will avoid we add it into napi_hash[], thus polluting this hash table. + */ +static inline void netif_tx_napi_add(struct net_device *dev, + struct napi_struct *napi, + int (*poll)(struct napi_struct *, int), + int weight) +{ + set_bit(NAPI_STATE_NO_BUSY_POLL, &napi->state); + netif_napi_add(dev, napi, poll, weight); +} + /** * netif_napi_del - remove a napi context * @napi: napi context diff --git a/net/core/dev.c b/net/core/dev.c index 83b48747928c..ff58a8bc5e3c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4737,7 +4737,8 @@ EXPORT_SYMBOL(sk_busy_loop); void napi_hash_add(struct napi_struct *napi) { - if (test_and_set_bit(NAPI_STATE_HASHED, &napi->state)) + if (test_bit(NAPI_STATE_NO_BUSY_POLL, &napi->state) || + test_and_set_bit(NAPI_STATE_HASHED, &napi->state)) return; spin_lock(&napi_hash_lock); -- cgit v1.2.3-59-g8ed1b From 6180d9de61a5c461f9e3efef5417a844701dbbb2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:31:01 -0800 Subject: net: move napi_hash[] into read mostly section We do not often add/delete a napi context. Moving napi_hash[] into read mostly section avoids potential false sharing. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/hashtable.h | 4 ++++ net/core/dev.c | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/hashtable.h b/include/linux/hashtable.h index 519b6e2d769e..661e5c2a8e2a 100644 --- a/include/linux/hashtable.h +++ b/include/linux/hashtable.h @@ -16,6 +16,10 @@ struct hlist_head name[1 << (bits)] = \ { [0 ... ((1 << (bits)) - 1)] = HLIST_HEAD_INIT } +#define DEFINE_READ_MOSTLY_HASHTABLE(name, bits) \ + struct hlist_head name[1 << (bits)] __read_mostly = \ + { [0 ... ((1 << (bits)) - 1)] = HLIST_HEAD_INIT } + #define DECLARE_HASHTABLE(name, bits) \ struct hlist_head name[1 << (bits)] diff --git a/net/core/dev.c b/net/core/dev.c index ff58a8bc5e3c..02dfbd91a8e4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -184,7 +184,7 @@ EXPORT_SYMBOL(dev_base_lock); static DEFINE_SPINLOCK(napi_hash_lock); static unsigned int napi_gen_id = NR_CPUS; -static DEFINE_HASHTABLE(napi_hash, 8); +static DEFINE_READ_MOSTLY_HASHTABLE(napi_hash, 8); static seqcount_t devnet_rename_seq; -- cgit v1.2.3-59-g8ed1b From 34cbe27e811c591c854a39c0dee1b461bb796953 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:31:02 -0800 Subject: net: napi_hash_del() returns a boolean status napi_hash_del() will soon be used from both drivers (if they want) or core networking stack. Callers are responsibles to ensure an RCU grace period is respected before freeing napi structure : napi_hash_del() can signal if this RCU grace period is needed or not. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++-- net/core/dev.c | 10 +++++++--- 2 files changed, 10 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 838935d1cdbb..e5c33b29471b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -474,9 +474,10 @@ void napi_hash_add(struct napi_struct *napi); * @napi: napi context * * Warning: caller must observe rcu grace period - * before freeing memory containing @napi + * before freeing memory containing @napi, if + * this function returns true. */ -void napi_hash_del(struct napi_struct *napi); +bool napi_hash_del(struct napi_struct *napi); /** * napi_disable - prevent NAPI from scheduling diff --git a/net/core/dev.c b/net/core/dev.c index 02dfbd91a8e4..59dddac1c2e7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4760,14 +4760,18 @@ EXPORT_SYMBOL_GPL(napi_hash_add); /* Warning : caller is responsible to make sure rcu grace period * is respected before freeing memory containing @napi */ -void napi_hash_del(struct napi_struct *napi) +bool napi_hash_del(struct napi_struct *napi) { + bool rcu_sync_needed = false; + spin_lock(&napi_hash_lock); - if (test_and_clear_bit(NAPI_STATE_HASHED, &napi->state)) + if (test_and_clear_bit(NAPI_STATE_HASHED, &napi->state)) { + rcu_sync_needed = true; hlist_del_rcu(&napi->napi_hash_node); - + } spin_unlock(&napi_hash_lock); + return rcu_sync_needed; } EXPORT_SYMBOL_GPL(napi_hash_del); -- cgit v1.2.3-59-g8ed1b From 93d05d4a320cb16712bb3d57a9658f395d8cecb9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Nov 2015 06:31:03 -0800 Subject: net: provide generic busy polling to all NAPI drivers NAPI drivers no longer need to observe a particular protocol to benefit from busy polling (CONFIG_NET_RX_BUSY_POLL=y) napi_hash_add() and napi_hash_del() are automatically called from core networking stack, respectively from netif_napi_add() and netif_napi_del() This patch depends on free_netdev() and netif_napi_del() being called from process context, which seems to be the norm. Drivers might still prefer to call napi_hash_del() on their own, since they might combine all the rcu grace periods into a single one, knowing their NAPI structures lifetime, while core networking stack has no idea of a possible combining. Once this patch proves to not bring serious regressions, we will cleanup drivers to either remove napi_hash_del() or provide appropriate rcu grace periods combining. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 2 -- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 -- drivers/net/ethernet/chelsio/cxgb4/sge.c | 1 - drivers/net/ethernet/cisco/enic/enic_main.c | 2 -- drivers/net/ethernet/emulex/benet/be_main.c | 1 - drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c | 1 - drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 3 --- drivers/net/ethernet/mellanox/mlx4/en_cq.c | 6 ++---- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 1 - drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 1 - drivers/net/ethernet/sfc/efx.c | 1 - drivers/net/virtio_net.c | 1 - include/linux/netdevice.h | 7 +++++++ net/core/dev.c | 7 +++++++ 14 files changed, 16 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index ab9222924bd9..d9add7c02e42 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -46,7 +46,6 @@ static void bnx2x_add_all_napi_cnic(struct bnx2x *bp) for_each_rx_queue_cnic(bp, i) { netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi), bnx2x_poll, NAPI_POLL_WEIGHT); - napi_hash_add(&bnx2x_fp(bp, i, napi)); } } @@ -58,7 +57,6 @@ static void bnx2x_add_all_napi(struct bnx2x *bp) for_each_eth_queue(bp, i) { netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi), bnx2x_poll, NAPI_POLL_WEIGHT); - napi_hash_add(&bnx2x_fp(bp, i, napi)); } } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index db15c5ee09c5..f2d0dc9b1c41 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4227,12 +4227,10 @@ static void bnxt_init_napi(struct bnxt *bp) bnapi = bp->bnapi[i]; netif_napi_add(bp->dev, &bnapi->napi, bnxt_poll, 64); - napi_hash_add(&bnapi->napi); } } else { bnapi = bp->bnapi[0]; netif_napi_add(bp->dev, &bnapi->napi, bnxt_poll, 64); - napi_hash_add(&bnapi->napi); } } diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index f650f295f264..48d8fbb1c220 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -2527,7 +2527,6 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, goto err; netif_napi_add(dev, &iq->napi, napi_rx_handler, 64); - napi_hash_add(&iq->napi); iq->cur_desc = iq->desc; iq->cidx = 0; iq->gen = 1; diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index b36643ef0593..b2182d3ba3cc 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -2458,13 +2458,11 @@ static int enic_dev_init(struct enic *enic) switch (vnic_dev_get_intr_mode(enic->vdev)) { default: netif_napi_add(netdev, &enic->napi[0], enic_poll, 64); - napi_hash_add(&enic->napi[0]); break; case VNIC_DEV_INTR_MODE_MSIX: for (i = 0; i < enic->rq_count; i++) { netif_napi_add(netdev, &enic->napi[i], enic_poll_msix_rq, NAPI_POLL_WEIGHT); - napi_hash_add(&enic->napi[i]); } for (i = 0; i < enic->wq_count; i++) netif_napi_add(netdev, &enic->napi[enic_cq_wq(enic, i)], diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index c29d62496ad9..4cab8879f5ae 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -2630,7 +2630,6 @@ static int be_evt_queues_create(struct be_adapter *adapter) eqo->affinity_mask); netif_napi_add(adapter->netdev, &eqo->napi, be_poll, BE_NAPI_WEIGHT); - napi_hash_add(&eqo->napi); } return 0; } diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c index f3168bcc7d87..e771e764daa3 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c @@ -844,7 +844,6 @@ static int ixgbe_alloc_q_vector(struct ixgbe_adapter *adapter, /* initialize NAPI */ netif_napi_add(adapter->netdev, &q_vector->napi, ixgbe_poll, 64); - napi_hash_add(&q_vector->napi); #ifdef CONFIG_NET_RX_BUSY_POLL /* initialize busy poll */ diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 592ff237d692..2955186cd4f6 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -2483,9 +2483,6 @@ static int ixgbevf_alloc_q_vectors(struct ixgbevf_adapter *adapter) q_vector->v_idx = q_idx; netif_napi_add(adapter->netdev, &q_vector->napi, ixgbevf_poll, 64); -#ifdef CONFIG_NET_RX_BUSY_POLL - napi_hash_add(&q_vector->napi); -#endif adapter->q_vector[q_idx] = q_vector; } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c index 3a6176fea78d..af975a2b74c6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c @@ -155,13 +155,11 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, cq->mcq.comp = cq->is_tx ? mlx4_en_tx_irq : mlx4_en_rx_irq; cq->mcq.event = mlx4_en_cq_event; - if (cq->is_tx) { + if (cq->is_tx) netif_tx_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq, NAPI_POLL_WEIGHT); - } else { + else netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64); - napi_hash_add(&cq->napi); - } napi_enable(&cq->napi); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index ffb1f9c1b973..f6a8cc787603 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -982,7 +982,6 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, mlx5e_build_channeltc_to_txq_map(priv, ix); netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); - napi_hash_add(&c->napi); err = mlx5e_open_tx_cqs(c, cparam); if (err) diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index acf866147d65..270c9eeb7ab6 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -3814,7 +3814,6 @@ static int myri10ge_alloc_slices(struct myri10ge_priv *mgp) ss->dev = mgp->dev; netif_napi_add(ss->dev, &ss->napi, myri10ge_poll, myri10ge_napi_weight); - napi_hash_add(&ss->napi); } return 0; abort: diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index a3c42a376741..4e82bcfbe3e0 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -2059,7 +2059,6 @@ static void efx_init_napi_channel(struct efx_channel *channel) channel->napi_dev = efx->net_dev; netif_napi_add(channel->napi_dev, &channel->napi_str, efx_poll, napi_weight); - napi_hash_add(&channel->napi_str); efx_channel_busy_poll_init(channel); } diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index d1d14cecf450..b1ae4cbf2453 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1610,7 +1610,6 @@ static int virtnet_alloc_queues(struct virtnet_info *vi) vi->rq[i].pages = NULL; netif_napi_add(vi->dev, &vi->rq[i].napi, virtnet_poll, napi_weight); - napi_hash_add(&vi->rq[i].napi); sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg)); ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e5c33b29471b..7d2d1d7aaec7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -466,6 +466,9 @@ static inline void napi_complete(struct napi_struct *n) * @napi: napi context * * generate a new napi_id and store a @napi under it in napi_hash + * Used for busy polling (CONFIG_NET_RX_BUSY_POLL) + * Note: This is normally automatically done from netif_napi_add(), + * so might disappear in a future linux version. */ void napi_hash_add(struct napi_struct *napi); @@ -476,6 +479,10 @@ void napi_hash_add(struct napi_struct *napi); * Warning: caller must observe rcu grace period * before freeing memory containing @napi, if * this function returns true. + * Note: core networking stack automatically calls it + * from netif_napi_del() + * Drivers might want to call this helper to combine all + * the needed rcu grace periods into a single one. */ bool napi_hash_del(struct napi_struct *napi); diff --git a/net/core/dev.c b/net/core/dev.c index 59dddac1c2e7..41cef3e3f558 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4807,6 +4807,7 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi, napi->poll_owner = -1; #endif set_bit(NAPI_STATE_SCHED, &napi->state); + napi_hash_add(napi); } EXPORT_SYMBOL(netif_napi_add); @@ -4826,8 +4827,12 @@ void napi_disable(struct napi_struct *n) } EXPORT_SYMBOL(napi_disable); +/* Must be called in process context */ void netif_napi_del(struct napi_struct *napi) { + might_sleep(); + if (napi_hash_del(napi)) + synchronize_net(); list_del_init(&napi->dev_list); napi_free_frags(napi); @@ -7227,11 +7232,13 @@ EXPORT_SYMBOL(alloc_netdev_mqs); * This function does the last stage of destroying an allocated device * interface. The reference to the device object is released. * If this is the last reference then it will be freed. + * Must be called in process context. */ void free_netdev(struct net_device *dev) { struct napi_struct *p, *n; + might_sleep(); netif_free_tx_queues(dev); #ifdef CONFIG_SYSFS kvfree(dev->_rx); -- cgit v1.2.3-59-g8ed1b From f5c4a42a7549cbc29dbac2b5ede05a3bc2bb4522 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Thu, 5 Nov 2015 07:09:59 +0100 Subject: Bluetooth: Add hci_skb_* helper wrappers for bt_cb(skb) access For all the HCI driver related variables accesssed via bt_cb(skb), provide helper wrappers. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg --- include/net/bluetooth/bluetooth.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 42844d7b154a..663e0ef1eaa0 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -316,6 +316,10 @@ struct bt_skb_cb { }; #define bt_cb(skb) ((struct bt_skb_cb *)((skb)->cb)) +#define hci_skb_pkt_type(skb) bt_cb((skb))->pkt_type +#define hci_skb_expect(skb) bt_cb((skb))->expect +#define hci_skb_opcode(skb) bt_cb((skb))->hci.opcode + static inline struct sk_buff *bt_skb_alloc(unsigned int len, gfp_t how) { struct sk_buff *skb; -- cgit v1.2.3-59-g8ed1b From 5fcc86bd2695d4ccad2d87cb424f4c01a4e544f3 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Thu, 5 Nov 2015 09:31:39 +0200 Subject: Bluetooth: Remove redundant setting to zero of bt_cb The socket allocation functions will always memset skb->cb to zero so there's no need to make other initializations needing the same thing. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/bluetooth.h | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 663e0ef1eaa0..a85e6d3d75ef 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -325,10 +325,8 @@ static inline struct sk_buff *bt_skb_alloc(unsigned int len, gfp_t how) struct sk_buff *skb; skb = alloc_skb(len + BT_SKB_RESERVE, how); - if (skb) { + if (skb) skb_reserve(skb, BT_SKB_RESERVE); - bt_cb(skb)->incoming = 0; - } return skb; } @@ -338,10 +336,8 @@ static inline struct sk_buff *bt_skb_send_alloc(struct sock *sk, struct sk_buff *skb; skb = sock_alloc_send_skb(sk, len + BT_SKB_RESERVE, nb, err); - if (skb) { + if (skb) skb_reserve(skb, BT_SKB_RESERVE); - bt_cb(skb)->incoming = 0; - } if (!skb && *err) return NULL; -- cgit v1.2.3-59-g8ed1b From 44d271377479c4d4fe7f2d07d188656684773fbd Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Thu, 5 Nov 2015 09:31:40 +0200 Subject: Bluetooth: Compress the size of struct hci_ctrl We can reduce the size of the hci_ctrl struct by converting 'bool req_start' to 'u8 req_flags' and making the two function pointers a union (since only one is ever set at a time). Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/bluetooth.h | 11 ++++++++--- net/bluetooth/hci_core.c | 14 +++++++------- net/bluetooth/hci_request.c | 10 +++++++--- net/bluetooth/hci_sock.c | 2 +- 4 files changed, 23 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index a85e6d3d75ef..8d38f411009c 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -296,12 +296,17 @@ typedef void (*hci_req_complete_t)(struct hci_dev *hdev, u8 status, u16 opcode); typedef void (*hci_req_complete_skb_t)(struct hci_dev *hdev, u8 status, u16 opcode, struct sk_buff *skb); +#define HCI_REQ_START BIT(0) +#define HCI_REQ_SKB BIT(1) + struct hci_ctrl { __u16 opcode; - bool req_start; + u8 req_flags; u8 req_event; - hci_req_complete_t req_complete; - hci_req_complete_skb_t req_complete_skb; + union { + hci_req_complete_t req_complete; + hci_req_complete_skb_t req_complete_skb; + }; }; struct bt_skb_cb { diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index db26cbd1cd9d..bc97fc6de876 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -3695,7 +3695,7 @@ int hci_send_cmd(struct hci_dev *hdev, __u16 opcode, __u32 plen, /* Stand-alone HCI commands must be flagged as * single-command requests. */ - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; skb_queue_tail(&hdev->cmd_q, skb); queue_work(hdev->workqueue, &hdev->cmd_work); @@ -4392,7 +4392,7 @@ static bool hci_req_is_complete(struct hci_dev *hdev) if (!skb) return true; - return bt_cb(skb)->hci.req_start; + return (bt_cb(skb)->hci.req_flags & HCI_REQ_START); } static void hci_resend_last(struct hci_dev *hdev) @@ -4452,20 +4452,20 @@ void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status, * callback would be found in hdev->sent_cmd instead of the * command queue (hdev->cmd_q). */ - if (bt_cb(hdev->sent_cmd)->hci.req_complete) { - *req_complete = bt_cb(hdev->sent_cmd)->hci.req_complete; + if (bt_cb(hdev->sent_cmd)->hci.req_flags & HCI_REQ_SKB) { + *req_complete_skb = bt_cb(hdev->sent_cmd)->hci.req_complete_skb; return; } - if (bt_cb(hdev->sent_cmd)->hci.req_complete_skb) { - *req_complete_skb = bt_cb(hdev->sent_cmd)->hci.req_complete_skb; + if (bt_cb(hdev->sent_cmd)->hci.req_complete) { + *req_complete = bt_cb(hdev->sent_cmd)->hci.req_complete; return; } /* Remove all pending commands belonging to this request */ spin_lock_irqsave(&hdev->cmd_q.lock, flags); while ((skb = __skb_dequeue(&hdev->cmd_q))) { - if (bt_cb(skb)->hci.req_start) { + if (bt_cb(skb)->hci.req_flags & HCI_REQ_START) { __skb_queue_head(&hdev->cmd_q, skb); break; } diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index bdb170995966..5ba27c30e8f2 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -56,8 +56,12 @@ static int req_run(struct hci_request *req, hci_req_complete_t complete, return -ENODATA; skb = skb_peek_tail(&req->cmd_q); - bt_cb(skb)->hci.req_complete = complete; - bt_cb(skb)->hci.req_complete_skb = complete_skb; + if (complete) { + bt_cb(skb)->hci.req_complete = complete; + } else if (complete_skb) { + bt_cb(skb)->hci.req_complete_skb = complete_skb; + bt_cb(skb)->hci.req_flags |= HCI_REQ_SKB; + } spin_lock_irqsave(&hdev->cmd_q.lock, flags); skb_queue_splice_tail(&req->cmd_q, &hdev->cmd_q); @@ -128,7 +132,7 @@ void hci_req_add_ev(struct hci_request *req, u16 opcode, u32 plen, } if (skb_queue_empty(&req->cmd_q)) - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; bt_cb(skb)->hci.req_event = event; diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 235ad0fa3571..19b23013c4f6 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -1249,7 +1249,7 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, /* Stand-alone HCI commands must be flagged as * single-command requests. */ - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; skb_queue_tail(&hdev->cmd_q, skb); queue_work(hdev->workqueue, &hdev->cmd_work); -- cgit v1.2.3-59-g8ed1b From dd31506d4aece48943802c2bca3f1f7d2e7266b4 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 8 Nov 2015 07:47:12 +0100 Subject: Bluetooth: Add support for sending system notes to monitor channel The monitor channel can be used to send generic system notes as text strings for debugging purposes. This adds the system note monitor code and uses it for including kernel and subsystem version into traces. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg --- include/net/bluetooth/bluetooth.h | 2 ++ include/net/bluetooth/hci_mon.h | 1 + net/bluetooth/af_bluetooth.c | 8 +++----- net/bluetooth/hci_sock.c | 29 +++++++++++++++++++++++++++++ 4 files changed, 35 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 8d38f411009c..bfd1590821d6 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -29,6 +29,8 @@ #include #include +#define BT_SUBSYS_VERSION "2.21" + #ifndef AF_BLUETOOTH #define AF_BLUETOOTH 31 #define PF_BLUETOOTH AF_BLUETOOTH diff --git a/include/net/bluetooth/hci_mon.h b/include/net/bluetooth/hci_mon.h index 2b67567cf28d..c91bb23eb29e 100644 --- a/include/net/bluetooth/hci_mon.h +++ b/include/net/bluetooth/hci_mon.h @@ -43,6 +43,7 @@ struct hci_mon_hdr { #define HCI_MON_CLOSE_INDEX 9 #define HCI_MON_INDEX_INFO 10 #define HCI_MON_VENDOR_DIAG 11 +#define HCI_MON_SYSTEM_NOTE 12 struct hci_mon_new_index { __u8 type; diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index a3bffd1ec2b4..34c53d5862f6 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -33,8 +33,6 @@ #include "selftest.h" -#define VERSION "2.21" - /* Bluetooth sockets */ #define BT_MAX_PROTO 8 static const struct net_proto_family *bt_proto[BT_MAX_PROTO]; @@ -715,7 +713,7 @@ static int __init bt_init(void) sock_skb_cb_check_size(sizeof(struct bt_skb_cb)); - BT_INFO("Core ver %s", VERSION); + BT_INFO("Core ver %s", BT_SUBSYS_VERSION); err = bt_selftest(); if (err < 0) @@ -789,7 +787,7 @@ subsys_initcall(bt_init); module_exit(bt_exit); MODULE_AUTHOR("Marcel Holtmann "); -MODULE_DESCRIPTION("Bluetooth Core ver " VERSION); -MODULE_VERSION(VERSION); +MODULE_DESCRIPTION("Bluetooth Core ver " BT_SUBSYS_VERSION); +MODULE_VERSION(BT_SUBSYS_VERSION); MODULE_LICENSE("GPL"); MODULE_ALIAS_NETPROTO(PF_BLUETOOTH); diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 18a41eae295c..710265c35d16 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -26,6 +26,8 @@ #include #include +#include +#include #include #include @@ -383,6 +385,29 @@ static struct sk_buff *create_monitor_event(struct hci_dev *hdev, int event) return skb; } +static void send_monitor_note(struct sock *sk, const char *text) +{ + size_t len = strlen(text); + struct hci_mon_hdr *hdr; + struct sk_buff *skb; + + skb = bt_skb_alloc(len + 1, GFP_ATOMIC); + if (!skb) + return; + + strcpy(skb_put(skb, len + 1), text); + + __net_timestamp(skb); + + hdr = (void *)skb_push(skb, HCI_MON_HDR_SIZE); + hdr->opcode = cpu_to_le16(HCI_MON_SYSTEM_NOTE); + hdr->index = cpu_to_le16(HCI_DEV_NONE); + hdr->len = cpu_to_le16(skb->len - HCI_MON_HDR_SIZE); + + if (sock_queue_rcv_skb(sk, skb)) + kfree_skb(skb); +} + static void send_monitor_replay(struct sock *sk) { struct hci_dev *hdev; @@ -872,6 +897,10 @@ static int hci_sock_bind(struct socket *sock, struct sockaddr *addr, */ hci_sock_set_flag(sk, HCI_SOCK_TRUSTED); + send_monitor_note(sk, "Linux version " UTS_RELEASE + " (" UTS_MACHINE ")"); + send_monitor_note(sk, "Bluetooth subsystem version " + BT_SUBSYS_VERSION); send_monitor_replay(sk); atomic_inc(&monitor_promisc); -- cgit v1.2.3-59-g8ed1b From ac71494934c475e3f51e5e3e64a12f57618d82a4 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 8 Nov 2015 07:47:13 +0100 Subject: Bluetooth: Add support for controller specific logging To enable controller specific logging, the userspace daemon has to have the ability to log per controller. To facilitate this support, provide a dedicated logging channel. Messages in this channel will be included in the monitor queue and with that also forwarded to monitoring tools along with the actual hardware traces. All messages from the logging channel are timestamped and with that allow an easy correlation between userspace messages and hardware events. This will increase the ability to debug problems faster. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg --- include/net/bluetooth/hci_mon.h | 1 + include/net/bluetooth/hci_sock.h | 1 + net/bluetooth/hci_sock.c | 102 +++++++++++++++++++++++++++++++++++++++ 3 files changed, 104 insertions(+) (limited to 'include') diff --git a/include/net/bluetooth/hci_mon.h b/include/net/bluetooth/hci_mon.h index c91bb23eb29e..587d0131b349 100644 --- a/include/net/bluetooth/hci_mon.h +++ b/include/net/bluetooth/hci_mon.h @@ -44,6 +44,7 @@ struct hci_mon_hdr { #define HCI_MON_INDEX_INFO 10 #define HCI_MON_VENDOR_DIAG 11 #define HCI_MON_SYSTEM_NOTE 12 +#define HCI_MON_USER_LOGGING 13 struct hci_mon_new_index { __u8 type; diff --git a/include/net/bluetooth/hci_sock.h b/include/net/bluetooth/hci_sock.h index 9a46d665c1b5..8e9138acdae1 100644 --- a/include/net/bluetooth/hci_sock.h +++ b/include/net/bluetooth/hci_sock.h @@ -45,6 +45,7 @@ struct sockaddr_hci { #define HCI_CHANNEL_USER 1 #define HCI_CHANNEL_MONITOR 2 #define HCI_CHANNEL_CONTROL 3 +#define HCI_CHANNEL_LOGGING 4 struct hci_filter { unsigned long type_mask; diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 710265c35d16..41f579ba447b 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -906,6 +906,18 @@ static int hci_sock_bind(struct socket *sock, struct sockaddr *addr, atomic_inc(&monitor_promisc); break; + case HCI_CHANNEL_LOGGING: + if (haddr.hci_dev != HCI_DEV_NONE) { + err = -EINVAL; + goto done; + } + + if (!capable(CAP_NET_ADMIN)) { + err = -EPERM; + goto done; + } + break; + default: if (!hci_mgmt_chan_find(haddr.hci_channel)) { err = -EINVAL; @@ -1033,6 +1045,9 @@ static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, if (flags & MSG_OOB) return -EOPNOTSUPP; + if (hci_pi(sk)->channel == HCI_CHANNEL_LOGGING) + return -EOPNOTSUPP; + if (sk->sk_state == BT_CLOSED) return 0; @@ -1179,6 +1194,90 @@ done: return err; } +static int hci_logging_frame(struct sock *sk, struct msghdr *msg, int len) +{ + struct hci_mon_hdr *hdr; + struct sk_buff *skb; + struct hci_dev *hdev; + u16 index; + int err; + + /* The logging frame consists at minimum of the standard header, + * the priority byte, the ident length byte and at least one string + * terminator NUL byte. Anything shorter are invalid packets. + */ + if (len < sizeof(*hdr) + 3) + return -EINVAL; + + skb = bt_skb_send_alloc(sk, len, msg->msg_flags & MSG_DONTWAIT, &err); + if (!skb) + return err; + + if (memcpy_from_msg(skb_put(skb, len), msg, len)) { + err = -EFAULT; + goto drop; + } + + hdr = (void *)skb->data; + + if (__le16_to_cpu(hdr->len) != len - sizeof(*hdr)) { + err = -EINVAL; + goto drop; + } + + if (__le16_to_cpu(hdr->opcode) == 0x0000) { + __u8 priority = skb->data[sizeof(*hdr)]; + __u8 ident_len = skb->data[sizeof(*hdr) + 1]; + + /* Only the priorities 0-7 are valid and with that any other + * value results in an invalid packet. + * + * The priority byte is followed by an ident length byte and + * the NUL terminated ident string. Check that the ident + * length is not overflowing the packet and also that the + * ident string itself is NUL terminated. In case the ident + * length is zero, the length value actually doubles as NUL + * terminator identifier. + * + * The message follows the ident string (if present) and + * must be NUL terminated. Otherwise it is not a valid packet. + */ + if (priority > 7 || skb->data[len - 1] != 0x00 || + ident_len > len - sizeof(*hdr) - 3 || + skb->data[sizeof(*hdr) + ident_len + 1] != 0x00) { + err = -EINVAL; + goto drop; + } + } else { + err = -EINVAL; + goto drop; + } + + index = __le16_to_cpu(hdr->index); + + if (index != MGMT_INDEX_NONE) { + hdev = hci_dev_get(index); + if (!hdev) { + err = -ENODEV; + goto drop; + } + } else { + hdev = NULL; + } + + hdr->opcode = cpu_to_le16(HCI_MON_USER_LOGGING); + + hci_send_to_channel(HCI_CHANNEL_MONITOR, skb, HCI_SOCK_TRUSTED, NULL); + err = len; + + if (hdev) + hci_dev_put(hdev); + +drop: + kfree_skb(skb); + return err; +} + static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { @@ -1208,6 +1307,9 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, case HCI_CHANNEL_MONITOR: err = -EOPNOTSUPP; goto done; + case HCI_CHANNEL_LOGGING: + err = hci_logging_frame(sk, msg, len); + goto done; default: mutex_lock(&mgmt_chan_list_lock); chan = __hci_mgmt_chan_find(hci_pi(sk)->channel); -- cgit v1.2.3-59-g8ed1b From 030e7f8141a262e32dc064d7cf12377d769d45c2 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Tue, 10 Nov 2015 09:44:53 +0200 Subject: Bluetooth: Remove unnecessary call to hci_update_background_scan The hci_conn_params_clear_all() function is only called from hci_unregister_dev() at which point it's completely futile to try to do any LE scanning updates. Simply remove this unnecessary function call. At the same time we can make the function static since it's only accessed from within the same c-file. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 - net/bluetooth/hci_core.c | 4 +--- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 1878d0a96333..15e6a2bffc2b 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1036,7 +1036,6 @@ struct hci_conn_params *hci_conn_params_lookup(struct hci_dev *hdev, struct hci_conn_params *hci_conn_params_add(struct hci_dev *hdev, bdaddr_t *addr, u8 addr_type); void hci_conn_params_del(struct hci_dev *hdev, bdaddr_t *addr, u8 addr_type); -void hci_conn_params_clear_all(struct hci_dev *hdev); void hci_conn_params_clear_disabled(struct hci_dev *hdev); struct hci_conn_params *hci_pend_le_action_lookup(struct list_head *list, diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index bc97fc6de876..ea648e9913f9 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -3070,15 +3070,13 @@ void hci_conn_params_clear_disabled(struct hci_dev *hdev) } /* This function requires the caller holds hdev->lock */ -void hci_conn_params_clear_all(struct hci_dev *hdev) +static void hci_conn_params_clear_all(struct hci_dev *hdev) { struct hci_conn_params *params, *tmp; list_for_each_entry_safe(params, tmp, &hdev->le_conn_params, list) hci_conn_params_free(params); - hci_update_background_scan(hdev); - BT_DBG("All LE connection parameters were removed"); } -- cgit v1.2.3-59-g8ed1b From 2e93e53b8f86fb38a9a3c3bd08e539c40b3f8d89 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 08:11:17 +0200 Subject: Bluetooth: Run all background scan updates through req_workqueue Instead of firing off a simple async request queue all background scan updates through req_workqueue and use hci_req_sync() there to ensure that no two updates overlap with each other. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 ++ net/bluetooth/hci_request.c | 39 +++++++++++++++++---------------------- net/bluetooth/hci_request.h | 6 +++++- net/bluetooth/mgmt.c | 2 +- 4 files changed, 25 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 15e6a2bffc2b..c2ca6a58d1e0 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -327,6 +327,8 @@ struct hci_dev { struct work_struct cmd_work; struct work_struct tx_work; + struct work_struct bg_scan_update; + struct sk_buff_head rx_q; struct sk_buff_head raw_q; struct sk_buff_head cmd_q; diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index d48206277fe4..0adbb59ec2f0 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -731,28 +731,6 @@ void __hci_update_background_scan(struct hci_request *req) } } -static void update_background_scan_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - if (status) - BT_DBG("HCI request failed to update background scanning: " - "status 0x%2.2x", status); -} - -void hci_update_background_scan(struct hci_dev *hdev) -{ - int err; - struct hci_request req; - - hci_req_init(&req, hdev); - - __hci_update_background_scan(&req); - - err = hci_req_run(&req, update_background_scan_complete); - if (err && err != -ENODATA) - BT_ERR("Failed to run HCI request: err %d", err); -} - void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason) { @@ -846,10 +824,27 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason) return 0; } +static void update_bg_scan(struct hci_request *req, unsigned long opt) +{ + hci_dev_lock(req->hdev); + __hci_update_background_scan(req); + hci_dev_unlock(req->hdev); +} + +static void bg_scan_update(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + bg_scan_update); + + hci_req_sync(hdev, update_bg_scan, 0, HCI_CMD_TIMEOUT); +} + void hci_request_setup(struct hci_dev *hdev) { + INIT_WORK(&hdev->bg_scan_update, bg_scan_update); } void hci_request_cancel_all(struct hci_dev *hdev) { + cancel_work_sync(&hdev->bg_scan_update); } diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 9759b7175f8e..983e687fee22 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -64,12 +64,16 @@ void __hci_update_page_scan(struct hci_request *req); int hci_update_random_address(struct hci_request *req, bool require_privacy, u8 *own_addr_type); -void hci_update_background_scan(struct hci_dev *hdev); void __hci_update_background_scan(struct hci_request *req); int hci_abort_conn(struct hci_conn *conn, u8 reason); void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason); +static inline void hci_update_background_scan(struct hci_dev *hdev) +{ + queue_work(hdev->req_workqueue, &hdev->bg_scan_update); +} + void hci_request_setup(struct hci_dev *hdev); void hci_request_cancel_all(struct hci_dev *hdev); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 7f22119276f3..29c9fec814b4 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -2510,8 +2510,8 @@ static void le_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode) hci_req_init(&req, hdev); update_adv_data(&req); update_scan_rsp_data(&req); - __hci_update_background_scan(&req); hci_req_run(&req, NULL); + hci_update_background_scan(hdev); } unlock: -- cgit v1.2.3-59-g8ed1b From 4ebeee2dff9815619be6ff9a845d33716f48468c Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 08:11:19 +0200 Subject: Bluetooth: Add HCI status return parameter to hci_req_sync() In some cases it may be important to get the exact HCI status rather than the converted HCI-to-errno value. Add an optional return parameter to the hci_req_sync() API to allow for this. Since there are no good HCI translation candidates for cancelation and timeout, use the "unknown" status code for those cases. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 3 ++- net/bluetooth/hci_core.c | 26 +++++++++++++------------- net/bluetooth/hci_request.c | 12 +++++++++--- net/bluetooth/hci_request.h | 4 ++-- 4 files changed, 26 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 0205b80cc90b..cc2216727655 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -452,7 +452,8 @@ enum { #define HCI_ERROR_REMOTE_POWER_OFF 0x15 #define HCI_ERROR_LOCAL_HOST_TERM 0x16 #define HCI_ERROR_PAIRING_NOT_ALLOWED 0x18 -#define HCI_ERROR_INVALID_LL_PARAMS 0x1E +#define HCI_ERROR_INVALID_LL_PARAMS 0x1e +#define HCI_ERROR_UNSPECIFIED 0x1f #define HCI_ERROR_ADVERTISING_TIMEOUT 0x3c /* Flow control modes */ diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 965bc01a0d91..029d7798cffa 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -768,14 +768,14 @@ static int __hci_init(struct hci_dev *hdev) { int err; - err = __hci_req_sync(hdev, hci_init1_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init1_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; if (hci_dev_test_flag(hdev, HCI_SETUP)) hci_debugfs_create_basic(hdev); - err = __hci_req_sync(hdev, hci_init2_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init2_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -786,11 +786,11 @@ static int __hci_init(struct hci_dev *hdev) if (hdev->dev_type != HCI_BREDR) return 0; - err = __hci_req_sync(hdev, hci_init3_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init3_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; - err = __hci_req_sync(hdev, hci_init4_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init4_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -846,7 +846,7 @@ static int __hci_unconf_init(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RAW_DEVICE, &hdev->quirks)) return 0; - err = __hci_req_sync(hdev, hci_init0_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init0_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -1204,7 +1204,7 @@ int hci_inquiry(void __user *arg) if (do_inquiry) { err = hci_req_sync(hdev, hci_inq_req, (unsigned long) &ir, - timeo); + timeo, NULL); if (err < 0) goto done; @@ -1570,7 +1570,7 @@ int hci_dev_do_close(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RESET_ON_CLOSE, &hdev->quirks) && !auto_off && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { set_bit(HCI_INIT, &hdev->flags); - __hci_req_sync(hdev, hci_reset_req, 0, HCI_CMD_TIMEOUT); + __hci_req_sync(hdev, hci_reset_req, 0, HCI_CMD_TIMEOUT, NULL); clear_bit(HCI_INIT, &hdev->flags); } @@ -1667,7 +1667,7 @@ static int hci_dev_do_reset(struct hci_dev *hdev) atomic_set(&hdev->cmd_cnt, 1); hdev->acl_cnt = 0; hdev->sco_cnt = 0; hdev->le_cnt = 0; - ret = __hci_req_sync(hdev, hci_reset_req, 0, HCI_INIT_TIMEOUT); + ret = __hci_req_sync(hdev, hci_reset_req, 0, HCI_INIT_TIMEOUT, NULL); hci_req_sync_unlock(hdev); return ret; @@ -1802,7 +1802,7 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) switch (cmd) { case HCISETAUTH: err = hci_req_sync(hdev, hci_auth_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETENCRYPT: @@ -1814,18 +1814,18 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) if (!test_bit(HCI_AUTH, &hdev->flags)) { /* Auth must be enabled first */ err = hci_req_sync(hdev, hci_auth_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); if (err) break; } err = hci_req_sync(hdev, hci_encrypt_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETSCAN: err = hci_req_sync(hdev, hci_scan_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); /* Ensure that the connectable and discoverable states * get correctly modified as this was a non-mgmt change. @@ -1836,7 +1836,7 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) case HCISETLINKPOL: err = hci_req_sync(hdev, hci_linkpol_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETLINKMODE: diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 0adbb59ec2f0..b1d4d5bba7c1 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -186,7 +186,7 @@ EXPORT_SYMBOL(__hci_cmd_sync); /* Execute request and wait for completion. */ int __hci_req_sync(struct hci_dev *hdev, void (*func)(struct hci_request *req, unsigned long opt), - unsigned long opt, __u32 timeout) + unsigned long opt, u32 timeout, u8 *hci_status) { struct hci_request req; DECLARE_WAITQUEUE(wait, current); @@ -231,14 +231,20 @@ int __hci_req_sync(struct hci_dev *hdev, void (*func)(struct hci_request *req, switch (hdev->req_status) { case HCI_REQ_DONE: err = -bt_to_errno(hdev->req_result); + if (hci_status) + *hci_status = hdev->req_result; break; case HCI_REQ_CANCELED: err = -hdev->req_result; + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; break; default: err = -ETIMEDOUT; + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; break; } @@ -251,7 +257,7 @@ int __hci_req_sync(struct hci_dev *hdev, void (*func)(struct hci_request *req, int hci_req_sync(struct hci_dev *hdev, void (*req)(struct hci_request *req, unsigned long opt), - unsigned long opt, __u32 timeout) + unsigned long opt, u32 timeout, u8 *hci_status) { int ret; @@ -260,7 +266,7 @@ int hci_req_sync(struct hci_dev *hdev, void (*req)(struct hci_request *req, /* Serialize all requests */ hci_req_sync_lock(hdev); - ret = __hci_req_sync(hdev, req, opt, timeout); + ret = __hci_req_sync(hdev, req, opt, timeout, hci_status); hci_req_sync_unlock(hdev); return ret; diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 983e687fee22..8441d12a62dd 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -46,10 +46,10 @@ void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status, int hci_req_sync(struct hci_dev *hdev, void (*req)(struct hci_request *req, unsigned long opt), - unsigned long opt, __u32 timeout); + unsigned long opt, u32 timeout, u8 *hci_status); int __hci_req_sync(struct hci_dev *hdev, void (*func)(struct hci_request *req, unsigned long opt), - unsigned long opt, __u32 timeout); + unsigned long opt, u32 timeout, u8 *hci_status); void hci_req_sync_cancel(struct hci_dev *hdev, int err); struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, -- cgit v1.2.3-59-g8ed1b From 7c1fbed23981faff2840ddc8909e7c78d80ade30 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 08:11:23 +0200 Subject: Bluetooth: Move LE scan disable/restart behind req_workqueue To avoid any risks of races, place also these LE scan modification work callbacks behind the same work queue as the other LE scan changes. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 5 +- net/bluetooth/hci_core.c | 168 ------------------------------------ net/bluetooth/hci_request.c | 179 +++++++++++++++++++++++++++++++++++++++ net/bluetooth/mgmt.c | 4 +- 4 files changed, 183 insertions(+), 173 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index c2ca6a58d1e0..1f75aebbd8c4 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -328,6 +328,8 @@ struct hci_dev { struct work_struct tx_work; struct work_struct bg_scan_update; + struct delayed_work le_scan_disable; + struct delayed_work le_scan_restart; struct sk_buff_head rx_q; struct sk_buff_head raw_q; @@ -372,9 +374,6 @@ struct hci_dev { DECLARE_BITMAP(dev_flags, __HCI_NUM_FLAGS); - struct delayed_work le_scan_disable; - struct delayed_work le_scan_restart; - __s8 adv_tx_power; __u8 adv_data[HCI_MAX_AD_LENGTH]; __u8 adv_data_len; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 029d7798cffa..0655521dd8bc 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1527,9 +1527,6 @@ int hci_dev_do_close(struct hci_dev *hdev) if (hci_dev_test_and_clear_flag(hdev, HCI_SERVICE_CACHE)) cancel_delayed_work(&hdev->service_cache); - cancel_delayed_work_sync(&hdev->le_scan_disable); - cancel_delayed_work_sync(&hdev->le_scan_restart); - if (hci_dev_test_flag(hdev, HCI_MGMT)) cancel_delayed_work_sync(&hdev->rpa_expired); @@ -2889,169 +2886,6 @@ static void hci_conn_params_clear_all(struct hci_dev *hdev) BT_DBG("All LE connection parameters were removed"); } -static void inquiry_complete(struct hci_dev *hdev, u8 status, u16 opcode) -{ - if (status) { - BT_ERR("Failed to start inquiry: status %d", status); - - hci_dev_lock(hdev); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); - return; - } -} - -static void le_scan_disable_work_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - /* General inquiry access code (GIAC) */ - u8 lap[3] = { 0x33, 0x8b, 0x9e }; - struct hci_cp_inquiry cp; - int err; - - if (status) { - BT_ERR("Failed to disable LE scanning: status %d", status); - return; - } - - hdev->discovery.scan_start = 0; - - switch (hdev->discovery.type) { - case DISCOV_TYPE_LE: - hci_dev_lock(hdev); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); - break; - - case DISCOV_TYPE_INTERLEAVED: - hci_dev_lock(hdev); - - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, - &hdev->quirks)) { - /* If we were running LE only scan, change discovery - * state. If we were running both LE and BR/EDR inquiry - * simultaneously, and BR/EDR inquiry is already - * finished, stop discovery, otherwise BR/EDR inquiry - * will stop discovery when finished. If we will resolve - * remote device name, do not change discovery state. - */ - if (!test_bit(HCI_INQUIRY, &hdev->flags) && - hdev->discovery.state != DISCOVERY_RESOLVING) - hci_discovery_set_state(hdev, - DISCOVERY_STOPPED); - } else { - struct hci_request req; - - hci_inquiry_cache_flush(hdev); - - hci_req_init(&req, hdev); - - memset(&cp, 0, sizeof(cp)); - memcpy(&cp.lap, lap, sizeof(cp.lap)); - cp.length = DISCOV_INTERLEAVED_INQUIRY_LEN; - hci_req_add(&req, HCI_OP_INQUIRY, sizeof(cp), &cp); - - err = hci_req_run(&req, inquiry_complete); - if (err) { - BT_ERR("Inquiry request failed: err %d", err); - hci_discovery_set_state(hdev, - DISCOVERY_STOPPED); - } - } - - hci_dev_unlock(hdev); - break; - } -} - -static void le_scan_disable_work(struct work_struct *work) -{ - struct hci_dev *hdev = container_of(work, struct hci_dev, - le_scan_disable.work); - struct hci_request req; - int err; - - BT_DBG("%s", hdev->name); - - cancel_delayed_work_sync(&hdev->le_scan_restart); - - hci_req_init(&req, hdev); - - hci_req_add_le_scan_disable(&req); - - err = hci_req_run(&req, le_scan_disable_work_complete); - if (err) - BT_ERR("Disable LE scanning request failed: err %d", err); -} - -static void le_scan_restart_work_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - unsigned long timeout, duration, scan_start, now; - - BT_DBG("%s", hdev->name); - - if (status) { - BT_ERR("Failed to restart LE scan: status %d", status); - return; - } - - if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) || - !hdev->discovery.scan_start) - return; - - /* When the scan was started, hdev->le_scan_disable has been queued - * after duration from scan_start. During scan restart this job - * has been canceled, and we need to queue it again after proper - * timeout, to make sure that scan does not run indefinitely. - */ - duration = hdev->discovery.scan_duration; - scan_start = hdev->discovery.scan_start; - now = jiffies; - if (now - scan_start <= duration) { - int elapsed; - - if (now >= scan_start) - elapsed = now - scan_start; - else - elapsed = ULONG_MAX - scan_start + now; - - timeout = duration - elapsed; - } else { - timeout = 0; - } - queue_delayed_work(hdev->workqueue, - &hdev->le_scan_disable, timeout); -} - -static void le_scan_restart_work(struct work_struct *work) -{ - struct hci_dev *hdev = container_of(work, struct hci_dev, - le_scan_restart.work); - struct hci_request req; - struct hci_cp_le_set_scan_enable cp; - int err; - - BT_DBG("%s", hdev->name); - - /* If controller is not scanning we are done. */ - if (!hci_dev_test_flag(hdev, HCI_LE_SCAN)) - return; - - hci_req_init(&req, hdev); - - hci_req_add_le_scan_disable(&req); - - memset(&cp, 0, sizeof(cp)); - cp.enable = LE_SCAN_ENABLE; - cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; - hci_req_add(&req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp); - - err = hci_req_run(&req, le_scan_restart_work_complete); - if (err) - BT_ERR("Restart LE scan request failed: err %d", err); -} - /* Copy the Identity Address of the controller. * * If the controller has a public BD_ADDR, then by default use that one. @@ -3151,8 +2985,6 @@ struct hci_dev *hci_alloc_dev(void) INIT_DELAYED_WORK(&hdev->power_off, hci_power_off); INIT_DELAYED_WORK(&hdev->discov_off, hci_discov_off); - INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); - INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); INIT_DELAYED_WORK(&hdev->adv_instance_expire, hci_adv_timeout_expire); skb_queue_head_init(&hdev->rx_q); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 8aa06cc545c3..4588fe2bfc0e 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -858,12 +858,191 @@ static void bg_scan_update(struct work_struct *work) hci_dev_unlock(hdev); } +static void inquiry_complete(struct hci_dev *hdev, u8 status, u16 opcode) +{ + if (status) { + BT_ERR("Failed to start inquiry: status %d", status); + + hci_dev_lock(hdev); + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + hci_dev_unlock(hdev); + return; + } +} + +static void le_scan_disable_work_complete(struct hci_dev *hdev, u8 status) +{ + /* General inquiry access code (GIAC) */ + u8 lap[3] = { 0x33, 0x8b, 0x9e }; + struct hci_cp_inquiry cp; + int err; + + if (status) { + BT_ERR("Failed to disable LE scanning: status %d", status); + return; + } + + hdev->discovery.scan_start = 0; + + switch (hdev->discovery.type) { + case DISCOV_TYPE_LE: + hci_dev_lock(hdev); + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + hci_dev_unlock(hdev); + break; + + case DISCOV_TYPE_INTERLEAVED: + hci_dev_lock(hdev); + + if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, + &hdev->quirks)) { + /* If we were running LE only scan, change discovery + * state. If we were running both LE and BR/EDR inquiry + * simultaneously, and BR/EDR inquiry is already + * finished, stop discovery, otherwise BR/EDR inquiry + * will stop discovery when finished. If we will resolve + * remote device name, do not change discovery state. + */ + if (!test_bit(HCI_INQUIRY, &hdev->flags) && + hdev->discovery.state != DISCOVERY_RESOLVING) + hci_discovery_set_state(hdev, + DISCOVERY_STOPPED); + } else { + struct hci_request req; + + hci_inquiry_cache_flush(hdev); + + hci_req_init(&req, hdev); + + memset(&cp, 0, sizeof(cp)); + memcpy(&cp.lap, lap, sizeof(cp.lap)); + cp.length = DISCOV_INTERLEAVED_INQUIRY_LEN; + hci_req_add(&req, HCI_OP_INQUIRY, sizeof(cp), &cp); + + err = hci_req_run(&req, inquiry_complete); + if (err) { + BT_ERR("Inquiry request failed: err %d", err); + hci_discovery_set_state(hdev, + DISCOVERY_STOPPED); + } + } + + hci_dev_unlock(hdev); + break; + } +} + +static void le_scan_disable(struct hci_request *req, unsigned long opt) +{ + hci_req_add_le_scan_disable(req); +} + +static void le_scan_disable_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + le_scan_disable.work); + u8 status; + int err; + + BT_DBG("%s", hdev->name); + + cancel_delayed_work(&hdev->le_scan_restart); + + err = hci_req_sync(hdev, le_scan_disable, 0, HCI_CMD_TIMEOUT, &status); + if (err) + return; + + le_scan_disable_work_complete(hdev, status); +} + +static void le_scan_restart_work_complete(struct hci_dev *hdev, u8 status) +{ + unsigned long timeout, duration, scan_start, now; + + BT_DBG("%s", hdev->name); + + if (status) { + BT_ERR("Failed to restart LE scan: status %d", status); + return; + } + + hci_dev_lock(hdev); + + if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) || + !hdev->discovery.scan_start) + goto unlock; + + /* When the scan was started, hdev->le_scan_disable has been queued + * after duration from scan_start. During scan restart this job + * has been canceled, and we need to queue it again after proper + * timeout, to make sure that scan does not run indefinitely. + */ + duration = hdev->discovery.scan_duration; + scan_start = hdev->discovery.scan_start; + now = jiffies; + if (now - scan_start <= duration) { + int elapsed; + + if (now >= scan_start) + elapsed = now - scan_start; + else + elapsed = ULONG_MAX - scan_start + now; + + timeout = duration - elapsed; + } else { + timeout = 0; + } + + queue_delayed_work(hdev->req_workqueue, + &hdev->le_scan_disable, timeout); + +unlock: + hci_dev_unlock(hdev); +} + +static void le_scan_restart(struct hci_request *req, unsigned long opt) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_scan_enable cp; + + /* If controller is not scanning we are done. */ + if (!hci_dev_test_flag(hdev, HCI_LE_SCAN)) + return; + + hci_req_add_le_scan_disable(req); + + memset(&cp, 0, sizeof(cp)); + cp.enable = LE_SCAN_ENABLE; + cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; + hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp); +} + +static void le_scan_restart_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + le_scan_restart.work); + u8 status; + int err; + + BT_DBG("%s", hdev->name); + + err = hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status); + if (err) + return; + + le_scan_restart_work_complete(hdev, status); +} + void hci_request_setup(struct hci_dev *hdev) { INIT_WORK(&hdev->bg_scan_update, bg_scan_update); + INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); + INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); } void hci_request_cancel_all(struct hci_dev *hdev) { cancel_work_sync(&hdev->bg_scan_update); + cancel_delayed_work_sync(&hdev->le_scan_disable); + cancel_delayed_work_sync(&hdev->le_scan_restart); } diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index bb870c3aadae..a229cfd0530e 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -4367,7 +4367,7 @@ static void start_discovery_complete(struct hci_dev *hdev, u8 status, hdev->discovery.scan_duration = timeout; } - queue_delayed_work(hdev->workqueue, + queue_delayed_work(hdev->req_workqueue, &hdev->le_scan_disable, timeout); } @@ -8389,7 +8389,7 @@ static void restart_le_scan(struct hci_dev *hdev) hdev->discovery.scan_duration)) return; - queue_delayed_work(hdev->workqueue, &hdev->le_scan_restart, + queue_delayed_work(hdev->req_workqueue, &hdev->le_scan_restart, DISCOV_LE_RESTART_DELAY); } -- cgit v1.2.3-59-g8ed1b From e68f072b7396574df5324e1cf93e4b0c92460735 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 08:30:30 +0200 Subject: Bluetooth: Move Start Discovery to req_workqueue Since discovery also deals with LE scanning it makes sense to move it behind the same req_workqueue as other LE scanning changes. This also simplifies the logic since we do many of the actions in a synchronous manner. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 + net/bluetooth/hci_request.c | 202 +++++++++++++++++++++++++++++++++ net/bluetooth/mgmt.c | 238 ++------------------------------------- 3 files changed, 213 insertions(+), 229 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 1f75aebbd8c4..72ea8a6d7d70 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -327,6 +327,7 @@ struct hci_dev { struct work_struct cmd_work; struct work_struct tx_work; + struct work_struct discov_update; struct work_struct bg_scan_update; struct delayed_work le_scan_disable; struct delayed_work le_scan_restart; @@ -1473,6 +1474,7 @@ void mgmt_ssp_enable_complete(struct hci_dev *hdev, u8 enable, u8 status); void mgmt_set_class_of_dev_complete(struct hci_dev *hdev, u8 *dev_class, u8 status); void mgmt_set_local_name_complete(struct hci_dev *hdev, u8 *name, u8 status); +void mgmt_start_discovery_complete(struct hci_dev *hdev, u8 status); void mgmt_device_found(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type, u8 addr_type, u8 *dev_class, s8 rssi, u32 flags, u8 *eir, u16 eir_len, u8 *scan_rsp, u8 scan_rsp_len); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index ecfa4105e00d..da1e30b85e77 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1042,8 +1042,209 @@ static void le_scan_restart_work(struct work_struct *work) le_scan_restart_work_complete(hdev, status); } +static int bredr_inquiry(struct hci_request *req, unsigned long opt) +{ + struct hci_cp_inquiry cp; + /* General inquiry access code (GIAC) */ + u8 lap[3] = { 0x33, 0x8b, 0x9e }; + + BT_DBG("%s", req->hdev->name); + + hci_dev_lock(req->hdev); + hci_inquiry_cache_flush(req->hdev); + hci_dev_unlock(req->hdev); + + memset(&cp, 0, sizeof(cp)); + memcpy(&cp.lap, lap, sizeof(cp.lap)); + cp.length = DISCOV_BREDR_INQUIRY_LEN; + + hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); + + return 0; +} + +static void cancel_adv_timeout(struct hci_dev *hdev) +{ + if (hdev->adv_instance_timeout) { + hdev->adv_instance_timeout = 0; + cancel_delayed_work(&hdev->adv_instance_expire); + } +} + +static void disable_advertising(struct hci_request *req) +{ + u8 enable = 0x00; + + hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); +} + +static int active_scan(struct hci_request *req, unsigned long opt) +{ + uint16_t interval = opt; + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_scan_param param_cp; + struct hci_cp_le_set_scan_enable enable_cp; + u8 own_addr_type; + int err; + + BT_DBG("%s", hdev->name); + + if (hci_dev_test_flag(hdev, HCI_LE_ADV)) { + hci_dev_lock(hdev); + + /* Don't let discovery abort an outgoing connection attempt + * that's using directed advertising. + */ + if (hci_lookup_le_connect(hdev)) { + hci_dev_unlock(hdev); + return -EBUSY; + } + + cancel_adv_timeout(hdev); + hci_dev_unlock(hdev); + + disable_advertising(req); + } + + /* If controller is scanning, it means the background scanning is + * running. Thus, we should temporarily stop it in order to set the + * discovery scanning parameters. + */ + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) + hci_req_add_le_scan_disable(req); + + /* All active scans will be done with either a resolvable private + * address (when privacy feature has been enabled) or non-resolvable + * private address. + */ + err = hci_update_random_address(req, true, &own_addr_type); + if (err < 0) + own_addr_type = ADDR_LE_DEV_PUBLIC; + + memset(¶m_cp, 0, sizeof(param_cp)); + param_cp.type = LE_SCAN_ACTIVE; + param_cp.interval = cpu_to_le16(interval); + param_cp.window = cpu_to_le16(DISCOV_LE_SCAN_WIN); + param_cp.own_address_type = own_addr_type; + + hci_req_add(req, HCI_OP_LE_SET_SCAN_PARAM, sizeof(param_cp), + ¶m_cp); + + memset(&enable_cp, 0, sizeof(enable_cp)); + enable_cp.enable = LE_SCAN_ENABLE; + enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; + + hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(enable_cp), + &enable_cp); + + return 0; +} + +static int interleaved_discov(struct hci_request *req, unsigned long opt) +{ + int err; + + BT_DBG("%s", req->hdev->name); + + err = active_scan(req, opt); + if (err) + return err; + + return bredr_inquiry(req, opt); +} + +static void start_discovery(struct hci_dev *hdev, u8 *status) +{ + unsigned long timeout; + + BT_DBG("%s type %u", hdev->name, hdev->discovery.type); + + switch (hdev->discovery.type) { + case DISCOV_TYPE_BREDR: + if (!hci_dev_test_flag(hdev, HCI_INQUIRY)) + hci_req_sync(hdev, bredr_inquiry, 0, HCI_CMD_TIMEOUT, + status); + return; + case DISCOV_TYPE_INTERLEAVED: + /* When running simultaneous discovery, the LE scanning time + * should occupy the whole discovery time sine BR/EDR inquiry + * and LE scanning are scheduled by the controller. + * + * For interleaving discovery in comparison, BR/EDR inquiry + * and LE scanning are done sequentially with separate + * timeouts. + */ + if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, + &hdev->quirks)) { + timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); + /* During simultaneous discovery, we double LE scan + * interval. We must leave some time for the controller + * to do BR/EDR inquiry. + */ + hci_req_sync(hdev, interleaved_discov, + DISCOV_LE_SCAN_INT * 2, HCI_CMD_TIMEOUT, + status); + break; + } + + timeout = msecs_to_jiffies(hdev->discov_interleaved_timeout); + hci_req_sync(hdev, active_scan, DISCOV_LE_SCAN_INT, + HCI_CMD_TIMEOUT, status); + break; + case DISCOV_TYPE_LE: + timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); + hci_req_sync(hdev, active_scan, DISCOV_LE_SCAN_INT, + HCI_CMD_TIMEOUT, status); + break; + default: + *status = HCI_ERROR_UNSPECIFIED; + return; + } + + if (*status) + return; + + BT_DBG("%s timeout %u ms", hdev->name, jiffies_to_msecs(timeout)); + + /* When service discovery is used and the controller has a + * strict duplicate filter, it is important to remember the + * start and duration of the scan. This is required for + * restarting scanning during the discovery phase. + */ + if (test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) && + hdev->discovery.result_filtering) { + hdev->discovery.scan_start = jiffies; + hdev->discovery.scan_duration = timeout; + } + + queue_delayed_work(hdev->req_workqueue, &hdev->le_scan_disable, + timeout); +} + +static void discov_update(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + discov_update); + u8 status = 0; + + switch (hdev->discovery.state) { + case DISCOVERY_STARTING: + start_discovery(hdev, &status); + mgmt_start_discovery_complete(hdev, status); + if (status) + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + else + hci_discovery_set_state(hdev, DISCOVERY_FINDING); + break; + case DISCOVERY_STOPPED: + default: + return; + } +} + void hci_request_setup(struct hci_dev *hdev) { + INIT_WORK(&hdev->discov_update, discov_update); INIT_WORK(&hdev->bg_scan_update, bg_scan_update); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); @@ -1051,6 +1252,7 @@ void hci_request_setup(struct hci_dev *hdev) void hci_request_cancel_all(struct hci_dev *hdev) { + cancel_work_sync(&hdev->discov_update); cancel_work_sync(&hdev->bg_scan_update); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index e634b4d85249..63b099471c92 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -4164,145 +4164,9 @@ done: return err; } -static bool trigger_bredr_inquiry(struct hci_request *req, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_inquiry cp; - /* General inquiry access code (GIAC) */ - u8 lap[3] = { 0x33, 0x8b, 0x9e }; - - *status = mgmt_bredr_support(hdev); - if (*status) - return false; - - if (hci_dev_test_flag(hdev, HCI_INQUIRY)) { - *status = MGMT_STATUS_BUSY; - return false; - } - - hci_inquiry_cache_flush(hdev); - - memset(&cp, 0, sizeof(cp)); - memcpy(&cp.lap, lap, sizeof(cp.lap)); - cp.length = DISCOV_BREDR_INQUIRY_LEN; - - hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); - - return true; -} - -static bool trigger_le_scan(struct hci_request *req, u16 interval, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_le_set_scan_param param_cp; - struct hci_cp_le_set_scan_enable enable_cp; - u8 own_addr_type; - int err; - - *status = mgmt_le_support(hdev); - if (*status) - return false; - - if (hci_dev_test_flag(hdev, HCI_LE_ADV)) { - /* Don't let discovery abort an outgoing connection attempt - * that's using directed advertising. - */ - if (hci_lookup_le_connect(hdev)) { - *status = MGMT_STATUS_REJECTED; - return false; - } - - cancel_adv_timeout(hdev); - disable_advertising(req); - } - - /* If controller is scanning, it means the background scanning is - * running. Thus, we should temporarily stop it in order to set the - * discovery scanning parameters. - */ - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) - hci_req_add_le_scan_disable(req); - - /* All active scans will be done with either a resolvable private - * address (when privacy feature has been enabled) or non-resolvable - * private address. - */ - err = hci_update_random_address(req, true, &own_addr_type); - if (err < 0) { - *status = MGMT_STATUS_FAILED; - return false; - } - - memset(¶m_cp, 0, sizeof(param_cp)); - param_cp.type = LE_SCAN_ACTIVE; - param_cp.interval = cpu_to_le16(interval); - param_cp.window = cpu_to_le16(DISCOV_LE_SCAN_WIN); - param_cp.own_address_type = own_addr_type; - - hci_req_add(req, HCI_OP_LE_SET_SCAN_PARAM, sizeof(param_cp), - ¶m_cp); - - memset(&enable_cp, 0, sizeof(enable_cp)); - enable_cp.enable = LE_SCAN_ENABLE; - enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; - - hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(enable_cp), - &enable_cp); - - return true; -} - -static bool trigger_discovery(struct hci_request *req, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - - switch (hdev->discovery.type) { - case DISCOV_TYPE_BREDR: - if (!trigger_bredr_inquiry(req, status)) - return false; - break; - - case DISCOV_TYPE_INTERLEAVED: - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, - &hdev->quirks)) { - /* During simultaneous discovery, we double LE scan - * interval. We must leave some time for the controller - * to do BR/EDR inquiry. - */ - if (!trigger_le_scan(req, DISCOV_LE_SCAN_INT * 2, - status)) - return false; - - if (!trigger_bredr_inquiry(req, status)) - return false; - - return true; - } - - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { - *status = MGMT_STATUS_NOT_SUPPORTED; - return false; - } - /* fall through */ - - case DISCOV_TYPE_LE: - if (!trigger_le_scan(req, DISCOV_LE_SCAN_INT, status)) - return false; - break; - - default: - *status = MGMT_STATUS_INVALID_PARAMS; - return false; - } - - return true; -} - -static void start_discovery_complete(struct hci_dev *hdev, u8 status, - u16 opcode) +void mgmt_start_discovery_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; - unsigned long timeout; BT_DBG("status %d", status); @@ -4317,61 +4181,6 @@ static void start_discovery_complete(struct hci_dev *hdev, u8 status, mgmt_pending_remove(cmd); } - if (status) { - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - goto unlock; - } - - hci_discovery_set_state(hdev, DISCOVERY_FINDING); - - /* If the scan involves LE scan, pick proper timeout to schedule - * hdev->le_scan_disable that will stop it. - */ - switch (hdev->discovery.type) { - case DISCOV_TYPE_LE: - timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); - break; - case DISCOV_TYPE_INTERLEAVED: - /* When running simultaneous discovery, the LE scanning time - * should occupy the whole discovery time sine BR/EDR inquiry - * and LE scanning are scheduled by the controller. - * - * For interleaving discovery in comparison, BR/EDR inquiry - * and LE scanning are done sequentially with separate - * timeouts. - */ - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks)) - timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); - else - timeout = msecs_to_jiffies(hdev->discov_interleaved_timeout); - break; - case DISCOV_TYPE_BREDR: - timeout = 0; - break; - default: - BT_ERR("Invalid discovery type %d", hdev->discovery.type); - timeout = 0; - break; - } - - if (timeout) { - /* When service discovery is used and the controller has - * a strict duplicate filter, it is important to remember - * the start and duration of the scan. This is required - * for restarting scanning during the discovery phase. - */ - if (test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, - &hdev->quirks) && - hdev->discovery.result_filtering) { - hdev->discovery.scan_start = jiffies; - hdev->discovery.scan_duration = timeout; - } - - queue_delayed_work(hdev->req_workqueue, - &hdev->le_scan_disable, timeout); - } - -unlock: hci_dev_unlock(hdev); } @@ -4407,7 +4216,6 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, { struct mgmt_cp_start_discovery *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; u8 status; int err; @@ -4436,14 +4244,6 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, goto failed; } - cmd = mgmt_pending_add(sk, MGMT_OP_START_DISCOVERY, hdev, data, len); - if (!cmd) { - err = -ENOMEM; - goto failed; - } - - cmd->cmd_complete = generic_cmd_complete; - /* Clear the discovery filter first to free any previously * allocated memory for the UUID list. */ @@ -4452,22 +4252,17 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, hdev->discovery.type = cp->type; hdev->discovery.report_invalid_rssi = false; - hci_req_init(&req, hdev); - - if (!trigger_discovery(&req, &status)) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, - status, &cp->type, sizeof(cp->type)); - mgmt_pending_remove(cmd); + cmd = mgmt_pending_add(sk, MGMT_OP_START_DISCOVERY, hdev, data, len); + if (!cmd) { + err = -ENOMEM; goto failed; } - err = hci_req_run(&req, start_discovery_complete); - if (err < 0) { - mgmt_pending_remove(cmd); - goto failed; - } + cmd->cmd_complete = generic_cmd_complete; hci_discovery_set_state(hdev, DISCOVERY_STARTING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; failed: hci_dev_unlock(hdev); @@ -4486,7 +4281,6 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, { struct mgmt_cp_start_service_discovery *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; const u16 max_uuid_count = ((U16_MAX - sizeof(*cp)) / 16); u16 uuid_count, expected_len; u8 status; @@ -4574,23 +4368,9 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, } } - hci_req_init(&req, hdev); - - if (!trigger_discovery(&req, &status)) { - err = mgmt_cmd_complete(sk, hdev->id, - MGMT_OP_START_SERVICE_DISCOVERY, - status, &cp->type, sizeof(cp->type)); - mgmt_pending_remove(cmd); - goto failed; - } - - err = hci_req_run(&req, start_discovery_complete); - if (err < 0) { - mgmt_pending_remove(cmd); - goto failed; - } - hci_discovery_set_state(hdev, DISCOVERY_STARTING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; failed: hci_dev_unlock(hdev); -- cgit v1.2.3-59-g8ed1b From 2154d3f4fb83c812a161c4910948dd876997e111 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 08:30:45 +0200 Subject: Bluetooth: Move Stop Discovery to req_workqueue Since discovery also deals with LE scanning it makes sense to move it behind the same req_workqueue as other LE scanning changes. This also simplifies the logic since we do many of the actions in a synchronous manner. Part of this refactoring is moving hci_req_stop_discovery() to hci_request.c. At the same time the function receives support for properly handling the STOPPING state since that's the state we'll be in when stopping through the req_workqueue. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_request.c | 62 ++++++++++++++++++++++++++++++++++ net/bluetooth/hci_request.h | 3 ++ net/bluetooth/mgmt.c | 72 +++------------------------------------- 4 files changed, 71 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 72ea8a6d7d70..609f4a03899c 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1475,6 +1475,7 @@ void mgmt_set_class_of_dev_complete(struct hci_dev *hdev, u8 *dev_class, u8 status); void mgmt_set_local_name_complete(struct hci_dev *hdev, u8 *name, u8 status); void mgmt_start_discovery_complete(struct hci_dev *hdev, u8 status); +void mgmt_stop_discovery_complete(struct hci_dev *hdev, u8 status); void mgmt_device_found(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type, u8 addr_type, u8 *dev_class, s8 rssi, u32 flags, u8 *eir, u16 eir_len, u8 *scan_rsp, u8 scan_rsp_len); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index da1e30b85e77..3219ee66faad 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1221,6 +1221,62 @@ static void start_discovery(struct hci_dev *hdev, u8 *status) timeout); } +bool hci_req_stop_discovery(struct hci_request *req) +{ + struct hci_dev *hdev = req->hdev; + struct discovery_state *d = &hdev->discovery; + struct hci_cp_remote_name_req_cancel cp; + struct inquiry_entry *e; + bool ret = false; + + BT_DBG("%s state %u", hdev->name, hdev->discovery.state); + + if (d->state == DISCOVERY_FINDING || d->state == DISCOVERY_STOPPING) { + if (test_bit(HCI_INQUIRY, &hdev->flags)) + hci_req_add(req, HCI_OP_INQUIRY_CANCEL, 0, NULL); + + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { + cancel_delayed_work(&hdev->le_scan_disable); + hci_req_add_le_scan_disable(req); + } + + ret = true; + } else { + /* Passive scanning */ + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { + hci_req_add_le_scan_disable(req); + ret = true; + } + } + + /* No further actions needed for LE-only discovery */ + if (d->type == DISCOV_TYPE_LE) + return ret; + + if (d->state == DISCOVERY_RESOLVING || d->state == DISCOVERY_STOPPING) { + e = hci_inquiry_cache_lookup_resolve(hdev, BDADDR_ANY, + NAME_PENDING); + if (!e) + return ret; + + bacpy(&cp.bdaddr, &e->data.bdaddr); + hci_req_add(req, HCI_OP_REMOTE_NAME_REQ_CANCEL, sizeof(cp), + &cp); + ret = true; + } + + return ret; +} + +static int stop_discovery(struct hci_request *req, unsigned long opt) +{ + hci_dev_lock(req->hdev); + hci_req_stop_discovery(req); + hci_dev_unlock(req->hdev); + + return 0; +} + static void discov_update(struct work_struct *work) { struct hci_dev *hdev = container_of(work, struct hci_dev, @@ -1236,6 +1292,12 @@ static void discov_update(struct work_struct *work) else hci_discovery_set_state(hdev, DISCOVERY_FINDING); break; + case DISCOVERY_STOPPING: + hci_req_sync(hdev, stop_discovery, 0, HCI_CMD_TIMEOUT, &status); + mgmt_stop_discovery_complete(hdev, status); + if (!status) + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + break; case DISCOVERY_STOPPED: default: return; diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 1927013f5e67..6b9e59f7f7a9 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -58,6 +58,9 @@ struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, void hci_req_add_le_scan_disable(struct hci_request *req); void hci_req_add_le_passive_scan(struct hci_request *req); +/* Returns true if HCI commands were queued */ +bool hci_req_stop_discovery(struct hci_request *req); + void hci_update_page_scan(struct hci_dev *hdev); void __hci_update_page_scan(struct hci_request *req); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 63b099471c92..8cdacef6b108 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1416,49 +1416,6 @@ static void clean_up_hci_complete(struct hci_dev *hdev, u8 status, u16 opcode) } } -static bool hci_stop_discovery(struct hci_request *req) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_remote_name_req_cancel cp; - struct inquiry_entry *e; - - switch (hdev->discovery.state) { - case DISCOVERY_FINDING: - if (test_bit(HCI_INQUIRY, &hdev->flags)) - hci_req_add(req, HCI_OP_INQUIRY_CANCEL, 0, NULL); - - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { - cancel_delayed_work(&hdev->le_scan_disable); - hci_req_add_le_scan_disable(req); - } - - return true; - - case DISCOVERY_RESOLVING: - e = hci_inquiry_cache_lookup_resolve(hdev, BDADDR_ANY, - NAME_PENDING); - if (!e) - break; - - bacpy(&cp.bdaddr, &e->data.bdaddr); - hci_req_add(req, HCI_OP_REMOTE_NAME_REQ_CANCEL, sizeof(cp), - &cp); - - return true; - - default: - /* Passive scanning */ - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { - hci_req_add_le_scan_disable(req); - return true; - } - - break; - } - - return false; -} - static void advertising_added(struct sock *sk, struct hci_dev *hdev, u8 instance) { @@ -1636,7 +1593,7 @@ static int clean_up_hci_state(struct hci_dev *hdev) if (hci_dev_test_flag(hdev, HCI_LE_ADV)) disable_advertising(&req); - discov_stopped = hci_stop_discovery(&req); + discov_stopped = hci_req_stop_discovery(&req); list_for_each_entry(conn, &hdev->conn_hash.list, list) { /* 0x15 == Terminated due to Power Off */ @@ -4377,7 +4334,7 @@ failed: return err; } -static void stop_discovery_complete(struct hci_dev *hdev, u8 status, u16 opcode) +void mgmt_stop_discovery_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; @@ -4391,9 +4348,6 @@ static void stop_discovery_complete(struct hci_dev *hdev, u8 status, u16 opcode) mgmt_pending_remove(cmd); } - if (!status) - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); } @@ -4402,7 +4356,6 @@ static int stop_discovery(struct sock *sk, struct hci_dev *hdev, void *data, { struct mgmt_cp_stop_discovery *mgmt_cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; int err; BT_DBG("%s", hdev->name); @@ -4431,24 +4384,9 @@ static int stop_discovery(struct sock *sk, struct hci_dev *hdev, void *data, cmd->cmd_complete = generic_cmd_complete; - hci_req_init(&req, hdev); - - hci_stop_discovery(&req); - - err = hci_req_run(&req, stop_discovery_complete); - if (!err) { - hci_discovery_set_state(hdev, DISCOVERY_STOPPING); - goto unlock; - } - - mgmt_pending_remove(cmd); - - /* If no HCI commands were sent we're done */ - if (err == -ENODATA) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_STOP_DISCOVERY, 0, - &mgmt_cp->type, sizeof(mgmt_cp->type)); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - } + hci_discovery_set_state(hdev, DISCOVERY_STOPPING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; unlock: hci_dev_unlock(hdev); -- cgit v1.2.3-59-g8ed1b From 0ad06aa6a7682319bb1adcc187a1fa8db6b2da2c Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 11 Nov 2015 14:44:57 +0200 Subject: Bluetooth: Fix specifying role for LE connections The hci_connect_le_scan() is (as the name implies) a master/central role API, so it makes no sense in passing a role parameter to it. At the same time this patch also fixes the direct advertising support for LE L2CAP sockets where we now call the more appropriate hci_le_connect() API if slave/peripheral role is desired. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 +- net/bluetooth/hci_conn.c | 4 ++-- net/bluetooth/l2cap_core.c | 15 +++++++-------- net/bluetooth/mgmt.c | 3 +-- 4 files changed, 11 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 609f4a03899c..55ce209157b1 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -877,7 +877,7 @@ struct hci_chan *hci_chan_lookup_handle(struct hci_dev *hdev, __u16 handle); struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, u8 dst_type, u8 sec_level, - u16 conn_timeout, u8 role); + u16 conn_timeout); struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, u8 dst_type, u8 sec_level, u16 conn_timeout, u8 role); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 1ed1e153b3fa..673c2254935b 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -988,7 +988,7 @@ static int hci_explicit_conn_params_set(struct hci_dev *hdev, /* This function requires the caller holds hdev->lock */ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, u8 dst_type, u8 sec_level, - u16 conn_timeout, u8 role) + u16 conn_timeout) { struct hci_conn *conn; @@ -1018,7 +1018,7 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, BT_DBG("requesting refresh of dst_addr"); - conn = hci_conn_add(hdev, LE_LINK, dst, role); + conn = hci_conn_add(hdev, LE_LINK, dst, HCI_ROLE_MASTER); if (!conn) return ERR_PTR(-ENOMEM); diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 66e8b6ee19a5..139da8106b04 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -7113,8 +7113,6 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, chan->dcid = cid; if (bdaddr_type_is_le(dst_type)) { - u8 role; - /* Convert from L2CAP channel address type to HCI address type */ if (dst_type == BDADDR_LE_PUBLIC) @@ -7123,14 +7121,15 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, dst_type = ADDR_LE_DEV_RANDOM; if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - role = HCI_ROLE_SLAVE; + hcon = hci_connect_le(hdev, dst, dst_type, + chan->sec_level, + HCI_LE_CONN_TIMEOUT, + HCI_ROLE_SLAVE); else - role = HCI_ROLE_MASTER; + hcon = hci_connect_le_scan(hdev, dst, dst_type, + chan->sec_level, + HCI_LE_CONN_TIMEOUT); - hcon = hci_connect_le_scan(hdev, dst, dst_type, - chan->sec_level, - HCI_LE_CONN_TIMEOUT, - role); } else { u8 auth_type = l2cap_get_auth_type(chan); hcon = hci_connect_acl(hdev, dst, chan->sec_level, auth_type); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index e4ad0457547a..eca203e891d2 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -3518,8 +3518,7 @@ static int pair_device(struct sock *sk, struct hci_dev *hdev, void *data, conn = hci_connect_le_scan(hdev, &cp->addr.bdaddr, addr_type, sec_level, - HCI_LE_CONN_TIMEOUT, - HCI_ROLE_MASTER); + HCI_LE_CONN_TIMEOUT); } if (IS_ERR(conn)) { -- cgit v1.2.3-59-g8ed1b From a8acce6aa584aa731a2bed240bcd8dc955f01414 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Thu, 19 Nov 2015 12:53:21 +0100 Subject: ppp: remove PPPOX_ZOMBIE socket state PPPOX_ZOMBIE is never set anymore. Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- drivers/net/ppp/pppoe.c | 4 ++-- drivers/net/ppp/pppox.c | 2 +- include/linux/if_pppox.h | 1 - 3 files changed, 3 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c index 1dedfbf1d423..277e6827d7cd 100644 --- a/drivers/net/ppp/pppoe.c +++ b/drivers/net/ppp/pppoe.c @@ -311,7 +311,7 @@ static void pppoe_flush_dev(struct net_device *dev) lock_sock(sk); if (po->pppoe_dev == dev && - sk->sk_state & (PPPOX_CONNECTED | PPPOX_BOUND | PPPOX_ZOMBIE)) { + sk->sk_state & (PPPOX_CONNECTED | PPPOX_BOUND)) { pppox_unbind_sock(sk); sk->sk_state_change(sk); po->pppoe_dev = NULL; @@ -775,7 +775,7 @@ static int pppoe_ioctl(struct socket *sock, unsigned int cmd, struct pppox_sock *relay_po; err = -EBUSY; - if (sk->sk_state & (PPPOX_BOUND | PPPOX_ZOMBIE | PPPOX_DEAD)) + if (sk->sk_state & (PPPOX_BOUND | PPPOX_DEAD)) break; err = -ENOTCONN; diff --git a/drivers/net/ppp/pppox.c b/drivers/net/ppp/pppox.c index 0e1b30622477..0200de74eebc 100644 --- a/drivers/net/ppp/pppox.c +++ b/drivers/net/ppp/pppox.c @@ -58,7 +58,7 @@ void pppox_unbind_sock(struct sock *sk) { /* Clear connection to ppp device, if attached. */ - if (sk->sk_state & (PPPOX_BOUND | PPPOX_CONNECTED | PPPOX_ZOMBIE)) { + if (sk->sk_state & (PPPOX_BOUND | PPPOX_CONNECTED)) { ppp_unregister_channel(&pppox_sk(sk)->chan); sk->sk_state = PPPOX_DEAD; } diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h index b49cf923becc..ba7a9b0c7c57 100644 --- a/include/linux/if_pppox.h +++ b/include/linux/if_pppox.h @@ -91,7 +91,6 @@ enum { PPPOX_CONNECTED = 1, /* connection established ==TCP_ESTABLISHED */ PPPOX_BOUND = 2, /* bound to ppp device */ PPPOX_RELAY = 4, /* forwarding is enabled */ - PPPOX_ZOMBIE = 8, /* dead, but still bound to ppp device */ PPPOX_DEAD = 16 /* dead, useless, please clean me up!*/ }; -- cgit v1.2.3-59-g8ed1b From b11cfb5807e30333b36c02701382b820b7dcf0d5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 20 Nov 2015 15:55:52 -0500 Subject: cgroup: record ancestor IDs and reimplement cgroup_is_descendant() using it cgroup_is_descendant() currently walks up the hierarchy and compares each ancestor to the cgroup in question. While enough for cgroup core usages, this can't be used in hot paths to test cgroup membership. This patch adds cgroup->ancestor_ids[] which records the IDs of all ancestors including self and cgroup->level for the nesting level. This allows testing whether a given cgroup is a descendant of another in three finite steps - testing whether the two belong to the same hierarchy, whether the descendant candidate is at the same or a higher level than the ancestor and comparing the recorded ancestor_id at the matching level. cgroup_is_descendant() is accordingly reimplmented and made inline. Signed-off-by: Tejun Heo --- include/linux/cgroup-defs.h | 14 ++++++++++++++ include/linux/cgroup.h | 18 +++++++++++++++++- kernel/cgroup.c | 32 ++++++++++---------------------- 3 files changed, 41 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 60d44b26276d..504d8591b6d3 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -234,6 +234,14 @@ struct cgroup { */ int id; + /* + * The depth this cgroup is at. The root is at depth zero and each + * step down the hierarchy increments the level. This along with + * ancestor_ids[] can determine whether a given cgroup is a + * descendant of another without traversing the hierarchy. + */ + int level; + /* * Each non-empty css_set associated with this cgroup contributes * one to populated_cnt. All children with non-zero popuplated_cnt @@ -289,6 +297,9 @@ struct cgroup { /* used to schedule release agent */ struct work_struct release_agent_work; + + /* ids of the ancestors at each level including self */ + int ancestor_ids[]; }; /* @@ -308,6 +319,9 @@ struct cgroup_root { /* The root cgroup. Root is destroyed on its release. */ struct cgroup cgrp; + /* for cgrp->ancestor_ids[0] */ + int cgrp_ancestor_id_storage; + /* Number of cgroups in the hierarchy, used only for /proc/cgroups */ atomic_t nr_cgrps; diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 22e3754f89c5..b5ee2c4210f9 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -81,7 +81,6 @@ struct cgroup_subsys_state *cgroup_get_e_css(struct cgroup *cgroup, struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry, struct cgroup_subsys *ss); -bool cgroup_is_descendant(struct cgroup *cgrp, struct cgroup *ancestor); int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from); @@ -459,6 +458,23 @@ static inline struct cgroup *task_cgroup(struct task_struct *task, return task_css(task, subsys_id)->cgroup; } +/** + * cgroup_is_descendant - test ancestry + * @cgrp: the cgroup to be tested + * @ancestor: possible ancestor of @cgrp + * + * Test whether @cgrp is a descendant of @ancestor. It also returns %true + * if @cgrp == @ancestor. This function is safe to call as long as @cgrp + * and @ancestor are accessible. + */ +static inline bool cgroup_is_descendant(struct cgroup *cgrp, + struct cgroup *ancestor) +{ + if (cgrp->root != ancestor->root || cgrp->level < ancestor->level) + return false; + return cgrp->ancestor_ids[ancestor->level] == ancestor->id; +} + /* no synchronization, the result can only be used as a hint */ static inline bool cgroup_is_populated(struct cgroup *cgrp) { diff --git a/kernel/cgroup.c b/kernel/cgroup.c index f1603c153890..3190040792c8 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -459,25 +459,6 @@ struct cgroup_subsys_state *of_css(struct kernfs_open_file *of) } EXPORT_SYMBOL_GPL(of_css); -/** - * cgroup_is_descendant - test ancestry - * @cgrp: the cgroup to be tested - * @ancestor: possible ancestor of @cgrp - * - * Test whether @cgrp is a descendant of @ancestor. It also returns %true - * if @cgrp == @ancestor. This function is safe to call as long as @cgrp - * and @ancestor are accessible. - */ -bool cgroup_is_descendant(struct cgroup *cgrp, struct cgroup *ancestor) -{ - while (cgrp) { - if (cgrp == ancestor) - return true; - cgrp = cgroup_parent(cgrp); - } - return false; -} - static int notify_on_release(const struct cgroup *cgrp) { return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); @@ -1903,6 +1884,7 @@ static int cgroup_setup_root(struct cgroup_root *root, unsigned long ss_mask) if (ret < 0) goto out; root_cgrp->id = ret; + root_cgrp->ancestor_ids[0] = ret; ret = percpu_ref_init(&root_cgrp->self.refcnt, css_release, 0, GFP_KERNEL); @@ -4846,11 +4828,11 @@ err_free_css: static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, umode_t mode) { - struct cgroup *parent, *cgrp; + struct cgroup *parent, *cgrp, *tcgrp; struct cgroup_root *root; struct cgroup_subsys *ss; struct kernfs_node *kn; - int ssid, ret; + int level, ssid, ret; /* Do not accept '\n' to prevent making /proc//cgroup unparsable. */ @@ -4861,9 +4843,11 @@ static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, if (!parent) return -ENODEV; root = parent->root; + level = parent->level + 1; /* allocate the cgroup and its ID, 0 is reserved for the root */ - cgrp = kzalloc(sizeof(*cgrp), GFP_KERNEL); + cgrp = kzalloc(sizeof(*cgrp) + + sizeof(cgrp->ancestor_ids[0]) * (level + 1), GFP_KERNEL); if (!cgrp) { ret = -ENOMEM; goto out_unlock; @@ -4887,6 +4871,10 @@ static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, cgrp->self.parent = &parent->self; cgrp->root = root; + cgrp->level = level; + + for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) + cgrp->ancestor_ids[tcgrp->level] = tcgrp->id; if (notify_on_release(parent)) set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); -- cgit v1.2.3-59-g8ed1b From bd96f76a2454c6b97d70945902e30b4c31510678 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 20 Nov 2015 15:55:52 -0500 Subject: kernfs: implement kernfs_walk_and_get() Implement kernfs_walk_and_get() which is similar to kernfs_find_and_get() but can walk a path instead of just a name. v2: Use strlcpy() instead of strlen() + memcpy() as suggested by David. Signed-off-by: Tejun Heo Acked-by: Greg Kroah-Hartman Cc: David Miller --- fs/kernfs/dir.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ include/linux/kernfs.h | 12 ++++++++++++ 2 files changed, 58 insertions(+) (limited to 'include') diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c index 91e004518237..742bf4a230e8 100644 --- a/fs/kernfs/dir.c +++ b/fs/kernfs/dir.c @@ -694,6 +694,29 @@ static struct kernfs_node *kernfs_find_ns(struct kernfs_node *parent, return NULL; } +static struct kernfs_node *kernfs_walk_ns(struct kernfs_node *parent, + const unsigned char *path, + const void *ns) +{ + static char path_buf[PATH_MAX]; /* protected by kernfs_mutex */ + size_t len = strlcpy(path_buf, path, PATH_MAX); + char *p = path_buf; + char *name; + + lockdep_assert_held(&kernfs_mutex); + + if (len >= PATH_MAX) + return NULL; + + while ((name = strsep(&p, "/")) && parent) { + if (*name == '\0') + continue; + parent = kernfs_find_ns(parent, name, ns); + } + + return parent; +} + /** * kernfs_find_and_get_ns - find and get kernfs_node with the given name * @parent: kernfs_node to search under @@ -718,6 +741,29 @@ struct kernfs_node *kernfs_find_and_get_ns(struct kernfs_node *parent, } EXPORT_SYMBOL_GPL(kernfs_find_and_get_ns); +/** + * kernfs_walk_and_get_ns - find and get kernfs_node with the given path + * @parent: kernfs_node to search under + * @path: path to look for + * @ns: the namespace tag to use + * + * Look for kernfs_node with path @path under @parent and get a reference + * if found. This function may sleep and returns pointer to the found + * kernfs_node on success, %NULL on failure. + */ +struct kernfs_node *kernfs_walk_and_get_ns(struct kernfs_node *parent, + const char *path, const void *ns) +{ + struct kernfs_node *kn; + + mutex_lock(&kernfs_mutex); + kn = kernfs_walk_ns(parent, path, ns); + kernfs_get(kn); + mutex_unlock(&kernfs_mutex); + + return kn; +} + /** * kernfs_create_root - create a new kernfs hierarchy * @scops: optional syscall operations for the hierarchy diff --git a/include/linux/kernfs.h b/include/linux/kernfs.h index 5d4e9c4b821d..af51df35d749 100644 --- a/include/linux/kernfs.h +++ b/include/linux/kernfs.h @@ -274,6 +274,8 @@ void pr_cont_kernfs_path(struct kernfs_node *kn); struct kernfs_node *kernfs_get_parent(struct kernfs_node *kn); struct kernfs_node *kernfs_find_and_get_ns(struct kernfs_node *parent, const char *name, const void *ns); +struct kernfs_node *kernfs_walk_and_get_ns(struct kernfs_node *parent, + const char *path, const void *ns); void kernfs_get(struct kernfs_node *kn); void kernfs_put(struct kernfs_node *kn); @@ -350,6 +352,10 @@ static inline struct kernfs_node * kernfs_find_and_get_ns(struct kernfs_node *parent, const char *name, const void *ns) { return NULL; } +static inline struct kernfs_node * +kernfs_walk_and_get_ns(struct kernfs_node *parent, const char *path, + const void *ns) +{ return NULL; } static inline void kernfs_get(struct kernfs_node *kn) { } static inline void kernfs_put(struct kernfs_node *kn) { } @@ -430,6 +436,12 @@ kernfs_find_and_get(struct kernfs_node *kn, const char *name) return kernfs_find_and_get_ns(kn, name, NULL); } +static inline struct kernfs_node * +kernfs_walk_and_get(struct kernfs_node *kn, const char *path) +{ + return kernfs_walk_and_get_ns(kn, path, NULL); +} + static inline struct kernfs_node * kernfs_create_dir(struct kernfs_node *parent, const char *name, umode_t mode, void *priv) -- cgit v1.2.3-59-g8ed1b From 16af439645455fbf36984ca5e72f31073ee19ab7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 20 Nov 2015 15:55:52 -0500 Subject: cgroup: implement cgroup_get_from_path() and expose cgroup_put() Implement cgroup_get_from_path() using kernfs_walk_and_get() which obtains a default hierarchy cgroup from its path. This will be used to allow cgroup path based matching from outside cgroup proper - e.g. networking and perf. v2: Add EXPORT_SYMBOL_GPL(cgroup_get_from_path). Signed-off-by: Tejun Heo --- include/linux/cgroup.h | 7 +++++++ kernel/cgroup.c | 39 ++++++++++++++++++++++++++++++++++----- 2 files changed, 41 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index b5ee2c4210f9..4c3ffab81ba7 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -81,6 +81,8 @@ struct cgroup_subsys_state *cgroup_get_e_css(struct cgroup *cgroup, struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry, struct cgroup_subsys *ss); +struct cgroup *cgroup_get_from_path(const char *path); + int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from); @@ -351,6 +353,11 @@ static inline void css_put_many(struct cgroup_subsys_state *css, unsigned int n) percpu_ref_put_many(&css->refcnt, n); } +static inline void cgroup_put(struct cgroup *cgrp) +{ + css_put(&cgrp->self); +} + /** * task_css_set_check - obtain a task's css_set with extra access conditions * @task: the task to obtain css_set for diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 3190040792c8..3db5e8f5b702 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -434,11 +434,6 @@ static bool cgroup_tryget(struct cgroup *cgrp) return css_tryget(&cgrp->self); } -static void cgroup_put(struct cgroup *cgrp) -{ - css_put(&cgrp->self); -} - struct cgroup_subsys_state *of_css(struct kernfs_open_file *of) { struct cgroup *cgrp = of->kn->parent->priv; @@ -5753,6 +5748,40 @@ struct cgroup_subsys_state *css_from_id(int id, struct cgroup_subsys *ss) return id > 0 ? idr_find(&ss->css_idr, id) : NULL; } +/** + * cgroup_get_from_path - lookup and get a cgroup from its default hierarchy path + * @path: path on the default hierarchy + * + * Find the cgroup at @path on the default hierarchy, increment its + * reference count and return it. Returns pointer to the found cgroup on + * success, ERR_PTR(-ENOENT) if @path doens't exist and ERR_PTR(-ENOTDIR) + * if @path points to a non-directory. + */ +struct cgroup *cgroup_get_from_path(const char *path) +{ + struct kernfs_node *kn; + struct cgroup *cgrp; + + mutex_lock(&cgroup_mutex); + + kn = kernfs_walk_and_get(cgrp_dfl_root.cgrp.kn, path); + if (kn) { + if (kernfs_type(kn) == KERNFS_DIR) { + cgrp = kn->priv; + cgroup_get(cgrp); + } else { + cgrp = ERR_PTR(-ENOTDIR); + } + kernfs_put(kn); + } else { + cgrp = ERR_PTR(-ENOENT); + } + + mutex_unlock(&cgroup_mutex); + return cgrp; +} +EXPORT_SYMBOL_GPL(cgroup_get_from_path); + #ifdef CONFIG_CGROUP_DEBUG static struct cgroup_subsys_state * debug_css_alloc(struct cgroup_subsys_state *parent_css) -- cgit v1.2.3-59-g8ed1b From 40b25fe5dc57a6557b96241b75ae63dce716a487 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Thu, 19 Nov 2015 16:16:43 +0100 Subject: Bluetooth: Add support for Get Advertising Size Information command The Get Advertising Size Information command allows to retrieve size information for advertising data and scan response data fields depending on the selected flags. This is useful if applications want to know the available size ahead of time. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg --- include/net/bluetooth/mgmt.h | 13 ++++++++++ net/bluetooth/mgmt.c | 58 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 71 insertions(+) (limited to 'include') diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index b831242d48a4..af17774c9416 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -571,6 +571,19 @@ struct mgmt_rp_remove_advertising { __u8 instance; } __packed; +#define MGMT_OP_GET_ADV_SIZE_INFO 0x0040 +struct mgmt_cp_get_adv_size_info { + __u8 instance; + __le32 flags; +} __packed; +#define MGMT_GET_ADV_SIZE_INFO_SIZE 5 +struct mgmt_rp_get_adv_size_info { + __u8 instance; + __le32 flags; + __u8 max_adv_data_len; + __u8 max_scan_rsp_len; +} __packed; + #define MGMT_EV_CMD_COMPLETE 0x0001 struct mgmt_ev_cmd_complete { __le16 opcode; diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 05370e76feb0..dc8e428050d9 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -102,6 +102,7 @@ static const u16 mgmt_commands[] = { MGMT_OP_READ_ADV_FEATURES, MGMT_OP_ADD_ADVERTISING, MGMT_OP_REMOVE_ADVERTISING, + MGMT_OP_GET_ADV_SIZE_INFO, }; static const u16 mgmt_events[] = { @@ -7059,6 +7060,62 @@ unlock: return err; } +static u8 tlv_data_max_len(u32 adv_flags, bool is_adv_data) +{ + u8 max_len = HCI_MAX_AD_LENGTH; + + if (is_adv_data) { + if (adv_flags & (MGMT_ADV_FLAG_DISCOV | + MGMT_ADV_FLAG_LIMITED_DISCOV | + MGMT_ADV_FLAG_MANAGED_FLAGS)) + max_len -= 3; + + if (adv_flags & MGMT_ADV_FLAG_TX_POWER) + max_len -= 3; + } + + return max_len; +} + +static int get_adv_size_info(struct sock *sk, struct hci_dev *hdev, + void *data, u16 data_len) +{ + struct mgmt_cp_get_adv_size_info *cp = data; + struct mgmt_rp_get_adv_size_info rp; + u32 flags, supported_flags; + int err; + + BT_DBG("%s", hdev->name); + + if (!lmp_le_capable(hdev)) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_REJECTED); + + if (cp->instance < 1 || cp->instance > HCI_MAX_ADV_INSTANCES) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_INVALID_PARAMS); + + flags = __le32_to_cpu(cp->flags); + + /* The current implementation only supports a subset of the specified + * flags. + */ + supported_flags = get_supported_adv_flags(hdev); + if (flags & ~supported_flags) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_INVALID_PARAMS); + + rp.instance = cp->instance; + rp.flags = cp->flags; + rp.max_adv_data_len = tlv_data_max_len(flags, true); + rp.max_scan_rsp_len = tlv_data_max_len(flags, false); + + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_SUCCESS, &rp, sizeof(rp)); + + return err; +} + static const struct hci_mgmt_handler mgmt_handlers[] = { { NULL }, /* 0x0000 (no command) */ { read_version, MGMT_READ_VERSION_SIZE, @@ -7146,6 +7203,7 @@ static const struct hci_mgmt_handler mgmt_handlers[] = { { add_advertising, MGMT_ADD_ADVERTISING_SIZE, HCI_MGMT_VAR_LEN }, { remove_advertising, MGMT_REMOVE_ADVERTISING_SIZE }, + { get_adv_size_info, MGMT_GET_ADV_SIZE_INFO_SIZE }, }; void mgmt_index_added(struct hci_dev *hdev) -- cgit v1.2.3-59-g8ed1b From b811580d91e9c0945b0a923dcec3e10cce04ac30 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 19 Nov 2015 12:24:22 -0800 Subject: net: IPv6 fib lookup tracepoint Add tracepoint to show fib6 table lookups and result. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/trace/events/fib6.h | 76 +++++++++++++++++++++++++++++++++++++++++++++ net/core/net-traces.c | 4 +++ net/ipv6/route.c | 10 ++++++ 3 files changed, 90 insertions(+) create mode 100644 include/trace/events/fib6.h (limited to 'include') diff --git a/include/trace/events/fib6.h b/include/trace/events/fib6.h new file mode 100644 index 000000000000..4cf6bac4686d --- /dev/null +++ b/include/trace/events/fib6.h @@ -0,0 +1,76 @@ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM fib6 + +#if !defined(_TRACE_FIB6_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_FIB6_H + +#include +#include +#include +#include + +TRACE_EVENT(fib6_table_lookup, + + TP_PROTO(const struct net *net, const struct rt6_info *rt, + u32 tb_id, const struct flowi6 *flp), + + TP_ARGS(net, rt, tb_id, flp), + + TP_STRUCT__entry( + __field( u32, tb_id ) + + __field( int, oif ) + __field( int, iif ) + __field( __u8, tos ) + __field( __u8, scope ) + __field( __u8, flags ) + __array( __u8, src, 16 ) + __array( __u8, dst, 16 ) + + __dynamic_array( char, name, IFNAMSIZ ) + __array( __u8, gw, 16 ) + ), + + TP_fast_assign( + struct in6_addr *in6; + + __entry->tb_id = tb_id; + __entry->oif = flp->flowi6_oif; + __entry->iif = flp->flowi6_iif; + __entry->tos = flp->flowi6_tos; + __entry->scope = flp->flowi6_scope; + __entry->flags = flp->flowi6_flags; + + in6 = (struct in6_addr *)__entry->src; + *in6 = flp->saddr; + + in6 = (struct in6_addr *)__entry->dst; + *in6 = flp->daddr; + + if (rt->rt6i_idev) { + __assign_str(name, rt->rt6i_idev->dev->name); + } else { + __assign_str(name, ""); + } + if (rt == net->ipv6.ip6_null_entry) { + struct in6_addr in6_zero = {}; + + in6 = (struct in6_addr *)__entry->gw; + *in6 = in6_zero; + + } else if (rt) { + in6 = (struct in6_addr *)__entry->gw; + *in6 = rt->rt6i_gateway; + } + ), + + TP_printk("table %3u oif %d iif %d src %pI6c dst %pI6c tos %d scope %d flags %x ==> dev %s gw %pI6c", + __entry->tb_id, __entry->oif, __entry->iif, + __entry->src, __entry->dst, __entry->tos, __entry->scope, + __entry->flags, __get_str(name), __entry->gw) +); + +#endif /* _TRACE_FIB6_H */ + +/* This part must be outside protection */ +#include diff --git a/net/core/net-traces.c b/net/core/net-traces.c index adef015b2f41..92da5e4ceb4f 100644 --- a/net/core/net-traces.c +++ b/net/core/net-traces.c @@ -32,6 +32,10 @@ #include #include #include +#if IS_ENABLED(CONFIG_IPV6) +#include +EXPORT_TRACEPOINT_SYMBOL_GPL(fib6_table_lookup); +#endif EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 6f01fe122abd..89758be9c6a6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -62,6 +62,7 @@ #include #include #include +#include #include @@ -865,6 +866,9 @@ restart: } dst_use(&rt->dst, jiffies); read_unlock_bh(&table->tb6_lock); + + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); + return rt; } @@ -1078,6 +1082,8 @@ redo_rt6_select: read_unlock_bh(&table->tb6_lock); rt6_dst_from_metrics_check(rt); + + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; } else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) && !(rt->rt6i_flags & RTF_GATEWAY))) { @@ -1101,6 +1107,8 @@ redo_rt6_select: uncached_rt = net->ipv6.ip6_null_entry; dst_hold(&uncached_rt->dst); + + trace_fib6_table_lookup(net, uncached_rt, table->tb6_id, fl6); return uncached_rt; } else { @@ -1125,6 +1133,7 @@ redo_rt6_select: dst_release(&rt->dst); } + trace_fib6_table_lookup(net, pcpu_rt, table->tb6_id, fl6); return pcpu_rt; } @@ -1474,6 +1483,7 @@ out: read_unlock_bh(&table->tb6_lock); + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; }; -- cgit v1.2.3-59-g8ed1b From cc30c16344fc3a25153175c7eb9037b2136cd466 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Fri, 20 Nov 2015 03:56:23 +0100 Subject: net: dsa: Add support for a switch reset gpio Some boards have a gpio line tied to the switch reset pin. Allow this gpio to be retrieved from the device tree, and take the switch out of reset before performing the probe. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/dsa/dsa.txt | 3 +++ include/net/dsa.h | 8 ++++++++ net/dsa/dsa.c | 17 +++++++++++++++++ 3 files changed, 28 insertions(+) (limited to 'include') diff --git a/Documentation/devicetree/bindings/net/dsa/dsa.txt b/Documentation/devicetree/bindings/net/dsa/dsa.txt index 04e6bef3ac3f..5fdbbcdf8c4b 100644 --- a/Documentation/devicetree/bindings/net/dsa/dsa.txt +++ b/Documentation/devicetree/bindings/net/dsa/dsa.txt @@ -31,6 +31,8 @@ A switch child node has the following optional property: switch. Must be set if the switch can not detect the presence and/or size of a connected EEPROM, otherwise optional. +- reset-gpios : phandle and specifier to a gpio line connected to + reset pin of the switch chip. A switch may have multiple "port" children nodes @@ -114,6 +116,7 @@ Example: #size-cells = <0>; reg = <17 1>; /* MDIO address 17, switch 1 in tree */ mii-bus = <&mii_bus1>; + reset-gpios = <&gpio5 1 GPIO_ACTIVE_LOW>; switch1port0: port@0 { reg = <0>; diff --git a/include/net/dsa.h b/include/net/dsa.h index 82a4c6011173..3f23dd9d6a69 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -64,6 +65,13 @@ struct dsa_chip_data { * NULL if there is only one switch chip. */ s8 *rtable; + + /* + * A switch may have a GPIO line tied to its reset pin. Parse + * this from the device tree, and use it before performing + * switch soft reset. + */ + struct gpio_desc *reset; }; struct dsa_platform_data { diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 1eba07feb34a..0b5565f923cc 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include "dsa_priv.h" @@ -688,6 +689,9 @@ static int dsa_of_probe(struct device *dev) const char *port_name; int chip_index, port_index; const unsigned int *sw_addr, *port_reg; + int gpio; + enum of_gpio_flags of_flags; + unsigned long flags; u32 eeprom_len; int ret; @@ -766,6 +770,19 @@ static int dsa_of_probe(struct device *dev) put_device(cd->host_dev); cd->host_dev = &mdio_bus_switch->dev; } + gpio = of_get_named_gpio_flags(child, "reset-gpios", 0, + &of_flags); + if (gpio_is_valid(gpio)) { + flags = (of_flags == OF_GPIO_ACTIVE_LOW ? + GPIOF_ACTIVE_LOW : 0); + ret = devm_gpio_request_one(dev, gpio, flags, + "switch_reset"); + if (ret) + goto out_free_chip; + + cd->reset = gpio_to_desc(gpio); + gpiod_direction_output(cd->reset, 0); + } for_each_available_child_of_node(child, port) { port_reg = of_get_property(port, "reg", NULL); -- cgit v1.2.3-59-g8ed1b From dad1581944139bb104965340804d1fb4518aab2c Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Thu, 15 Oct 2015 07:55:59 +0200 Subject: netfilter: ebtables: use __u64 from linux/types.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compilation error: linux/netfilter_bridge/ebtables.h:38:2: error: unknown type name ‘uint64_t’ Signed-off-by: Mikko Rapeli Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter_bridge/ebtables.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/netfilter_bridge/ebtables.h b/include/uapi/linux/netfilter_bridge/ebtables.h index fd2ee501726d..e3cdf9f1a259 100644 --- a/include/uapi/linux/netfilter_bridge/ebtables.h +++ b/include/uapi/linux/netfilter_bridge/ebtables.h @@ -12,6 +12,8 @@ #ifndef _UAPI__LINUX_BRIDGE_EFF_H #define _UAPI__LINUX_BRIDGE_EFF_H +#include +#include #include #define EBT_TABLE_MAXNAMELEN 32 @@ -33,8 +35,8 @@ struct xt_match; struct xt_target; struct ebt_counter { - uint64_t pcnt; - uint64_t bcnt; + __u64 pcnt; + __u64 bcnt; }; struct ebt_replace { -- cgit v1.2.3-59-g8ed1b From 1ffad83dffd675cd742286ae82dca7d746cb0da8 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Thu, 15 Oct 2015 07:56:30 +0200 Subject: netfilter: fix include files for compilation Add missing header dependencies and other small changes so that each file compiles alone in userspace. Signed-off-by: Mikko Rapeli Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/ipset/ip_set_bitmap.h | 2 ++ include/uapi/linux/netfilter/ipset/ip_set_hash.h | 2 ++ include/uapi/linux/netfilter/ipset/ip_set_list.h | 2 ++ include/uapi/linux/netfilter/nf_conntrack_tuple_common.h | 3 +++ include/uapi/linux/netfilter/xt_HMARK.h | 1 + include/uapi/linux/netfilter/xt_RATEEST.h | 1 + include/uapi/linux/netfilter/xt_TEE.h | 2 ++ include/uapi/linux/netfilter/xt_TPROXY.h | 1 + include/uapi/linux/netfilter/xt_hashlimit.h | 1 + include/uapi/linux/netfilter/xt_ipvs.h | 1 + include/uapi/linux/netfilter/xt_mac.h | 2 ++ include/uapi/linux/netfilter/xt_osf.h | 2 ++ include/uapi/linux/netfilter/xt_physdev.h | 2 +- include/uapi/linux/netfilter/xt_policy.h | 2 ++ include/uapi/linux/netfilter/xt_rateest.h | 1 + include/uapi/linux/netfilter/xt_recent.h | 1 + include/uapi/linux/netfilter/xt_sctp.h | 12 ++++++------ include/uapi/linux/netfilter_arp/arp_tables.h | 1 + include/uapi/linux/netfilter_bridge.h | 1 + include/uapi/linux/netfilter_bridge/ebt_arp.h | 1 + include/uapi/linux/netfilter_bridge/ebt_arpreply.h | 2 ++ include/uapi/linux/netfilter_bridge/ebt_ip6.h | 1 + include/uapi/linux/netfilter_bridge/ebt_nat.h | 2 ++ include/uapi/linux/netfilter_ipv4/ip_tables.h | 1 + include/uapi/linux/netfilter_ipv6/ip6_tables.h | 1 + include/uapi/linux/netfilter_ipv6/ip6t_rt.h | 2 +- 26 files changed, 42 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/netfilter/ipset/ip_set_bitmap.h b/include/uapi/linux/netfilter/ipset/ip_set_bitmap.h index 6a2c038d1888..fd5024d26269 100644 --- a/include/uapi/linux/netfilter/ipset/ip_set_bitmap.h +++ b/include/uapi/linux/netfilter/ipset/ip_set_bitmap.h @@ -1,6 +1,8 @@ #ifndef _UAPI__IP_SET_BITMAP_H #define _UAPI__IP_SET_BITMAP_H +#include + /* Bitmap type specific error codes */ enum { /* The element is out of the range of the set */ diff --git a/include/uapi/linux/netfilter/ipset/ip_set_hash.h b/include/uapi/linux/netfilter/ipset/ip_set_hash.h index 352eeccdc7f2..82deeb883ac4 100644 --- a/include/uapi/linux/netfilter/ipset/ip_set_hash.h +++ b/include/uapi/linux/netfilter/ipset/ip_set_hash.h @@ -1,6 +1,8 @@ #ifndef _UAPI__IP_SET_HASH_H #define _UAPI__IP_SET_HASH_H +#include + /* Hash type specific error codes */ enum { /* Hash is full */ diff --git a/include/uapi/linux/netfilter/ipset/ip_set_list.h b/include/uapi/linux/netfilter/ipset/ip_set_list.h index a44efaa98213..84d430368266 100644 --- a/include/uapi/linux/netfilter/ipset/ip_set_list.h +++ b/include/uapi/linux/netfilter/ipset/ip_set_list.h @@ -1,6 +1,8 @@ #ifndef _UAPI__IP_SET_LIST_H #define _UAPI__IP_SET_LIST_H +#include + /* List type specific error codes */ enum { /* Set name to be added/deleted/tested does not exist. */ diff --git a/include/uapi/linux/netfilter/nf_conntrack_tuple_common.h b/include/uapi/linux/netfilter/nf_conntrack_tuple_common.h index 2f6bbc5b8125..a9c3834abdd4 100644 --- a/include/uapi/linux/netfilter/nf_conntrack_tuple_common.h +++ b/include/uapi/linux/netfilter/nf_conntrack_tuple_common.h @@ -1,6 +1,9 @@ #ifndef _NF_CONNTRACK_TUPLE_COMMON_H #define _NF_CONNTRACK_TUPLE_COMMON_H +#include +#include + enum ip_conntrack_dir { IP_CT_DIR_ORIGINAL, IP_CT_DIR_REPLY, diff --git a/include/uapi/linux/netfilter/xt_HMARK.h b/include/uapi/linux/netfilter/xt_HMARK.h index 826fc5807577..3fb48c8d8d78 100644 --- a/include/uapi/linux/netfilter/xt_HMARK.h +++ b/include/uapi/linux/netfilter/xt_HMARK.h @@ -2,6 +2,7 @@ #define XT_HMARK_H_ #include +#include enum { XT_HMARK_SADDR_MASK, diff --git a/include/uapi/linux/netfilter/xt_RATEEST.h b/include/uapi/linux/netfilter/xt_RATEEST.h index 6605e20ad8cf..ec1b57047e03 100644 --- a/include/uapi/linux/netfilter/xt_RATEEST.h +++ b/include/uapi/linux/netfilter/xt_RATEEST.h @@ -2,6 +2,7 @@ #define _XT_RATEEST_TARGET_H #include +#include struct xt_rateest_target_info { char name[IFNAMSIZ]; diff --git a/include/uapi/linux/netfilter/xt_TEE.h b/include/uapi/linux/netfilter/xt_TEE.h index 5c21d5c829af..01092023404b 100644 --- a/include/uapi/linux/netfilter/xt_TEE.h +++ b/include/uapi/linux/netfilter/xt_TEE.h @@ -1,6 +1,8 @@ #ifndef _XT_TEE_TARGET_H #define _XT_TEE_TARGET_H +#include + struct xt_tee_tginfo { union nf_inet_addr gw; char oif[16]; diff --git a/include/uapi/linux/netfilter/xt_TPROXY.h b/include/uapi/linux/netfilter/xt_TPROXY.h index 902043c2073f..8d693eefdc1f 100644 --- a/include/uapi/linux/netfilter/xt_TPROXY.h +++ b/include/uapi/linux/netfilter/xt_TPROXY.h @@ -2,6 +2,7 @@ #define _XT_TPROXY_H #include +#include /* TPROXY target is capable of marking the packet to perform * redirection. We can get rid of that whenever we get support for diff --git a/include/uapi/linux/netfilter/xt_hashlimit.h b/include/uapi/linux/netfilter/xt_hashlimit.h index cbfc43d1af68..6db90372f09c 100644 --- a/include/uapi/linux/netfilter/xt_hashlimit.h +++ b/include/uapi/linux/netfilter/xt_hashlimit.h @@ -2,6 +2,7 @@ #define _UAPI_XT_HASHLIMIT_H #include +#include /* timings are in milliseconds. */ #define XT_HASHLIMIT_SCALE 10000 diff --git a/include/uapi/linux/netfilter/xt_ipvs.h b/include/uapi/linux/netfilter/xt_ipvs.h index eff34ac18808..e03b9c31a39d 100644 --- a/include/uapi/linux/netfilter/xt_ipvs.h +++ b/include/uapi/linux/netfilter/xt_ipvs.h @@ -2,6 +2,7 @@ #define _XT_IPVS_H #include +#include enum { XT_IPVS_IPVS_PROPERTY = 1 << 0, /* all other options imply this one */ diff --git a/include/uapi/linux/netfilter/xt_mac.h b/include/uapi/linux/netfilter/xt_mac.h index b892cdc67e06..9a19a08a9181 100644 --- a/include/uapi/linux/netfilter/xt_mac.h +++ b/include/uapi/linux/netfilter/xt_mac.h @@ -1,6 +1,8 @@ #ifndef _XT_MAC_H #define _XT_MAC_H +#include + struct xt_mac_info { unsigned char srcaddr[ETH_ALEN]; int invert; diff --git a/include/uapi/linux/netfilter/xt_osf.h b/include/uapi/linux/netfilter/xt_osf.h index 5d66caeba3ee..e6159958b2fb 100644 --- a/include/uapi/linux/netfilter/xt_osf.h +++ b/include/uapi/linux/netfilter/xt_osf.h @@ -20,6 +20,8 @@ #define _XT_OSF_H #include +#include +#include #define MAXGENRELEN 32 diff --git a/include/uapi/linux/netfilter/xt_physdev.h b/include/uapi/linux/netfilter/xt_physdev.h index db7a2982e9c0..ccdde87da214 100644 --- a/include/uapi/linux/netfilter/xt_physdev.h +++ b/include/uapi/linux/netfilter/xt_physdev.h @@ -2,7 +2,7 @@ #define _UAPI_XT_PHYSDEV_H #include - +#include #define XT_PHYSDEV_OP_IN 0x01 #define XT_PHYSDEV_OP_OUT 0x02 diff --git a/include/uapi/linux/netfilter/xt_policy.h b/include/uapi/linux/netfilter/xt_policy.h index be8ead05c316..d8a9800dce61 100644 --- a/include/uapi/linux/netfilter/xt_policy.h +++ b/include/uapi/linux/netfilter/xt_policy.h @@ -2,6 +2,8 @@ #define _XT_POLICY_H #include +#include +#include #define XT_POLICY_MAX_ELEM 4 diff --git a/include/uapi/linux/netfilter/xt_rateest.h b/include/uapi/linux/netfilter/xt_rateest.h index d40a6196842a..13fe50d4e4b3 100644 --- a/include/uapi/linux/netfilter/xt_rateest.h +++ b/include/uapi/linux/netfilter/xt_rateest.h @@ -2,6 +2,7 @@ #define _XT_RATEEST_MATCH_H #include +#include enum xt_rateest_match_flags { XT_RATEEST_MATCH_INVERT = 1<<0, diff --git a/include/uapi/linux/netfilter/xt_recent.h b/include/uapi/linux/netfilter/xt_recent.h index 6ef36c113e89..955d562031cc 100644 --- a/include/uapi/linux/netfilter/xt_recent.h +++ b/include/uapi/linux/netfilter/xt_recent.h @@ -2,6 +2,7 @@ #define _LINUX_NETFILTER_XT_RECENT_H 1 #include +#include enum { XT_RECENT_CHECK = 1 << 0, diff --git a/include/uapi/linux/netfilter/xt_sctp.h b/include/uapi/linux/netfilter/xt_sctp.h index 29287be696a2..58ffcfb7978e 100644 --- a/include/uapi/linux/netfilter/xt_sctp.h +++ b/include/uapi/linux/netfilter/xt_sctp.h @@ -66,26 +66,26 @@ struct xt_sctp_info { #define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) \ __sctp_chunkmap_is_clear((chunkmap), ARRAY_SIZE(chunkmap)) -static inline bool +static inline _Bool __sctp_chunkmap_is_clear(const __u32 *chunkmap, unsigned int n) { unsigned int i; for (i = 0; i < n; ++i) if (chunkmap[i]) - return false; - return true; + return 0; + return 1; } #define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) \ __sctp_chunkmap_is_all_set((chunkmap), ARRAY_SIZE(chunkmap)) -static inline bool +static inline _Bool __sctp_chunkmap_is_all_set(const __u32 *chunkmap, unsigned int n) { unsigned int i; for (i = 0; i < n; ++i) if (chunkmap[i] != ~0U) - return false; - return true; + return 0; + return 1; } #endif /* _XT_SCTP_H_ */ diff --git a/include/uapi/linux/netfilter_arp/arp_tables.h b/include/uapi/linux/netfilter_arp/arp_tables.h index a5a86a4db6b3..ece3ad4eecda 100644 --- a/include/uapi/linux/netfilter_arp/arp_tables.h +++ b/include/uapi/linux/netfilter_arp/arp_tables.h @@ -11,6 +11,7 @@ #include #include +#include #include #include diff --git a/include/uapi/linux/netfilter_bridge.h b/include/uapi/linux/netfilter_bridge.h index a5eda6db8d79..514519b47651 100644 --- a/include/uapi/linux/netfilter_bridge.h +++ b/include/uapi/linux/netfilter_bridge.h @@ -4,6 +4,7 @@ /* bridge-specific defines for netfilter. */ +#include #include #include #include diff --git a/include/uapi/linux/netfilter_bridge/ebt_arp.h b/include/uapi/linux/netfilter_bridge/ebt_arp.h index 522f3e427f49..dd4df25330e8 100644 --- a/include/uapi/linux/netfilter_bridge/ebt_arp.h +++ b/include/uapi/linux/netfilter_bridge/ebt_arp.h @@ -2,6 +2,7 @@ #define __LINUX_BRIDGE_EBT_ARP_H #include +#include #define EBT_ARP_OPCODE 0x01 #define EBT_ARP_HTYPE 0x02 diff --git a/include/uapi/linux/netfilter_bridge/ebt_arpreply.h b/include/uapi/linux/netfilter_bridge/ebt_arpreply.h index 7e77896e1fbf..6fee3402e307 100644 --- a/include/uapi/linux/netfilter_bridge/ebt_arpreply.h +++ b/include/uapi/linux/netfilter_bridge/ebt_arpreply.h @@ -1,6 +1,8 @@ #ifndef __LINUX_BRIDGE_EBT_ARPREPLY_H #define __LINUX_BRIDGE_EBT_ARPREPLY_H +#include + struct ebt_arpreply_info { unsigned char mac[ETH_ALEN]; int target; diff --git a/include/uapi/linux/netfilter_bridge/ebt_ip6.h b/include/uapi/linux/netfilter_bridge/ebt_ip6.h index 42b889682721..a062f0ce95f9 100644 --- a/include/uapi/linux/netfilter_bridge/ebt_ip6.h +++ b/include/uapi/linux/netfilter_bridge/ebt_ip6.h @@ -13,6 +13,7 @@ #define __LINUX_BRIDGE_EBT_IP6_H #include +#include #define EBT_IP6_SOURCE 0x01 #define EBT_IP6_DEST 0x02 diff --git a/include/uapi/linux/netfilter_bridge/ebt_nat.h b/include/uapi/linux/netfilter_bridge/ebt_nat.h index 5e74e3b03bd6..c990d74ee966 100644 --- a/include/uapi/linux/netfilter_bridge/ebt_nat.h +++ b/include/uapi/linux/netfilter_bridge/ebt_nat.h @@ -1,6 +1,8 @@ #ifndef __LINUX_BRIDGE_EBT_NAT_H #define __LINUX_BRIDGE_EBT_NAT_H +#include + #define NAT_ARP_BIT (0x00000010) struct ebt_nat_info { unsigned char mac[ETH_ALEN]; diff --git a/include/uapi/linux/netfilter_ipv4/ip_tables.h b/include/uapi/linux/netfilter_ipv4/ip_tables.h index f1e6ef256034..d0da53d96d93 100644 --- a/include/uapi/linux/netfilter_ipv4/ip_tables.h +++ b/include/uapi/linux/netfilter_ipv4/ip_tables.h @@ -17,6 +17,7 @@ #include #include +#include #include #include diff --git a/include/uapi/linux/netfilter_ipv6/ip6_tables.h b/include/uapi/linux/netfilter_ipv6/ip6_tables.h index 649c68062dca..d1b22653daf2 100644 --- a/include/uapi/linux/netfilter_ipv6/ip6_tables.h +++ b/include/uapi/linux/netfilter_ipv6/ip6_tables.h @@ -17,6 +17,7 @@ #include #include +#include #include #include diff --git a/include/uapi/linux/netfilter_ipv6/ip6t_rt.h b/include/uapi/linux/netfilter_ipv6/ip6t_rt.h index 7605a5ff81cd..558f81e46fb9 100644 --- a/include/uapi/linux/netfilter_ipv6/ip6t_rt.h +++ b/include/uapi/linux/netfilter_ipv6/ip6t_rt.h @@ -2,7 +2,7 @@ #define _IP6T_RT_H #include -/*#include */ +#include #define IP6T_RT_HOPS 16 -- cgit v1.2.3-59-g8ed1b From f7ccdb96fa31305d480678b1ba81225907dd81ef Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 11 Nov 2015 20:17:37 -0200 Subject: netfilter: nf_ct_sctp: move ip_ct_sctp away from UAPI ip_ct_sctp is an internal structure, embedded by the union nf_conntrack_proto to store sctp-specific information at conntrack entries. It has no business with UAPI. This patch moves it from UAPI to a saner place, together with similar structs for other protocols. Signed-off-by: Marcelo Ricardo Leitner Acked-by: Neil Horman Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_sctp.h | 13 +++++++++++++ include/uapi/linux/netfilter/nf_conntrack_sctp.h | 12 +++--------- 2 files changed, 16 insertions(+), 9 deletions(-) create mode 100644 include/linux/netfilter/nf_conntrack_sctp.h (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h new file mode 100644 index 000000000000..22a16a23cd8a --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_sctp.h @@ -0,0 +1,13 @@ +#ifndef _NF_CONNTRACK_SCTP_H +#define _NF_CONNTRACK_SCTP_H +/* SCTP tracking. */ + +#include + +struct ip_ct_sctp { + enum sctp_conntrack state; + + __be32 vtag[IP_CT_DIR_MAX]; +}; + +#endif /* _NF_CONNTRACK_SCTP_H */ diff --git a/include/uapi/linux/netfilter/nf_conntrack_sctp.h b/include/uapi/linux/netfilter/nf_conntrack_sctp.h index ed4e776e1242..2cbc366c3fb4 100644 --- a/include/uapi/linux/netfilter/nf_conntrack_sctp.h +++ b/include/uapi/linux/netfilter/nf_conntrack_sctp.h @@ -1,5 +1,5 @@ -#ifndef _NF_CONNTRACK_SCTP_H -#define _NF_CONNTRACK_SCTP_H +#ifndef _UAPI_NF_CONNTRACK_SCTP_H +#define _UAPI_NF_CONNTRACK_SCTP_H /* SCTP tracking. */ #include @@ -18,10 +18,4 @@ enum sctp_conntrack { SCTP_CONNTRACK_MAX }; -struct ip_ct_sctp { - enum sctp_conntrack state; - - __be32 vtag[IP_CT_DIR_MAX]; -}; - -#endif /* _NF_CONNTRACK_SCTP_H */ +#endif /* _UAPI_NF_CONNTRACK_SCTP_H */ -- cgit v1.2.3-59-g8ed1b From 029f7f3b8701cc7aca8bdb31f0c7edd6a479e357 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 18 Nov 2015 23:32:39 +0100 Subject: netfilter: ipv6: nf_defrag: avoid/free clone operations commit 6aafeef03b9d9ecf ("netfilter: push reasm skb through instead of original frag skbs") changed ipv6 defrag to not use the original skbs anymore. So rather than keeping the original skbs around just to discard them afterwards just use the original skbs directly for the fraglist of the newly assembled skb and remove the extra clone/free operations. The skb that completes the fragment queue is morphed into a the reassembled one instead, just like ipv4 defrag. openvswitch doesn't need any additional skb_morph magic anymore to deal with this situation so just remove that. A followup patch can then also remove the NF_HOOK (re)invocation in the ipv6 netfilter defrag hook. Cc: Joe Stringer Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/ipv6/nf_defrag_ipv6.h | 1 - net/ipv6/netfilter/nf_conntrack_reasm.c | 105 +++++++++++----------------- net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 6 -- net/openvswitch/conntrack.c | 14 ---- 4 files changed, 40 insertions(+), 86 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/ipv6/nf_defrag_ipv6.h b/include/net/netfilter/ipv6/nf_defrag_ipv6.h index fb7da5bb76cc..fcd20cf8f5d5 100644 --- a/include/net/netfilter/ipv6/nf_defrag_ipv6.h +++ b/include/net/netfilter/ipv6/nf_defrag_ipv6.h @@ -6,7 +6,6 @@ void nf_defrag_ipv6_enable(void); int nf_ct_frag6_init(void); void nf_ct_frag6_cleanup(void); struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user); -void nf_ct_frag6_consume_orig(struct sk_buff *skb); struct inet_frags_ctl; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index d5efeb87350e..1a86a08adbe5 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -56,7 +56,6 @@ struct nf_ct_frag6_skb_cb { struct inet6_skb_parm h; int offset; - struct sk_buff *orig; }; #define NFCT_FRAG6_CB(skb) ((struct nf_ct_frag6_skb_cb *)((skb)->cb)) @@ -170,12 +169,6 @@ static unsigned int nf_hashfn(const struct inet_frag_queue *q) return nf_hash_frag(nq->id, &nq->saddr, &nq->daddr); } -static void nf_skb_free(struct sk_buff *skb) -{ - if (NFCT_FRAG6_CB(skb)->orig) - kfree_skb(NFCT_FRAG6_CB(skb)->orig); -} - static void nf_ct_frag6_expire(unsigned long data) { struct frag_queue *fq; @@ -376,9 +369,9 @@ err: * the last and the first frames arrived and all the bits are here. */ static struct sk_buff * -nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) +nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_device *dev) { - struct sk_buff *fp, *op, *head = fq->q.fragments; + struct sk_buff *fp, *head = fq->q.fragments; int payload_len; u8 ecn; @@ -429,10 +422,38 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) clone->csum = 0; clone->ip_summed = head->ip_summed; - NFCT_FRAG6_CB(clone)->orig = NULL; add_frag_mem_limit(fq->q.net, clone->truesize); } + /* morph head into last received skb: prev. + * + * This allows callers of ipv6 conntrack defrag to continue + * to use the last skb(frag) passed into the reasm engine. + * The last skb frag 'silently' turns into the full reassembled skb. + * + * Since prev is also part of q->fragments we have to clone it first. + */ + if (head != prev) { + struct sk_buff *iter; + + fp = skb_clone(prev, GFP_ATOMIC); + if (!fp) + goto out_oom; + + fp->next = prev->next; + skb_queue_walk(head, iter) { + if (iter->next != prev) + continue; + iter->next = fp; + break; + } + + skb_morph(prev, head); + prev->next = head->next; + consume_skb(head); + head = prev; + } + /* We have to remove fragment header from datagram and to relocate * header in order to calculate ICV correctly. */ skb_network_header(head)[fq->nhoffset] = skb_transport_header(head)[0]; @@ -473,21 +494,6 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) fq->q.fragments = NULL; fq->q.fragments_tail = NULL; - /* all original skbs are linked into the NFCT_FRAG6_CB(head).orig */ - fp = skb_shinfo(head)->frag_list; - if (fp && NFCT_FRAG6_CB(fp)->orig == NULL) - /* at above code, head skb is divided into two skbs. */ - fp = fp->next; - - op = NFCT_FRAG6_CB(head)->orig; - for (; fp; fp = fp->next) { - struct sk_buff *orig = NFCT_FRAG6_CB(fp)->orig; - - op->next = orig; - op = orig; - NFCT_FRAG6_CB(fp)->orig = NULL; - } - return head; out_oversize: @@ -565,7 +571,6 @@ find_prev_fhdr(struct sk_buff *skb, u8 *prevhdrp, int *prevhoff, int *fhoff) struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user) { - struct sk_buff *clone; struct net_device *dev = skb->dev; struct frag_hdr *fhdr; struct frag_queue *fq; @@ -583,42 +588,30 @@ struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 use if (find_prev_fhdr(skb, &prevhdr, &nhoff, &fhoff) < 0) return skb; - clone = skb_clone(skb, GFP_ATOMIC); - if (clone == NULL) { - pr_debug("Can't clone skb\n"); + if (!pskb_may_pull(skb, fhoff + sizeof(*fhdr))) return skb; - } - NFCT_FRAG6_CB(clone)->orig = skb; - - if (!pskb_may_pull(clone, fhoff + sizeof(*fhdr))) { - pr_debug("message is too short.\n"); - goto ret_orig; - } - - skb_set_transport_header(clone, fhoff); - hdr = ipv6_hdr(clone); - fhdr = (struct frag_hdr *)skb_transport_header(clone); + skb_set_transport_header(skb, fhoff); + hdr = ipv6_hdr(skb); + fhdr = (struct frag_hdr *)skb_transport_header(skb); fq = fq_find(net, fhdr->identification, user, &hdr->saddr, &hdr->daddr, ip6_frag_ecn(hdr)); - if (fq == NULL) { - pr_debug("Can't find and can't create new queue\n"); - goto ret_orig; - } + if (fq == NULL) + return skb; spin_lock_bh(&fq->q.lock); - if (nf_ct_frag6_queue(fq, clone, fhdr, nhoff) < 0) { + if (nf_ct_frag6_queue(fq, skb, fhdr, nhoff) < 0) { spin_unlock_bh(&fq->q.lock); pr_debug("Can't insert skb to queue\n"); inet_frag_put(&fq->q, &nf_frags); - goto ret_orig; + return skb; } if (fq->q.flags == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && fq->q.meat == fq->q.len) { - ret_skb = nf_ct_frag6_reasm(fq, dev); + ret_skb = nf_ct_frag6_reasm(fq, skb, dev); if (ret_skb == NULL) pr_debug("Can't reassemble fragmented packets\n"); } @@ -626,26 +619,9 @@ struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 use inet_frag_put(&fq->q, &nf_frags); return ret_skb; - -ret_orig: - kfree_skb(clone); - return skb; } EXPORT_SYMBOL_GPL(nf_ct_frag6_gather); -void nf_ct_frag6_consume_orig(struct sk_buff *skb) -{ - struct sk_buff *s, *s2; - - for (s = NFCT_FRAG6_CB(skb)->orig; s;) { - s2 = s->next; - s->next = NULL; - consume_skb(s); - s = s2; - } -} -EXPORT_SYMBOL_GPL(nf_ct_frag6_consume_orig); - static int nf_ct_net_init(struct net *net) { int res; @@ -680,7 +656,6 @@ int nf_ct_frag6_init(void) nf_frags.hashfn = nf_hashfn; nf_frags.constructor = ip6_frag_init; nf_frags.destructor = NULL; - nf_frags.skb_free = nf_skb_free; nf_frags.qsize = sizeof(struct frag_queue); nf_frags.match = ip6_frag_match; nf_frags.frag_expire = nf_ct_frag6_expire; diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index 4fdbed5ebfb6..fb96b1018884 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -69,12 +69,6 @@ static unsigned int ipv6_defrag(void *priv, if (reasm == NULL) return NF_STOLEN; - /* error occurred or not fragmented */ - if (reasm == skb) - return NF_ACCEPT; - - nf_ct_frag6_consume_orig(reasm); - NF_HOOK_THRESH(NFPROTO_IPV6, state->hook, state->net, state->sk, reasm, state->in, state->out, state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1); diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index c2cc11168fd5..cac2169f2909 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -321,21 +321,7 @@ static int handle_fragments(struct net *net, struct sw_flow_key *key, if (!reasm) return -EINPROGRESS; - if (skb == reasm) { - kfree_skb(skb); - return -EINVAL; - } - - /* Don't free 'skb' even though it is one of the original - * fragments, as we're going to morph it into the head. - */ - skb_get(skb); - nf_ct_frag6_consume_orig(reasm); - key->ip.proto = ipv6_hdr(reasm)->nexthdr; - skb_morph(skb, reasm); - skb->next = reasm->next; - consume_skb(reasm); ovs_cb.mru = IP6CB(skb)->frag_max_size; #endif } else { -- cgit v1.2.3-59-g8ed1b From daaa7d647f81f3f1494d9a9029d611b666d63181 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 18 Nov 2015 23:32:40 +0100 Subject: netfilter: ipv6: avoid nf_iterate recursion The previous patch changed nf_ct_frag6_gather() to morph reassembled skb with the previous one. This means that the return value is always NULL or the skb argument. So change it to an err value. Instead of invoking NF_HOOK recursively with threshold to skip already-called hooks we can now just return NF_ACCEPT to move on to the next hook except for -EINPROGRESS (which means skb has been queued for reassembly), in which case we return NF_STOLEN. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/ipv6/nf_defrag_ipv6.h | 2 +- net/ipv6/netfilter/nf_conntrack_reasm.c | 71 +++++++++++++---------------- net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 14 ++---- net/openvswitch/conntrack.c | 11 ++--- 4 files changed, 42 insertions(+), 56 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/ipv6/nf_defrag_ipv6.h b/include/net/netfilter/ipv6/nf_defrag_ipv6.h index fcd20cf8f5d5..ddf162f7966f 100644 --- a/include/net/netfilter/ipv6/nf_defrag_ipv6.h +++ b/include/net/netfilter/ipv6/nf_defrag_ipv6.h @@ -5,7 +5,7 @@ void nf_defrag_ipv6_enable(void); int nf_ct_frag6_init(void); void nf_ct_frag6_cleanup(void); -struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user); +int nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user); struct inet_frags_ctl; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 1a86a08adbe5..912bc3afc183 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -361,14 +361,15 @@ err: /* * Check if this packet is complete. - * Returns NULL on failure by any reason, and pointer - * to current nexthdr field in reassembled frame. * * It is called with locked fq, and caller must check that * queue is eligible for reassembly i.e. it is not COMPLETE, * the last and the first frames arrived and all the bits are here. + * + * returns true if *prev skb has been transformed into the reassembled + * skb, false otherwise. */ -static struct sk_buff * +static bool nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_device *dev) { struct sk_buff *fp, *head = fq->q.fragments; @@ -382,22 +383,21 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_devic ecn = ip_frag_ecn_table[fq->ecn]; if (unlikely(ecn == 0xff)) - goto out_fail; + return false; /* Unfragmented part is taken from the first segment. */ payload_len = ((head->data - skb_network_header(head)) - sizeof(struct ipv6hdr) + fq->q.len - sizeof(struct frag_hdr)); if (payload_len > IPV6_MAXPLEN) { - pr_debug("payload len is too large.\n"); - goto out_oversize; + net_dbg_ratelimited("nf_ct_frag6_reasm: payload len = %d\n", + payload_len); + return false; } /* Head of list must not be cloned. */ - if (skb_unclone(head, GFP_ATOMIC)) { - pr_debug("skb is cloned but can't expand head"); - goto out_oom; - } + if (skb_unclone(head, GFP_ATOMIC)) + return false; /* If the first fragment is fragmented itself, we split * it to two chunks: the first with data and paged part @@ -408,7 +408,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_devic clone = alloc_skb(0, GFP_ATOMIC); if (clone == NULL) - goto out_oom; + return false; clone->next = head->next; head->next = clone; @@ -438,7 +438,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_devic fp = skb_clone(prev, GFP_ATOMIC); if (!fp) - goto out_oom; + return false; fp->next = prev->next; skb_queue_walk(head, iter) { @@ -494,16 +494,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct sk_buff *prev, struct net_devic fq->q.fragments = NULL; fq->q.fragments_tail = NULL; - return head; - -out_oversize: - net_dbg_ratelimited("nf_ct_frag6_reasm: payload len = %d\n", - payload_len); - goto out_fail; -out_oom: - net_dbg_ratelimited("nf_ct_frag6_reasm: no memory for reassembly\n"); -out_fail: - return NULL; + return true; } /* @@ -569,27 +560,26 @@ find_prev_fhdr(struct sk_buff *skb, u8 *prevhdrp, int *prevhoff, int *fhoff) return 0; } -struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user) +int nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user) { struct net_device *dev = skb->dev; + int fhoff, nhoff, ret; struct frag_hdr *fhdr; struct frag_queue *fq; struct ipv6hdr *hdr; - int fhoff, nhoff; u8 prevhdr; - struct sk_buff *ret_skb = NULL; /* Jumbo payload inhibits frag. header */ if (ipv6_hdr(skb)->payload_len == 0) { pr_debug("payload len = 0\n"); - return skb; + return -EINVAL; } if (find_prev_fhdr(skb, &prevhdr, &nhoff, &fhoff) < 0) - return skb; + return -EINVAL; if (!pskb_may_pull(skb, fhoff + sizeof(*fhdr))) - return skb; + return -ENOMEM; skb_set_transport_header(skb, fhoff); hdr = ipv6_hdr(skb); @@ -598,27 +588,28 @@ struct sk_buff *nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 use fq = fq_find(net, fhdr->identification, user, &hdr->saddr, &hdr->daddr, ip6_frag_ecn(hdr)); if (fq == NULL) - return skb; + return -ENOMEM; spin_lock_bh(&fq->q.lock); if (nf_ct_frag6_queue(fq, skb, fhdr, nhoff) < 0) { - spin_unlock_bh(&fq->q.lock); - pr_debug("Can't insert skb to queue\n"); - inet_frag_put(&fq->q, &nf_frags); - return skb; + ret = -EINVAL; + goto out_unlock; } + /* after queue has assumed skb ownership, only 0 or -EINPROGRESS + * must be returned. + */ + ret = -EINPROGRESS; if (fq->q.flags == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && - fq->q.meat == fq->q.len) { - ret_skb = nf_ct_frag6_reasm(fq, skb, dev); - if (ret_skb == NULL) - pr_debug("Can't reassemble fragmented packets\n"); - } - spin_unlock_bh(&fq->q.lock); + fq->q.meat == fq->q.len && + nf_ct_frag6_reasm(fq, skb, dev)) + ret = 0; +out_unlock: + spin_unlock_bh(&fq->q.lock); inet_frag_put(&fq->q, &nf_frags); - return ret_skb; + return ret; } EXPORT_SYMBOL_GPL(nf_ct_frag6_gather); diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index fb96b1018884..f7aab5ab93a5 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -55,7 +55,7 @@ static unsigned int ipv6_defrag(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) { - struct sk_buff *reasm; + int err; #if IS_ENABLED(CONFIG_NF_CONNTRACK) /* Previously seen (loopback)? */ @@ -63,17 +63,13 @@ static unsigned int ipv6_defrag(void *priv, return NF_ACCEPT; #endif - reasm = nf_ct_frag6_gather(state->net, skb, - nf_ct6_defrag_user(state->hook, skb)); + err = nf_ct_frag6_gather(state->net, skb, + nf_ct6_defrag_user(state->hook, skb)); /* queued */ - if (reasm == NULL) + if (err == -EINPROGRESS) return NF_STOLEN; - NF_HOOK_THRESH(NFPROTO_IPV6, state->hook, state->net, state->sk, reasm, - state->in, state->out, - state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1); - - return NF_STOLEN; + return NF_ACCEPT; } static struct nf_hook_ops ipv6_defrag_ops[] = { diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index cac2169f2909..0c68c8e46d0b 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -300,10 +300,10 @@ static int handle_fragments(struct net *net, struct sw_flow_key *key, u16 zone, struct sk_buff *skb) { struct ovs_skb_cb ovs_cb = *OVS_CB(skb); + int err; if (key->eth.type == htons(ETH_P_IP)) { enum ip_defrag_users user = IP_DEFRAG_CONNTRACK_IN + zone; - int err; memset(IPCB(skb), 0, sizeof(struct inet_skb_parm)); err = ip_defrag(net, skb, user); @@ -314,14 +314,13 @@ static int handle_fragments(struct net *net, struct sw_flow_key *key, #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) } else if (key->eth.type == htons(ETH_P_IPV6)) { enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone; - struct sk_buff *reasm; memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm)); - reasm = nf_ct_frag6_gather(net, skb, user); - if (!reasm) - return -EINPROGRESS; + err = nf_ct_frag6_gather(net, skb, user); + if (err) + return err; - key->ip.proto = ipv6_hdr(reasm)->nexthdr; + key->ip.proto = ipv6_hdr(skb)->nexthdr; ovs_cb.mru = IP6CB(skb)->frag_max_size; #endif } else { -- cgit v1.2.3-59-g8ed1b From 7ef8f65df976369588fa1b6466668b1b6a26eb3c Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Sat, 21 Nov 2015 15:57:27 +0100 Subject: net: ipmr: fix code and comment style Trivial code and comment style fixes, also removed some extra newlines, spaces and tabs. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/mroute.h | 59 ++++++------------ net/ipv4/ipmr.c | 142 ++++++++++++-------------------------------- 2 files changed, 54 insertions(+), 147 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/mroute.h b/include/uapi/linux/mroute.h index a382d2c04a42..cf943016930f 100644 --- a/include/uapi/linux/mroute.h +++ b/include/uapi/linux/mroute.h @@ -4,15 +4,13 @@ #include #include -/* - * Based on the MROUTING 3.5 defines primarily to keep - * source compatibility with BSD. +/* Based on the MROUTING 3.5 defines primarily to keep + * source compatibility with BSD. * - * See the mrouted code for the original history. - * - * Protocol Independent Multicast (PIM) data structures included - * Carlos Picoto (cap@di.fc.ul.pt) + * See the mrouted code for the original history. * + * Protocol Independent Multicast (PIM) data structures included + * Carlos Picoto (cap@di.fc.ul.pt) */ #define MRT_BASE 200 @@ -34,15 +32,13 @@ #define SIOCGETSGCNT (SIOCPROTOPRIVATE+1) #define SIOCGETRPF (SIOCPROTOPRIVATE+2) -#define MAXVIFS 32 +#define MAXVIFS 32 typedef unsigned long vifbitmap_t; /* User mode code depends on this lot */ typedef unsigned short vifi_t; #define ALL_VIFS ((vifi_t)(-1)) -/* - * Same idea as select - */ - +/* Same idea as select */ + #define VIFM_SET(n,m) ((m)|=(1<<(n))) #define VIFM_CLR(n,m) ((m)&=~(1<<(n))) #define VIFM_ISSET(n,m) ((m)&(1<<(n))) @@ -50,11 +46,9 @@ typedef unsigned short vifi_t; #define VIFM_COPY(mfrom,mto) ((mto)=(mfrom)) #define VIFM_SAME(m1,m2) ((m1)==(m2)) -/* - * Passed by mrouted for an MRT_ADD_VIF - again we use the - * mrouted 3.6 structures for compatibility +/* Passed by mrouted for an MRT_ADD_VIF - again we use the + * mrouted 3.6 structures for compatibility */ - struct vifctl { vifi_t vifc_vifi; /* Index of VIF */ unsigned char vifc_flags; /* VIFF_ flags */ @@ -73,10 +67,7 @@ struct vifctl { #define VIFF_USE_IFINDEX 0x8 /* use vifc_lcl_ifindex instead of vifc_lcl_addr to find an interface */ -/* - * Cache manipulation structures for mrouted and PIMd - */ - +/* Cache manipulation structures for mrouted and PIMd */ struct mfcctl { struct in_addr mfcc_origin; /* Origin of mcast */ struct in_addr mfcc_mcastgrp; /* Group in question */ @@ -88,10 +79,7 @@ struct mfcctl { int mfcc_expire; }; -/* - * Group count retrieval for mrouted - */ - +/* Group count retrieval for mrouted */ struct sioc_sg_req { struct in_addr src; struct in_addr grp; @@ -100,10 +88,7 @@ struct sioc_sg_req { unsigned long wrong_if; }; -/* - * To get vif packet counts - */ - +/* To get vif packet counts */ struct sioc_vif_req { vifi_t vifi; /* Which iface */ unsigned long icount; /* In packets */ @@ -112,11 +97,9 @@ struct sioc_vif_req { unsigned long obytes; /* Out bytes */ }; -/* - * This is the format the mroute daemon expects to see IGMP control - * data. Magically happens to be like an IP packet as per the original +/* This is the format the mroute daemon expects to see IGMP control + * data. Magically happens to be like an IP packet as per the original */ - struct igmpmsg { __u32 unused1,unused2; unsigned char im_msgtype; /* What is this */ @@ -126,21 +109,13 @@ struct igmpmsg { struct in_addr im_src,im_dst; }; -/* - * That's all usermode folks - */ - - +/* That's all usermode folks */ #define MFC_ASSERT_THRESH (3*HZ) /* Maximal freq. of asserts */ -/* - * Pseudo messages used by mrouted - */ - +/* Pseudo messages used by mrouted */ #define IGMPMSG_NOCACHE 1 /* Kern cache fill request to mrouted */ #define IGMPMSG_WRONGVIF 2 /* For PIM assert processing (unused) */ #define IGMPMSG_WHOLEPKT 3 /* For PIM Register processing */ - #endif /* _UAPI__LINUX_MROUTE_H */ diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index e153ab7b17a1..286ede3716ee 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -102,9 +102,7 @@ static inline bool pimsm_enabled(void) static DEFINE_RWLOCK(mrt_lock); -/* - * Multicast router control variables - */ +/* Multicast router control variables */ #define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) @@ -393,8 +391,7 @@ static void ipmr_del_tunnel(struct net_device *dev, struct vifctl *v) } } -static -struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) +static struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) { struct net_device *dev; @@ -561,8 +558,7 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, struct iphdr *encap; encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); - /* - * Check that: + /* Check that: * a. packet is really sent to a multicast group * b. packet is not a NULL-REGISTER * c. packet is not truncated @@ -603,7 +599,6 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) * vif_delete - Delete a VIF entry * @notify: Set to 1, if the caller is a notifier_call */ - static int vif_delete(struct mr_table *mrt, int vifi, int notify, struct list_head *head) { @@ -673,7 +668,6 @@ static inline void ipmr_cache_free(struct mfc_cache *c) /* Destroy an unresolved cache entry, killing queued skbs * and reporting error to netlink readers. */ - static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) { struct net *net = read_pnet(&mrt->net); @@ -701,9 +695,7 @@ static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) ipmr_cache_free(c); } - /* Timer process for the unresolved queue. */ - static void ipmr_expire_process(unsigned long arg) { struct mr_table *mrt = (struct mr_table *)arg; @@ -743,7 +735,6 @@ out: } /* Fill oifs list. It is called under write locked mrt_lock. */ - static void ipmr_update_thresholds(struct mr_table *mrt, struct mfc_cache *cache, unsigned char *ttls) { @@ -808,7 +799,6 @@ static int vif_add(struct net *net, struct mr_table *mrt, return err; } break; - case VIFF_USE_IFINDEX: case 0: if (vifc->vifc_flags == VIFF_USE_IFINDEX) { @@ -928,9 +918,7 @@ skip: return ipmr_cache_find_any_parent(mrt, vifi); } -/* - * Allocate a multicast cache entry - */ +/* Allocate a multicast cache entry */ static struct mfc_cache *ipmr_cache_alloc(void) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_KERNEL); @@ -951,10 +939,7 @@ static struct mfc_cache *ipmr_cache_alloc_unres(void) return c; } -/* - * A cache entry has gone into a resolved state from queued - */ - +/* A cache entry has gone into a resolved state from queued */ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, struct mfc_cache *uc, struct mfc_cache *c) { @@ -962,7 +947,6 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, struct nlmsgerr *e; /* Play the pending entries through our router */ - while ((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) { if (ip_hdr(skb)->version == 0) { struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct iphdr)); @@ -1064,12 +1048,9 @@ static int ipmr_cache_report(struct mr_table *mrt, return ret; } -/* - * Queue a packet for resolution. It gets locked cache entry! - */ - -static int -ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) +/* Queue a packet for resolution. It gets locked cache entry! */ +static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, + struct sk_buff *skb) { bool found = false; int err; @@ -1087,7 +1068,6 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) if (!found) { /* Create a new entry if allowable */ - if (atomic_read(&mrt->cache_resolve_queue_len) >= 10 || (c = ipmr_cache_alloc_unres()) == NULL) { spin_unlock_bh(&mfc_unres_lock); @@ -1097,13 +1077,11 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) } /* Fill in the new cache entry */ - c->mfc_parent = -1; c->mfc_origin = iph->saddr; c->mfc_mcastgrp = iph->daddr; /* Reflect first query at mrouted. */ - err = ipmr_cache_report(mrt, skb, vifi, IGMPMSG_NOCACHE); if (err < 0) { /* If the report failed throw the cache entry @@ -1125,7 +1103,6 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) } /* See if we can append the packet */ - if (c->mfc_un.unres.unresolved.qlen > 3) { kfree_skb(skb); err = -ENOBUFS; @@ -1138,9 +1115,7 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) return err; } -/* - * MFC cache manipulation by user space mroute daemon - */ +/* MFC cache manipulation by user space mroute daemon */ static int ipmr_mfc_delete(struct mr_table *mrt, struct mfcctl *mfc, int parent) { @@ -1211,9 +1186,8 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, list_add_rcu(&c->list, &mrt->mfc_cache_array[line]); - /* - * Check to see if we resolved a queued list. If so we - * need to send on the frames and tidy up. + /* Check to see if we resolved a queued list. If so we + * need to send on the frames and tidy up. */ found = false; spin_lock_bh(&mfc_unres_lock); @@ -1238,10 +1212,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, return 0; } -/* - * Close the multicast socket, and clear the vif tables etc - */ - +/* Close the multicast socket, and clear the vif tables etc */ static void mroute_clean_tables(struct mr_table *mrt) { int i; @@ -1249,7 +1220,6 @@ static void mroute_clean_tables(struct mr_table *mrt) struct mfc_cache *c, *next; /* Shut down all active vif entries */ - for (i = 0; i < mrt->maxvif; i++) { if (!(mrt->vif_table[i].flags & VIFF_STATIC)) vif_delete(mrt, i, 0, &list); @@ -1257,7 +1227,6 @@ static void mroute_clean_tables(struct mr_table *mrt) unregister_netdevice_many(&list); /* Wipe the cache */ - for (i = 0; i < MFC_LINES; i++) { list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[i], list) { if (c->mfc_flags & MFC_STATIC) @@ -1301,11 +1270,10 @@ static void mrtsock_destruct(struct sock *sk) rtnl_unlock(); } -/* - * Socket options and virtual interface manipulation. The whole - * virtual interface system is a complete heap, but unfortunately - * that's how BSD mrouted happens to think. Maybe one day with a proper - * MOSPF/PIM router set up we can clean this up. +/* Socket options and virtual interface manipulation. The whole + * virtual interface system is a complete heap, but unfortunately + * that's how BSD mrouted happens to think. Maybe one day with a proper + * MOSPF/PIM router set up we can clean this up. */ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsigned int optlen) @@ -1373,10 +1341,9 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi rtnl_unlock(); return ret; - /* - * Manipulate the forwarding caches. These live - * in a sort of kernel/user symbiosis. - */ + /* Manipulate the forwarding caches. These live + * in a sort of kernel/user symbiosis. + */ case MRT_ADD_MFC: case MRT_DEL_MFC: parent = -1; @@ -1397,9 +1364,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi parent); rtnl_unlock(); return ret; - /* - * Control PIM assert. - */ + /* Control PIM assert. */ case MRT_ASSERT: { int v; @@ -1456,19 +1421,13 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi return ret; } #endif - /* - * Spurious command, or MRT_VERSION which you cannot - * set. - */ + /* Spurious command, or MRT_VERSION which you cannot set. */ default: return -ENOPROTOOPT; } } -/* - * Getsock opt support for the multicast routing system. - */ - +/* Getsock opt support for the multicast routing system. */ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int __user *optlen) { int olr; @@ -1512,10 +1471,7 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int return 0; } -/* - * The IP multicast ioctl support routines. - */ - +/* The IP multicast ioctl support routines. */ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) { struct sioc_sg_req sr; @@ -1648,7 +1604,6 @@ int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg) } #endif - static int ipmr_device_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); @@ -1670,17 +1625,14 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v return NOTIFY_DONE; } - static struct notifier_block ip_mr_notifier = { .notifier_call = ipmr_device_event, }; -/* - * Encapsulate a packet by attaching a valid IPIP header to it. - * This avoids tunnel drivers and other mess and gives us the speed so - * important for multicast video. +/* Encapsulate a packet by attaching a valid IPIP header to it. + * This avoids tunnel drivers and other mess and gives us the speed so + * important for multicast video. */ - static void ip_encap(struct net *net, struct sk_buff *skb, __be32 saddr, __be32 daddr) { @@ -1722,9 +1674,7 @@ static inline int ipmr_forward_finish(struct net *net, struct sock *sk, return dst_output(net, sk, skb); } -/* - * Processing handlers for ipmr_forward - */ +/* Processing handlers for ipmr_forward */ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *c, int vifi) @@ -1773,7 +1723,6 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, * allow to send ICMP, so that packets will disappear * to blackhole. */ - IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); ip_rt_put(rt); goto out_free; @@ -1805,8 +1754,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, IPCB(skb)->flags |= IPSKB_FORWARDED; - /* - * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally + /* RFC1584 teaches, that DVMRP/PIM router must deliver packets locally * not only before forwarding, but after forwarding on all output * interfaces. It is clear, if mrouter runs a multicasting * program, it should receive packets not depending to what interface @@ -1837,7 +1785,6 @@ static int ipmr_find_vif(struct mr_table *mrt, struct net_device *dev) } /* "local" means that we should preserve one skb (for local delivery) */ - static void ip_mr_forward(struct net *net, struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *cache, int local) @@ -1862,9 +1809,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, goto forward; } - /* - * Wrong interface: drop packet and (maybe) send PIM assert. - */ + /* Wrong interface: drop packet and (maybe) send PIM assert. */ if (mrt->vif_table[vif].dev != skb->dev) { if (rt_is_output_route(skb_rtable(skb))) { /* It is our own packet, looped back. @@ -1903,9 +1848,7 @@ forward: mrt->vif_table[vif].pkt_in++; mrt->vif_table[vif].bytes_in += skb->len; - /* - * Forward the frame - */ + /* Forward the frame */ if (cache->mfc_origin == htonl(INADDR_ANY) && cache->mfc_mcastgrp == htonl(INADDR_ANY)) { if (true_vifi >= 0 && @@ -1979,11 +1922,9 @@ static struct mr_table *ipmr_rt_fib_lookup(struct net *net, struct sk_buff *skb) return mrt; } -/* - * Multicast packets for forwarding arrive here - * Called with rcu_read_lock(); +/* Multicast packets for forwarding arrive here + * Called with rcu_read_lock(); */ - int ip_mr_input(struct sk_buff *skb) { struct mfc_cache *cache; @@ -2034,9 +1975,7 @@ int ip_mr_input(struct sk_buff *skb) vif); } - /* - * No usable cache entry - */ + /* No usable cache entry */ if (!cache) { int vif; @@ -2078,10 +2017,7 @@ dont_forward: } #ifdef CONFIG_IP_PIMSM_V1 -/* - * Handle IGMP messages of PIMv1 - */ - +/* Handle IGMP messages of PIMv1 */ int pim_rcv_v1(struct sk_buff *skb) { struct igmphdr *pim; @@ -2406,9 +2342,8 @@ done: } #ifdef CONFIG_PROC_FS -/* - * The /proc interfaces to multicast routing : - * /proc/net/ip_mr_cache & /proc/net/ip_mr_vif +/* The /proc interfaces to multicast routing : + * /proc/net/ip_mr_cache & /proc/net/ip_mr_vif */ struct ipmr_vif_iter { struct seq_net_private p; @@ -2692,10 +2627,7 @@ static const struct net_protocol pim_protocol = { }; #endif - -/* - * Setup for IP multicast routing - */ +/* Setup for IP multicast routing */ static int __net_init ipmr_net_init(struct net *net) { int err; -- cgit v1.2.3-59-g8ed1b From a9ecfbe7fcf2f600718c3f995cbb46043f7a7a5d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 24 Nov 2015 00:03:28 +0100 Subject: netfilter: nf_tables: remove unused struct members Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 4bd7508bedc9..101d7d7ec243 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -19,8 +19,6 @@ struct nft_pktinfo { const struct net_device *out; u8 pf; u8 hook; - u8 nhoff; - u8 thoff; u8 tprot; /* for x_tables compatibility */ struct xt_action_param xt; -- cgit v1.2.3-59-g8ed1b From 7ec3f7b47b8d9ad7ba425726f2c58f9ddce040df Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 24 Nov 2015 10:00:22 +0000 Subject: netfilter: nft_payload: add packet mangling support Add support for mangling packet payload. Checksum for the specified base header is updated automatically if requested, however no updates for any kind of pseudo headers are supported, meaning no stateless NAT is supported. For checksum updates different checksumming methods can be specified. The currently supported methods are NONE for no checksum updates, and INET for internet type checksums. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables_core.h | 9 +++ include/uapi/linux/netfilter/nf_tables.h | 17 ++++ net/netfilter/nft_payload.c | 135 +++++++++++++++++++++++++++++-- 3 files changed, 155 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h index c6f400cfaac8..4ff5424909aa 100644 --- a/include/net/netfilter/nf_tables_core.h +++ b/include/net/netfilter/nf_tables_core.h @@ -47,6 +47,15 @@ struct nft_payload { enum nft_registers dreg:8; }; +struct nft_payload_set { + enum nft_payload_bases base:8; + u8 offset; + u8 len; + enum nft_registers sreg:8; + u8 csum_type; + u8 csum_offset; +}; + extern const struct nft_expr_ops nft_payload_fast_ops; int nft_payload_module_init(void); diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index d8c8a7c9d88a..5f3ececf84b3 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -597,6 +597,17 @@ enum nft_payload_bases { NFT_PAYLOAD_TRANSPORT_HEADER, }; +/** + * enum nft_payload_csum_types - nf_tables payload expression checksum types + * + * @NFT_PAYLOAD_CSUM_NONE: no checksumming + * @NFT_PAYLOAD_CSUM_INET: internet checksum (RFC 791) + */ +enum nft_payload_csum_types { + NFT_PAYLOAD_CSUM_NONE, + NFT_PAYLOAD_CSUM_INET, +}; + /** * enum nft_payload_attributes - nf_tables payload expression netlink attributes * @@ -604,6 +615,9 @@ enum nft_payload_bases { * @NFTA_PAYLOAD_BASE: payload base (NLA_U32: nft_payload_bases) * @NFTA_PAYLOAD_OFFSET: payload offset relative to base (NLA_U32) * @NFTA_PAYLOAD_LEN: payload length (NLA_U32) + * @NFTA_PAYLOAD_SREG: source register to load data from (NLA_U32: nft_registers) + * @NFTA_PAYLOAD_CSUM_TYPE: checksum type (NLA_U32) + * @NFTA_PAYLOAD_CSUM_OFFSET: checksum offset relative to base (NLA_U32) */ enum nft_payload_attributes { NFTA_PAYLOAD_UNSPEC, @@ -611,6 +625,9 @@ enum nft_payload_attributes { NFTA_PAYLOAD_BASE, NFTA_PAYLOAD_OFFSET, NFTA_PAYLOAD_LEN, + NFTA_PAYLOAD_SREG, + NFTA_PAYLOAD_CSUM_TYPE, + NFTA_PAYLOAD_CSUM_OFFSET, __NFTA_PAYLOAD_MAX }; #define NFTA_PAYLOAD_MAX (__NFTA_PAYLOAD_MAX - 1) diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index 09b4b07eb676..12cd4bf16d17 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -107,10 +107,13 @@ err: } static const struct nla_policy nft_payload_policy[NFTA_PAYLOAD_MAX + 1] = { - [NFTA_PAYLOAD_DREG] = { .type = NLA_U32 }, - [NFTA_PAYLOAD_BASE] = { .type = NLA_U32 }, - [NFTA_PAYLOAD_OFFSET] = { .type = NLA_U32 }, - [NFTA_PAYLOAD_LEN] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_SREG] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_DREG] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_BASE] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_OFFSET] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_LEN] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_CSUM_TYPE] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_CSUM_OFFSET] = { .type = NLA_U32 }, }; static int nft_payload_init(const struct nft_ctx *ctx, @@ -160,6 +163,118 @@ const struct nft_expr_ops nft_payload_fast_ops = { .dump = nft_payload_dump, }; +static void nft_payload_set_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + const struct nft_payload_set *priv = nft_expr_priv(expr); + struct sk_buff *skb = pkt->skb; + const u32 *src = ®s->data[priv->sreg]; + int offset, csum_offset; + __wsum fsum, tsum; + __sum16 sum; + + switch (priv->base) { + case NFT_PAYLOAD_LL_HEADER: + if (!skb_mac_header_was_set(skb)) + goto err; + offset = skb_mac_header(skb) - skb->data; + break; + case NFT_PAYLOAD_NETWORK_HEADER: + offset = skb_network_offset(skb); + break; + case NFT_PAYLOAD_TRANSPORT_HEADER: + offset = pkt->xt.thoff; + break; + default: + BUG(); + } + + csum_offset = offset + priv->csum_offset; + offset += priv->offset; + + if (priv->csum_type == NFT_PAYLOAD_CSUM_INET && + (priv->base != NFT_PAYLOAD_TRANSPORT_HEADER || + skb->ip_summed != CHECKSUM_PARTIAL)) { + if (skb_copy_bits(skb, csum_offset, &sum, sizeof(sum)) < 0) + goto err; + + fsum = skb_checksum(skb, offset, priv->len, 0); + tsum = csum_partial(src, priv->len, 0); + sum = csum_fold(csum_add(csum_sub(~csum_unfold(sum), fsum), + tsum)); + if (sum == 0) + sum = CSUM_MANGLED_0; + + if (!skb_make_writable(skb, csum_offset + sizeof(sum)) || + skb_store_bits(skb, csum_offset, &sum, sizeof(sum)) < 0) + goto err; + } + + if (!skb_make_writable(skb, max(offset + priv->len, 0)) || + skb_store_bits(skb, offset, src, priv->len) < 0) + goto err; + + return; +err: + regs->verdict.code = NFT_BREAK; +} + +static int nft_payload_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_payload_set *priv = nft_expr_priv(expr); + + priv->base = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_BASE])); + priv->offset = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_OFFSET])); + priv->len = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_LEN])); + priv->sreg = nft_parse_register(tb[NFTA_PAYLOAD_SREG]); + + if (tb[NFTA_PAYLOAD_CSUM_TYPE]) + priv->csum_type = + ntohl(nla_get_be32(tb[NFTA_PAYLOAD_CSUM_TYPE])); + if (tb[NFTA_PAYLOAD_CSUM_OFFSET]) + priv->csum_offset = + ntohl(nla_get_be32(tb[NFTA_PAYLOAD_CSUM_OFFSET])); + + switch (priv->csum_type) { + case NFT_PAYLOAD_CSUM_NONE: + case NFT_PAYLOAD_CSUM_INET: + break; + default: + return -EOPNOTSUPP; + } + + return nft_validate_register_load(priv->sreg, priv->len); +} + +static int nft_payload_set_dump(struct sk_buff *skb, const struct nft_expr *expr) +{ + const struct nft_payload_set *priv = nft_expr_priv(expr); + + if (nft_dump_register(skb, NFTA_PAYLOAD_SREG, priv->sreg) || + nla_put_be32(skb, NFTA_PAYLOAD_BASE, htonl(priv->base)) || + nla_put_be32(skb, NFTA_PAYLOAD_OFFSET, htonl(priv->offset)) || + nla_put_be32(skb, NFTA_PAYLOAD_LEN, htonl(priv->len)) || + nla_put_be32(skb, NFTA_PAYLOAD_CSUM_TYPE, htonl(priv->csum_type)) || + nla_put_be32(skb, NFTA_PAYLOAD_CSUM_OFFSET, + htonl(priv->csum_offset))) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -1; +} + +static const struct nft_expr_ops nft_payload_set_ops = { + .type = &nft_payload_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_payload_set)), + .eval = nft_payload_set_eval, + .init = nft_payload_set_init, + .dump = nft_payload_set_dump, +}; + static const struct nft_expr_ops * nft_payload_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[]) @@ -167,8 +282,7 @@ nft_payload_select_ops(const struct nft_ctx *ctx, enum nft_payload_bases base; unsigned int offset, len; - if (tb[NFTA_PAYLOAD_DREG] == NULL || - tb[NFTA_PAYLOAD_BASE] == NULL || + if (tb[NFTA_PAYLOAD_BASE] == NULL || tb[NFTA_PAYLOAD_OFFSET] == NULL || tb[NFTA_PAYLOAD_LEN] == NULL) return ERR_PTR(-EINVAL); @@ -183,6 +297,15 @@ nft_payload_select_ops(const struct nft_ctx *ctx, return ERR_PTR(-EOPNOTSUPP); } + if (tb[NFTA_PAYLOAD_SREG] != NULL) { + if (tb[NFTA_PAYLOAD_DREG] != NULL) + return ERR_PTR(-EINVAL); + return &nft_payload_set_ops; + } + + if (tb[NFTA_PAYLOAD_DREG] == NULL) + return ERR_PTR(-EINVAL); + offset = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_OFFSET])); len = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_LEN])); -- cgit v1.2.3-59-g8ed1b From 9458ceab49179b7fd2d5192fd9dcf316ca195dc0 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 24 Nov 2015 15:30:21 -0800 Subject: net: phy: bcm7xxx: Add entry for Broadcom BCM7435 Add a PHY entry for the Broadcom BCM7435 chips, this is a 40nm generation Ethernet PHY which is analogous to its 7425 and 7429 counter parts. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm7xxx.c | 14 ++++++++++++++ include/linux/brcmphy.h | 1 + 2 files changed, 15 insertions(+) (limited to 'include') diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c index 03d4809a9126..d4083c381cd1 100644 --- a/drivers/net/phy/bcm7xxx.c +++ b/drivers/net/phy/bcm7xxx.c @@ -360,6 +360,19 @@ static struct phy_driver bcm7xxx_driver[] = { .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, .driver = { .owner = THIS_MODULE }, +}, { + .phy_id = PHY_ID_BCM7435, + .phy_id_mask = 0xfffffff0, + .name = "Broadcom BCM7435", + .features = PHY_GBIT_FEATURES | + SUPPORTED_Pause | SUPPORTED_Asym_Pause, + .flags = PHY_IS_INTERNAL, + .config_init = bcm7xxx_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .suspend = bcm7xxx_suspend, + .resume = bcm7xxx_config_init, + .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_BCM_OUI_4, .phy_id_mask = 0xffff0000, @@ -395,6 +408,7 @@ static struct mdio_device_id __maybe_unused bcm7xxx_tbl[] = { { PHY_ID_BCM7425, 0xfffffff0, }, { PHY_ID_BCM7429, 0xfffffff0, }, { PHY_ID_BCM7439, 0xfffffff0, }, + { PHY_ID_BCM7435, 0xfffffff0, }, { PHY_ID_BCM7445, 0xfffffff0, }, { PHY_BCM_OUI_4, 0xffff0000 }, { PHY_BCM_OUI_5, 0xffffff00 }, diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h index 59f4a7304419..f0ba9c2ec639 100644 --- a/include/linux/brcmphy.h +++ b/include/linux/brcmphy.h @@ -26,6 +26,7 @@ #define PHY_ID_BCM7366 0x600d8490 #define PHY_ID_BCM7425 0x600d86b0 #define PHY_ID_BCM7429 0x600d8730 +#define PHY_ID_BCM7435 0x600d8750 #define PHY_ID_BCM7439 0x600d8480 #define PHY_ID_BCM7439_2 0xae025080 #define PHY_ID_BCM7445 0x600d8510 -- cgit v1.2.3-59-g8ed1b From 1ce0bf50ae2233c7115a18c0c623662d177b434c Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 26 Nov 2015 13:55:39 +0800 Subject: net: Generalise wq_has_sleeper helper The memory barrier in the helper wq_has_sleeper is needed by just about every user of waitqueue_active. This patch generalises it by making it take a wait_queue_head_t directly. The existing helper is renamed to skwq_has_sleeper. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- crypto/algif_aead.c | 4 ++-- crypto/algif_skcipher.c | 4 ++-- include/linux/wait.h | 21 +++++++++++++++++++++ include/net/sock.h | 15 +++++---------- net/atm/common.c | 4 ++-- net/core/sock.c | 8 ++++---- net/core/stream.c | 2 +- net/dccp/output.c | 2 +- net/iucv/af_iucv.c | 2 +- net/rxrpc/af_rxrpc.c | 2 +- net/sctp/socket.c | 2 +- net/tipc/socket.c | 4 ++-- net/unix/af_unix.c | 2 +- 13 files changed, 44 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/crypto/algif_aead.c b/crypto/algif_aead.c index 0aa6fdfb448a..fb99f30849d2 100644 --- a/crypto/algif_aead.c +++ b/crypto/algif_aead.c @@ -106,7 +106,7 @@ static void aead_wmem_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); @@ -157,7 +157,7 @@ static void aead_data_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLRDNORM | POLLRDBAND); diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c index af31a0ee4057..0e6702e41472 100644 --- a/crypto/algif_skcipher.c +++ b/crypto/algif_skcipher.c @@ -238,7 +238,7 @@ static void skcipher_wmem_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); @@ -288,7 +288,7 @@ static void skcipher_data_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLRDNORM | POLLRDBAND); diff --git a/include/linux/wait.h b/include/linux/wait.h index 1e1bf9f963a9..6aa09a875fbd 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -107,6 +107,27 @@ static inline int waitqueue_active(wait_queue_head_t *q) return !list_empty(&q->task_list); } +/** + * wq_has_sleeper - check if there are any waiting processes + * @wq: wait queue head + * + * Returns true if wq has waiting processes + * + * Please refer to the comment for waitqueue_active. + */ +static inline bool wq_has_sleeper(wait_queue_head_t *wq) +{ + /* + * We need to be sure we are in sync with the + * add_wait_queue modifications to the wait queue. + * + * This memory barrier should be paired with one on the + * waiting side. + */ + smp_mb(); + return waitqueue_active(wq); +} + extern void add_wait_queue(wait_queue_head_t *q, wait_queue_t *wait); extern void add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t *wait); extern void remove_wait_queue(wait_queue_head_t *q, wait_queue_t *wait); diff --git a/include/net/sock.h b/include/net/sock.h index 7f89e4ba18d1..62d35afcb3ac 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -58,6 +58,7 @@ #include #include #include +#include #include #include @@ -1879,12 +1880,12 @@ static inline bool sk_has_allocations(const struct sock *sk) } /** - * wq_has_sleeper - check if there are any waiting processes + * skwq_has_sleeper - check if there are any waiting processes * @wq: struct socket_wq * * Returns true if socket_wq has waiting processes * - * The purpose of the wq_has_sleeper and sock_poll_wait is to wrap the memory + * The purpose of the skwq_has_sleeper and sock_poll_wait is to wrap the memory * barrier call. They were added due to the race found within the tcp code. * * Consider following tcp code paths: @@ -1910,15 +1911,9 @@ static inline bool sk_has_allocations(const struct sock *sk) * data on the socket. * */ -static inline bool wq_has_sleeper(struct socket_wq *wq) +static inline bool skwq_has_sleeper(struct socket_wq *wq) { - /* We need to be sure we are in sync with the - * add_wait_queue modifications to the wait queue. - * - * This memory barrier is paired in the sock_poll_wait. - */ - smp_mb(); - return wq && waitqueue_active(&wq->wait); + return wq && wq_has_sleeper(&wq->wait); } /** diff --git a/net/atm/common.c b/net/atm/common.c index 49a872db7e42..6dc12305799e 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -96,7 +96,7 @@ static void vcc_def_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up(&wq->wait); rcu_read_unlock(); } @@ -117,7 +117,7 @@ static void vcc_write_space(struct sock *sk) if (vcc_writable(sk)) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); diff --git a/net/core/sock.c b/net/core/sock.c index 1e4dd54bfb5a..2769bd3a4d7c 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2283,7 +2283,7 @@ static void sock_def_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); rcu_read_unlock(); } @@ -2294,7 +2294,7 @@ static void sock_def_error_report(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_poll(&wq->wait, POLLERR); sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR); rcu_read_unlock(); @@ -2306,7 +2306,7 @@ static void sock_def_readable(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND); sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); @@ -2324,7 +2324,7 @@ static void sock_def_write_space(struct sock *sk) */ if ((atomic_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); diff --git a/net/core/stream.c b/net/core/stream.c index d70f77a0c889..8ff9d63b4265 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -35,7 +35,7 @@ void sk_stream_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); if (wq && wq->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN)) diff --git a/net/dccp/output.c b/net/dccp/output.c index 4ce912e691d0..b66c84db0766 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -201,7 +201,7 @@ void dccp_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); /* Should agree with poll, otherwise some programs break */ if (sock_writeable(sk)) diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index fcb2752419c6..4f0aa91470c6 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -303,7 +303,7 @@ static void iucv_sock_wake_msglim(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); rcu_read_unlock(); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 1f8a144a5dc2..7e2d1057d8bc 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -67,7 +67,7 @@ static void rxrpc_write_space(struct sock *sk) if (rxrpc_writable(sk)) { struct socket_wq *wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 897c01c029ca..ec10b66354b8 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6978,7 +6978,7 @@ void sctp_data_ready(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 552dbaba9cf3..525acf6dd1c6 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1492,7 +1492,7 @@ static void tipc_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); rcu_read_unlock(); @@ -1509,7 +1509,7 @@ static void tipc_data_ready(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); rcu_read_unlock(); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 955ec152cb71..efb706e1d1c0 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -339,7 +339,7 @@ static void unix_write_space(struct sock *sk) rcu_read_lock(); if (unix_writable(sk)) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); -- cgit v1.2.3-59-g8ed1b From 06bd6c0370bb88a2256c6763a32bc4e4ade06521 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 26 Nov 2015 15:23:45 +0100 Subject: net: ipmr: remove unused MFC_NOTIFY flag and make the flags enum MFC_NOTIFY was introduced in kernel 2.1.68 but afaik it hasn't been used and I couldn't find any users currently so just remove it. Only MFC_STATIC is left, so move it into an enum, add a description and use BIT(). Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 10 +++++++--- net/ipv4/ipmr.c | 2 -- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 79aaa9fc1a15..fa66ebc1fed6 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -64,6 +64,13 @@ struct vif_device { #define VIFF_STATIC 0x8000 +/* mfc_flags: + * MFC_STATIC - the entry was added statically (not by a routing daemon) + */ +enum { + MFC_STATIC = BIT(0), +}; + struct mfc_cache { struct list_head list; __be32 mfc_mcastgrp; /* Group the entry belongs to */ @@ -89,9 +96,6 @@ struct mfc_cache { struct rcu_head rcu; }; -#define MFC_STATIC 1 -#define MFC_NOTIFY 2 - #define MFC_LINES 64 #ifdef __BIG_ENDIAN diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index a2d248d9c35c..a74e61883b8f 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -2199,8 +2199,6 @@ int ipmr_get_route(struct net *net, struct sk_buff *skb, } read_lock(&mrt_lock); - if (!nowait && (rtm->rtm_flags & RTM_F_NOTIFY)) - cache->mfc_flags |= MFC_NOTIFY; err = __ipmr_fill_mroute(mrt, skb, cache, rtm); read_unlock(&mrt_lock); rcu_read_unlock(); -- cgit v1.2.3-59-g8ed1b From 520191bb404c4b7b4cdb70a5480ada974b0c2d60 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 26 Nov 2015 15:23:46 +0100 Subject: net: ipmr: adjust mroute.h style and drop extern Remove extra spaces and tabs, adjust function definitions, remove an unnecessary ifdef (already used below, just move code) and drop extern from the functions. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 47 +++++++++++++++++++++-------------------------- 1 file changed, 21 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index fa66ebc1fed6..7c567a2679ce 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -9,38 +9,28 @@ #ifdef CONFIG_IP_MROUTE static inline int ip_mroute_opt(int opt) { - return (opt >= MRT_BASE) && (opt <= MRT_MAX); + return opt >= MRT_BASE && opt <= MRT_MAX; } -#else -static inline int ip_mroute_opt(int opt) -{ - return 0; -} -#endif -#ifdef CONFIG_IP_MROUTE -extern int ip_mroute_setsockopt(struct sock *, int, char __user *, unsigned int); -extern int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *); -extern int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg); -extern int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg); -extern int ip_mr_init(void); +int ip_mroute_setsockopt(struct sock *, int, char __user *, unsigned int); +int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *); +int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg); +int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg); +int ip_mr_init(void); #else -static inline -int ip_mroute_setsockopt(struct sock *sock, - int optname, char __user *optval, unsigned int optlen) +static inline int ip_mroute_setsockopt(struct sock *sock, int optname, + char __user *optval, unsigned int optlen) { return -ENOPROTOOPT; } -static inline -int ip_mroute_getsockopt(struct sock *sock, - int optname, char __user *optval, int __user *optlen) +static inline int ip_mroute_getsockopt(struct sock *sock, int optname, + char __user *optval, int __user *optlen) { return -ENOPROTOOPT; } -static inline -int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) +static inline int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) { return -ENOIOCTLCMD; } @@ -49,6 +39,11 @@ static inline int ip_mr_init(void) { return 0; } + +static inline int ip_mroute_opt(int opt) +{ + return 0; +} #endif struct vif_device { @@ -96,16 +91,16 @@ struct mfc_cache { struct rcu_head rcu; }; -#define MFC_LINES 64 +#define MFC_LINES 64 #ifdef __BIG_ENDIAN #define MFC_HASH(a,b) (((((__force u32)(__be32)a)>>24)^(((__force u32)(__be32)b)>>26))&(MFC_LINES-1)) #else #define MFC_HASH(a,b) ((((__force u32)(__be32)a)^(((__force u32)(__be32)b)>>2))&(MFC_LINES-1)) -#endif +#endif struct rtmsg; -extern int ipmr_get_route(struct net *net, struct sk_buff *skb, - __be32 saddr, __be32 daddr, - struct rtmsg *rtm, int nowait); +int ipmr_get_route(struct net *net, struct sk_buff *skb, + __be32 saddr, __be32 daddr, + struct rtmsg *rtm, int nowait); #endif -- cgit v1.2.3-59-g8ed1b From 5ea1f13299d8b8edcb2969eda4c81f8e3264b706 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 26 Nov 2015 15:23:47 +0100 Subject: net: ipmr: move struct mr_table and VIF_EXISTS to mroute.h Move the definitions of VIF_EXISTS() and struct mr_table to mroute.h Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 21 +++++++++++++++++++-- net/ipv4/ipmr.c | 18 ------------------ 2 files changed, 19 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 7c567a2679ce..bf9b322cb0b0 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -59,6 +59,25 @@ struct vif_device { #define VIFF_STATIC 0x8000 +#define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) +#define MFC_LINES 64 + +struct mr_table { + struct list_head list; + possible_net_t net; + u32 id; + struct sock __rcu *mroute_sk; + struct timer_list ipmr_expire_timer; + struct list_head mfc_unres_queue; + struct list_head mfc_cache_array[MFC_LINES]; + struct vif_device vif_table[MAXVIFS]; + int maxvif; + atomic_t cache_resolve_queue_len; + bool mroute_do_assert; + bool mroute_do_pim; + int mroute_reg_vif_num; +}; + /* mfc_flags: * MFC_STATIC - the entry was added statically (not by a routing daemon) */ @@ -91,8 +110,6 @@ struct mfc_cache { struct rcu_head rcu; }; -#define MFC_LINES 64 - #ifdef __BIG_ENDIAN #define MFC_HASH(a,b) (((((__force u32)(__be32)a)>>24)^(((__force u32)(__be32)b)>>26))&(MFC_LINES-1)) #else diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index a74e61883b8f..ff3dbbb9f11c 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -67,22 +67,6 @@ #include #include -struct mr_table { - struct list_head list; - possible_net_t net; - u32 id; - struct sock __rcu *mroute_sk; - struct timer_list ipmr_expire_timer; - struct list_head mfc_unres_queue; - struct list_head mfc_cache_array[MFC_LINES]; - struct vif_device vif_table[MAXVIFS]; - int maxvif; - atomic_t cache_resolve_queue_len; - bool mroute_do_assert; - bool mroute_do_pim; - int mroute_reg_vif_num; -}; - struct ipmr_rule { struct fib_rule common; }; @@ -104,8 +88,6 @@ static DEFINE_RWLOCK(mrt_lock); /* Multicast router control variables */ -#define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) - /* Special spinlock for queue of unresolved entries */ static DEFINE_SPINLOCK(mfc_unres_lock); -- cgit v1.2.3-59-g8ed1b From 1973a4ea6ceaa47671227c3077f90508ea30897b Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 26 Nov 2015 15:23:48 +0100 Subject: net: ipmr: move pimsm_enabled to pim.h and rename Move the inline pimsm_enabled() to pim.h and rename it to ipmr_pimsm_enabled to show it's for the ipv4 ipmr code since pim.h is used by IPv6 too. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/pim.h | 5 +++++ net/ipv4/ipmr.c | 11 +++-------- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/pim.h b/include/linux/pim.h index 252bf6644c51..e1d756f81348 100644 --- a/include/linux/pim.h +++ b/include/linux/pim.h @@ -13,6 +13,11 @@ #define PIM_NULL_REGISTER cpu_to_be32(0x40000000) +static inline bool ipmr_pimsm_enabled(void) +{ + return IS_BUILTIN(CONFIG_IP_PIMSM_V1) || IS_BUILTIN(CONFIG_IP_PIMSM_V2); +} + /* PIMv2 register message header layout (ietf-draft-idmr-pimvsm-v2-00.ps */ struct pimreghdr { diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index ff3dbbb9f11c..322fdc6ac75b 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -75,11 +75,6 @@ struct ipmr_result { struct mr_table *mrt; }; -static inline bool pimsm_enabled(void) -{ - return IS_BUILTIN(CONFIG_IP_PIMSM_V1) || IS_BUILTIN(CONFIG_IP_PIMSM_V2); -} - /* Big lock, protecting vif table, mrt cache and mroute socket state. * Note that the changes are semaphored via rtnl_lock. */ @@ -751,7 +746,7 @@ static int vif_add(struct net *net, struct mr_table *mrt, switch (vifc->vifc_flags) { case VIFF_REGISTER: - if (!pimsm_enabled()) + if (!ipmr_pimsm_enabled()) return -EINVAL; /* Special Purpose VIF in PIM * All the packets will be sent to the daemon @@ -1377,7 +1372,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, mrt->mroute_do_assert = val; break; case MRT_PIM: - if (!pimsm_enabled()) { + if (!ipmr_pimsm_enabled()) { ret = -ENOPROTOOPT; break; } @@ -1451,7 +1446,7 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int val = 0x0305; break; case MRT_PIM: - if (!pimsm_enabled()) + if (!ipmr_pimsm_enabled()) return -ENOPROTOOPT; val = mrt->mroute_do_pim; break; -- cgit v1.2.3-59-g8ed1b From 91420b83baa046ada1a899c97f3b2c52a9045705 Mon Sep 17 00:00:00 2001 From: Sudarsana Kalluru Date: Mon, 30 Nov 2015 12:25:03 +0200 Subject: qed: Add support for changing LED state Physical LEDs are being controlled by the management FW. This adds the qed functionality required to request management FW to change the LED configuration, as well as the necessary APIs for this functionality to later be used by the protocol drivers. Signed-off-by: Sudarsana Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_hsi.h | 6 ++++++ drivers/net/ethernet/qlogic/qed/qed_main.c | 18 ++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.c | 27 +++++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.h | 13 +++++++++++++ include/linux/qed/qed_if.h | 17 +++++++++++++++++ 5 files changed, 81 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index b2f8e854dfd1..264e954675d1 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -3993,6 +3993,8 @@ struct public_drv_mb { #define DRV_MSG_CODE_PHY_CORE_WRITE 0x000e0000 #define DRV_MSG_CODE_SET_VERSION 0x000f0000 +#define DRV_MSG_CODE_SET_LED_MODE 0x00200000 + #define DRV_MSG_SEQ_NUMBER_MASK 0x0000ffff u32 drv_mb_param; @@ -4044,6 +4046,10 @@ struct public_drv_mb { #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 +#define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 +#define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 +#define DRV_MB_PARAM_SET_LED_MODE_OFF 0x2 + u32 fw_mb_header; #define FW_MSG_CODE_MASK 0xffff0000 #define FW_MSG_CODE_DRV_LOAD_ENGINE 0x10100000 diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 947c7af72b25..6b02e1134360 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1135,6 +1135,23 @@ static int qed_drain(struct qed_dev *cdev) return 0; } +static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_ptt *ptt; + int status = 0; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EAGAIN; + + status = qed_mcp_set_led(hwfn, ptt, mode); + + qed_ptt_release(hwfn, ptt); + + return status; +} + const struct qed_common_ops qed_common_ops_pass = { .probe = &qed_probe, .remove = &qed_remove, @@ -1155,6 +1172,7 @@ const struct qed_common_ops qed_common_ops_pass = { .update_msglvl = &qed_init_dp, .chain_alloc = &qed_chain_alloc, .chain_free = &qed_chain_free, + .set_led = &qed_set_led, }; u32 qed_get_protocol_version(enum qed_protocol protocol) diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 20d048cdcb88..ba1b1f1ef789 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -858,3 +858,30 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, return 0; } + +int qed_mcp_set_led(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + enum qed_led_mode mode) +{ + u32 resp = 0, param = 0, drv_mb_param; + int rc; + + switch (mode) { + case QED_LED_MODE_ON: + drv_mb_param = DRV_MB_PARAM_SET_LED_MODE_ON; + break; + case QED_LED_MODE_OFF: + drv_mb_param = DRV_MB_PARAM_SET_LED_MODE_OFF; + break; + case QED_LED_MODE_RESTORE: + drv_mb_param = DRV_MB_PARAM_SET_LED_MODE_OPER; + break; + default: + DP_NOTICE(p_hwfn, "Invalid LED mode %d\n", mode); + return -EINVAL; + } + + rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_SET_LED_MODE, + drv_mb_param, &resp, ¶m); + + return rc; +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index dbaae586b4a7..506197d5c3dd 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -224,6 +224,19 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_mcp_drv_version *p_ver); +/** + * @brief Set LED status + * + * @param p_hwfn + * @param p_ptt + * @param mode - LED mode + * + * @return int - 0 - operation was successful. + */ +int qed_mcp_set_led(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + enum qed_led_mode mode); + /* Using hwfn number (and not pf_num) is required since in CMT mode, * same pf_num may be used by two different hwfn * TODO - this shouldn't really be in .h file, but until all fields diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index dc9a1353f971..d4a32e878180 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -25,6 +25,12 @@ #include #include +enum qed_led_mode { + QED_LED_MODE_OFF, + QED_LED_MODE_ON, + QED_LED_MODE_RESTORE +}; + #define DIRECT_REG_WR(reg_addr, val) writel((u32)val, \ (void __iomem *)(reg_addr)) @@ -252,6 +258,17 @@ struct qed_common_ops { void (*chain_free)(struct qed_dev *cdev, struct qed_chain *p_chain); + +/** + * @brief set_led - Configure LED mode + * + * @param cdev + * @param mode - LED mode + * + * @return 0 on success, error otherwise. + */ + int (*set_led)(struct qed_dev *cdev, + enum qed_led_mode mode); }; /** -- cgit v1.2.3-59-g8ed1b From c0eb454034aab783dc602739237a63b30867f5bd Mon Sep 17 00:00:00 2001 From: KY Srinivasan Date: Tue, 1 Dec 2015 16:43:10 -0800 Subject: hv_netvsc: Don't ask for additional head room in the skb The rndis header is 116 bytes big and can be placed in the default head room that will be available in the skb. Since the netvsc packet is less than 48 bytes, we can use the skb control buffer for the netvsc packet. With these changes we don't need to ask for additional head room. Signed-off-by: K. Y. Srinivasan Reviewed-by: Haiyang Zhang Signed-off-by: David S. Miller --- drivers/net/hyperv/hyperv_net.h | 3 +++ drivers/net/hyperv/netvsc_drv.c | 30 +++++++++++------------------- include/linux/netdevice.h | 4 +++- 3 files changed, 17 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index fc6d0c6de741..731054ef6da5 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -124,6 +124,9 @@ struct ndis_tcp_ip_checksum_info; /* * Represent netvsc packet which contains 1 RNDIS and 1 ethernet frame * within the RNDIS + * + * The size of this structure is less than 48 bytes and we can now + * place this structure in the skb->cb field. */ struct hv_netvsc_packet { /* Bookkeeping stuff */ diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 7e356a11c1d7..b820888409bc 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -433,7 +433,6 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) u32 net_trans_info; u32 hash; u32 skb_length; - u32 pkt_sz; struct hv_page_buffer page_buf[MAX_PAGE_BUFFER_COUNT]; struct netvsc_stats *tx_stats = this_cpu_ptr(net_device_ctx->tx_stats); @@ -461,16 +460,21 @@ check_size: goto check_size; } - pkt_sz = sizeof(struct hv_netvsc_packet) + RNDIS_AND_PPI_SIZE; - - ret = skb_cow_head(skb, pkt_sz); + /* + * Place the rndis header in the skb head room and + * the skb->cb will be used for hv_netvsc_packet + * structure. + */ + ret = skb_cow_head(skb, RNDIS_AND_PPI_SIZE); if (ret) { netdev_err(net, "unable to alloc hv_netvsc_packet\n"); ret = -ENOMEM; goto drop; } - /* Use the headroom for building up the packet */ - packet = (struct hv_netvsc_packet *)skb->head; + /* Use the skb control buffer for building up the packet */ + BUILD_BUG_ON(sizeof(struct hv_netvsc_packet) > + FIELD_SIZEOF(struct sk_buff, cb)); + packet = (struct hv_netvsc_packet *)skb->cb; packet->status = 0; packet->xmit_more = skb->xmit_more; @@ -483,8 +487,7 @@ check_size: packet->is_data_pkt = true; packet->total_data_buflen = skb->len; - rndis_msg = (struct rndis_message *)((unsigned long)packet + - sizeof(struct hv_netvsc_packet)); + rndis_msg = (struct rndis_message *)skb->head; memset(rndis_msg, 0, RNDIS_AND_PPI_SIZE); @@ -1118,16 +1121,12 @@ static int netvsc_probe(struct hv_device *dev, struct netvsc_device_info device_info; struct netvsc_device *nvdev; int ret; - u32 max_needed_headroom; net = alloc_etherdev_mq(sizeof(struct net_device_context), num_online_cpus()); if (!net) return -ENOMEM; - max_needed_headroom = sizeof(struct hv_netvsc_packet) + - RNDIS_AND_PPI_SIZE; - netif_carrier_off(net); net_device_ctx = netdev_priv(net); @@ -1166,13 +1165,6 @@ static int netvsc_probe(struct hv_device *dev, net->ethtool_ops = ðtool_ops; SET_NETDEV_DEV(net, &dev->device); - /* - * Request additional head room in the skb. - * We will use this space to build the rndis - * heaser and other state we need to maintain. - */ - net->needed_headroom = max_needed_headroom; - /* Notify the netvsc driver of the new device */ memset(&device_info, 0, sizeof(device_info)); device_info.ring_size = ring_size; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7d2d1d7aaec7..fcbc5259c630 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -132,7 +132,9 @@ static inline bool dev_xmit_complete(int rc) * used. */ -#if defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25) +#if defined(CONFIG_HYPERV_NET) +# define LL_MAX_HEADER 128 +#elif defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25) # if defined(CONFIG_MAC80211_MESH) # define LL_MAX_HEADER 128 # else -- cgit v1.2.3-59-g8ed1b From 7450aaf61f0ae2ee6cc6491138d11df2c25e7609 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 30 Nov 2015 08:57:28 -0800 Subject: tcp: suppress too verbose messages in tcp_send_ack() If tcp_send_ack() can not allocate skb, we properly handle this and setup a timer to try later. Use __GFP_NOWARN to avoid polluting syslog in the case host is under memory pressure, so that pertinent messages are not lost under a flood of useless information. sk_gfp_atomic() can use its gfp_mask argument (all callers currently were using GFP_ATOMIC before this patch) We rename sk_gfp_atomic() to sk_gfp_mask() to clearly express this function now takes into account its second argument (gfp_mask) Note that when tcp_transmit_skb() is called with clone_it set to false, we do not attempt memory allocations, so can pass a 0 gfp_mask, which most compilers can emit faster than a non zero or constant value. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 4 ++-- net/ipv4/tcp_output.c | 14 ++++++++------ net/ipv6/tcp_ipv6.c | 6 +++--- 3 files changed, 13 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 62d35afcb3ac..9065f8b7e646 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -775,9 +775,9 @@ static inline int sk_memalloc_socks(void) #endif -static inline gfp_t sk_gfp_atomic(const struct sock *sk, gfp_t gfp_mask) +static inline gfp_t sk_gfp_mask(const struct sock *sk, gfp_t gfp_mask) { - return GFP_ATOMIC | (sk->sk_allocation & __GFP_MEMALLOC); + return gfp_mask | (sk->sk_allocation & __GFP_MEMALLOC); } static inline void sk_acceptq_removed(struct sock *sk) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index cb7ca569052c..a800cee88035 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2296,7 +2296,7 @@ void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, return; if (tcp_write_xmit(sk, cur_mss, nonagle, 0, - sk_gfp_atomic(sk, GFP_ATOMIC))) + sk_gfp_mask(sk, GFP_ATOMIC))) tcp_check_probe_timer(sk); } @@ -3352,8 +3352,9 @@ void tcp_send_ack(struct sock *sk) * tcp_transmit_skb() will set the ownership to this * sock. */ - buff = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC)); - if (!buff) { + buff = alloc_skb(MAX_TCP_HEADER, + sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN)); + if (unlikely(!buff)) { inet_csk_schedule_ack(sk); inet_csk(sk)->icsk_ack.ato = TCP_ATO_MIN; inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, @@ -3375,7 +3376,7 @@ void tcp_send_ack(struct sock *sk) /* Send it off, this clears delayed acks for us. */ skb_mstamp_get(&buff->skb_mstamp); - tcp_transmit_skb(sk, buff, 0, sk_gfp_atomic(sk, GFP_ATOMIC)); + tcp_transmit_skb(sk, buff, 0, (__force gfp_t)0); } EXPORT_SYMBOL_GPL(tcp_send_ack); @@ -3396,7 +3397,8 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib) struct sk_buff *skb; /* We don't queue it, tcp_transmit_skb() sets ownership. */ - skb = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC)); + skb = alloc_skb(MAX_TCP_HEADER, + sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN)); if (!skb) return -1; @@ -3409,7 +3411,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib) tcp_init_nondata_skb(skb, tp->snd_una - !urgent, TCPHDR_ACK); skb_mstamp_get(&skb->skb_mstamp); NET_INC_STATS(sock_net(sk), mib); - return tcp_transmit_skb(sk, skb, 0, GFP_ATOMIC); + return tcp_transmit_skb(sk, skb, 0, (__force gfp_t)0); } void tcp_send_window_probe(struct sock *sk) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index c5429a636f1a..41bcd59a2ac7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1130,7 +1130,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * */ tcp_md5_do_add(newsk, (union tcp_md5_addr *)&newsk->sk_v6_daddr, AF_INET6, key->key, key->keylen, - sk_gfp_atomic(sk, GFP_ATOMIC)); + sk_gfp_mask(sk, GFP_ATOMIC)); } #endif @@ -1146,7 +1146,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * /* Clone pktoptions received with SYN, if we own the req */ if (ireq->pktopts) { newnp->pktoptions = skb_clone(ireq->pktopts, - sk_gfp_atomic(sk, GFP_ATOMIC)); + sk_gfp_mask(sk, GFP_ATOMIC)); consume_skb(ireq->pktopts); ireq->pktopts = NULL; if (newnp->pktoptions) @@ -1212,7 +1212,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) --ANK (980728) */ if (np->rxopt.all) - opt_skb = skb_clone(skb, sk_gfp_atomic(sk, GFP_ATOMIC)); + opt_skb = skb_clone(skb, sk_gfp_mask(sk, GFP_ATOMIC)); if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ struct dst_entry *dst = sk->sk_rx_dst; -- cgit v1.2.3-59-g8ed1b From c981e4213e9d2d4ec79501bd607722ec712742a2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:06 +0100 Subject: net: add netif_is_team_master helper Similar to other helpers, caller can use this to find out if device is team master. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/team/team.c | 1 + include/linux/netdevice.h | 8 ++++++++ 2 files changed, 9 insertions(+) (limited to 'include') diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 651d35ea22c5..d2f3ee832c47 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2054,6 +2054,7 @@ static void team_setup(struct net_device *dev) dev->flags |= IFF_MULTICAST; dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING); dev->priv_flags |= IFF_NO_QUEUE; + dev->priv_flags |= IFF_TEAM; /* * Indicate we support unicast address filtering. That way core won't diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index fcbc5259c630..2b889be65d88 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1273,6 +1273,7 @@ struct net_device_ops { * @IFF_NO_QUEUE: device can run without qdisc attached * @IFF_OPENVSWITCH: device is a Open vSwitch master * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device + * @IFF_TEAM: device is a team device */ enum netdev_priv_flags { IFF_802_1Q_VLAN = 1<<0, @@ -1299,6 +1300,7 @@ enum netdev_priv_flags { IFF_NO_QUEUE = 1<<21, IFF_OPENVSWITCH = 1<<22, IFF_L3MDEV_SLAVE = 1<<23, + IFF_TEAM = 1<<24, }; #define IFF_802_1Q_VLAN IFF_802_1Q_VLAN @@ -1325,6 +1327,7 @@ enum netdev_priv_flags { #define IFF_NO_QUEUE IFF_NO_QUEUE #define IFF_OPENVSWITCH IFF_OPENVSWITCH #define IFF_L3MDEV_SLAVE IFF_L3MDEV_SLAVE +#define IFF_TEAM IFF_TEAM /** * struct net_device - The DEVICE structure. @@ -3889,6 +3892,11 @@ static inline bool netif_is_ovs_master(const struct net_device *dev) return dev->priv_flags & IFF_OPENVSWITCH; } +static inline bool netif_is_team_master(struct net_device *dev) +{ + return dev->priv_flags & IFF_TEAM; +} + /* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */ static inline void netif_keep_dst(struct net_device *dev) { -- cgit v1.2.3-59-g8ed1b From f7f019ee6d117de5007d0b10e7960696bbf111eb Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:07 +0100 Subject: net: add netif_is_team_port helper Similar to other helpers, caller can use this to find out if device is team port. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2b889be65d88..b3601f8a9b42 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3897,6 +3897,11 @@ static inline bool netif_is_team_master(struct net_device *dev) return dev->priv_flags & IFF_TEAM; } +static inline bool netif_is_team_port(struct net_device *dev) +{ + return dev->priv_flags & IFF_TEAM_PORT; +} + /* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */ static inline void netif_keep_dst(struct net_device *dev) { -- cgit v1.2.3-59-g8ed1b From 7be61833042e7757745345eedc7b0efee240c189 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:08 +0100 Subject: net: add netif_is_lag_master helper Some code does not mind if the master is bond or team and treats them the same, as generic LAG. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b3601f8a9b42..3ca083efa560 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3902,6 +3902,11 @@ static inline bool netif_is_team_port(struct net_device *dev) return dev->priv_flags & IFF_TEAM_PORT; } +static inline bool netif_is_lag_master(struct net_device *dev) +{ + return netif_is_bond_master(dev) || netif_is_team_master(dev); +} + /* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */ static inline void netif_keep_dst(struct net_device *dev) { -- cgit v1.2.3-59-g8ed1b From e0ba1414f310c83bf425fe26fa2cd5f1befcd6dc Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:09 +0100 Subject: net: add netif_is_lag_port helper Some code does not mind if a device is bond slave or team port and treats them the same, as generic LAG ports. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3ca083efa560..1506be58c59a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3907,6 +3907,11 @@ static inline bool netif_is_lag_master(struct net_device *dev) return netif_is_bond_master(dev) || netif_is_team_master(dev); } +static inline bool netif_is_lag_port(struct net_device *dev) +{ + return netif_is_bond_slave(dev) || netif_is_team_port(dev); +} + /* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */ static inline void netif_keep_dst(struct net_device *dev) { -- cgit v1.2.3-59-g8ed1b From 6dffb0447c25476f499d205dfceb1972e8dae919 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:10 +0100 Subject: net: propagate upper priv via netdev_master_upper_dev_link Eliminate netdev_master_upper_dev_link_private and pass priv directly as a parameter of netdev_master_upper_dev_link. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 +- drivers/net/team/team.c | 2 +- drivers/net/vrf.c | 2 +- include/linux/netdevice.h | 6 ++---- net/batman-adv/hard-interface.c | 3 ++- net/bridge/br_if.c | 2 +- net/core/dev.c | 18 ++++++------------ net/openvswitch/vport-netdev.c | 2 +- 8 files changed, 15 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 9e0f8a7ef8b1..924015729b2d 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1204,7 +1204,7 @@ static int bond_master_upper_dev_link(struct net_device *bond_dev, { int err; - err = netdev_master_upper_dev_link_private(slave_dev, bond_dev, slave); + err = netdev_master_upper_dev_link(slave_dev, bond_dev, slave); if (err) return err; slave_dev->flags |= IFF_SLAVE; diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index d2f3ee832c47..b37f8d14dca0 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1083,7 +1083,7 @@ static int team_upper_dev_link(struct net_device *dev, { int err; - err = netdev_master_upper_dev_link(port_dev, dev); + err = netdev_master_upper_dev_link(port_dev, dev, NULL); if (err) return err; port_dev->priv_flags |= IFF_TEAM_PORT; diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index c2d54c4ed556..59c5bddeaedd 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -624,7 +624,7 @@ static int do_vrf_add_slave(struct net_device *dev, struct net_device *port_dev) goto out_fail; } - ret = netdev_master_upper_dev_link(port_dev, dev); + ret = netdev_master_upper_dev_link(port_dev, dev, NULL); if (ret < 0) goto out_unregister; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1506be58c59a..939b8f3de810 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3619,10 +3619,8 @@ struct net_device *netdev_master_upper_dev_get(struct net_device *dev); struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev); int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev); int netdev_master_upper_dev_link(struct net_device *dev, - struct net_device *upper_dev); -int netdev_master_upper_dev_link_private(struct net_device *dev, - struct net_device *upper_dev, - void *private); + struct net_device *upper_dev, + void *upper_priv); void netdev_upper_dev_unlink(struct net_device *dev, struct net_device *upper_dev); void netdev_adjacent_rename_links(struct net_device *dev, char *oldname); diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index f11345e163d7..a7f4f1085dbb 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -464,7 +464,8 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, hard_iface->soft_iface = soft_iface; bat_priv = netdev_priv(hard_iface->soft_iface); - ret = netdev_master_upper_dev_link(hard_iface->net_dev, soft_iface); + ret = netdev_master_upper_dev_link(hard_iface->net_dev, + soft_iface, NULL); if (ret) goto err_dev; diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index ec02f5869a78..781abc34667a 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -493,7 +493,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) dev->priv_flags |= IFF_BRIDGE_PORT; - err = netdev_master_upper_dev_link(dev, br->dev); + err = netdev_master_upper_dev_link(dev, br->dev, NULL); if (err) goto err5; diff --git a/net/core/dev.c b/net/core/dev.c index 939cd1b1da15..27d052bb78bc 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5421,7 +5421,7 @@ static void __netdev_adjacent_dev_unlink_neighbour(struct net_device *dev, static int __netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev, bool master, - void *private) + void *upper_priv) { struct netdev_notifier_changeupper_info changeupper_info; struct netdev_adjacent *i, *j, *to_i, *to_j; @@ -5452,7 +5452,7 @@ static int __netdev_upper_dev_link(struct net_device *dev, if (ret) return ret; - ret = __netdev_adjacent_dev_link_neighbour(dev, upper_dev, private, + ret = __netdev_adjacent_dev_link_neighbour(dev, upper_dev, upper_priv, master); if (ret) return ret; @@ -5557,6 +5557,7 @@ EXPORT_SYMBOL(netdev_upper_dev_link); * netdev_master_upper_dev_link - Add a master link to the upper device * @dev: device * @upper_dev: new upper device + * @upper_priv: upper device private * * Adds a link to device which is upper to this one. In this case, only * one master upper device can be linked, although other non-master devices @@ -5565,20 +5566,13 @@ EXPORT_SYMBOL(netdev_upper_dev_link); * counts are adjusted and the function returns zero. */ int netdev_master_upper_dev_link(struct net_device *dev, - struct net_device *upper_dev) + struct net_device *upper_dev, + void *upper_priv) { - return __netdev_upper_dev_link(dev, upper_dev, true, NULL); + return __netdev_upper_dev_link(dev, upper_dev, true, upper_priv); } EXPORT_SYMBOL(netdev_master_upper_dev_link); -int netdev_master_upper_dev_link_private(struct net_device *dev, - struct net_device *upper_dev, - void *private) -{ - return __netdev_upper_dev_link(dev, upper_dev, true, private); -} -EXPORT_SYMBOL(netdev_master_upper_dev_link_private); - /** * netdev_upper_dev_unlink - Removes a link to upper device * @dev: device diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index b327368a3848..3ee3df1edeae 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -105,7 +105,7 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) rtnl_lock(); err = netdev_master_upper_dev_link(vport->dev, - get_dpdev(vport->dp)); + get_dpdev(vport->dp), NULL); if (err) goto error_unlock; -- cgit v1.2.3-59-g8ed1b From 29bf24afb29042f568fa67b1b0eee46796725ed2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:11 +0100 Subject: net: add possibility to pass information about upper device via notifier Sometimes the drivers and other code would find it handy to know some internal information about upper device being changed. So allow upper-code to pass information down to notifier listeners during linking. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 +- drivers/net/team/team.c | 2 +- drivers/net/vrf.c | 2 +- include/linux/netdevice.h | 3 ++- net/batman-adv/hard-interface.c | 2 +- net/bridge/br_if.c | 2 +- net/core/dev.c | 11 +++++++---- net/openvswitch/vport-netdev.c | 2 +- 8 files changed, 15 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 924015729b2d..fa3ed1d8a12d 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1204,7 +1204,7 @@ static int bond_master_upper_dev_link(struct net_device *bond_dev, { int err; - err = netdev_master_upper_dev_link(slave_dev, bond_dev, slave); + err = netdev_master_upper_dev_link(slave_dev, bond_dev, slave, NULL); if (err) return err; slave_dev->flags |= IFF_SLAVE; diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index b37f8d14dca0..f7b6ff7948b8 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1083,7 +1083,7 @@ static int team_upper_dev_link(struct net_device *dev, { int err; - err = netdev_master_upper_dev_link(port_dev, dev, NULL); + err = netdev_master_upper_dev_link(port_dev, dev, NULL, NULL); if (err) return err; port_dev->priv_flags |= IFF_TEAM_PORT; diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 59c5bddeaedd..8944a49cda15 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -624,7 +624,7 @@ static int do_vrf_add_slave(struct net_device *dev, struct net_device *port_dev) goto out_fail; } - ret = netdev_master_upper_dev_link(port_dev, dev, NULL); + ret = netdev_master_upper_dev_link(port_dev, dev, NULL, NULL); if (ret < 0) goto out_unregister; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 939b8f3de810..aea556c64f2c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2163,6 +2163,7 @@ struct netdev_notifier_changeupper_info { struct net_device *upper_dev; /* new upper dev */ bool master; /* is upper dev master */ bool linking; /* is the nofication for link or unlink */ + void *upper_info; /* upper dev info */ }; static inline void netdev_notifier_info_init(struct netdev_notifier_info *info, @@ -3620,7 +3621,7 @@ struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev); int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev); int netdev_master_upper_dev_link(struct net_device *dev, struct net_device *upper_dev, - void *upper_priv); + void *upper_priv, void *upper_info); void netdev_upper_dev_unlink(struct net_device *dev, struct net_device *upper_dev); void netdev_adjacent_rename_links(struct net_device *dev, char *oldname); diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index a7f4f1085dbb..aa8867e1d983 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -465,7 +465,7 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, bat_priv = netdev_priv(hard_iface->soft_iface); ret = netdev_master_upper_dev_link(hard_iface->net_dev, - soft_iface, NULL); + soft_iface, NULL, NULL); if (ret) goto err_dev; diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 781abc34667a..8d1d4a22c50d 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -493,7 +493,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) dev->priv_flags |= IFF_BRIDGE_PORT; - err = netdev_master_upper_dev_link(dev, br->dev, NULL); + err = netdev_master_upper_dev_link(dev, br->dev, NULL, NULL); if (err) goto err5; diff --git a/net/core/dev.c b/net/core/dev.c index 27d052bb78bc..8ed886663c6d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5421,7 +5421,7 @@ static void __netdev_adjacent_dev_unlink_neighbour(struct net_device *dev, static int __netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev, bool master, - void *upper_priv) + void *upper_priv, void *upper_info) { struct netdev_notifier_changeupper_info changeupper_info; struct netdev_adjacent *i, *j, *to_i, *to_j; @@ -5445,6 +5445,7 @@ static int __netdev_upper_dev_link(struct net_device *dev, changeupper_info.upper_dev = upper_dev; changeupper_info.master = master; changeupper_info.linking = true; + changeupper_info.upper_info = upper_info; ret = call_netdevice_notifiers_info(NETDEV_PRECHANGEUPPER, dev, &changeupper_info.info); @@ -5549,7 +5550,7 @@ rollback_mesh: int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev) { - return __netdev_upper_dev_link(dev, upper_dev, false, NULL); + return __netdev_upper_dev_link(dev, upper_dev, false, NULL, NULL); } EXPORT_SYMBOL(netdev_upper_dev_link); @@ -5558,6 +5559,7 @@ EXPORT_SYMBOL(netdev_upper_dev_link); * @dev: device * @upper_dev: new upper device * @upper_priv: upper device private + * @upper_info: upper info to be passed down via notifier * * Adds a link to device which is upper to this one. In this case, only * one master upper device can be linked, although other non-master devices @@ -5567,9 +5569,10 @@ EXPORT_SYMBOL(netdev_upper_dev_link); */ int netdev_master_upper_dev_link(struct net_device *dev, struct net_device *upper_dev, - void *upper_priv) + void *upper_priv, void *upper_info) { - return __netdev_upper_dev_link(dev, upper_dev, true, upper_priv); + return __netdev_upper_dev_link(dev, upper_dev, true, + upper_priv, upper_info); } EXPORT_SYMBOL(netdev_master_upper_dev_link); diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 3ee3df1edeae..8f4dd4c39bfe 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -105,7 +105,7 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) rtnl_lock(); err = netdev_master_upper_dev_link(vport->dev, - get_dpdev(vport->dp), NULL); + get_dpdev(vport->dp), NULL, NULL); if (err) goto error_unlock; -- cgit v1.2.3-59-g8ed1b From 764f5e544118508add420724789f46e04dba91eb Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:12 +0100 Subject: net: add info struct for LAG changeupper This struct will be shared by bonding and team to pass internal information to notifier listeners. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index aea556c64f2c..3ab90ea0ed03 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2110,6 +2110,19 @@ struct pcpu_sw_netstats { #define netdev_alloc_pcpu_stats(type) \ __netdev_alloc_pcpu_stats(type, GFP_KERNEL); +enum netdev_lag_tx_type { + NETDEV_LAG_TX_TYPE_UNKNOWN, + NETDEV_LAG_TX_TYPE_RANDOM, + NETDEV_LAG_TX_TYPE_BROADCAST, + NETDEV_LAG_TX_TYPE_ROUNDROBIN, + NETDEV_LAG_TX_TYPE_ACTIVEBACKUP, + NETDEV_LAG_TX_TYPE_HASH, +}; + +struct netdev_lag_upper_info { + enum netdev_lag_tx_type tx_type; +}; + #include /* netdevice notifier chain. Please remember to update the rtnetlink -- cgit v1.2.3-59-g8ed1b From 8fd728566a354f7bc9cb6e781f185b8c39cf505b Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:13 +0100 Subject: team: fill-up LAG changeupper info struct and pass it along Initialize netdev_lag_upper_info structure by TX type according to current team mode and pass it along via netdev_master_upper_dev_link. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/team/team.c | 23 ++++++++++++----------- drivers/net/team/team_mode_activebackup.c | 1 + drivers/net/team/team_mode_broadcast.c | 1 + drivers/net/team/team_mode_loadbalance.c | 1 + drivers/net/team/team_mode_random.c | 1 + drivers/net/team/team_mode_roundrobin.c | 1 + include/linux/if_team.h | 1 + 7 files changed, 18 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index f7b6ff7948b8..dd1504bbb4a7 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1078,23 +1078,24 @@ static void team_port_disable_netpoll(struct team_port *port) } #endif -static int team_upper_dev_link(struct net_device *dev, - struct net_device *port_dev) +static int team_upper_dev_link(struct team *team, struct team_port *port) { + struct netdev_lag_upper_info lag_upper_info; int err; - err = netdev_master_upper_dev_link(port_dev, dev, NULL, NULL); + lag_upper_info.tx_type = team->mode->lag_tx_type; + err = netdev_master_upper_dev_link(port->dev, team->dev, NULL, + &lag_upper_info); if (err) return err; - port_dev->priv_flags |= IFF_TEAM_PORT; + port->dev->priv_flags |= IFF_TEAM_PORT; return 0; } -static void team_upper_dev_unlink(struct net_device *dev, - struct net_device *port_dev) +static void team_upper_dev_unlink(struct team *team, struct team_port *port) { - netdev_upper_dev_unlink(port_dev, dev); - port_dev->priv_flags &= ~IFF_TEAM_PORT; + netdev_upper_dev_unlink(port->dev, team->dev); + port->dev->priv_flags &= ~IFF_TEAM_PORT; } static void __team_port_change_port_added(struct team_port *port, bool linkup); @@ -1194,7 +1195,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev) goto err_handler_register; } - err = team_upper_dev_link(dev, port_dev); + err = team_upper_dev_link(team, port); if (err) { netdev_err(dev, "Device %s failed to set upper link\n", portname); @@ -1220,7 +1221,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev) return 0; err_option_port_add: - team_upper_dev_unlink(dev, port_dev); + team_upper_dev_unlink(team, port); err_set_upper_link: netdev_rx_handler_unregister(port_dev); @@ -1264,7 +1265,7 @@ static int team_port_del(struct team *team, struct net_device *port_dev) team_port_disable(team, port); list_del_rcu(&port->list); - team_upper_dev_unlink(dev, port_dev); + team_upper_dev_unlink(team, port); netdev_rx_handler_unregister(port_dev); team_port_disable_netpoll(port); vlan_vids_del_by_dev(port_dev, dev); diff --git a/drivers/net/team/team_mode_activebackup.c b/drivers/net/team/team_mode_activebackup.c index 40fd3381b693..3f189823ba3b 100644 --- a/drivers/net/team/team_mode_activebackup.c +++ b/drivers/net/team/team_mode_activebackup.c @@ -127,6 +127,7 @@ static const struct team_mode ab_mode = { .owner = THIS_MODULE, .priv_size = sizeof(struct ab_priv), .ops = &ab_mode_ops, + .lag_tx_type = NETDEV_LAG_TX_TYPE_ACTIVEBACKUP, }; static int __init ab_init_module(void) diff --git a/drivers/net/team/team_mode_broadcast.c b/drivers/net/team/team_mode_broadcast.c index c366cd299c06..302ff35b0cbc 100644 --- a/drivers/net/team/team_mode_broadcast.c +++ b/drivers/net/team/team_mode_broadcast.c @@ -56,6 +56,7 @@ static const struct team_mode bc_mode = { .kind = "broadcast", .owner = THIS_MODULE, .ops = &bc_mode_ops, + .lag_tx_type = NETDEV_LAG_TX_TYPE_BROADCAST, }; static int __init bc_init_module(void) diff --git a/drivers/net/team/team_mode_loadbalance.c b/drivers/net/team/team_mode_loadbalance.c index a1536d0d83a9..cdb19b385d42 100644 --- a/drivers/net/team/team_mode_loadbalance.c +++ b/drivers/net/team/team_mode_loadbalance.c @@ -661,6 +661,7 @@ static const struct team_mode lb_mode = { .priv_size = sizeof(struct lb_priv), .port_priv_size = sizeof(struct lb_port_priv), .ops = &lb_mode_ops, + .lag_tx_type = NETDEV_LAG_TX_TYPE_HASH, }; static int __init lb_init_module(void) diff --git a/drivers/net/team/team_mode_random.c b/drivers/net/team/team_mode_random.c index cd2f692b8074..215f845782db 100644 --- a/drivers/net/team/team_mode_random.c +++ b/drivers/net/team/team_mode_random.c @@ -46,6 +46,7 @@ static const struct team_mode rnd_mode = { .kind = "random", .owner = THIS_MODULE, .ops = &rnd_mode_ops, + .lag_tx_type = NETDEV_LAG_TX_TYPE_RANDOM, }; static int __init rnd_init_module(void) diff --git a/drivers/net/team/team_mode_roundrobin.c b/drivers/net/team/team_mode_roundrobin.c index 53665850b59e..0aa234118c03 100644 --- a/drivers/net/team/team_mode_roundrobin.c +++ b/drivers/net/team/team_mode_roundrobin.c @@ -58,6 +58,7 @@ static const struct team_mode rr_mode = { .owner = THIS_MODULE, .priv_size = sizeof(struct rr_priv), .ops = &rr_mode_ops, + .lag_tx_type = NETDEV_LAG_TX_TYPE_ROUNDROBIN, }; static int __init rr_init_module(void) diff --git a/include/linux/if_team.h b/include/linux/if_team.h index a6aa970758a2..b84e49c3a738 100644 --- a/include/linux/if_team.h +++ b/include/linux/if_team.h @@ -164,6 +164,7 @@ struct team_mode { size_t priv_size; size_t port_priv_size; const struct team_mode_ops *ops; + enum netdev_lag_tx_type lag_tx_type; }; #define TEAM_PORT_HASHBITS 4 -- cgit v1.2.3-59-g8ed1b From 04d482660a07039fc4e9a42bb3517db236d98f96 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:15 +0100 Subject: net: introduce change lower state notifier When lower device like bonding slave, team/bridge port, etc changes its state, it is useful for others to notice this change. Currently this is implemented specificly for bonding as NETDEV_BONDING_INFO notifier. This patch aims to replace this specific usage and make this more generic to be used for all upper-lower devices. Introduce NETDEV_CHANGELOWERSTATE netdev notifier type and netdev_lower_state_changed() helper. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 8 ++++++++ net/core/dev.c | 20 ++++++++++++++++++++ 2 files changed, 28 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3ab90ea0ed03..ad69f237aa78 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2158,6 +2158,7 @@ struct netdev_lag_upper_info { #define NETDEV_CHANGEINFODATA 0x0018 #define NETDEV_BONDING_INFO 0x0019 #define NETDEV_PRECHANGEUPPER 0x001A +#define NETDEV_CHANGELOWERSTATE 0x001B int register_netdevice_notifier(struct notifier_block *nb); int unregister_netdevice_notifier(struct notifier_block *nb); @@ -2179,6 +2180,11 @@ struct netdev_notifier_changeupper_info { void *upper_info; /* upper dev info */ }; +struct netdev_notifier_changelowerstate_info { + struct netdev_notifier_info info; /* must be first */ + void *lower_state_info; /* is lower dev state */ +}; + static inline void netdev_notifier_info_init(struct netdev_notifier_info *info, struct net_device *dev) { @@ -3640,6 +3646,8 @@ void netdev_upper_dev_unlink(struct net_device *dev, void netdev_adjacent_rename_links(struct net_device *dev, char *oldname); void *netdev_lower_dev_get_private(struct net_device *dev, struct net_device *lower_dev); +void netdev_lower_state_changed(struct net_device *lower_dev, + void *lower_state_info); /* RSS keys are 40 or 52 bytes long */ #define NETDEV_RSS_KEY_LEN 52 diff --git a/net/core/dev.c b/net/core/dev.c index 8ed886663c6d..d1706e88fbeb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5756,6 +5756,26 @@ int dev_get_nest_level(struct net_device *dev, } EXPORT_SYMBOL(dev_get_nest_level); +/** + * netdev_lower_change - Dispatch event about lower device state change + * @lower_dev: device + * @lower_state_info: state to dispatch + * + * Send NETDEV_CHANGELOWERSTATE to netdev notifiers with info. + * The caller must hold the RTNL lock. + */ +void netdev_lower_state_changed(struct net_device *lower_dev, + void *lower_state_info) +{ + struct netdev_notifier_changelowerstate_info changelowerstate_info; + + ASSERT_RTNL(); + changelowerstate_info.lower_state_info = lower_state_info; + call_netdevice_notifiers_info(NETDEV_CHANGELOWERSTATE, lower_dev, + &changelowerstate_info.info); +} +EXPORT_SYMBOL(netdev_lower_state_changed); + static void dev_change_rx_flags(struct net_device *dev, int flags) { const struct net_device_ops *ops = dev->netdev_ops; -- cgit v1.2.3-59-g8ed1b From fb1b2e3ce53aef80b3cef71f3885d584cdbdc6b8 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:16 +0100 Subject: net: introduce lower state changed info structure for LAG lowers This is shared info structure for bonding and team. Serves to pass down info about link state and port activity to notification listeners. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ad69f237aa78..fa84b59eb197 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2123,6 +2123,11 @@ struct netdev_lag_upper_info { enum netdev_lag_tx_type tx_type; }; +struct netdev_lag_lower_state_info { + u8 link_up : 1, + tx_enabled : 1; +}; + #include /* netdevice notifier chain. Please remember to update the rtnetlink -- cgit v1.2.3-59-g8ed1b From 5d397061ca2081d8a99e4bee5792122faa6aaf86 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:19 +0100 Subject: bonding: allow notifications for bond_set_slave_link_state Similar to state notifications. We allow caller to indicate if the notification should happen now or later, depending on if he holds rtnl mutex or not. Introduce bond_slave_link_notify function (similar to bond_slave_state_notify) which is later on called with rtnl mutex and goes over slaves and executes delayed notification. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 54 +++++++++++++++++++++++++++-------------- include/net/bonding.h | 32 +++++++++++++++++++++--- 2 files changed, 65 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index e94e79ad757c..7695490061ec 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -830,7 +830,8 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) } new_active->delay = 0; - bond_set_slave_link_state(new_active, BOND_LINK_UP); + bond_set_slave_link_state(new_active, BOND_LINK_UP, + BOND_SLAVE_NOTIFY_NOW); if (BOND_MODE(bond) == BOND_MODE_8023AD) bond_3ad_handle_link_change(new_active, BOND_LINK_UP); @@ -1580,21 +1581,26 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) if (bond_check_dev_link(bond, slave_dev, 0) == BMSR_LSTATUS) { if (bond->params.updelay) { bond_set_slave_link_state(new_slave, - BOND_LINK_BACK); + BOND_LINK_BACK, + BOND_SLAVE_NOTIFY_NOW); new_slave->delay = bond->params.updelay; } else { bond_set_slave_link_state(new_slave, - BOND_LINK_UP); + BOND_LINK_UP, + BOND_SLAVE_NOTIFY_NOW); } } else { - bond_set_slave_link_state(new_slave, BOND_LINK_DOWN); + bond_set_slave_link_state(new_slave, BOND_LINK_DOWN, + BOND_SLAVE_NOTIFY_NOW); } } else if (bond->params.arp_interval) { bond_set_slave_link_state(new_slave, (netif_carrier_ok(slave_dev) ? - BOND_LINK_UP : BOND_LINK_DOWN)); + BOND_LINK_UP : BOND_LINK_DOWN), + BOND_SLAVE_NOTIFY_NOW); } else { - bond_set_slave_link_state(new_slave, BOND_LINK_UP); + bond_set_slave_link_state(new_slave, BOND_LINK_UP, + BOND_SLAVE_NOTIFY_NOW); } if (new_slave->link != BOND_LINK_DOWN) @@ -2013,7 +2019,8 @@ static int bond_miimon_inspect(struct bonding *bond) if (link_state) continue; - bond_set_slave_link_state(slave, BOND_LINK_FAIL); + bond_set_slave_link_state(slave, BOND_LINK_FAIL, + BOND_SLAVE_NOTIFY_LATER); slave->delay = bond->params.downdelay; if (slave->delay) { netdev_info(bond->dev, "link status down for %sinterface %s, disabling it in %d ms\n", @@ -2028,7 +2035,8 @@ static int bond_miimon_inspect(struct bonding *bond) case BOND_LINK_FAIL: if (link_state) { /* recovered before downdelay expired */ - bond_set_slave_link_state(slave, BOND_LINK_UP); + bond_set_slave_link_state(slave, BOND_LINK_UP, + BOND_SLAVE_NOTIFY_LATER); slave->last_link_up = jiffies; netdev_info(bond->dev, "link status up again after %d ms for interface %s\n", (bond->params.downdelay - slave->delay) * @@ -2050,7 +2058,8 @@ static int bond_miimon_inspect(struct bonding *bond) if (!link_state) continue; - bond_set_slave_link_state(slave, BOND_LINK_BACK); + bond_set_slave_link_state(slave, BOND_LINK_BACK, + BOND_SLAVE_NOTIFY_LATER); slave->delay = bond->params.updelay; if (slave->delay) { @@ -2064,7 +2073,8 @@ static int bond_miimon_inspect(struct bonding *bond) case BOND_LINK_BACK: if (!link_state) { bond_set_slave_link_state(slave, - BOND_LINK_DOWN); + BOND_LINK_DOWN, + BOND_SLAVE_NOTIFY_LATER); netdev_info(bond->dev, "link status down again after %d ms for interface %s\n", (bond->params.updelay - slave->delay) * bond->params.miimon, @@ -2102,7 +2112,8 @@ static void bond_miimon_commit(struct bonding *bond) continue; case BOND_LINK_UP: - bond_set_slave_link_state(slave, BOND_LINK_UP); + bond_set_slave_link_state(slave, BOND_LINK_UP, + BOND_SLAVE_NOTIFY_NOW); slave->last_link_up = jiffies; primary = rtnl_dereference(bond->primary_slave); @@ -2142,7 +2153,8 @@ static void bond_miimon_commit(struct bonding *bond) if (slave->link_failure_count < UINT_MAX) slave->link_failure_count++; - bond_set_slave_link_state(slave, BOND_LINK_DOWN); + bond_set_slave_link_state(slave, BOND_LINK_DOWN, + BOND_SLAVE_NOTIFY_NOW); if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP || BOND_MODE(bond) == BOND_MODE_8023AD) @@ -2725,7 +2737,8 @@ static void bond_ab_arp_commit(struct bonding *bond) struct slave *current_arp_slave; current_arp_slave = rtnl_dereference(bond->current_arp_slave); - bond_set_slave_link_state(slave, BOND_LINK_UP); + bond_set_slave_link_state(slave, BOND_LINK_UP, + BOND_SLAVE_NOTIFY_NOW); if (current_arp_slave) { bond_set_slave_inactive_flags( current_arp_slave, @@ -2748,7 +2761,8 @@ static void bond_ab_arp_commit(struct bonding *bond) if (slave->link_failure_count < UINT_MAX) slave->link_failure_count++; - bond_set_slave_link_state(slave, BOND_LINK_DOWN); + bond_set_slave_link_state(slave, BOND_LINK_DOWN, + BOND_SLAVE_NOTIFY_NOW); bond_set_slave_inactive_flags(slave, BOND_SLAVE_NOTIFY_NOW); @@ -2827,7 +2841,8 @@ static bool bond_ab_arp_probe(struct bonding *bond) * up when it is actually down */ if (!bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) { - bond_set_slave_link_state(slave, BOND_LINK_DOWN); + bond_set_slave_link_state(slave, BOND_LINK_DOWN, + BOND_SLAVE_NOTIFY_LATER); if (slave->link_failure_count < UINT_MAX) slave->link_failure_count++; @@ -2847,7 +2862,8 @@ static bool bond_ab_arp_probe(struct bonding *bond) if (!new_slave) goto check_state; - bond_set_slave_link_state(new_slave, BOND_LINK_BACK); + bond_set_slave_link_state(new_slave, BOND_LINK_BACK, + BOND_SLAVE_NOTIFY_LATER); bond_set_slave_active_flags(new_slave, BOND_SLAVE_NOTIFY_LATER); bond_arp_send_all(bond, new_slave); new_slave->last_link_up = jiffies; @@ -2855,7 +2871,7 @@ static bool bond_ab_arp_probe(struct bonding *bond) check_state: bond_for_each_slave_rcu(bond, slave, iter) { - if (slave->should_notify) { + if (slave->should_notify || slave->should_notify_link) { should_notify_rtnl = BOND_SLAVE_NOTIFY_NOW; break; } @@ -2910,8 +2926,10 @@ re_arm: if (should_notify_peers) call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, bond->dev); - if (should_notify_rtnl) + if (should_notify_rtnl) { bond_slave_state_notify(bond); + bond_slave_link_notify(bond); + } rtnl_unlock(); } diff --git a/include/net/bonding.h b/include/net/bonding.h index c1740a2794a3..1df437715e2f 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -165,7 +165,8 @@ struct slave { u8 backup:1, /* indicates backup slave. Value corresponds with BOND_STATE_ACTIVE and BOND_STATE_BACKUP */ inactive:1, /* indicates inactive slave */ - should_notify:1; /* indicateds whether the state changed */ + should_notify:1, /* indicates whether the state changed */ + should_notify_link:1; /* indicates whether the link changed */ u8 duplex; u32 original_mtu; u32 link_failure_count; @@ -504,10 +505,35 @@ static inline bool bond_is_slave_inactive(struct slave *slave) return slave->inactive; } -static inline void bond_set_slave_link_state(struct slave *slave, int state) +static inline void bond_set_slave_link_state(struct slave *slave, int state, + bool notify) { + if (slave->link == state) + return; + slave->link = state; - bond_queue_slave_event(slave); + if (notify) { + bond_queue_slave_event(slave); + slave->should_notify_link = 0; + } else { + if (slave->should_notify_link) + slave->should_notify_link = 0; + else + slave->should_notify_link = 1; + } +} + +static inline void bond_slave_link_notify(struct bonding *bond) +{ + struct list_head *iter; + struct slave *tmp; + + bond_for_each_slave(bond, tmp, iter) { + if (tmp->should_notify_link) { + bond_queue_slave_event(tmp); + tmp->should_notify_link = 0; + } + } } static inline __be32 bond_confirm_addr(struct net_device *dev, __be32 dst, __be32 local) -- cgit v1.2.3-59-g8ed1b From f7c7eb7f7af7f87e0fc150994785fd139576e43a Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 3 Dec 2015 12:12:20 +0100 Subject: bonding: implement lower state change propagation Let netdev notifier listeners know about link and slave state change. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 10 ++++++++++ include/net/bonding.h | 7 +++++++ 2 files changed, 17 insertions(+) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 7695490061ec..2f1145063c60 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1317,6 +1317,16 @@ void bond_queue_slave_event(struct slave *slave) queue_delayed_work(slave->bond->wq, &nnw->work, 0); } +void bond_lower_state_changed(struct slave *slave) +{ + struct netdev_lag_lower_state_info info; + + info.link_up = slave->link == BOND_LINK_UP || + slave->link == BOND_LINK_FAIL; + info.tx_enabled = bond_is_active_slave(slave); + netdev_lower_state_changed(slave->dev, &info); +} + /* enslave device to bond device */ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) { diff --git a/include/net/bonding.h b/include/net/bonding.h index 1df437715e2f..ee6c52053aa3 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -247,6 +247,7 @@ struct bonding { ((struct slave *) rtnl_dereference(dev->rx_handler_data)) void bond_queue_slave_event(struct slave *slave); +void bond_lower_state_changed(struct slave *slave); struct bond_vlan_tag { __be16 vlan_proto; @@ -328,6 +329,7 @@ static inline void bond_set_active_slave(struct slave *slave) if (slave->backup) { slave->backup = 0; bond_queue_slave_event(slave); + bond_lower_state_changed(slave); rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); } } @@ -337,6 +339,7 @@ static inline void bond_set_backup_slave(struct slave *slave) if (!slave->backup) { slave->backup = 1; bond_queue_slave_event(slave); + bond_lower_state_changed(slave); rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); } } @@ -349,6 +352,7 @@ static inline void bond_set_slave_state(struct slave *slave, slave->backup = slave_state; if (notify) { + bond_lower_state_changed(slave); rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); bond_queue_slave_event(slave); slave->should_notify = 0; @@ -380,6 +384,7 @@ static inline void bond_slave_state_notify(struct bonding *bond) bond_for_each_slave(bond, tmp, iter) { if (tmp->should_notify) { + bond_lower_state_changed(tmp); rtmsg_ifinfo(RTM_NEWLINK, tmp->dev, 0, GFP_ATOMIC); tmp->should_notify = 0; } @@ -514,6 +519,7 @@ static inline void bond_set_slave_link_state(struct slave *slave, int state, slave->link = state; if (notify) { bond_queue_slave_event(slave); + bond_lower_state_changed(slave); slave->should_notify_link = 0; } else { if (slave->should_notify_link) @@ -531,6 +537,7 @@ static inline void bond_slave_link_notify(struct bonding *bond) bond_for_each_slave(bond, tmp, iter) { if (tmp->should_notify_link) { bond_queue_slave_event(tmp); + bond_lower_state_changed(tmp); tmp->should_notify_link = 0; } } -- cgit v1.2.3-59-g8ed1b From fc50db98ff872372f266695858f87a12eb1b4f05 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 1 Dec 2015 18:03:09 +0200 Subject: net/mlx5_core: Add base sriov support This patch adds SRIOV base support for mlx5 supported devices. The same driver is used for both PFs and VFs; VFs are identified by the driver through the flag MLX5_PCI_DEV_IS_VF added to the pci table entries. Virtual functions are created as usual through writing a value to the sriov_numvs sysfs file of the PF device. Upon instantiating VFs, they will all be probed by the driver on the hypervisor. One can gracefully unbind them through /sys/bus/pci/drivers/mlx5_core/unbind. mlx5_wait_for_vf_pages() was added to ensure that when a VF dies without executing proper teardown, the hypervisor driver waits till all of the pages that were allocated at the hypervisor to maintain its operation are returned. In order for the VF to be operational, the PF needs to call enable_hca for it. This can be done before the VFs are created through a call to pci_enable_sriov. If the there are VFs assigned to a VMs when the driver of the PF is unloaded, all the VF will experience system error and PF driver unloads cleanly; in this case pci_disable_sriov is not called and the devices will show when running lspci. Once the PF driver is reloaded, it will sync its data structures which maintain state on its VFs. Signed-off-by: Eli Cohen Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 36 +++- .../net/ethernet/mellanox/mlx5/core/mlx5_core.h | 2 + .../net/ethernet/mellanox/mlx5/core/pagealloc.c | 38 ++++ drivers/net/ethernet/mellanox/mlx5/core/sriov.c | 221 +++++++++++++++++++++ include/linux/mlx5/driver.h | 24 +++ include/linux/mlx5/mlx5_ifc.h | 4 +- 7 files changed, 318 insertions(+), 9 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/sriov.c (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 26a68b8af2c5..4d5103911527 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -2,7 +2,7 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ - mad.o transobj.o vport.o + mad.o transobj.o vport.o sriov.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o \ en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \ en_txrx.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index f2e64dc1a443..66e2b37cfbbf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -454,6 +454,9 @@ static int set_hca_ctrl(struct mlx5_core_dev *dev) struct mlx5_reg_host_endianess he_out; int err; + if (!mlx5_core_is_pf(dev)) + return 0; + memset(&he_in, 0, sizeof(he_in)); he_in.he = MLX5_SET_HOST_ENDIANNESS; err = mlx5_core_access_reg(dev, &he_in, sizeof(he_in), @@ -1049,6 +1052,12 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_init_srq_table(dev); mlx5_init_mr_table(dev); + err = mlx5_sriov_init(dev); + if (err) { + dev_err(&pdev->dev, "sriov init failed %d\n", err); + goto err_sriov; + } + err = mlx5_register_device(dev); if (err) { dev_err(&pdev->dev, "mlx5_register_device failed %d\n", err); @@ -1065,6 +1074,10 @@ out: return 0; +err_sriov: + if (mlx5_sriov_cleanup(dev)) + dev_err(&dev->pdev->dev, "sriov cleanup failed\n"); + err_reg_dev: mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); @@ -1120,6 +1133,13 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) { int err = 0; + err = mlx5_sriov_cleanup(dev); + if (err) { + dev_warn(&dev->pdev->dev, "%s: sriov cleanup failed - abort\n", + __func__); + return err; + } + mutex_lock(&dev->intf_state_mutex); if (dev->interface_state == MLX5_INTERFACE_STATE_DOWN) { dev_warn(&dev->pdev->dev, "%s: interface is down, NOP\n", @@ -1192,6 +1212,7 @@ static int init_one(struct pci_dev *pdev, return -ENOMEM; } priv = &dev->priv; + priv->pci_dev_data = id->driver_data; pci_set_drvdata(pdev, dev); @@ -1362,12 +1383,12 @@ static const struct pci_error_handlers mlx5_err_handler = { }; static const struct pci_device_id mlx5_core_pci_table[] = { - { PCI_VDEVICE(MELLANOX, 0x1011) }, /* Connect-IB */ - { PCI_VDEVICE(MELLANOX, 0x1012) }, /* Connect-IB VF */ - { PCI_VDEVICE(MELLANOX, 0x1013) }, /* ConnectX-4 */ - { PCI_VDEVICE(MELLANOX, 0x1014) }, /* ConnectX-4 VF */ - { PCI_VDEVICE(MELLANOX, 0x1015) }, /* ConnectX-4LX */ - { PCI_VDEVICE(MELLANOX, 0x1016) }, /* ConnectX-4LX VF */ + { PCI_VDEVICE(MELLANOX, 0x1011) }, /* Connect-IB */ + { PCI_VDEVICE(MELLANOX, 0x1012), MLX5_PCI_DEV_IS_VF}, /* Connect-IB VF */ + { PCI_VDEVICE(MELLANOX, 0x1013) }, /* ConnectX-4 */ + { PCI_VDEVICE(MELLANOX, 0x1014), MLX5_PCI_DEV_IS_VF}, /* ConnectX-4 VF */ + { PCI_VDEVICE(MELLANOX, 0x1015) }, /* ConnectX-4LX */ + { PCI_VDEVICE(MELLANOX, 0x1016), MLX5_PCI_DEV_IS_VF}, /* ConnectX-4LX VF */ { 0, } }; @@ -1378,7 +1399,8 @@ static struct pci_driver mlx5_core_driver = { .id_table = mlx5_core_pci_table, .probe = init_one, .remove = remove_one, - .err_handler = &mlx5_err_handler + .err_handler = &mlx5_err_handler, + .sriov_configure = mlx5_core_sriov_configure, }; static int __init init(void) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 1ed2239a6a6d..1649d5cf9e29 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -90,8 +90,10 @@ void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event, unsigned long param); void mlx5_enter_error_state(struct mlx5_core_dev *dev); void mlx5_disable_device(struct mlx5_core_dev *dev); +int mlx5_core_sriov_configure(struct pci_dev *dev, int num_vfs); int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id); int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id); +int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev); void mlx5e_init(void); void mlx5e_cleanup(void); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index 4d3377b12657..9eeee0545f1c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include "mlx5_core.h" @@ -95,6 +96,7 @@ struct mlx5_manage_pages_outbox { enum { MAX_RECLAIM_TIME_MSECS = 5000, + MAX_RECLAIM_VFS_PAGES_TIME_MSECS = 2 * 1000 * 60, }; enum { @@ -352,6 +354,10 @@ retry: goto out_4k; } + dev->priv.fw_pages += npages; + if (func_id) + dev->priv.vfs_pages += npages; + mlx5_core_dbg(dev, "err %d\n", err); kvfree(in); @@ -405,6 +411,12 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages, } num_claimed = be32_to_cpu(out->num_entries); + if (num_claimed > npages) { + mlx5_core_warn(dev, "fw returned %d, driver asked %d => corruption\n", + num_claimed, npages); + err = -EINVAL; + goto out_free; + } if (nclaimed) *nclaimed = num_claimed; @@ -412,6 +424,9 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages, addr = be64_to_cpu(out->pas[i]); free_4k(dev, addr); } + dev->priv.fw_pages -= num_claimed; + if (func_id) + dev->priv.vfs_pages -= num_claimed; out_free: kvfree(out); @@ -548,3 +563,26 @@ void mlx5_pagealloc_stop(struct mlx5_core_dev *dev) { destroy_workqueue(dev->priv.pg_wq); } + +int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev) +{ + unsigned long end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS); + int prev_vfs_pages = dev->priv.vfs_pages; + + mlx5_core_dbg(dev, "Waiting for %d pages from %s\n", prev_vfs_pages, + dev->priv.name); + while (dev->priv.vfs_pages) { + if (time_after(jiffies, end)) { + mlx5_core_warn(dev, "aborting while there are %d pending pages\n", dev->priv.vfs_pages); + return -ETIMEDOUT; + } + if (dev->priv.vfs_pages < prev_vfs_pages) { + end = jiffies + msecs_to_jiffies(MAX_RECLAIM_VFS_PAGES_TIME_MSECS); + prev_vfs_pages = dev->priv.vfs_pages; + } + msleep(50); + } + + mlx5_core_dbg(dev, "All pages received from %s\n", dev->priv.name); + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c new file mode 100644 index 000000000000..19a43240e359 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c @@ -0,0 +1,221 @@ +/* + * Copyright (c) 2014, Mellanox Technologies inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include "mlx5_core.h" + +static void enable_vfs(struct mlx5_core_dev *dev, int num_vfs) +{ + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + int err; + int vf; + + for (vf = 1; vf <= num_vfs; vf++) { + err = mlx5_core_enable_hca(dev, vf); + if (err) { + mlx5_core_warn(dev, "failed to enable VF %d\n", vf - 1); + } else { + sriov->vfs_ctx[vf - 1].enabled = 1; + mlx5_core_dbg(dev, "successfully enabled VF %d\n", vf - 1); + } + } +} + +static void disable_vfs(struct mlx5_core_dev *dev, int num_vfs) +{ + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + int vf; + + for (vf = 1; vf <= num_vfs; vf++) { + if (sriov->vfs_ctx[vf - 1].enabled) { + if (mlx5_core_disable_hca(dev, vf)) + mlx5_core_warn(dev, "failed to disable VF %d\n", vf - 1); + else + sriov->vfs_ctx[vf - 1].enabled = 0; + } + } +} + +static int mlx5_core_create_vfs(struct pci_dev *pdev, int num_vfs) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + int err; + + if (pci_num_vf(pdev)) + pci_disable_sriov(pdev); + + enable_vfs(dev, num_vfs); + + err = pci_enable_sriov(pdev, num_vfs); + if (err) { + dev_warn(&pdev->dev, "enable sriov failed %d\n", err); + goto ex; + } + + return 0; + +ex: + disable_vfs(dev, num_vfs); + return err; +} + +static int mlx5_core_sriov_enable(struct pci_dev *pdev, int num_vfs) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + int err; + + kfree(sriov->vfs_ctx); + sriov->vfs_ctx = kcalloc(num_vfs, sizeof(*sriov->vfs_ctx), GFP_ATOMIC); + if (!sriov->vfs_ctx) + return -ENOMEM; + + sriov->enabled_vfs = num_vfs; + err = mlx5_core_create_vfs(pdev, num_vfs); + if (err) { + kfree(sriov->vfs_ctx); + sriov->vfs_ctx = NULL; + return err; + } + + return 0; +} + +static void mlx5_core_init_vfs(struct mlx5_core_dev *dev, int num_vfs) +{ + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + + sriov->num_vfs = num_vfs; +} + +static void mlx5_core_cleanup_vfs(struct mlx5_core_dev *dev) +{ + struct mlx5_core_sriov *sriov; + + sriov = &dev->priv.sriov; + disable_vfs(dev, sriov->num_vfs); + + if (mlx5_wait_for_vf_pages(dev)) + mlx5_core_warn(dev, "timeout claiming VFs pages\n"); + + sriov->num_vfs = 0; +} + +int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + int err; + + mlx5_core_dbg(dev, "requsted num_vfs %d\n", num_vfs); + if (!mlx5_core_is_pf(dev)) + return -EPERM; + + mlx5_core_cleanup_vfs(dev); + + if (!num_vfs) { + kfree(sriov->vfs_ctx); + sriov->vfs_ctx = NULL; + if (!pci_vfs_assigned(pdev)) + pci_disable_sriov(pdev); + else + pr_info("unloading PF driver while leaving orphan VFs\n"); + + return 0; + } + + err = mlx5_core_sriov_enable(pdev, num_vfs); + if (err) { + dev_warn(&pdev->dev, "mlx5_core_sriov_enable failed %d\n", err); + return err; + } + + mlx5_core_init_vfs(dev, num_vfs); + + return num_vfs; +} + +static int sync_required(struct pci_dev *pdev) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + int cur_vfs = pci_num_vf(pdev); + + if (cur_vfs != sriov->num_vfs) { + pr_info("current VFs %d, registered %d - sync needed\n", cur_vfs, sriov->num_vfs); + return 1; + } + + return 0; +} + +int mlx5_sriov_init(struct mlx5_core_dev *dev) +{ + struct mlx5_core_sriov *sriov = &dev->priv.sriov; + struct pci_dev *pdev = dev->pdev; + int cur_vfs; + + if (!mlx5_core_is_pf(dev)) + return 0; + + if (!sync_required(dev->pdev)) + return 0; + + cur_vfs = pci_num_vf(pdev); + sriov->vfs_ctx = kcalloc(cur_vfs, sizeof(*sriov->vfs_ctx), GFP_KERNEL); + if (!sriov->vfs_ctx) + return -ENOMEM; + + sriov->enabled_vfs = cur_vfs; + + mlx5_core_init_vfs(dev, cur_vfs); + + enable_vfs(dev, cur_vfs); + + return 0; +} + +int mlx5_sriov_cleanup(struct mlx5_core_dev *dev) +{ + struct pci_dev *pdev = dev->pdev; + int err; + + if (!mlx5_core_is_pf(dev)) + return 0; + + err = mlx5_core_sriov_configure(pdev, 0); + if (err) + return err; + + return 0; +} diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 5c857f2a20d7..efebb87163c8 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -426,6 +426,16 @@ struct mlx5_mr_table { struct radix_tree_root tree; }; +struct mlx5_vf_context { + int enabled; +}; + +struct mlx5_core_sriov { + struct mlx5_vf_context *vfs_ctx; + int num_vfs; + int enabled_vfs; +}; + struct mlx5_irq_info { cpumask_var_t mask; char name[MLX5_MAX_IRQ_NAME]; @@ -447,6 +457,7 @@ struct mlx5_priv { int fw_pages; atomic_t reg_pages; struct list_head free_list; + int vfs_pages; struct mlx5_core_health health; @@ -485,6 +496,8 @@ struct mlx5_priv { struct list_head dev_list; struct list_head ctx_list; spinlock_t ctx_lock; + struct mlx5_core_sriov sriov; + unsigned long pci_dev_data; }; enum mlx5_device_state { @@ -739,6 +752,8 @@ void mlx5_pagealloc_init(struct mlx5_core_dev *dev); void mlx5_pagealloc_cleanup(struct mlx5_core_dev *dev); int mlx5_pagealloc_start(struct mlx5_core_dev *dev); void mlx5_pagealloc_stop(struct mlx5_core_dev *dev); +int mlx5_sriov_init(struct mlx5_core_dev *dev); +int mlx5_sriov_cleanup(struct mlx5_core_dev *dev); void mlx5_core_req_pages_handler(struct mlx5_core_dev *dev, u16 func_id, s32 npages); int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot); @@ -884,6 +899,15 @@ struct mlx5_profile { } mr_cache[MAX_MR_CACHE_ENTRIES]; }; +enum { + MLX5_PCI_DEV_IS_VF = 1 << 0, +}; + +static inline int mlx5_core_is_pf(struct mlx5_core_dev *dev) +{ + return !(dev->priv.pci_dev_data & MLX5_PCI_DEV_IS_VF); +} + static inline int mlx5_get_gid_table_len(u16 param) { if (param > 4) { diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 1565324eb620..9b76fddd696b 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -665,7 +665,9 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_17[0x1]; u8 ets[0x1]; u8 nic_flow_table[0x1]; - u8 reserved_18[0x4]; + u8 reserved_18_0; + u8 early_vf_enable; + u8 reserved_18[0x2]; u8 local_ca_ack_delay[0x5]; u8 reserved_19[0x6]; u8 port_type[0x2]; -- cgit v1.2.3-59-g8ed1b From 54f0a411ec72cb437d57d0c9654dcbd0f198ff3a Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:10 +0200 Subject: net/mlx5: Add HW capabilities and structs for SR-IOV E-Switch Update HCA capabilities and HW struct to include needed capabilities for upcoming Ethernet Switch (SR-IOV E-Switch). Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- include/linux/mlx5/mlx5_ifc.h | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 9b76fddd696b..836cf0e43174 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -665,7 +665,7 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_17[0x1]; u8 ets[0x1]; u8 nic_flow_table[0x1]; - u8 reserved_18_0; + u8 eswitch_flow_table[0x1]; u8 early_vf_enable; u8 reserved_18[0x2]; u8 local_ca_ack_delay[0x5]; @@ -789,22 +789,30 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_60[0x1b]; u8 log_max_wq_sz[0x5]; - u8 reserved_61[0xa0]; - + u8 nic_vport_change_event[0x1]; + u8 reserved_61[0xa]; + u8 log_max_vlan_list[0x5]; u8 reserved_62[0x3]; + u8 log_max_current_mc_list[0x5]; + u8 reserved_63[0x3]; + u8 log_max_current_uc_list[0x5]; + + u8 reserved_64[0x80]; + + u8 reserved_65[0x3]; u8 log_max_l2_table[0x5]; - u8 reserved_63[0x8]; + u8 reserved_66[0x8]; u8 log_uar_page_sz[0x10]; - u8 reserved_64[0x100]; + u8 reserved_67[0xe0]; - u8 reserved_65[0x1f]; + u8 reserved_68[0x1f]; u8 cqe_zip[0x1]; u8 cqe_zip_timeout[0x10]; u8 cqe_zip_max_num[0x10]; - u8 reserved_66[0x220]; + u8 reserved_69[0x220]; }; enum { @@ -2135,10 +2143,6 @@ struct mlx5_ifc_rmpc_bits { struct mlx5_ifc_wq_bits wq; }; -enum { - MLX5_NIC_VPORT_CONTEXT_ALLOWED_LIST_TYPE_CURRENT_UC_MAC_ADDRESS = 0x0, -}; - struct mlx5_ifc_nic_vport_context_bits { u8 reserved_0[0x1f]; u8 roce_en[0x1]; -- cgit v1.2.3-59-g8ed1b From e1d7d349c69d12721c420f1fe673ce9aa462aadd Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:11 +0200 Subject: net/mlx5: Update access functions to Query/Modify vport MAC address In preparation for SR-IOV we add here an API to enable each e-switch client (PF/VF) to configure its L2 MAC addresses and for the e-switch manager (usually the PF) to access them in order to be able to configure them into the e-switch. Therefore we now pass vport num parameter to mlx5_query_nic_vport_context, so PF can access other vports contexts. preperation for ethernet sriov and l2 table management. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 87 ++++++++++++++++++++--- include/linux/mlx5/vport.h | 5 +- 3 files changed, 81 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index f6a8cc787603..2ef717f98b0d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -2028,7 +2028,7 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); - mlx5_query_nic_vport_mac_address(priv->mdev, netdev->dev_addr); + mlx5_query_nic_vport_mac_address(priv->mdev, 0, netdev->dev_addr); } static void mlx5e_build_netdev(struct net_device *netdev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index b94177ebcf3a..442916e98724 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -57,33 +57,98 @@ u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod) } EXPORT_SYMBOL(mlx5_query_vport_state); -void mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u8 *addr) +static int mlx5_query_nic_vport_context(struct mlx5_core_dev *mdev, u16 vport, + u32 *out, int outlen) +{ + u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(query_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); + + MLX5_SET(query_nic_vport_context_in, in, vport_number, vport); + if (vport) + MLX5_SET(query_nic_vport_context_in, in, other_vport, 1); + + return mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen); +} + +static int mlx5_modify_nic_vport_context(struct mlx5_core_dev *mdev, void *in, + int inlen) +{ + u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)]; + + MLX5_SET(modify_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT); + + memset(out, 0, sizeof(out)); + return mlx5_cmd_exec_check_status(mdev, in, inlen, out, sizeof(out)); +} + +int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, + u16 vport, u8 *addr) { - u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)]; u32 *out; int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out); u8 *out_addr; + int err; out = mlx5_vzalloc(outlen); if (!out) - return; + return -ENOMEM; out_addr = MLX5_ADDR_OF(query_nic_vport_context_out, out, nic_vport_context.permanent_address); - memset(in, 0, sizeof(in)); - - MLX5_SET(query_nic_vport_context_in, in, opcode, - MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); - - memset(out, 0, outlen); - mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen); + err = mlx5_query_nic_vport_context(mdev, vport, out, outlen); + if (err) + goto out; ether_addr_copy(addr, &out_addr[2]); +out: kvfree(out); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_mac_address); + +int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *mdev, + u16 vport, u8 *addr) +{ + void *in; + int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in); + int err; + void *nic_vport_ctx; + u8 *perm_mac; + + in = mlx5_vzalloc(inlen); + if (!in) { + mlx5_core_warn(mdev, "failed to allocate inbox\n"); + return -ENOMEM; + } + + MLX5_SET(modify_nic_vport_context_in, in, + field_select.permanent_address, 1); + MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport); + + if (vport) + MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1); + + nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, + in, nic_vport_context); + perm_mac = MLX5_ADDR_OF(nic_vport_context, nic_vport_ctx, + permanent_address); + + ether_addr_copy(&perm_mac[2], addr); + + err = mlx5_modify_nic_vport_context(mdev, in, inlen); + + kvfree(in); + + return err; } -EXPORT_SYMBOL(mlx5_query_nic_vport_mac_address); +EXPORT_SYMBOL(mlx5_modify_nic_vport_mac_address); int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport, u8 port_num, u16 vf_num, u16 gid_index, diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index 967e0fd06e89..43e82d9f5463 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -36,7 +36,10 @@ #include u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod); -void mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u8 *addr); +int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, + u16 vport, u8 *addr); +int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *dev, + u16 vport, u8 *addr); int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport, u8 port_num, u16 vf_num, u16 gid_index, union ib_gid *gid); -- cgit v1.2.3-59-g8ed1b From e16aea2744abea612c27ee0eef606c6a6a8204de Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:12 +0200 Subject: net/mlx5: Introduce access functions to modify/query vport mac lists Those functions are needed to notify the upcoming L2 table and SR-IOV E-Switch(FDB) manager(PF), of the NIC vport (vf) UC/MC mac lists changes. preperation for ethernet sriov and l2 table management. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 119 ++++++++++++++++++++++++ include/linux/mlx5/device.h | 6 ++ include/linux/mlx5/vport.h | 10 ++ 3 files changed, 135 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 442916e98724..986d0d364df7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -150,6 +150,125 @@ int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *mdev, } EXPORT_SYMBOL(mlx5_modify_nic_vport_mac_address); +int mlx5_query_nic_vport_mac_list(struct mlx5_core_dev *dev, + u32 vport, + enum mlx5_list_type list_type, + u8 addr_list[][ETH_ALEN], + int *list_size) +{ + u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)]; + void *nic_vport_ctx; + int max_list_size; + int req_list_size; + int out_sz; + void *out; + int err; + int i; + + req_list_size = *list_size; + + max_list_size = list_type == MLX5_NVPRT_LIST_TYPE_UC ? + 1 << MLX5_CAP_GEN(dev, log_max_current_uc_list) : + 1 << MLX5_CAP_GEN(dev, log_max_current_mc_list); + + if (req_list_size > max_list_size) { + mlx5_core_warn(dev, "Requested list size (%d) > (%d) max_list_size\n", + req_list_size, max_list_size); + req_list_size = max_list_size; + } + + out_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) + + req_list_size * MLX5_ST_SZ_BYTES(mac_address_layout); + + memset(in, 0, sizeof(in)); + out = kzalloc(out_sz, GFP_KERNEL); + if (!out) + return -ENOMEM; + + MLX5_SET(query_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); + MLX5_SET(query_nic_vport_context_in, in, allowed_list_type, list_type); + MLX5_SET(query_nic_vport_context_in, in, vport_number, vport); + + if (vport) + MLX5_SET(query_nic_vport_context_in, in, other_vport, 1); + + err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, out_sz); + if (err) + goto out; + + nic_vport_ctx = MLX5_ADDR_OF(query_nic_vport_context_out, out, + nic_vport_context); + req_list_size = MLX5_GET(nic_vport_context, nic_vport_ctx, + allowed_list_size); + + *list_size = req_list_size; + for (i = 0; i < req_list_size; i++) { + u8 *mac_addr = MLX5_ADDR_OF(nic_vport_context, + nic_vport_ctx, + current_uc_mac_address[i]) + 2; + ether_addr_copy(addr_list[i], mac_addr); + } +out: + kfree(out); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_mac_list); + +int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev, + enum mlx5_list_type list_type, + u8 addr_list[][ETH_ALEN], + int list_size) +{ + u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)]; + void *nic_vport_ctx; + int max_list_size; + int in_sz; + void *in; + int err; + int i; + + max_list_size = list_type == MLX5_NVPRT_LIST_TYPE_UC ? + 1 << MLX5_CAP_GEN(dev, log_max_current_uc_list) : + 1 << MLX5_CAP_GEN(dev, log_max_current_mc_list); + + if (list_size > max_list_size) + return -ENOSPC; + + in_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) + + list_size * MLX5_ST_SZ_BYTES(mac_address_layout); + + memset(out, 0, sizeof(out)); + in = kzalloc(in_sz, GFP_KERNEL); + if (!in) + return -ENOMEM; + + MLX5_SET(modify_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT); + MLX5_SET(modify_nic_vport_context_in, in, + field_select.addresses_list, 1); + + nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, in, + nic_vport_context); + + MLX5_SET(nic_vport_context, nic_vport_ctx, + allowed_list_type, list_type); + MLX5_SET(nic_vport_context, nic_vport_ctx, + allowed_list_size, list_size); + + for (i = 0; i < list_size; i++) { + u8 *curr_mac = MLX5_ADDR_OF(nic_vport_context, + nic_vport_ctx, + current_uc_mac_address[i]) + 2; + ether_addr_copy(curr_mac, addr_list[i]); + } + + err = mlx5_cmd_exec_check_status(dev, in, in_sz, out, sizeof(out)); + kfree(in); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_list); + int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport, u8 port_num, u16 vf_num, u16 gid_index, union ib_gid *gid) diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 0b473cbfa7ef..0d2f0435a9f0 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1102,6 +1102,12 @@ enum { MLX5_FLOW_CONTEXT_DEST_TYPE_TIR = 2, }; +enum mlx5_list_type { + MLX5_NVPRT_LIST_TYPE_UC = 0x0, + MLX5_NVPRT_LIST_TYPE_MC = 0x1, + MLX5_NVPRT_LIST_TYPE_VLAN = 0x2, +}; + enum { MLX5_RQC_RQ_TYPE_MEMORY_RQ_INLINE = 0x0, MLX5_RQC_RQ_TYPE_MEMORY_RQ_RPM = 0x1, diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index 43e82d9f5463..00bbec8d9527 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -34,6 +34,7 @@ #define __MLX5_VPORT_H__ #include +#include u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod); int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, @@ -54,5 +55,14 @@ int mlx5_query_hca_vport_system_image_guid(struct mlx5_core_dev *dev, u64 *sys_image_guid); int mlx5_query_hca_vport_node_guid(struct mlx5_core_dev *dev, u64 *node_guid); +int mlx5_query_nic_vport_mac_list(struct mlx5_core_dev *dev, + u32 vport, + enum mlx5_list_type list_type, + u8 addr_list[][ETH_ALEN], + int *list_size); +int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev, + enum mlx5_list_type list_type, + u8 addr_list[][ETH_ALEN], + int list_size); #endif /* __MLX5_VPORT_H__ */ -- cgit v1.2.3-59-g8ed1b From e75465148b7df7f2796c75bf98bf33f171edeb2b Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:13 +0200 Subject: net/mlx5: Introduce access functions to modify/query vport state In preparation for SR-IOV we add here an API to enable each e-switch manager (PF) to configure its VFs link states in e-switch preparation for ethernet sriov. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 61 ++++++++++++++++++++--- include/linux/mlx5/mlx5_ifc.h | 1 + include/linux/mlx5/vport.h | 6 ++- 4 files changed, 62 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 2ef717f98b0d..007e464b3e58 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -63,7 +63,7 @@ static void mlx5e_update_carrier(struct mlx5e_priv *priv) u8 port_state; port_state = mlx5_query_vport_state(mdev, - MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT); + MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT, 0); if (port_state == VPORT_STATE_UP) netif_carrier_on(priv->netdev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 986d0d364df7..b017a7e68b28 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -36,26 +36,75 @@ #include #include "mlx5_core.h" -u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod) +static int _mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod, + u16 vport, u32 *out, int outlen) { - u32 in[MLX5_ST_SZ_DW(query_vport_state_in)]; - u32 out[MLX5_ST_SZ_DW(query_vport_state_out)]; int err; + u32 in[MLX5_ST_SZ_DW(query_vport_state_in)]; memset(in, 0, sizeof(in)); MLX5_SET(query_vport_state_in, in, opcode, MLX5_CMD_OP_QUERY_VPORT_STATE); MLX5_SET(query_vport_state_in, in, op_mod, opmod); + MLX5_SET(query_vport_state_in, in, vport_number, vport); + if (vport) + MLX5_SET(query_vport_state_in, in, other_vport, 1); - err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, - sizeof(out)); + err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen); if (err) mlx5_core_warn(mdev, "MLX5_CMD_OP_QUERY_VPORT_STATE failed\n"); + return err; +} + +u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport) +{ + u32 out[MLX5_ST_SZ_DW(query_vport_state_out)] = {0}; + + _mlx5_query_vport_state(mdev, opmod, vport, out, sizeof(out)); + return MLX5_GET(query_vport_state_out, out, state); } -EXPORT_SYMBOL(mlx5_query_vport_state); +EXPORT_SYMBOL_GPL(mlx5_query_vport_state); + +u8 mlx5_query_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport) +{ + u32 out[MLX5_ST_SZ_DW(query_vport_state_out)] = {0}; + + _mlx5_query_vport_state(mdev, opmod, vport, out, sizeof(out)); + + return MLX5_GET(query_vport_state_out, out, admin_state); +} +EXPORT_SYMBOL(mlx5_query_vport_admin_state); + +int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, + u16 vport, u8 state) +{ + u32 in[MLX5_ST_SZ_DW(modify_vport_state_in)]; + u32 out[MLX5_ST_SZ_DW(modify_vport_state_out)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(modify_vport_state_in, in, opcode, + MLX5_CMD_OP_MODIFY_VPORT_STATE); + MLX5_SET(modify_vport_state_in, in, op_mod, opmod); + MLX5_SET(modify_vport_state_in, in, vport_number, vport); + + if (vport) + MLX5_SET(modify_vport_state_in, in, other_vport, 1); + + MLX5_SET(modify_vport_state_in, in, admin_state, state); + + err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, + sizeof(out)); + if (err) + mlx5_core_warn(mdev, "MLX5_CMD_OP_MODIFY_VPORT_STATE failed\n"); + + return err; +} +EXPORT_SYMBOL(mlx5_modify_vport_admin_state); static int mlx5_query_nic_vport_context(struct mlx5_core_dev *mdev, u16 vport, u32 *out, int outlen) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 836cf0e43174..655184702ea2 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -2946,6 +2946,7 @@ struct mlx5_ifc_query_vport_state_out_bits { enum { MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT = 0x0, + MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT = 0x1, }; struct mlx5_ifc_query_vport_state_in_bits { diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index 00bbec8d9527..c1bba5948851 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -36,7 +36,11 @@ #include #include -u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod); +u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport); +u8 mlx5_query_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, + u16 vport); +int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, + u16 vport, u8 state); int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u16 vport, u8 *addr); int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *dev, -- cgit v1.2.3-59-g8ed1b From d82b73186dab70d6d332dd2afdb48608be2e5230 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:14 +0200 Subject: net/mlx5: Introduce access functions to modify/query vport promisc mode Those functions are needed to notify the upcoming SR-IOV E-Switch(FDB) manager(PF), of the NIC vport (vf) promisc mode changes. Preperation for ethernet sriov and l2 table management. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 62 +++++++++++++++++++++++++ include/linux/mlx5/mlx5_ifc.h | 28 +++++++++-- include/linux/mlx5/vport.h | 9 ++++ 3 files changed, 94 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index b017a7e68b28..68aa51df29c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -576,3 +576,65 @@ int mlx5_query_hca_vport_node_guid(struct mlx5_core_dev *dev, return err; } EXPORT_SYMBOL_GPL(mlx5_query_hca_vport_node_guid); + +int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev, + u32 vport, + int *promisc_uc, + int *promisc_mc, + int *promisc_all) +{ + u32 *out; + int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out); + int err; + + out = kzalloc(outlen, GFP_KERNEL); + if (!out) + return -ENOMEM; + + err = mlx5_query_nic_vport_context(mdev, vport, out, outlen); + if (err) + goto out; + + *promisc_uc = MLX5_GET(query_nic_vport_context_out, out, + nic_vport_context.promisc_uc); + *promisc_mc = MLX5_GET(query_nic_vport_context_out, out, + nic_vport_context.promisc_mc); + *promisc_all = MLX5_GET(query_nic_vport_context_out, out, + nic_vport_context.promisc_all); + +out: + kfree(out); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_promisc); + +int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev, + int promisc_uc, + int promisc_mc, + int promisc_all) +{ + void *in; + int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in); + int err; + + in = mlx5_vzalloc(inlen); + if (!in) { + mlx5_core_err(mdev, "failed to allocate inbox\n"); + return -ENOMEM; + } + + MLX5_SET(modify_nic_vport_context_in, in, field_select.promisc, 1); + MLX5_SET(modify_nic_vport_context_in, in, + nic_vport_context.promisc_uc, promisc_uc); + MLX5_SET(modify_nic_vport_context_in, in, + nic_vport_context.promisc_mc, promisc_mc); + MLX5_SET(modify_nic_vport_context_in, in, + nic_vport_context.promisc_all, promisc_all); + + err = mlx5_modify_nic_vport_context(mdev, in, inlen); + + kvfree(in); + + return err; +} +EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_promisc); diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 655184702ea2..2728b5f6c017 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -2147,16 +2147,31 @@ struct mlx5_ifc_nic_vport_context_bits { u8 reserved_0[0x1f]; u8 roce_en[0x1]; - u8 reserved_1[0x760]; + u8 arm_change_event[0x1]; + u8 reserved_1[0x1a]; + u8 event_on_mtu[0x1]; + u8 event_on_promisc_change[0x1]; + u8 event_on_vlan_change[0x1]; + u8 event_on_mc_address_change[0x1]; + u8 event_on_uc_address_change[0x1]; - u8 reserved_2[0x5]; + u8 reserved_2[0xf0]; + + u8 mtu[0x10]; + + u8 reserved_3[0x640]; + + u8 promisc_uc[0x1]; + u8 promisc_mc[0x1]; + u8 promisc_all[0x1]; + u8 reserved_4[0x2]; u8 allowed_list_type[0x3]; - u8 reserved_3[0xc]; + u8 reserved_5[0xc]; u8 allowed_list_size[0xc]; struct mlx5_ifc_mac_address_layout_bits permanent_address; - u8 reserved_4[0x20]; + u8 reserved_6[0x20]; u8 current_uc_mac_address[0][0x40]; }; @@ -4235,7 +4250,10 @@ struct mlx5_ifc_modify_nic_vport_context_out_bits { }; struct mlx5_ifc_modify_nic_vport_field_select_bits { - u8 reserved_0[0x1c]; + u8 reserved_0[0x19]; + u8 mtu[0x1]; + u8 change_event[0x1]; + u8 promisc[0x1]; u8 permanent_address[0x1]; u8 addresses_list[0x1]; u8 roce_en[0x1]; diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index c1bba5948851..dbbaed9f975a 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -68,5 +68,14 @@ int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev, enum mlx5_list_type list_type, u8 addr_list[][ETH_ALEN], int list_size); +int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev, + u32 vport, + int *promisc_uc, + int *promisc_mc, + int *promisc_all); +int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev, + int promisc_uc, + int promisc_mc, + int promisc_all); #endif /* __MLX5_VPORT_H__ */ -- cgit v1.2.3-59-g8ed1b From c0046cf7b81ac55b8bf056c71918ec04edd99379 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:15 +0200 Subject: net/mlx5: Introduce access functions to modify/query vport vlans Those functions are needed to notify the upcoming L2 table and SR-IOV E-Switch(FDB) manager(PF), of the NIC vport (vf) vlan table changes. preperation for ethernet sriov and l2 table management. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/vport.c | 112 ++++++++++++++++++++++++ include/linux/mlx5/mlx5_ifc.h | 7 ++ include/linux/mlx5/vport.h | 7 ++ 3 files changed, 126 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 68aa51df29c1..076197efea9b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -318,6 +318,118 @@ int mlx5_modify_nic_vport_mac_list(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_mac_list); +int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev, + u32 vport, + u16 vlans[], + int *size) +{ + u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)]; + void *nic_vport_ctx; + int req_list_size; + int max_list_size; + int out_sz; + void *out; + int err; + int i; + + req_list_size = *size; + max_list_size = 1 << MLX5_CAP_GEN(dev, log_max_vlan_list); + if (req_list_size > max_list_size) { + mlx5_core_warn(dev, "Requested list size (%d) > (%d) max list size\n", + req_list_size, max_list_size); + req_list_size = max_list_size; + } + + out_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) + + req_list_size * MLX5_ST_SZ_BYTES(vlan_layout); + + memset(in, 0, sizeof(in)); + out = kzalloc(out_sz, GFP_KERNEL); + if (!out) + return -ENOMEM; + + MLX5_SET(query_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); + MLX5_SET(query_nic_vport_context_in, in, allowed_list_type, + MLX5_NVPRT_LIST_TYPE_VLAN); + MLX5_SET(query_nic_vport_context_in, in, vport_number, vport); + + if (vport) + MLX5_SET(query_nic_vport_context_in, in, other_vport, 1); + + err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, out_sz); + if (err) + goto out; + + nic_vport_ctx = MLX5_ADDR_OF(query_nic_vport_context_out, out, + nic_vport_context); + req_list_size = MLX5_GET(nic_vport_context, nic_vport_ctx, + allowed_list_size); + + *size = req_list_size; + for (i = 0; i < req_list_size; i++) { + void *vlan_addr = MLX5_ADDR_OF(nic_vport_context, + nic_vport_ctx, + current_uc_mac_address[i]); + vlans[i] = MLX5_GET(vlan_layout, vlan_addr, vlan); + } +out: + kfree(out); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_vlans); + +int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev, + u16 vlans[], + int list_size) +{ + u32 out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)]; + void *nic_vport_ctx; + int max_list_size; + int in_sz; + void *in; + int err; + int i; + + max_list_size = 1 << MLX5_CAP_GEN(dev, log_max_vlan_list); + + if (list_size > max_list_size) + return -ENOSPC; + + in_sz = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in) + + list_size * MLX5_ST_SZ_BYTES(vlan_layout); + + memset(out, 0, sizeof(out)); + in = kzalloc(in_sz, GFP_KERNEL); + if (!in) + return -ENOMEM; + + MLX5_SET(modify_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT); + MLX5_SET(modify_nic_vport_context_in, in, + field_select.addresses_list, 1); + + nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, in, + nic_vport_context); + + MLX5_SET(nic_vport_context, nic_vport_ctx, + allowed_list_type, MLX5_NVPRT_LIST_TYPE_VLAN); + MLX5_SET(nic_vport_context, nic_vport_ctx, + allowed_list_size, list_size); + + for (i = 0; i < list_size; i++) { + void *vlan_addr = MLX5_ADDR_OF(nic_vport_context, + nic_vport_ctx, + current_uc_mac_address[i]); + MLX5_SET(vlan_layout, vlan_addr, vlan, vlans[i]); + } + + err = mlx5_cmd_exec_check_status(dev, in, in_sz, out, sizeof(out)); + kfree(in); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_modify_nic_vport_vlans); + int mlx5_query_hca_vport_gid(struct mlx5_core_dev *dev, u8 other_vport, u8 port_num, u16 vf_num, u16 gid_index, union ib_gid *gid) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 2728b5f6c017..39487d0c305d 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -910,6 +910,13 @@ struct mlx5_ifc_mac_address_layout_bits { u8 mac_addr_31_0[0x20]; }; +struct mlx5_ifc_vlan_layout_bits { + u8 reserved_0[0x14]; + u8 vlan[0x0c]; + + u8 reserved_1[0x20]; +}; + struct mlx5_ifc_cong_control_r_roce_ecn_np_bits { u8 reserved_0[0xa0]; diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index dbbaed9f975a..638f2ca7a527 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -77,5 +77,12 @@ int mlx5_modify_nic_vport_promisc(struct mlx5_core_dev *mdev, int promisc_uc, int promisc_mc, int promisc_all); +int mlx5_query_nic_vport_vlans(struct mlx5_core_dev *dev, + u32 vport, + u16 vlans[], + int *size); +int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev, + u16 vlans[], + int list_size); #endif /* __MLX5_VPORT_H__ */ -- cgit v1.2.3-59-g8ed1b From 073bb189a41d7bbad509b576a690611c46c4858f Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:18 +0200 Subject: net/mlx5: Introducing E-Switch and l2 table E-Switch is the software entity that represents and manages ConnectX4 inter-HCA ethernet l2 switching. E-Switch has its own Virtual Ports, each Vport/vNIC/VF can be connected to the device through a vport of an e-switch. Each e-switch is managed by one vNIC identified by HCA_CAP.vport_group_manager (usually it is the PF/vport[0]), and its main responsibility is to forward each packet to the right vport. e-Switch needs to manage its own l2-table and FDB tables. L2 table is a flow table that is managed by FW, it is needed for Multi-host (Multi PF) configuration for inter HCA switching between PFs. FDB table is a flow table that is totally managed by e-Switch driver, its main responsibility is to switch packets between e-Swtich internal vports and uplink vport that belong to the same. This patch introduces only e-Swtich l2 table management, FDB managemnt will come later when ethernet SRIOV/VFs will be enabled. preperation for ethernet sriov and l2 table management. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +- drivers/net/ethernet/mellanox/mlx5/core/eq.c | 13 + drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 500 ++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 122 ++++++ drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 + include/linux/mlx5/device.h | 8 + include/linux/mlx5/driver.h | 4 + 7 files changed, 666 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 4d5103911527..a0755919ccaf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -3,6 +3,6 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ mad.o transobj.o vport.o sriov.o -mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o \ +mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o eswitch.o \ en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \ en_txrx.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 713ead583347..23c244a7e5d7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -35,6 +35,9 @@ #include #include #include "mlx5_core.h" +#ifdef CONFIG_MLX5_CORE_EN +#include "eswitch.h" +#endif enum { MLX5_EQE_SIZE = sizeof(struct mlx5_eqe), @@ -287,6 +290,11 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq) break; #endif +#ifdef CONFIG_MLX5_CORE_EN + case MLX5_EVENT_TYPE_NIC_VPORT_CHANGE: + mlx5_eswitch_vport_event(dev->priv.eswitch, eqe); + break; +#endif default: mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n", eqe->type, eq->eqn); @@ -459,6 +467,11 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev) if (MLX5_CAP_GEN(dev, pg)) async_event_mask |= (1ull << MLX5_EVENT_TYPE_PAGE_FAULT); + if (MLX5_CAP_GEN(dev, port_type) == MLX5_CAP_PORT_TYPE_ETH && + MLX5_CAP_GEN(dev, vport_group_manager) && + mlx5_core_is_pf(dev)) + async_event_mask |= (1ull << MLX5_EVENT_TYPE_NIC_VPORT_CHANGE); + err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD, MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD, "mlx5_cmd_eq", &dev->priv.uuari.uars[0]); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c new file mode 100644 index 000000000000..1f2f804bde3e --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -0,0 +1,500 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "mlx5_core.h" +#include "eswitch.h" + +#define MLX5_DEBUG_ESWITCH_MASK BIT(3) + +#define esw_info(dev, format, ...) \ + pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_warn(dev, format, ...) \ + pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_debug(dev, format, ...) \ + mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) + +enum { + MLX5_ACTION_NONE = 0, + MLX5_ACTION_ADD = 1, + MLX5_ACTION_DEL = 2, +}; + +/* HW UC L2 table hash node */ +struct mlx5_uc_l2addr { + struct l2addr_node node; + u8 action; + u32 table_index; + u32 vport; +}; + +/* Vport UC L2 table hash node */ +struct mlx5_vport_addr { + struct l2addr_node node; + u8 action; +}; + +enum { + UC_ADDR_CHANGE = BIT(0), + MC_ADDR_CHANGE = BIT(1), +}; + +static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, int vport, + u32 events_mask) +{ + int in[MLX5_ST_SZ_DW(modify_nic_vport_context_in)]; + int out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)]; + void *nic_vport_ctx; + int err; + + memset(out, 0, sizeof(out)); + memset(in, 0, sizeof(in)); + + MLX5_SET(modify_nic_vport_context_in, in, + opcode, MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT); + MLX5_SET(modify_nic_vport_context_in, in, field_select.change_event, 1); + MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport); + if (vport) + MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1); + nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, + in, nic_vport_context); + + MLX5_SET(nic_vport_context, nic_vport_ctx, arm_change_event, 1); + + if (events_mask & UC_ADDR_CHANGE) + MLX5_SET(nic_vport_context, nic_vport_ctx, + event_on_uc_address_change, 1); + if (events_mask & MC_ADDR_CHANGE) + MLX5_SET(nic_vport_context, nic_vport_ctx, + event_on_mc_address_change, 1); + + err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); + if (err) + goto ex; + err = mlx5_cmd_status_to_err_v2(out); + if (err) + goto ex; + return 0; +ex: + return err; +} + +/* HW L2 Table (MPFS) management */ +static int set_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index, + u8 *mac, u8 vlan_valid, u16 vlan) +{ + u32 in[MLX5_ST_SZ_DW(set_l2_table_entry_in)]; + u32 out[MLX5_ST_SZ_DW(set_l2_table_entry_out)]; + u8 *in_mac_addr; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(set_l2_table_entry_in, in, opcode, + MLX5_CMD_OP_SET_L2_TABLE_ENTRY); + MLX5_SET(set_l2_table_entry_in, in, table_index, index); + MLX5_SET(set_l2_table_entry_in, in, vlan_valid, vlan_valid); + MLX5_SET(set_l2_table_entry_in, in, vlan, vlan); + + in_mac_addr = MLX5_ADDR_OF(set_l2_table_entry_in, in, mac_address); + ether_addr_copy(&in_mac_addr[2], mac); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + out, sizeof(out)); +} + +static int del_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index) +{ + u32 in[MLX5_ST_SZ_DW(delete_l2_table_entry_in)]; + u32 out[MLX5_ST_SZ_DW(delete_l2_table_entry_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(delete_l2_table_entry_in, in, opcode, + MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY); + MLX5_SET(delete_l2_table_entry_in, in, table_index, index); + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + out, sizeof(out)); +} + +static int alloc_l2_table_index(struct mlx5_l2_table *l2_table, u32 *ix) +{ + int err = 0; + + *ix = find_first_zero_bit(l2_table->bitmap, l2_table->size); + if (*ix >= l2_table->size) + err = -ENOSPC; + else + __set_bit(*ix, l2_table->bitmap); + + return err; +} + +static void free_l2_table_index(struct mlx5_l2_table *l2_table, u32 ix) +{ + __clear_bit(ix, l2_table->bitmap); +} + +static int set_l2_table_entry(struct mlx5_core_dev *dev, u8 *mac, + u8 vlan_valid, u16 vlan, + u32 *index) +{ + struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table; + int err; + + err = alloc_l2_table_index(l2_table, index); + if (err) + return err; + + err = set_l2_table_entry_cmd(dev, *index, mac, vlan_valid, vlan); + if (err) + free_l2_table_index(l2_table, *index); + + return err; +} + +static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index) +{ + struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table; + + del_l2_table_entry_cmd(dev, index); + free_l2_table_index(l2_table, index); +} + +/* SW E-Switch L2 Table management */ +static int l2_table_addr_add(struct mlx5_eswitch *esw, + u8 mac[ETH_ALEN], u32 vport) +{ + struct hlist_head *hash; + struct mlx5_uc_l2addr *addr; + int err; + + hash = esw->l2_table.l2_hash; + addr = l2addr_hash_find(hash, mac, struct mlx5_uc_l2addr); + if (addr) { + esw_warn(esw->dev, + "Failed to set L2 mac(%pM) for vport(%d), mac is already in use by vport(%d)\n", + mac, vport, addr->vport); + return -EEXIST; + } + + addr = l2addr_hash_add(hash, mac, struct mlx5_uc_l2addr, + GFP_KERNEL); + if (!addr) + return -ENOMEM; + + addr->vport = vport; + addr->action = MLX5_ACTION_NONE; + err = set_l2_table_entry(esw->dev, mac, 0, 0, &addr->table_index); + if (err) + l2addr_hash_del(addr); + else + esw_debug(esw->dev, "\tADDED L2 MAC: vport[%d] %pM index:%d\n", + vport, addr->node.addr, addr->table_index); + return err; +} + +static int l2_table_addr_del(struct mlx5_eswitch *esw, + u8 mac[ETH_ALEN], u32 vport) +{ + struct hlist_head *hash; + struct mlx5_uc_l2addr *addr; + + hash = esw->l2_table.l2_hash; + addr = l2addr_hash_find(hash, mac, struct mlx5_uc_l2addr); + if (!addr || addr->vport != vport) { + esw_warn(esw->dev, "MAC(%pM) doesn't belong to vport (%d)\n", + mac, vport); + return -EINVAL; + } + + esw_debug(esw->dev, "\tDELETE L2 MAC: vport[%d] %pM index:%d\n", + vport, addr->node.addr, addr->table_index); + del_l2_table_entry(esw->dev, addr->table_index); + l2addr_hash_del(addr); + return 0; +} + +/* E-Switch vport uc list management */ + +/* Apply vport uc list to HW l2 table */ +static void esw_apply_vport_uc_list(struct mlx5_eswitch *esw, + u32 vport_num) +{ + struct mlx5_vport *vport = &esw->vports[vport_num]; + struct mlx5_vport_addr *addr; + struct l2addr_node *node; + struct hlist_head *hash; + struct hlist_node *tmp; + int hi; + + hash = vport->uc_list; + for_each_l2hash_node(node, tmp, hash, hi) { + addr = container_of(node, struct mlx5_vport_addr, node); + switch (addr->action) { + case MLX5_ACTION_ADD: + l2_table_addr_add(esw, addr->node.addr, vport_num); + addr->action = MLX5_ACTION_NONE; + break; + case MLX5_ACTION_DEL: + l2_table_addr_del(esw, addr->node.addr, vport_num); + l2addr_hash_del(addr); + break; + } + } +} + +/* Sync vport uc list from vport context */ +static void esw_update_vport_uc_list(struct mlx5_eswitch *esw, + u32 vport_num) +{ + struct mlx5_vport *vport = &esw->vports[vport_num]; + struct mlx5_vport_addr *addr; + struct l2addr_node *node; + u8 (*mac_list)[ETH_ALEN]; + struct hlist_head *hash; + struct hlist_node *tmp; + int size; + int err; + int hi; + int i; + + size = MLX5_MAX_UC_PER_VPORT(esw->dev); + + mac_list = kcalloc(size, ETH_ALEN, GFP_KERNEL); + if (!mac_list) + return; + + hash = vport->uc_list; + + for_each_l2hash_node(node, tmp, hash, hi) { + addr = container_of(node, struct mlx5_vport_addr, node); + addr->action = MLX5_ACTION_DEL; + } + + err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, + MLX5_NVPRT_LIST_TYPE_UC, + mac_list, &size); + if (err) + return; + esw_debug(esw->dev, "vport[%d] context update UC list size (%d)\n", + vport_num, size); + + for (i = 0; i < size; i++) { + if (!is_valid_ether_addr(mac_list[i])) + continue; + + addr = l2addr_hash_find(hash, mac_list[i], + struct mlx5_vport_addr); + if (addr) { + addr->action = MLX5_ACTION_NONE; + continue; + } + + addr = l2addr_hash_add(hash, mac_list[i], + struct mlx5_vport_addr, + GFP_KERNEL); + if (!addr) { + esw_warn(esw->dev, + "Failed to add MAC(%pM) to vport[%d] DB\n", + mac_list[i], vport_num); + continue; + } + addr->action = MLX5_ACTION_ADD; + } + kfree(mac_list); +} + +static void esw_vport_change_handler(struct work_struct *work) +{ + struct mlx5_vport *vport = + container_of(work, struct mlx5_vport, vport_change_handler); + struct mlx5_core_dev *dev = vport->dev; + u8 mac[ETH_ALEN]; + + mlx5_query_nic_vport_mac_address(dev, vport->vport, mac); + + if (!is_valid_ether_addr(mac)) + goto out; + + esw_update_vport_uc_list(dev->priv.eswitch, vport->vport); + esw_apply_vport_uc_list(dev->priv.eswitch, vport->vport); + +out: + if (vport->enabled) + arm_vport_context_events_cmd(dev, vport->vport, + UC_ADDR_CHANGE); +} + +static void esw_enable_vport(struct mlx5_eswitch *esw, int vport_num) +{ + struct mlx5_vport *vport = &esw->vports[vport_num]; + unsigned long flags; + + spin_lock_irqsave(&vport->lock, flags); + vport->enabled = true; + spin_unlock_irqrestore(&vport->lock, flags); + + arm_vport_context_events_cmd(esw->dev, vport_num, UC_ADDR_CHANGE); +} + +static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) +{ + struct mlx5_vport *vport = &esw->vports[vport_num]; + unsigned long flags; + + if (!vport->enabled) + return; + + /* Mark this vport as disabled to discard new events */ + spin_lock_irqsave(&vport->lock, flags); + vport->enabled = false; + spin_unlock_irqrestore(&vport->lock, flags); + + /* Wait for current already scheduled events to complete */ + flush_workqueue(esw->work_queue); + + /* Disable events from this vport */ + arm_vport_context_events_cmd(esw->dev, vport->vport, 0); +} + +/* Public E-Switch API */ +int mlx5_eswitch_init(struct mlx5_core_dev *dev) +{ + int l2_table_size = 1 << MLX5_CAP_GEN(dev, log_max_l2_table); + int total_vports = 1 + pci_sriov_get_totalvfs(dev->pdev); + struct mlx5_eswitch *esw; + int vport_num; + int err; + + if (!MLX5_CAP_GEN(dev, vport_group_manager) || + MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return 0; + + esw_info(dev, + "Total vports %d, l2 table size(%d), per vport: max uc(%d) max mc(%d)\n", + total_vports, l2_table_size, + MLX5_MAX_UC_PER_VPORT(dev), + MLX5_MAX_MC_PER_VPORT(dev)); + + esw = kzalloc(sizeof(*esw), GFP_KERNEL); + if (!esw) + return -ENOMEM; + + esw->dev = dev; + + esw->l2_table.bitmap = kcalloc(BITS_TO_LONGS(l2_table_size), + sizeof(uintptr_t), GFP_KERNEL); + if (!esw->l2_table.bitmap) { + err = -ENOMEM; + goto abort; + } + esw->l2_table.size = l2_table_size; + + esw->work_queue = create_singlethread_workqueue("mlx5_esw_wq"); + if (!esw->work_queue) { + err = -ENOMEM; + goto abort; + } + + esw->vports = kcalloc(total_vports, sizeof(struct mlx5_vport), + GFP_KERNEL); + if (!esw->vports) { + err = -ENOMEM; + goto abort; + } + + esw->total_vports = total_vports; + for (vport_num = 0; vport_num < total_vports; vport_num++) { + struct mlx5_vport *vport = &esw->vports[vport_num]; + + vport->vport = vport_num; + vport->dev = dev; + INIT_WORK(&vport->vport_change_handler, + esw_vport_change_handler); + spin_lock_init(&vport->lock); + } + + dev->priv.eswitch = esw; + + esw_enable_vport(esw, 0); + /* VF Vports will be enabled when SRIOV is enabled */ + return 0; +abort: + if (esw->work_queue) + destroy_workqueue(esw->work_queue); + kfree(esw->l2_table.bitmap); + kfree(esw->vports); + kfree(esw); + return err; +} + +void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) +{ + if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || + MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return; + + esw_info(esw->dev, "cleanup\n"); + esw_disable_vport(esw, 0); + + esw->dev->priv.eswitch = NULL; + destroy_workqueue(esw->work_queue); + kfree(esw->l2_table.bitmap); + kfree(esw->vports); + kfree(esw); +} + +void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe) +{ + struct mlx5_eqe_vport_change *vc_eqe = &eqe->data.vport_change; + u16 vport_num = be16_to_cpu(vc_eqe->vport_num); + struct mlx5_vport *vport; + + if (!esw) { + pr_warn("MLX5 E-Switch: vport %d got an event while eswitch is not initialized\n", + vport_num); + return; + } + + vport = &esw->vports[vport_num]; + spin_lock(&vport->lock); + if (vport->enabled) + queue_work(esw->work_queue, &vport->vport_change_handler); + spin_unlock(&vport->lock); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h new file mode 100644 index 000000000000..0c41f2657824 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -0,0 +1,122 @@ +/* + * Copyright (c) 2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef __MLX5_ESWITCH_H__ +#define __MLX5_ESWITCH_H__ + +#include + +#define MLX5_MAX_UC_PER_VPORT(dev) \ + (1 << MLX5_CAP_GEN(dev, log_max_current_uc_list)) + +#define MLX5_MAX_MC_PER_VPORT(dev) \ + (1 << MLX5_CAP_GEN(dev, log_max_current_mc_list)) + +#define MLX5_L2_ADDR_HASH_SIZE (BIT(BITS_PER_BYTE)) +#define MLX5_L2_ADDR_HASH(addr) (addr[5]) + +/* L2 -mac address based- hash helpers */ +struct l2addr_node { + struct hlist_node hlist; + u8 addr[ETH_ALEN]; +}; + +#define for_each_l2hash_node(hn, tmp, hash, i) \ + for (i = 0; i < MLX5_L2_ADDR_HASH_SIZE; i++) \ + hlist_for_each_entry_safe(hn, tmp, &hash[i], hlist) + +#define l2addr_hash_find(hash, mac, type) ({ \ + int ix = MLX5_L2_ADDR_HASH(mac); \ + bool found = false; \ + type *ptr = NULL; \ + \ + hlist_for_each_entry(ptr, &hash[ix], node.hlist) \ + if (ether_addr_equal(ptr->node.addr, mac)) {\ + found = true; \ + break; \ + } \ + if (!found) \ + ptr = NULL; \ + ptr; \ +}) + +#define l2addr_hash_add(hash, mac, type, gfp) ({ \ + int ix = MLX5_L2_ADDR_HASH(mac); \ + type *ptr = NULL; \ + \ + ptr = kzalloc(sizeof(type), gfp); \ + if (ptr) { \ + ether_addr_copy(ptr->node.addr, mac); \ + hlist_add_head(&ptr->node.hlist, &hash[ix]);\ + } \ + ptr; \ +}) + +#define l2addr_hash_del(ptr) ({ \ + hlist_del(&ptr->node.hlist); \ + kfree(ptr); \ +}) + +struct mlx5_vport { + struct mlx5_core_dev *dev; + int vport; + struct hlist_head uc_list[MLX5_L2_ADDR_HASH_SIZE]; + struct work_struct vport_change_handler; + + /* This spinlock protects access to vport data, between + * "esw_vport_disable" and ongoing interrupt "mlx5_eswitch_vport_event" + * once vport marked as disabled new interrupts are discarded. + */ + spinlock_t lock; /* vport events sync */ + bool enabled; +}; + +struct mlx5_l2_table { + struct hlist_head l2_hash[MLX5_L2_ADDR_HASH_SIZE]; + u32 size; + unsigned long *bitmap; +}; + +struct mlx5_eswitch { + struct mlx5_core_dev *dev; + struct mlx5_l2_table l2_table; + struct workqueue_struct *work_queue; + struct mlx5_vport *vports; + int total_vports; +}; + +/* E-Switch API */ +int mlx5_eswitch_init(struct mlx5_core_dev *dev); +void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); +void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe); + +#endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 66e2b37cfbbf..c6de3240f76f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -49,6 +49,9 @@ #include #include #include "mlx5_core.h" +#ifdef CONFIG_MLX5_CORE_EN +#include "eswitch.h" +#endif MODULE_AUTHOR("Eli Cohen "); MODULE_DESCRIPTION("Mellanox Connect-IB, ConnectX-4 core driver"); @@ -1052,6 +1055,14 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_init_srq_table(dev); mlx5_init_mr_table(dev); +#ifdef CONFIG_MLX5_CORE_EN + err = mlx5_eswitch_init(dev); + if (err) { + dev_err(&pdev->dev, "eswitch init failed %d\n", err); + goto err_reg_dev; + } +#endif + err = mlx5_sriov_init(dev); if (err) { dev_err(&pdev->dev, "sriov init failed %d\n", err); @@ -1078,6 +1089,9 @@ err_sriov: if (mlx5_sriov_cleanup(dev)) dev_err(&dev->pdev->dev, "sriov cleanup failed\n"); +#ifdef CONFIG_MLX5_CORE_EN + mlx5_eswitch_cleanup(dev->priv.eswitch); +#endif err_reg_dev: mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); @@ -1147,6 +1161,10 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) goto out; } mlx5_unregister_device(dev); +#ifdef CONFIG_MLX5_CORE_EN + mlx5_eswitch_cleanup(dev->priv.eswitch); +#endif + mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 0d2f0435a9f0..90a4cb6dc4cb 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -251,6 +251,7 @@ enum mlx5_event { MLX5_EVENT_TYPE_PAGE_REQUEST = 0xb, MLX5_EVENT_TYPE_PAGE_FAULT = 0xc, + MLX5_EVENT_TYPE_NIC_VPORT_CHANGE = 0xd, }; enum { @@ -520,6 +521,12 @@ struct mlx5_eqe_page_fault { __be32 flags_qpn; } __packed; +struct mlx5_eqe_vport_change { + u8 rsvd0[2]; + __be16 vport_num; + __be32 rsvd1[6]; +} __packed; + union ev_data { __be32 raw[7]; struct mlx5_eqe_cmd cmd; @@ -532,6 +539,7 @@ union ev_data { struct mlx5_eqe_stall_vl stall_vl; struct mlx5_eqe_page_req req_pages; struct mlx5_eqe_page_fault page_fault; + struct mlx5_eqe_vport_change vport_change; } __packed; struct mlx5_eqe { diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index efebb87163c8..ac098b6b97bf 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -441,6 +441,8 @@ struct mlx5_irq_info { char name[MLX5_MAX_IRQ_NAME]; }; +struct mlx5_eswitch; + struct mlx5_priv { char name[MLX5_MAX_NAME_LEN]; struct mlx5_eq_table eq_table; @@ -496,6 +498,8 @@ struct mlx5_priv { struct list_head dev_list; struct list_head ctx_list; spinlock_t ctx_lock; + + struct mlx5_eswitch *eswitch; struct mlx5_core_sriov sriov; unsigned long pci_dev_data; }; -- cgit v1.2.3-59-g8ed1b From 495716b191f607b2cb2175f7499966daef79f663 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:19 +0200 Subject: net/mlx5: E-Switch, Introduce FDB hardware capabilities Define needed hardware structures and capabilities needed for E-Switch FDB flow tables and read them on driver load. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fw.c | 13 +++++++++++++ include/linux/mlx5/device.h | 15 +++++++++++++++ include/linux/mlx5/mlx5_ifc.h | 13 +++++++++++++ 3 files changed, 41 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 9335e5ae18cc..bf6e3dfcef51 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -160,6 +160,19 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) if (err) return err; } + + if (MLX5_CAP_GEN(dev, vport_group_manager) && + MLX5_CAP_GEN(dev, eswitch_flow_table)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH_FLOW_TABLE, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH_FLOW_TABLE, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } + return 0; } diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 90a4cb6dc4cb..bce9caed1eed 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1138,6 +1138,7 @@ enum mlx5_cap_type { MLX5_CAP_IPOIB_OFFLOADS, MLX5_CAP_EOIB_OFFLOADS, MLX5_CAP_FLOW_TABLE, + MLX5_CAP_ESWITCH_FLOW_TABLE, /* NUM OF CAP Types */ MLX5_CAP_NUM }; @@ -1175,6 +1176,20 @@ enum mlx5_cap_type { #define MLX5_CAP_FLOWTABLE_MAX(mdev, cap) \ MLX5_GET(flow_table_nic_cap, mdev->hca_caps_max[MLX5_CAP_FLOW_TABLE], cap) +#define MLX5_CAP_ESW_FLOWTABLE(mdev, cap) \ + MLX5_GET(flow_table_eswitch_cap, \ + mdev->hca_caps_cur[MLX5_CAP_ESWITCH_FLOW_TABLE], cap) + +#define MLX5_CAP_ESW_FLOWTABLE_MAX(mdev, cap) \ + MLX5_GET(flow_table_eswitch_cap, \ + mdev->hca_caps_max[MLX5_CAP_ESWITCH_FLOW_TABLE], cap) + +#define MLX5_CAP_ESW_FLOWTABLE_FDB(mdev, cap) \ + MLX5_CAP_ESW_FLOWTABLE(mdev, flow_table_properties_nic_esw_fdb.cap) + +#define MLX5_CAP_ESW_FLOWTABLE_FDB_MAX(mdev, cap) \ + MLX5_CAP_ESW_FLOWTABLE_MAX(mdev, flow_table_properties_nic_esw_fdb.cap) + #define MLX5_CAP_ODP(mdev, cap)\ MLX5_GET(odp_cap, mdev->hca_caps_cur[MLX5_CAP_ODP], cap) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 39487d0c305d..ae7c08adba4a 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -447,6 +447,18 @@ struct mlx5_ifc_flow_table_nic_cap_bits { u8 reserved_3[0x7200]; }; +struct mlx5_ifc_flow_table_eswitch_cap_bits { + u8 reserved_0[0x200]; + + struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_esw_fdb; + + struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_esw_acl_ingress; + + struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_esw_acl_egress; + + u8 reserved_1[0x7800]; +}; + struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 csum_cap[0x1]; u8 vlan_cap[0x1]; @@ -1846,6 +1858,7 @@ union mlx5_ifc_hca_cap_union_bits { struct mlx5_ifc_roce_cap_bits roce_cap; struct mlx5_ifc_per_protocol_networking_offload_caps_bits per_protocol_networking_offload_caps; struct mlx5_ifc_flow_table_nic_cap_bits flow_table_nic_cap; + struct mlx5_ifc_flow_table_eswitch_cap_bits flow_table_eswitch_cap; u8 reserved_0[0x8000]; }; -- cgit v1.2.3-59-g8ed1b From 81848731ff4070a3e4136efa6a99d507177a53fe Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:20 +0200 Subject: net/mlx5: E-Switch, Add SR-IOV (FDB) support Enabling E-Switch SRIOV for nvfs+1 vports. Create E-Switch FDB for L2 UC/MC mac steering between VFs/PF and external vport (Uplink). FDB contains forwarding rules such as: UC MAC0 -> vport0(PF). UC MAC1 -> vport1. UC MAC2 -> vport2. MC MACX -> vport0, vport2, Uplink. MC MACY -> vport1, Uplink. For unmatched traffic FDB has the following default rules: Unmached Traffic (src vport != Uplink) -> Uplink. Unmached Traffic (src vport == Uplink) -> vport0(PF). FDB rules population: Each NIC vport (VF) will notify E-Switch manager of its UC/MC vport context changes via modify vport context command, which will be translated to an event that will be handled by E-Switch manager (PF) which will update FDB table accordingly. Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 682 ++++++++++++++++++--- drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 25 + .../net/ethernet/mellanox/mlx5/core/mlx5_core.h | 1 + drivers/net/ethernet/mellanox/mlx5/core/sriov.c | 14 +- include/linux/mlx5/device.h | 6 + include/linux/mlx5/flow_table.h | 9 + include/linux/mlx5/mlx5_ifc.h | 7 +- 7 files changed, 661 insertions(+), 83 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 1f2f804bde3e..6fa6f013d2c5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -34,9 +34,12 @@ #include #include #include +#include #include "mlx5_core.h" #include "eswitch.h" +#define UPLINK_VPORT 0xFFFF + #define MLX5_DEBUG_ESWITCH_MASK BIT(3) #define esw_info(dev, format, ...) \ @@ -54,18 +57,26 @@ enum { MLX5_ACTION_DEL = 2, }; -/* HW UC L2 table hash node */ -struct mlx5_uc_l2addr { +/* E-Switch UC L2 table hash node */ +struct esw_uc_addr { struct l2addr_node node; - u8 action; u32 table_index; u32 vport; }; -/* Vport UC L2 table hash node */ -struct mlx5_vport_addr { - struct l2addr_node node; - u8 action; +/* E-Switch MC FDB table hash node */ +struct esw_mc_addr { /* SRIOV only */ + struct l2addr_node node; + struct mlx5_flow_rule *uplink_rule; /* Forward to uplink rule */ + u32 refcnt; +}; + +/* Vport UC/MC hash node */ +struct vport_addr { + struct l2addr_node node; + u8 action; + u32 vport; + struct mlx5_flow_rule *flow_rule; /* SRIOV only */ }; enum { @@ -73,7 +84,11 @@ enum { MC_ADDR_CHANGE = BIT(1), }; -static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, int vport, +/* Vport context events */ +#define SRIOV_VPORT_EVENTS (UC_ADDR_CHANGE | \ + MC_ADDR_CHANGE) + +static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport, u32 events_mask) { int in[MLX5_ST_SZ_DW(modify_nic_vport_context_in)]; @@ -196,97 +211,492 @@ static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index) free_l2_table_index(l2_table, index); } -/* SW E-Switch L2 Table management */ -static int l2_table_addr_add(struct mlx5_eswitch *esw, - u8 mac[ETH_ALEN], u32 vport) +/* E-Switch FDB flow steering */ +struct dest_node { + struct list_head list; + struct mlx5_flow_destination dest; +}; + +static int _mlx5_flow_rule_apply(struct mlx5_flow_rule *fr) { - struct hlist_head *hash; - struct mlx5_uc_l2addr *addr; + bool was_valid = fr->valid; + struct dest_node *dest_n; + u32 dest_list_size = 0; + void *in_match_value; + u32 *flow_context; + u32 flow_index; int err; + int i; + + if (list_empty(&fr->dest_list)) { + if (fr->valid) + mlx5_del_flow_table_entry(fr->ft, fr->fi); + fr->valid = false; + return 0; + } + + list_for_each_entry(dest_n, &fr->dest_list, list) + dest_list_size++; - hash = esw->l2_table.l2_hash; - addr = l2addr_hash_find(hash, mac, struct mlx5_uc_l2addr); - if (addr) { + flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + + MLX5_ST_SZ_BYTES(dest_format_struct) * + dest_list_size); + if (!flow_context) + return -ENOMEM; + + MLX5_SET(flow_context, flow_context, flow_tag, fr->flow_tag); + MLX5_SET(flow_context, flow_context, action, fr->action); + MLX5_SET(flow_context, flow_context, destination_list_size, + dest_list_size); + + i = 0; + list_for_each_entry(dest_n, &fr->dest_list, list) { + void *dest_addr = MLX5_ADDR_OF(flow_context, flow_context, + destination[i++]); + + MLX5_SET(dest_format_struct, dest_addr, destination_type, + dest_n->dest.type); + MLX5_SET(dest_format_struct, dest_addr, destination_id, + dest_n->dest.vport_num); + } + + in_match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); + memcpy(in_match_value, fr->match_value, MLX5_ST_SZ_BYTES(fte_match_param)); + + err = mlx5_add_flow_table_entry(fr->ft, fr->match_criteria_enable, + fr->match_criteria, flow_context, + &flow_index); + if (!err) { + if (was_valid) + mlx5_del_flow_table_entry(fr->ft, fr->fi); + fr->fi = flow_index; + fr->valid = true; + } + kfree(flow_context); + return err; +} + +static int mlx5_flow_rule_add_dest(struct mlx5_flow_rule *fr, + struct mlx5_flow_destination *new_dest) +{ + struct dest_node *dest_n; + int err; + + dest_n = kzalloc(sizeof(*dest_n), GFP_KERNEL); + if (!dest_n) + return -ENOMEM; + + memcpy(&dest_n->dest, new_dest, sizeof(dest_n->dest)); + mutex_lock(&fr->mutex); + list_add(&dest_n->list, &fr->dest_list); + err = _mlx5_flow_rule_apply(fr); + if (err) { + list_del(&dest_n->list); + kfree(dest_n); + } + mutex_unlock(&fr->mutex); + return err; +} + +static int mlx5_flow_rule_del_dest(struct mlx5_flow_rule *fr, + struct mlx5_flow_destination *dest) +{ + struct dest_node *dest_n; + struct dest_node *n; + int err; + + mutex_lock(&fr->mutex); + list_for_each_entry_safe(dest_n, n, &fr->dest_list, list) { + if (dest->vport_num == dest_n->dest.vport_num) + goto found; + } + mutex_unlock(&fr->mutex); + return -ENOENT; + +found: + list_del(&dest_n->list); + err = _mlx5_flow_rule_apply(fr); + mutex_unlock(&fr->mutex); + kfree(dest_n); + + return err; +} + +static struct mlx5_flow_rule *find_fr(struct mlx5_eswitch *esw, + u8 match_criteria_enable, + u32 *match_value) +{ + struct hlist_head *hash = esw->mc_table; + struct esw_mc_addr *esw_mc; + u8 *dmac_v; + + dmac_v = MLX5_ADDR_OF(fte_match_param, match_value, + outer_headers.dmac_47_16); + + /* UNICAST FULL MATCH */ + if (!is_multicast_ether_addr(dmac_v)) + return NULL; + + /* MULTICAST FULL MATCH */ + esw_mc = l2addr_hash_find(hash, dmac_v, struct esw_mc_addr); + + return esw_mc ? esw_mc->uplink_rule : NULL; +} + +static struct mlx5_flow_rule *alloc_fr(void *ft, + u8 match_criteria_enable, + u32 *match_criteria, + u32 *match_value, + u32 action, + u32 flow_tag) +{ + struct mlx5_flow_rule *fr = kzalloc(sizeof(*fr), GFP_KERNEL); + + if (!fr) + return NULL; + + fr->match_criteria = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + fr->match_value = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + if (!fr->match_criteria || !fr->match_value) { + kfree(fr->match_criteria); + kfree(fr->match_value); + kfree(fr); + return NULL; + } + + memcpy(fr->match_criteria, match_criteria, MLX5_ST_SZ_BYTES(fte_match_param)); + memcpy(fr->match_value, match_value, MLX5_ST_SZ_BYTES(fte_match_param)); + fr->match_criteria_enable = match_criteria_enable; + fr->flow_tag = flow_tag; + fr->action = action; + + mutex_init(&fr->mutex); + INIT_LIST_HEAD(&fr->dest_list); + atomic_set(&fr->refcount, 0); + fr->ft = ft; + return fr; +} + +static void deref_fr(struct mlx5_flow_rule *fr) +{ + if (!atomic_dec_and_test(&fr->refcount)) + return; + + kfree(fr->match_criteria); + kfree(fr->match_value); + kfree(fr); +} + +static struct mlx5_flow_rule * +mlx5_add_flow_rule(struct mlx5_eswitch *esw, + u8 match_criteria_enable, + u32 *match_criteria, + u32 *match_value, + u32 action, + u32 flow_tag, + struct mlx5_flow_destination *dest) +{ + struct mlx5_flow_rule *fr; + int err; + + fr = find_fr(esw, match_criteria_enable, match_value); + fr = fr ? fr : alloc_fr(esw->fdb_table.fdb, match_criteria_enable, match_criteria, + match_value, action, flow_tag); + if (!fr) + return NULL; + + atomic_inc(&fr->refcount); + + err = mlx5_flow_rule_add_dest(fr, dest); + if (err) { + deref_fr(fr); + return NULL; + } + + return fr; +} + +static void mlx5_del_flow_rule(struct mlx5_flow_rule *fr, u32 vport) +{ + struct mlx5_flow_destination dest; + + dest.vport_num = vport; + mlx5_flow_rule_del_dest(fr, &dest); + deref_fr(fr); +} + +/* E-Switch FDB */ +static struct mlx5_flow_rule * +esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport) +{ + int match_header = MLX5_MATCH_OUTER_HEADERS; + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule = NULL; + u32 *match_v; + u32 *match_c; + u8 *dmac_v; + u8 *dmac_c; + + match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + if (!match_v || !match_c) { + pr_warn("FDB: Failed to alloc match parameters\n"); + goto out; + } + dmac_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers.dmac_47_16); + dmac_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers.dmac_47_16); + + ether_addr_copy(dmac_v, mac); + /* Match criteria mask */ + memset(dmac_c, 0xff, 6); + + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = vport; + + esw_debug(esw->dev, + "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n", + dmac_v, dmac_c, vport); + flow_rule = + mlx5_add_flow_rule(esw, + match_header, + match_c, + match_v, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR_OR_NULL(flow_rule)) { + pr_warn( + "FDB: Failed to add flow rule: dmac_v(%pM) dmac_c(%pM) -> vport(%d), err(%ld)\n", + dmac_v, dmac_c, vport, PTR_ERR(flow_rule)); + flow_rule = NULL; + } +out: + kfree(match_v); + kfree(match_c); + return flow_rule; +} + +static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) +{ + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_table_group g; + struct mlx5_flow_table *fdb; + u8 *dmac; + + esw_debug(dev, "Create FDB log_max_size(%d)\n", + MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); + + memset(&g, 0, sizeof(g)); + /* UC MC Full match rules*/ + g.log_sz = MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size); + g.match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g.match_criteria, + outer_headers.dmac_47_16); + /* Match criteria mask */ + memset(dmac, 0xff, 6); + + fdb = mlx5_create_flow_table(dev, 0, + MLX5_FLOW_TABLE_TYPE_ESWITCH, + 1, &g); + if (fdb) + esw_debug(dev, "ESW: FDB Table created fdb->id %d\n", mlx5_get_flow_table_id(fdb)); + else + esw_warn(dev, "ESW: Failed to create FDB Table\n"); + + esw->fdb_table.fdb = fdb; + return fdb ? 0 : -ENOMEM; +} + +static void esw_destroy_fdb_table(struct mlx5_eswitch *esw) +{ + if (!esw->fdb_table.fdb) + return; + + esw_debug(esw->dev, "Destroy FDB Table fdb(%d)\n", + mlx5_get_flow_table_id(esw->fdb_table.fdb)); + mlx5_destroy_flow_table(esw->fdb_table.fdb); + esw->fdb_table.fdb = NULL; +} + +/* E-Switch vport UC/MC lists management */ +typedef int (*vport_addr_action)(struct mlx5_eswitch *esw, + struct vport_addr *vaddr); + +static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) +{ + struct hlist_head *hash = esw->l2_table.l2_hash; + struct esw_uc_addr *esw_uc; + u8 *mac = vaddr->node.addr; + u32 vport = vaddr->vport; + int err; + + esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr); + if (esw_uc) { esw_warn(esw->dev, "Failed to set L2 mac(%pM) for vport(%d), mac is already in use by vport(%d)\n", - mac, vport, addr->vport); + mac, vport, esw_uc->vport); return -EEXIST; } - addr = l2addr_hash_add(hash, mac, struct mlx5_uc_l2addr, - GFP_KERNEL); - if (!addr) + esw_uc = l2addr_hash_add(hash, mac, struct esw_uc_addr, GFP_KERNEL); + if (!esw_uc) return -ENOMEM; + esw_uc->vport = vport; - addr->vport = vport; - addr->action = MLX5_ACTION_NONE; - err = set_l2_table_entry(esw->dev, mac, 0, 0, &addr->table_index); + err = set_l2_table_entry(esw->dev, mac, 0, 0, &esw_uc->table_index); if (err) - l2addr_hash_del(addr); - else - esw_debug(esw->dev, "\tADDED L2 MAC: vport[%d] %pM index:%d\n", - vport, addr->node.addr, addr->table_index); + goto abort; + + if (esw->fdb_table.fdb) /* SRIOV is enabled: Forward UC MAC to vport */ + vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport); + + esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM index:%d fr(%p)\n", + vport, mac, esw_uc->table_index, vaddr->flow_rule); + return err; +abort: + l2addr_hash_del(esw_uc); return err; } -static int l2_table_addr_del(struct mlx5_eswitch *esw, - u8 mac[ETH_ALEN], u32 vport) +static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) { - struct hlist_head *hash; - struct mlx5_uc_l2addr *addr; + struct hlist_head *hash = esw->l2_table.l2_hash; + struct esw_uc_addr *esw_uc; + u8 *mac = vaddr->node.addr; + u32 vport = vaddr->vport; + + esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr); + if (!esw_uc || esw_uc->vport != vport) { + esw_debug(esw->dev, + "MAC(%pM) doesn't belong to vport (%d)\n", + mac, vport); + return -EINVAL; + } + esw_debug(esw->dev, "\tDELETE UC MAC: vport[%d] %pM index:%d fr(%p)\n", + vport, mac, esw_uc->table_index, vaddr->flow_rule); + + del_l2_table_entry(esw->dev, esw_uc->table_index); + + if (vaddr->flow_rule) + mlx5_del_flow_rule(vaddr->flow_rule, vport); + vaddr->flow_rule = NULL; + + l2addr_hash_del(esw_uc); + return 0; +} + +static int esw_add_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) +{ + struct hlist_head *hash = esw->mc_table; + struct esw_mc_addr *esw_mc; + u8 *mac = vaddr->node.addr; + u32 vport = vaddr->vport; + + if (!esw->fdb_table.fdb) + return 0; + + esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr); + if (esw_mc) + goto add; + + esw_mc = l2addr_hash_add(hash, mac, struct esw_mc_addr, GFP_KERNEL); + if (!esw_mc) + return -ENOMEM; + + esw_mc->uplink_rule = /* Forward MC MAC to Uplink */ + esw_fdb_set_vport_rule(esw, mac, UPLINK_VPORT); +add: + esw_mc->refcnt++; + /* Forward MC MAC to vport */ + vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport); + esw_debug(esw->dev, + "\tADDED MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n", + vport, mac, vaddr->flow_rule, + esw_mc->refcnt, esw_mc->uplink_rule); + return 0; +} + +static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) +{ + struct hlist_head *hash = esw->mc_table; + struct esw_mc_addr *esw_mc; + u8 *mac = vaddr->node.addr; + u32 vport = vaddr->vport; - hash = esw->l2_table.l2_hash; - addr = l2addr_hash_find(hash, mac, struct mlx5_uc_l2addr); - if (!addr || addr->vport != vport) { - esw_warn(esw->dev, "MAC(%pM) doesn't belong to vport (%d)\n", + if (!esw->fdb_table.fdb) + return 0; + + esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr); + if (!esw_mc) { + esw_warn(esw->dev, + "Failed to find eswitch MC addr for MAC(%pM) vport(%d)", mac, vport); return -EINVAL; } + esw_debug(esw->dev, + "\tDELETE MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n", + vport, mac, vaddr->flow_rule, esw_mc->refcnt, + esw_mc->uplink_rule); - esw_debug(esw->dev, "\tDELETE L2 MAC: vport[%d] %pM index:%d\n", - vport, addr->node.addr, addr->table_index); - del_l2_table_entry(esw->dev, addr->table_index); - l2addr_hash_del(addr); + if (vaddr->flow_rule) + mlx5_del_flow_rule(vaddr->flow_rule, vport); + vaddr->flow_rule = NULL; + + if (--esw_mc->refcnt) + return 0; + + if (esw_mc->uplink_rule) + mlx5_del_flow_rule(esw_mc->uplink_rule, UPLINK_VPORT); + + l2addr_hash_del(esw_mc); return 0; } -/* E-Switch vport uc list management */ - -/* Apply vport uc list to HW l2 table */ -static void esw_apply_vport_uc_list(struct mlx5_eswitch *esw, - u32 vport_num) +/* Apply vport UC/MC list to HW l2 table and FDB table */ +static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw, + u32 vport_num, int list_type) { struct mlx5_vport *vport = &esw->vports[vport_num]; - struct mlx5_vport_addr *addr; + bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC; + vport_addr_action vport_addr_add; + vport_addr_action vport_addr_del; + struct vport_addr *addr; struct l2addr_node *node; struct hlist_head *hash; struct hlist_node *tmp; int hi; - hash = vport->uc_list; + vport_addr_add = is_uc ? esw_add_uc_addr : + esw_add_mc_addr; + vport_addr_del = is_uc ? esw_del_uc_addr : + esw_del_mc_addr; + + hash = is_uc ? vport->uc_list : vport->mc_list; for_each_l2hash_node(node, tmp, hash, hi) { - addr = container_of(node, struct mlx5_vport_addr, node); + addr = container_of(node, struct vport_addr, node); switch (addr->action) { case MLX5_ACTION_ADD: - l2_table_addr_add(esw, addr->node.addr, vport_num); + vport_addr_add(esw, addr); addr->action = MLX5_ACTION_NONE; break; case MLX5_ACTION_DEL: - l2_table_addr_del(esw, addr->node.addr, vport_num); + vport_addr_del(esw, addr); l2addr_hash_del(addr); break; } } } -/* Sync vport uc list from vport context */ -static void esw_update_vport_uc_list(struct mlx5_eswitch *esw, - u32 vport_num) +/* Sync vport UC/MC list from vport context */ +static void esw_update_vport_addr_list(struct mlx5_eswitch *esw, + u32 vport_num, int list_type) { struct mlx5_vport *vport = &esw->vports[vport_num]; - struct mlx5_vport_addr *addr; - struct l2addr_node *node; + bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC; u8 (*mac_list)[ETH_ALEN]; + struct l2addr_node *node; + struct vport_addr *addr; struct hlist_head *hash; struct hlist_node *tmp; int size; @@ -294,40 +704,41 @@ static void esw_update_vport_uc_list(struct mlx5_eswitch *esw, int hi; int i; - size = MLX5_MAX_UC_PER_VPORT(esw->dev); + size = is_uc ? MLX5_MAX_UC_PER_VPORT(esw->dev) : + MLX5_MAX_MC_PER_VPORT(esw->dev); mac_list = kcalloc(size, ETH_ALEN, GFP_KERNEL); if (!mac_list) return; - hash = vport->uc_list; + hash = is_uc ? vport->uc_list : vport->mc_list; for_each_l2hash_node(node, tmp, hash, hi) { - addr = container_of(node, struct mlx5_vport_addr, node); + addr = container_of(node, struct vport_addr, node); addr->action = MLX5_ACTION_DEL; } - err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, - MLX5_NVPRT_LIST_TYPE_UC, + err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, list_type, mac_list, &size); if (err) return; - esw_debug(esw->dev, "vport[%d] context update UC list size (%d)\n", - vport_num, size); + esw_debug(esw->dev, "vport[%d] context update %s list size (%d)\n", + vport_num, is_uc ? "UC" : "MC", size); for (i = 0; i < size; i++) { - if (!is_valid_ether_addr(mac_list[i])) + if (is_uc && !is_valid_ether_addr(mac_list[i])) + continue; + + if (!is_uc && !is_multicast_ether_addr(mac_list[i])) continue; - addr = l2addr_hash_find(hash, mac_list[i], - struct mlx5_vport_addr); + addr = l2addr_hash_find(hash, mac_list[i], struct vport_addr); if (addr) { addr->action = MLX5_ACTION_NONE; continue; } - addr = l2addr_hash_add(hash, mac_list[i], - struct mlx5_vport_addr, + addr = l2addr_hash_add(hash, mac_list[i], struct vport_addr, GFP_KERNEL); if (!addr) { esw_warn(esw->dev, @@ -335,6 +746,7 @@ static void esw_update_vport_uc_list(struct mlx5_eswitch *esw, mac_list[i], vport_num); continue; } + addr->vport = vport_num; addr->action = MLX5_ACTION_ADD; } kfree(mac_list); @@ -345,32 +757,80 @@ static void esw_vport_change_handler(struct work_struct *work) struct mlx5_vport *vport = container_of(work, struct mlx5_vport, vport_change_handler); struct mlx5_core_dev *dev = vport->dev; + struct mlx5_eswitch *esw = dev->priv.eswitch; u8 mac[ETH_ALEN]; mlx5_query_nic_vport_mac_address(dev, vport->vport, mac); + esw_debug(dev, "vport[%d] Context Changed: perm mac: %pM\n", + vport->vport, mac); + + if (vport->enabled_events & UC_ADDR_CHANGE) { + esw_update_vport_addr_list(esw, vport->vport, + MLX5_NVPRT_LIST_TYPE_UC); + esw_apply_vport_addr_list(esw, vport->vport, + MLX5_NVPRT_LIST_TYPE_UC); + } - if (!is_valid_ether_addr(mac)) - goto out; - - esw_update_vport_uc_list(dev->priv.eswitch, vport->vport); - esw_apply_vport_uc_list(dev->priv.eswitch, vport->vport); + if (vport->enabled_events & MC_ADDR_CHANGE) { + esw_update_vport_addr_list(esw, vport->vport, + MLX5_NVPRT_LIST_TYPE_MC); + esw_apply_vport_addr_list(esw, vport->vport, + MLX5_NVPRT_LIST_TYPE_MC); + } -out: + esw_debug(esw->dev, "vport[%d] Context Changed: Done\n", vport->vport); if (vport->enabled) arm_vport_context_events_cmd(dev, vport->vport, - UC_ADDR_CHANGE); + vport->enabled_events); } -static void esw_enable_vport(struct mlx5_eswitch *esw, int vport_num) +static void esw_enable_vport(struct mlx5_eswitch *esw, int vport_num, + int enable_events) { struct mlx5_vport *vport = &esw->vports[vport_num]; unsigned long flags; + WARN_ON(vport->enabled); + + esw_debug(esw->dev, "Enabling VPORT(%d)\n", vport_num); + mlx5_modify_vport_admin_state(esw->dev, + MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT, + vport_num, + MLX5_ESW_VPORT_ADMIN_STATE_AUTO); + + /* Sync with current vport context */ + vport->enabled_events = enable_events; + esw_vport_change_handler(&vport->vport_change_handler); + spin_lock_irqsave(&vport->lock, flags); vport->enabled = true; spin_unlock_irqrestore(&vport->lock, flags); - arm_vport_context_events_cmd(esw->dev, vport_num, UC_ADDR_CHANGE); + arm_vport_context_events_cmd(esw->dev, vport_num, enable_events); + + esw->enabled_vports++; + esw_debug(esw->dev, "Enabled VPORT(%d)\n", vport_num); +} + +static void esw_cleanup_vport(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_vport *vport = &esw->vports[vport_num]; + struct l2addr_node *node; + struct vport_addr *addr; + struct hlist_node *tmp; + int hi; + + for_each_l2hash_node(node, tmp, vport->uc_list, hi) { + addr = container_of(node, struct vport_addr, node); + addr->action = MLX5_ACTION_DEL; + } + esw_apply_vport_addr_list(esw, vport_num, MLX5_NVPRT_LIST_TYPE_UC); + + for_each_l2hash_node(node, tmp, vport->mc_list, hi) { + addr = container_of(node, struct vport_addr, node); + addr->action = MLX5_ACTION_DEL; + } + esw_apply_vport_addr_list(esw, vport_num, MLX5_NVPRT_LIST_TYPE_MC); } static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) @@ -381,19 +841,82 @@ static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) if (!vport->enabled) return; + esw_debug(esw->dev, "Disabling vport(%d)\n", vport_num); /* Mark this vport as disabled to discard new events */ spin_lock_irqsave(&vport->lock, flags); vport->enabled = false; + vport->enabled_events = 0; spin_unlock_irqrestore(&vport->lock, flags); + mlx5_modify_vport_admin_state(esw->dev, + MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT, + vport_num, + MLX5_ESW_VPORT_ADMIN_STATE_DOWN); /* Wait for current already scheduled events to complete */ flush_workqueue(esw->work_queue); - /* Disable events from this vport */ arm_vport_context_events_cmd(esw->dev, vport->vport, 0); + /* We don't assume VFs will cleanup after themselves */ + esw_cleanup_vport(esw, vport_num); + esw->enabled_vports--; } /* Public E-Switch API */ +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs) +{ + int err; + int i; + + if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || + MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return 0; + + if (!MLX5_CAP_GEN(esw->dev, eswitch_flow_table) || + !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ft_support)) { + esw_warn(esw->dev, "E-Switch FDB is not supported, aborting ...\n"); + return -ENOTSUPP; + } + + esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d)\n", nvfs); + + esw_disable_vport(esw, 0); + + err = esw_create_fdb_table(esw, nvfs + 1); + if (err) + goto abort; + + for (i = 0; i <= nvfs; i++) + esw_enable_vport(esw, i, SRIOV_VPORT_EVENTS); + + esw_info(esw->dev, "SRIOV enabled: active vports(%d)\n", + esw->enabled_vports); + return 0; + +abort: + esw_enable_vport(esw, 0, UC_ADDR_CHANGE); + return err; +} + +void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) +{ + int i; + + if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || + MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return; + + esw_info(esw->dev, "disable SRIOV: active vports(%d)\n", + esw->enabled_vports); + + for (i = 0; i < esw->total_vports; i++) + esw_disable_vport(esw, i); + + esw_destroy_fdb_table(esw); + + /* VPORT 0 (PF) must be enabled back with non-sriov configuration */ + esw_enable_vport(esw, 0, UC_ADDR_CHANGE); +} + int mlx5_eswitch_init(struct mlx5_core_dev *dev) { int l2_table_size = 1 << MLX5_CAP_GEN(dev, log_max_l2_table); @@ -439,7 +962,6 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) goto abort; } - esw->total_vports = total_vports; for (vport_num = 0; vport_num < total_vports; vport_num++) { struct mlx5_vport *vport = &esw->vports[vport_num]; @@ -450,9 +972,11 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) spin_lock_init(&vport->lock); } - dev->priv.eswitch = esw; + esw->total_vports = total_vports; + esw->enabled_vports = 0; - esw_enable_vport(esw, 0); + dev->priv.eswitch = esw; + esw_enable_vport(esw, 0, UC_ADDR_CHANGE); /* VF Vports will be enabled when SRIOV is enabled */ return 0; abort: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 0c41f2657824..f222e336f34f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -86,10 +86,25 @@ struct l2addr_node { kfree(ptr); \ }) +struct mlx5_flow_rule { + void *ft; + u32 fi; + u8 match_criteria_enable; + u32 *match_criteria; + u32 *match_value; + u32 action; + u32 flow_tag; + bool valid; + atomic_t refcount; + struct mutex mutex; /* protect flow rule updates */ + struct list_head dest_list; +}; + struct mlx5_vport { struct mlx5_core_dev *dev; int vport; struct hlist_head uc_list[MLX5_L2_ADDR_HASH_SIZE]; + struct hlist_head mc_list[MLX5_L2_ADDR_HASH_SIZE]; struct work_struct vport_change_handler; /* This spinlock protects access to vport data, between @@ -98,6 +113,7 @@ struct mlx5_vport { */ spinlock_t lock; /* vport events sync */ bool enabled; + u16 enabled_events; }; struct mlx5_l2_table { @@ -106,17 +122,26 @@ struct mlx5_l2_table { unsigned long *bitmap; }; +struct mlx5_eswitch_fdb { + void *fdb; +}; + struct mlx5_eswitch { struct mlx5_core_dev *dev; struct mlx5_l2_table l2_table; + struct mlx5_eswitch_fdb fdb_table; + struct hlist_head mc_table[MLX5_L2_ADDR_HASH_SIZE]; struct workqueue_struct *work_queue; struct mlx5_vport *vports; int total_vports; + int enabled_vports; }; /* E-Switch API */ int mlx5_eswitch_init(struct mlx5_core_dev *dev); void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe); +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs); +void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw); #endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 1649d5cf9e29..bee7da822dfe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -36,6 +36,7 @@ #include #include #include +#include #define DRIVER_NAME "mlx5_core" #define DRIVER_VERSION "3.0-1" diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c index 19a43240e359..7b24386794f9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c @@ -33,6 +33,9 @@ #include #include #include "mlx5_core.h" +#ifdef CONFIG_MLX5_CORE_EN +#include "eswitch.h" +#endif static void enable_vfs(struct mlx5_core_dev *dev, int num_vfs) { @@ -144,13 +147,15 @@ int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs) mlx5_core_cleanup_vfs(dev); if (!num_vfs) { +#ifdef CONFIG_MLX5_CORE_EN + mlx5_eswitch_disable_sriov(dev->priv.eswitch); +#endif kfree(sriov->vfs_ctx); sriov->vfs_ctx = NULL; if (!pci_vfs_assigned(pdev)) pci_disable_sriov(pdev); else pr_info("unloading PF driver while leaving orphan VFs\n"); - return 0; } @@ -161,6 +166,9 @@ int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs) } mlx5_core_init_vfs(dev, num_vfs); +#ifdef CONFIG_MLX5_CORE_EN + mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs); +#endif return num_vfs; } @@ -199,6 +207,10 @@ int mlx5_sriov_init(struct mlx5_core_dev *dev) sriov->enabled_vfs = cur_vfs; mlx5_core_init_vfs(dev, cur_vfs); +#ifdef CONFIG_MLX5_CORE_EN + if (cur_vfs) + mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs); +#endif enable_vfs(dev, cur_vfs); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index bce9caed1eed..88eb4490a8b3 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1074,6 +1074,12 @@ enum { VPORT_STATE_UP = 0x1, }; +enum { + MLX5_ESW_VPORT_ADMIN_STATE_DOWN = 0x0, + MLX5_ESW_VPORT_ADMIN_STATE_UP = 0x1, + MLX5_ESW_VPORT_ADMIN_STATE_AUTO = 0x2, +}; + enum { MLX5_L3_PROT_TYPE_IPV4 = 0, MLX5_L3_PROT_TYPE_IPV6 = 1, diff --git a/include/linux/mlx5/flow_table.h b/include/linux/mlx5/flow_table.h index 5f922c6d4fc2..0f2a15cf3317 100644 --- a/include/linux/mlx5/flow_table.h +++ b/include/linux/mlx5/flow_table.h @@ -41,6 +41,15 @@ struct mlx5_flow_table_group { u32 match_criteria[MLX5_ST_SZ_DW(fte_match_param)]; }; +struct mlx5_flow_destination { + enum mlx5_flow_destination_type type; + union { + u32 tir_num; + void *ft; + u32 vport_num; + }; +}; + void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type, u16 num_groups, struct mlx5_flow_table_group *group); diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index ae7c08adba4a..a81b008738fd 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -827,9 +827,10 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_69[0x220]; }; -enum { - MLX5_DEST_FORMAT_STRUCT_DESTINATION_TYPE_FLOW_TABLE_ = 0x1, - MLX5_DEST_FORMAT_STRUCT_DESTINATION_TYPE_TIR = 0x2, +enum mlx5_flow_destination_type { + MLX5_FLOW_DESTINATION_TYPE_VPORT = 0x0, + MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE = 0x1, + MLX5_FLOW_DESTINATION_TYPE_TIR = 0x2, }; struct mlx5_ifc_dest_format_struct_bits { -- cgit v1.2.3-59-g8ed1b From d6666753c6e85834f1669c7b831cc2b7fc9e4390 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Tue, 1 Dec 2015 18:03:22 +0200 Subject: net/mlx5: E-Switch, Introduce HCA cap and E-Switch vport context E-Switch vport context is unlike NIC vport context, managed by the E-Switch manager or vport_group_manager and not by the NIC(VF) driver. The E-Switch manager can access (read/modify) any of its vports E-Switch context. Currently E-Switch vport context includes only clietnt and server vlan insertion and striping data (for later support of VST mode). Signed-off-by: Saeed Mahameed Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fw.c | 11 ++++ include/linux/mlx5/device.h | 9 +++ include/linux/mlx5/mlx5_ifc.h | 90 ++++++++++++++++++++++++++++ 3 files changed, 110 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index bf6e3dfcef51..1c9f9a54a873 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -173,6 +173,17 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) return err; } + if (MLX5_CAP_GEN(dev, vport_group_manager)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ESWITCH, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } + return 0; } diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 88eb4490a8b3..7d3a85faefb7 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1145,6 +1145,7 @@ enum mlx5_cap_type { MLX5_CAP_EOIB_OFFLOADS, MLX5_CAP_FLOW_TABLE, MLX5_CAP_ESWITCH_FLOW_TABLE, + MLX5_CAP_ESWITCH, /* NUM OF CAP Types */ MLX5_CAP_NUM }; @@ -1196,6 +1197,14 @@ enum mlx5_cap_type { #define MLX5_CAP_ESW_FLOWTABLE_FDB_MAX(mdev, cap) \ MLX5_CAP_ESW_FLOWTABLE_MAX(mdev, flow_table_properties_nic_esw_fdb.cap) +#define MLX5_CAP_ESW(mdev, cap) \ + MLX5_GET(e_switch_cap, \ + mdev->hca_caps_cur[MLX5_CAP_ESWITCH], cap) + +#define MLX5_CAP_ESW_MAX(mdev, cap) \ + MLX5_GET(e_switch_cap, \ + mdev->hca_caps_max[MLX5_CAP_ESWITCH], cap) + #define MLX5_CAP_ODP(mdev, cap)\ MLX5_GET(odp_cap, mdev->hca_caps_cur[MLX5_CAP_ODP], cap) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index a81b008738fd..f5d94495758a 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -459,6 +459,17 @@ struct mlx5_ifc_flow_table_eswitch_cap_bits { u8 reserved_1[0x7800]; }; +struct mlx5_ifc_e_switch_cap_bits { + u8 vport_svlan_strip[0x1]; + u8 vport_cvlan_strip[0x1]; + u8 vport_svlan_insert[0x1]; + u8 vport_cvlan_insert_if_not_exist[0x1]; + u8 vport_cvlan_insert_overwrite[0x1]; + u8 reserved_0[0x1b]; + + u8 reserved_1[0x7e0]; +}; + struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 csum_cap[0x1]; u8 vlan_cap[0x1]; @@ -1860,6 +1871,7 @@ union mlx5_ifc_hca_cap_union_bits { struct mlx5_ifc_per_protocol_networking_offload_caps_bits per_protocol_networking_offload_caps; struct mlx5_ifc_flow_table_nic_cap_bits flow_table_nic_cap; struct mlx5_ifc_flow_table_eswitch_cap_bits flow_table_eswitch_cap; + struct mlx5_ifc_e_switch_cap_bits e_switch_cap; u8 reserved_0[0x8000]; }; @@ -2305,6 +2317,26 @@ struct mlx5_ifc_hca_vport_context_bits { u8 reserved_6[0xca0]; }; +struct mlx5_ifc_esw_vport_context_bits { + u8 reserved_0[0x3]; + u8 vport_svlan_strip[0x1]; + u8 vport_cvlan_strip[0x1]; + u8 vport_svlan_insert[0x1]; + u8 vport_cvlan_insert[0x2]; + u8 reserved_1[0x18]; + + u8 reserved_2[0x20]; + + u8 svlan_cfi[0x1]; + u8 svlan_pcp[0x3]; + u8 svlan_id[0xc]; + u8 cvlan_cfi[0x1]; + u8 cvlan_pcp[0x3]; + u8 cvlan_id[0xc]; + + u8 reserved_3[0x7a0]; +}; + enum { MLX5_EQC_STATUS_OK = 0x0, MLX5_EQC_STATUS_EQ_WRITE_FAILURE = 0xa, @@ -3743,6 +3775,64 @@ struct mlx5_ifc_query_flow_group_in_bits { u8 reserved_5[0x120]; }; +struct mlx5_ifc_query_esw_vport_context_out_bits { + u8 status[0x8]; + u8 reserved_0[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_1[0x40]; + + struct mlx5_ifc_esw_vport_context_bits esw_vport_context; +}; + +struct mlx5_ifc_query_esw_vport_context_in_bits { + u8 opcode[0x10]; + u8 reserved_0[0x10]; + + u8 reserved_1[0x10]; + u8 op_mod[0x10]; + + u8 other_vport[0x1]; + u8 reserved_2[0xf]; + u8 vport_number[0x10]; + + u8 reserved_3[0x20]; +}; + +struct mlx5_ifc_modify_esw_vport_context_out_bits { + u8 status[0x8]; + u8 reserved_0[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_1[0x40]; +}; + +struct mlx5_ifc_esw_vport_context_fields_select_bits { + u8 reserved[0x1c]; + u8 vport_cvlan_insert[0x1]; + u8 vport_svlan_insert[0x1]; + u8 vport_cvlan_strip[0x1]; + u8 vport_svlan_strip[0x1]; +}; + +struct mlx5_ifc_modify_esw_vport_context_in_bits { + u8 opcode[0x10]; + u8 reserved_0[0x10]; + + u8 reserved_1[0x10]; + u8 op_mod[0x10]; + + u8 other_vport[0x1]; + u8 reserved_2[0xf]; + u8 vport_number[0x10]; + + struct mlx5_ifc_esw_vport_context_fields_select_bits field_select; + + struct mlx5_ifc_esw_vport_context_bits esw_vport_context; +}; + struct mlx5_ifc_query_eq_out_bits { u8 status[0x8]; u8 reserved_0[0x18]; -- cgit v1.2.3-59-g8ed1b From 2d1e0254ef8310e4f0756130a7ffc007ad1d58df Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 1 Dec 2015 14:55:21 +0000 Subject: pci_ids: add Netronome Systems vendor Add PCI vendor id for Netronome Systems. Signed-off-by: Jakub Kicinski Signed-off-by: Rolf Neugebauer Signed-off-by: David S. Miller --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index d9ba49cedc5d..1acbefc4bbda 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2495,6 +2495,8 @@ #define PCI_DEVICE_ID_KORENIX_JETCARDF2 0x1700 #define PCI_DEVICE_ID_KORENIX_JETCARDF3 0x17ff +#define PCI_VENDOR_ID_NETRONOME 0x19ee + #define PCI_VENDOR_ID_QMI 0x1a32 #define PCI_VENDOR_ID_AZWAVE 0x1a3b -- cgit v1.2.3-59-g8ed1b From 357ab2234d57f6c74386f64ded42dff8e3c0500b Mon Sep 17 00:00:00 2001 From: Asias He Date: Wed, 2 Dec 2015 14:43:59 +0800 Subject: VSOCK: Introduce vsock_find_unbound_socket and vsock_bind_dgram_generic Signed-off-by: Asias He Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- include/net/af_vsock.h | 2 ++ net/vmw_vsock/af_vsock.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 72 insertions(+) (limited to 'include') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index e9eb2d6791b3..a0c8fa2ababf 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -175,8 +175,10 @@ void vsock_insert_connected(struct vsock_sock *vsk); void vsock_remove_bound(struct vsock_sock *vsk); void vsock_remove_connected(struct vsock_sock *vsk); struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr); +struct sock *vsock_find_unbound_socket(struct sockaddr_vm *addr); struct sock *vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst); void vsock_for_each_connected_socket(void (*fn)(struct sock *sk)); +int vsock_bind_dgram_generic(struct vsock_sock *vsk, struct sockaddr_vm *addr); #endif /* __AF_VSOCK_H__ */ diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 7fd1220fbfa0..77247a2b670b 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -223,6 +223,17 @@ static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) return NULL; } +static struct sock *__vsock_find_unbound_socket(struct sockaddr_vm *addr) +{ + struct vsock_sock *vsk; + + list_for_each_entry(vsk, vsock_unbound_sockets, bound_table) + if (addr->svm_port == vsk->local_addr.svm_port) + return sk_vsock(vsk); + + return NULL; +} + static struct sock *__vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -298,6 +309,21 @@ struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr) } EXPORT_SYMBOL_GPL(vsock_find_bound_socket); +struct sock *vsock_find_unbound_socket(struct sockaddr_vm *addr) +{ + struct sock *sk; + + spin_lock_bh(&vsock_table_lock); + sk = __vsock_find_unbound_socket(addr); + if (sk) + sock_hold(sk); + + spin_unlock_bh(&vsock_table_lock); + + return sk; +} +EXPORT_SYMBOL_GPL(vsock_find_unbound_socket); + struct sock *vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -532,6 +558,50 @@ static int __vsock_bind_stream(struct vsock_sock *vsk, return 0; } +int vsock_bind_dgram_generic(struct vsock_sock *vsk, struct sockaddr_vm *addr) +{ + static u32 port = LAST_RESERVED_PORT + 1; + struct sockaddr_vm new_addr; + + vsock_addr_init(&new_addr, addr->svm_cid, addr->svm_port); + + if (addr->svm_port == VMADDR_PORT_ANY) { + bool found = false; + unsigned int i; + + for (i = 0; i < MAX_PORT_RETRIES; i++) { + if (port <= LAST_RESERVED_PORT) + port = LAST_RESERVED_PORT + 1; + + new_addr.svm_port = port++; + + if (!__vsock_find_unbound_socket(&new_addr)) { + found = true; + break; + } + } + + if (!found) + return -EADDRNOTAVAIL; + } else { + /* If port is in reserved range, ensure caller + * has necessary privileges. + */ + if (addr->svm_port <= LAST_RESERVED_PORT && + !capable(CAP_NET_BIND_SERVICE)) { + return -EACCES; + } + + if (__vsock_find_unbound_socket(&new_addr)) + return -EADDRINUSE; + } + + vsock_addr_init(&vsk->local_addr, new_addr.svm_cid, new_addr.svm_port); + + return 0; +} +EXPORT_SYMBOL_GPL(vsock_bind_dgram_generic); + static int __vsock_bind_dgram(struct vsock_sock *vsk, struct sockaddr_vm *addr) { -- cgit v1.2.3-59-g8ed1b From 80a19e338d458abb5a700df3fd00795c51361f06 Mon Sep 17 00:00:00 2001 From: Asias He Date: Wed, 2 Dec 2015 14:44:00 +0800 Subject: VSOCK: Introduce virtio-vsock-common.ko This module contains the common code and header files for the following virtio-vsock and virtio-vhost kernel modules. Signed-off-by: Asias He Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- include/linux/virtio_vsock.h | 209 +++++ include/uapi/linux/virtio_ids.h | 1 + include/uapi/linux/virtio_vsock.h | 89 +++ net/vmw_vsock/virtio_transport_common.c | 1272 +++++++++++++++++++++++++++++++ 4 files changed, 1571 insertions(+) create mode 100644 include/linux/virtio_vsock.h create mode 100644 include/uapi/linux/virtio_vsock.h create mode 100644 net/vmw_vsock/virtio_transport_common.c (limited to 'include') diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h new file mode 100644 index 000000000000..a5f3ecc038f7 --- /dev/null +++ b/include/linux/virtio_vsock.h @@ -0,0 +1,209 @@ +/* + * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so + * anyone can use the definitions to implement compatible drivers/servers: + * + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of IBM nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * Copyright (C) Red Hat, Inc., 2013-2015 + * Copyright (C) Asias He , 2013 + * Copyright (C) Stefan Hajnoczi , 2015 + */ + +#ifndef _LINUX_VIRTIO_VSOCK_H +#define _LINUX_VIRTIO_VSOCK_H + +#include +#include +#include + +#define VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE 128 +#define VIRTIO_VSOCK_DEFAULT_BUF_SIZE (1024 * 256) +#define VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE (1024 * 256) +#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE (1024 * 4) +#define VIRTIO_VSOCK_MAX_BUF_SIZE 0xFFFFFFFFUL +#define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE (1024 * 64) +#define VIRTIO_VSOCK_MAX_TX_BUF_SIZE (1024 * 1024 * 16) +#define VIRTIO_VSOCK_MAX_DGRAM_SIZE (1024 * 64) + +struct vsock_transport_recv_notify_data; +struct vsock_transport_send_notify_data; +struct sockaddr_vm; +struct vsock_sock; + +enum { + VSOCK_VQ_CTRL = 0, + VSOCK_VQ_RX = 1, /* for host to guest data */ + VSOCK_VQ_TX = 2, /* for guest to host data */ + VSOCK_VQ_MAX = 3, +}; + +/* virtio transport socket state */ +struct virtio_transport { + struct virtio_transport_pkt_ops *ops; + struct vsock_sock *vsk; + + u32 buf_size; + u32 buf_size_min; + u32 buf_size_max; + + struct mutex tx_lock; + struct mutex rx_lock; + + struct list_head rx_queue; + u32 rx_bytes; + + /* Protected by trans->tx_lock */ + u32 tx_cnt; + u32 buf_alloc; + u32 peer_fwd_cnt; + u32 peer_buf_alloc; + /* Protected by trans->rx_lock */ + u32 fwd_cnt; + + /* Protected by sk_lock */ + u16 dgram_id; + struct list_head incomplete_dgrams; /* dgram fragments */ +}; + +struct virtio_vsock_pkt { + struct virtio_vsock_hdr hdr; + struct virtio_transport *trans; + struct work_struct work; + struct list_head list; + void *buf; + u32 len; + u32 off; +}; + +struct virtio_vsock_pkt_info { + u32 remote_cid, remote_port; + struct msghdr *msg; + u32 pkt_len; + u16 type; + u16 op; + u32 flags; + u16 dgram_id; + u16 dgram_len; +}; + +struct virtio_transport_pkt_ops { + int (*send_pkt)(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info); +}; + +void virtio_vsock_dumppkt(const char *func, + const struct virtio_vsock_pkt *pkt); + +struct sock * +virtio_transport_get_pending(struct sock *listener, + struct virtio_vsock_pkt *pkt); +struct virtio_vsock_pkt * +virtio_transport_alloc_pkt(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info, + size_t len, + u32 src_cid, + u32 src_port, + u32 dst_cid, + u32 dst_port); +ssize_t +virtio_transport_stream_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, + int type); +int +virtio_transport_dgram_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags); + +s64 virtio_transport_stream_has_data(struct vsock_sock *vsk); +s64 virtio_transport_stream_has_space(struct vsock_sock *vsk); + +int virtio_transport_do_socket_init(struct vsock_sock *vsk, + struct vsock_sock *psk); +u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk); +u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk); +u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk); +void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val); +void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val); +void virtio_transport_set_max_buffer_size(struct vsock_sock *vs, u64 val); +int +virtio_transport_notify_poll_in(struct vsock_sock *vsk, + size_t target, + bool *data_ready_now); +int +virtio_transport_notify_poll_out(struct vsock_sock *vsk, + size_t target, + bool *space_available_now); + +int virtio_transport_notify_recv_init(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, + size_t target, ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, + ssize_t written, struct vsock_transport_send_notify_data *data); + +u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk); +bool virtio_transport_stream_is_active(struct vsock_sock *vsk); +bool virtio_transport_stream_allow(u32 cid, u32 port); +int virtio_transport_dgram_bind(struct vsock_sock *vsk, + struct sockaddr_vm *addr); +bool virtio_transport_dgram_allow(u32 cid, u32 port); + +int virtio_transport_connect(struct vsock_sock *vsk); + +int virtio_transport_shutdown(struct vsock_sock *vsk, int mode); + +void virtio_transport_release(struct vsock_sock *vsk); + +ssize_t +virtio_transport_stream_enqueue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len); +int +virtio_transport_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote_addr, + struct msghdr *msg, + size_t len); + +void virtio_transport_destruct(struct vsock_sock *vsk); + +void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt); +void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt); +void virtio_transport_inc_tx_pkt(struct virtio_vsock_pkt *pkt); +void virtio_transport_dec_tx_pkt(struct virtio_vsock_pkt *pkt); +u32 virtio_transport_get_credit(struct virtio_transport *trans, u32 wanted); +void virtio_transport_put_credit(struct virtio_transport *trans, u32 credit); +#endif /* _LINUX_VIRTIO_VSOCK_H */ diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h index 77925f587b15..16dcf5d06cd7 100644 --- a/include/uapi/linux/virtio_ids.h +++ b/include/uapi/linux/virtio_ids.h @@ -39,6 +39,7 @@ #define VIRTIO_ID_9P 9 /* 9p virtio console */ #define VIRTIO_ID_RPROC_SERIAL 11 /* virtio remoteproc serial link */ #define VIRTIO_ID_CAIF 12 /* Virtio caif */ +#define VIRTIO_ID_VSOCK 13 /* virtio vsock transport */ #define VIRTIO_ID_GPU 16 /* virtio GPU */ #define VIRTIO_ID_INPUT 18 /* virtio input */ diff --git a/include/uapi/linux/virtio_vsock.h b/include/uapi/linux/virtio_vsock.h new file mode 100644 index 000000000000..8cf9b5682628 --- /dev/null +++ b/include/uapi/linux/virtio_vsock.h @@ -0,0 +1,89 @@ +/* + * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so + * anyone can use the definitions to implement compatible drivers/servers: + * + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of IBM nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * Copyright (C) Red Hat, Inc., 2013-2015 + * Copyright (C) Asias He , 2013 + * Copyright (C) Stefan Hajnoczi , 2015 + */ + +#ifndef _UAPI_LINUX_VIRTIO_VSOCK_H +#define _UAPI_LINUX_VIRTIO_VOSCK_H + +#include +#include +#include + +struct virtio_vsock_config { + __le32 guest_cid; + __le32 max_virtqueue_pairs; +}; + +struct virtio_vsock_hdr { + __le32 src_cid; + __le32 src_port; + __le32 dst_cid; + __le32 dst_port; + __le32 len; + __le16 type; /* enum virtio_vsock_type */ + __le16 op; /* enum virtio_vsock_op */ + __le32 flags; + __le32 buf_alloc; + __le32 fwd_cnt; +}; + +enum virtio_vsock_type { + VIRTIO_VSOCK_TYPE_STREAM = 1, + VIRTIO_VSOCK_TYPE_DGRAM = 2, +}; + +enum virtio_vsock_op { + VIRTIO_VSOCK_OP_INVALID = 0, + + /* Connect operations */ + VIRTIO_VSOCK_OP_REQUEST = 1, + VIRTIO_VSOCK_OP_RESPONSE = 2, + VIRTIO_VSOCK_OP_ACK = 3, + VIRTIO_VSOCK_OP_RST = 4, + VIRTIO_VSOCK_OP_SHUTDOWN = 5, + + /* To send payload */ + VIRTIO_VSOCK_OP_RW = 6, + + /* Tell the peer our credit info */ + VIRTIO_VSOCK_OP_CREDIT_UPDATE = 7, + /* Request the peer to send the credit info to us */ + VIRTIO_VSOCK_OP_CREDIT_REQUEST = 8, +}; + +/* VIRTIO_VSOCK_OP_SHUTDOWN flags values */ +enum virtio_vsock_shutdown { + VIRTIO_VSOCK_SHUTDOWN_RCV = 1, + VIRTIO_VSOCK_SHUTDOWN_SEND = 2, +}; + +#endif /* _UAPI_LINUX_VIRTIO_VSOCK_H */ diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c new file mode 100644 index 000000000000..28f790da6f15 --- /dev/null +++ b/net/vmw_vsock/virtio_transport_common.c @@ -0,0 +1,1272 @@ +/* + * common code for virtio vsock + * + * Copyright (C) 2013-2015 Red Hat, Inc. + * Author: Asias He + * Stefan Hajnoczi + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#define COOKIEBITS 24 +#define COOKIEMASK (((u32)1 << COOKIEBITS) - 1) +#define VSOCK_TIMEOUT_INIT 4 + +#define SHA_MESSAGE_WORDS 16 +#define SHA_VSOCK_WORDS 5 + +static u32 vsockcookie_secret[2][SHA_MESSAGE_WORDS - SHA_VSOCK_WORDS + + SHA_DIGEST_WORDS]; + +static DEFINE_PER_CPU(__u32[SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS + + SHA_WORKSPACE_WORDS], vsock_cookie_scratch); + +static u32 cookie_hash(u32 saddr, u32 daddr, u16 sport, u16 dport, + u32 count, int c) +{ + __u32 *tmp = this_cpu_ptr(vsock_cookie_scratch); + + memcpy(tmp + SHA_VSOCK_WORDS, vsockcookie_secret[c], + sizeof(vsockcookie_secret[c])); + tmp[0] = saddr; + tmp[1] = daddr; + tmp[2] = sport; + tmp[3] = dport; + tmp[4] = count; + sha_transform(tmp + SHA_MESSAGE_WORDS, (__u8 *)tmp, + tmp + SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS); + + return tmp[17]; +} + +static u32 +virtio_vsock_secure_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, + u32 count) +{ + u32 h1, h2; + + h1 = cookie_hash(saddr, daddr, sport, dport, 0, 0); + h2 = cookie_hash(saddr, daddr, sport, dport, count, 1); + + return h1 + (count << COOKIEBITS) + (h2 & COOKIEMASK); +} + +static u32 +virtio_vsock_check_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, + u32 count, u32 cookie, u32 maxdiff) +{ + u32 diff; + u32 ret; + + cookie -= cookie_hash(saddr, daddr, sport, dport, 0, 0); + + diff = (count - (cookie >> COOKIEBITS)) & ((u32)-1 >> COOKIEBITS); + pr_debug("%s: diff=%x\n", __func__, diff); + if (diff >= maxdiff) + return (u32)-1; + + ret = (cookie - + cookie_hash(saddr, daddr, sport, dport, count - diff, 1)) + & COOKIEMASK; + pr_debug("%s: ret=%x\n", __func__, diff); + + return ret; +} + +void virtio_vsock_dumppkt(const char *func, const struct virtio_vsock_pkt *pkt) +{ + pr_debug("%s: pkt=%p, op=%d, len=%d, %d:%d---%d:%d, len=%d\n", + func, pkt, + le16_to_cpu(pkt->hdr.op), + le32_to_cpu(pkt->hdr.len), + le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port), + pkt->len); +} +EXPORT_SYMBOL_GPL(virtio_vsock_dumppkt); + +struct virtio_vsock_pkt * +virtio_transport_alloc_pkt(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info, + size_t len, + u32 src_cid, + u32 src_port, + u32 dst_cid, + u32 dst_port) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt *pkt; + int err; + + BUG_ON(!trans); + + pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); + if (!pkt) + return NULL; + + pkt->hdr.type = cpu_to_le16(info->type); + pkt->hdr.op = cpu_to_le16(info->op); + pkt->hdr.src_cid = cpu_to_le32(src_cid); + pkt->hdr.src_port = cpu_to_le32(src_port); + pkt->hdr.dst_cid = cpu_to_le32(dst_cid); + pkt->hdr.dst_port = cpu_to_le32(dst_port); + pkt->hdr.flags = cpu_to_le32(info->flags); + pkt->len = len; + pkt->trans = trans; + if (info->type == VIRTIO_VSOCK_TYPE_DGRAM) + pkt->hdr.len = cpu_to_le32(len + (info->dgram_len << 16)); + else if (info->type == VIRTIO_VSOCK_TYPE_STREAM) + pkt->hdr.len = cpu_to_le32(len); + + if (info->msg && len > 0) { + pkt->buf = kmalloc(len, GFP_KERNEL); + if (!pkt->buf) + goto out_pkt; + err = memcpy_from_msg(pkt->buf, info->msg, len); + if (err) + goto out; + } + + return pkt; + +out: + kfree(pkt->buf); +out_pkt: + kfree(pkt); + return NULL; +} +EXPORT_SYMBOL_GPL(virtio_transport_alloc_pkt); + +struct sock * +virtio_transport_get_pending(struct sock *listener, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vlistener; + struct vsock_sock *vpending; + struct sockaddr_vm src; + struct sockaddr_vm dst; + struct sock *pending; + + vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); + vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); + + vlistener = vsock_sk(listener); + list_for_each_entry(vpending, &vlistener->pending_links, + pending_links) { + if (vsock_addr_equals_addr(&src, &vpending->remote_addr) && + vsock_addr_equals_addr(&dst, &vpending->local_addr)) { + pending = sk_vsock(vpending); + sock_hold(pending); + return pending; + } + } + + return NULL; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_pending); + +static void virtio_transport_inc_rx_pkt(struct virtio_vsock_pkt *pkt) +{ + pkt->trans->rx_bytes += pkt->len; +} + +static void virtio_transport_dec_rx_pkt(struct virtio_vsock_pkt *pkt) +{ + pkt->trans->rx_bytes -= pkt->len; + pkt->trans->fwd_cnt += pkt->len; +} + +void virtio_transport_inc_tx_pkt(struct virtio_vsock_pkt *pkt) +{ + mutex_lock(&pkt->trans->tx_lock); + pkt->hdr.fwd_cnt = cpu_to_le32(pkt->trans->fwd_cnt); + pkt->hdr.buf_alloc = cpu_to_le32(pkt->trans->buf_alloc); + mutex_unlock(&pkt->trans->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt); + +void virtio_transport_dec_tx_pkt(struct virtio_vsock_pkt *pkt) +{ +} +EXPORT_SYMBOL_GPL(virtio_transport_dec_tx_pkt); + +u32 virtio_transport_get_credit(struct virtio_transport *trans, u32 credit) +{ + u32 ret; + + mutex_lock(&trans->tx_lock); + ret = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); + if (ret > credit) + ret = credit; + trans->tx_cnt += ret; + mutex_unlock(&trans->tx_lock); + + pr_debug("%s: ret=%d, buf_alloc=%d, peer_buf_alloc=%d," + "tx_cnt=%d, fwd_cnt=%d, peer_fwd_cnt=%d\n", __func__, + ret, trans->buf_alloc, trans->peer_buf_alloc, + trans->tx_cnt, trans->fwd_cnt, trans->peer_fwd_cnt); + + return ret; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_credit); + +void virtio_transport_put_credit(struct virtio_transport *trans, u32 credit) +{ + mutex_lock(&trans->tx_lock); + trans->tx_cnt -= credit; + mutex_unlock(&trans->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_put_credit); + +static int virtio_transport_send_credit_update(struct vsock_sock *vsk, int type, struct virtio_vsock_hdr *hdr) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_CREDIT_UPDATE, + .type = type, + }; + + if (hdr && type == VIRTIO_VSOCK_TYPE_DGRAM) { + info.remote_cid = le32_to_cpu(hdr->src_cid); + info.remote_port = le32_to_cpu(hdr->src_port); + } + + pr_debug("%s: sk=%p send_credit_update\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static int virtio_transport_send_credit_request(struct vsock_sock *vsk, int type) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_CREDIT_REQUEST, + .type = type, + }; + + pr_debug("%s: sk=%p send_credit_request\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static ssize_t +virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt *pkt; + size_t bytes, total = 0; + int err = -EFAULT; + + mutex_lock(&trans->rx_lock); + while (total < len && trans->rx_bytes > 0 && + !list_empty(&trans->rx_queue)) { + pkt = list_first_entry(&trans->rx_queue, + struct virtio_vsock_pkt, list); + + bytes = len - total; + if (bytes > pkt->len - pkt->off) + bytes = pkt->len - pkt->off; + + err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes); + if (err) + goto out; + total += bytes; + pkt->off += bytes; + if (pkt->off == pkt->len) { + virtio_transport_dec_rx_pkt(pkt); + list_del(&pkt->list); + virtio_transport_free_pkt(pkt); + } + } + mutex_unlock(&trans->rx_lock); + + /* Send a credit pkt to peer */ + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM, + NULL); + + return total; + +out: + mutex_unlock(&trans->rx_lock); + if (total) + err = total; + return err; +} + +ssize_t +virtio_transport_stream_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + if (flags & MSG_PEEK) + return -EOPNOTSUPP; + + return virtio_transport_stream_do_dequeue(vsk, msg, len); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue); + +struct dgram_skb { + struct list_head list; + struct sk_buff *skb; + u16 id; +}; + +static struct dgram_skb *dgram_id_to_skb(struct virtio_transport *trans, + u16 id) +{ + struct dgram_skb *dgram_skb; + + list_for_each_entry(dgram_skb, &trans->incomplete_dgrams, list) { + if (dgram_skb->id == id) + return dgram_skb; + } + + return NULL; +} + +static void +virtio_transport_recv_dgram(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct sk_buff *skb = NULL; + struct vsock_sock *vsk; + struct virtio_transport *trans; + size_t size; + u16 dgram_id, pkt_off, dgram_len, pkt_len; + u32 flags, len; + struct dgram_skb *dgram_skb; + + vsk = vsock_sk(sk); + trans = vsk->trans; + + /* len: dgram_len | pkt_len */ + len = le32_to_cpu(pkt->hdr.len); + dgram_len = len >> 16; + pkt_len = len & 0xFFFF; + + /* flags: dgram_id | pkt_off */ + flags = le32_to_cpu(pkt->hdr.flags); + dgram_id = flags >> 16; + pkt_off = flags & 0xFFFF; + + pr_debug("%s: dgram_len=%d, pkt_len=%d, id=%d, off=%d\n", __func__, + dgram_len, pkt_len, dgram_id, pkt_off); + + dgram_skb = dgram_id_to_skb(trans, dgram_id); + if (dgram_skb) { + /* This pkt is for a existing dgram */ + skb = dgram_skb->skb; + pr_debug("%s:found skb\n", __func__); + } + + /* Packet payload must be within datagram bounds */ + if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) + goto drop; + if (pkt_len > dgram_len) + goto drop; + if (pkt_off > dgram_len) + goto drop; + if (dgram_len - pkt_off < pkt_len) + goto drop; + + if (!skb) { + /* This pkt is for a new dgram */ + pr_debug("%s:create skb\n", __func__); + + size = sizeof(pkt->hdr) + dgram_len; + /* Attach the packet to the socket's receive queue as an sk_buff. */ + dgram_skb = kzalloc(sizeof(struct dgram_skb), GFP_ATOMIC); + if (!dgram_skb) + goto drop; + + skb = alloc_skb(size, GFP_ATOMIC); + if (!skb) { + kfree(dgram_skb); + dgram_skb = NULL; + goto drop; + } + dgram_skb->id = dgram_id; + dgram_skb->skb = skb; + list_add_tail(&dgram_skb->list, &trans->incomplete_dgrams); + + /* sk_receive_skb() will do a sock_put(), so hold here. */ + sock_hold(sk); + skb_put(skb, size); + memcpy(skb->data, &pkt->hdr, sizeof(pkt->hdr)); + } + + memcpy(skb->data + sizeof(pkt->hdr) + pkt_off, pkt->buf, pkt_len); + + pr_debug("%s:C, off=%d, pkt_len=%d, dgram_len=%d\n", __func__, + pkt_off, pkt_len, dgram_len); + + /* We are done with this dgram */ + if (pkt_off + pkt_len == dgram_len) { + pr_debug("%s:dgram_id=%d is done\n", __func__, dgram_id); + list_del(&dgram_skb->list); + kfree(dgram_skb); + sk_receive_skb(sk, skb, 0); + } + virtio_transport_free_pkt(pkt); + return; + +drop: + if (dgram_skb) { + list_del(&dgram_skb->list); + kfree(dgram_skb); + kfree_skb(skb); + sock_put(sk); + } + virtio_transport_free_pkt(pkt); +} + +int +virtio_transport_dgram_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + struct virtio_vsock_hdr *hdr; + struct sk_buff *skb; + int noblock; + int err; + int dgram_len; + + noblock = flags & MSG_DONTWAIT; + + if (flags & MSG_OOB || flags & MSG_ERRQUEUE) + return -EOPNOTSUPP; + + /* Retrieve the head sk_buff from the socket's receive queue. */ + err = 0; + skb = skb_recv_datagram(&vsk->sk, flags, noblock, &err); + if (err) + return err; + if (!skb) + return -EAGAIN; + + hdr = (struct virtio_vsock_hdr *)skb->data; + if (!hdr) + goto out; + + dgram_len = le32_to_cpu(hdr->len) >> 16; + /* Place the datagram payload in the user's iovec. */ + err = skb_copy_datagram_msg(skb, sizeof(*hdr), msg, dgram_len); + if (err) + goto out; + + if (msg->msg_name) { + /* Provide the address of the sender. */ + DECLARE_SOCKADDR(struct sockaddr_vm *, vm_addr, msg->msg_name); + vsock_addr_init(vm_addr, le32_to_cpu(hdr->src_cid), le32_to_cpu(hdr->src_port)); + msg->msg_namelen = sizeof(*vm_addr); + } + err = dgram_len; + + /* Send a credit pkt to peer */ + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, hdr); + + pr_debug("%s:done, recved =%d\n", __func__, dgram_len); +out: + skb_free_datagram(&vsk->sk, skb); + return err; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue); + +s64 virtio_transport_stream_has_data(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + mutex_lock(&trans->rx_lock); + bytes = trans->rx_bytes; + mutex_unlock(&trans->rx_lock); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data); + +static s64 virtio_transport_has_space(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + bytes = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); + if (bytes < 0) + bytes = 0; + + return bytes; +} + +s64 virtio_transport_stream_has_space(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + mutex_lock(&trans->tx_lock); + bytes = virtio_transport_has_space(vsk); + mutex_unlock(&trans->tx_lock); + + pr_debug("%s: bytes=%lld\n", __func__, bytes); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space); + +int virtio_transport_do_socket_init(struct vsock_sock *vsk, + struct vsock_sock *psk) +{ + struct virtio_transport *trans; + + trans = kzalloc(sizeof(*trans), GFP_KERNEL); + if (!trans) + return -ENOMEM; + + vsk->trans = trans; + trans->vsk = vsk; + if (psk) { + struct virtio_transport *ptrans = psk->trans; + trans->buf_size = ptrans->buf_size; + trans->buf_size_min = ptrans->buf_size_min; + trans->buf_size_max = ptrans->buf_size_max; + trans->peer_buf_alloc = ptrans->peer_buf_alloc; + } else { + trans->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE; + trans->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE; + trans->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE; + } + + trans->buf_alloc = trans->buf_size; + + pr_debug("%s: trans->buf_alloc=%d\n", __func__, trans->buf_alloc); + + mutex_init(&trans->rx_lock); + mutex_init(&trans->tx_lock); + INIT_LIST_HEAD(&trans->rx_queue); + INIT_LIST_HEAD(&trans->incomplete_dgrams); + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init); + +u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size); + +u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size_min; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size); + +u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size_max; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size); + +void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < trans->buf_size_min) + trans->buf_size_min = val; + if (val > trans->buf_size_max) + trans->buf_size_max = val; + trans->buf_size = val; + trans->buf_alloc = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size); + +void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val > trans->buf_size) + trans->buf_size = val; + trans->buf_size_min = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size); + +void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < trans->buf_size) + trans->buf_size = val; + trans->buf_size_max = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size); + +int +virtio_transport_notify_poll_in(struct vsock_sock *vsk, + size_t target, + bool *data_ready_now) +{ + if (vsock_stream_has_data(vsk)) + *data_ready_now = true; + else + *data_ready_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in); + +int +virtio_transport_notify_poll_out(struct vsock_sock *vsk, + size_t target, + bool *space_avail_now) +{ + s64 free_space; + + free_space = vsock_stream_has_space(vsk); + if (free_space > 0) + *space_avail_now = true; + else if (free_space == 0) + *space_avail_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out); + +int virtio_transport_notify_recv_init(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init); + +int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block); + +int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue); + +int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, + size_t target, ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue); + +int virtio_transport_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init); + +int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block); + +int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue); + +int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, + ssize_t written, struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue); + +u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat); + +bool virtio_transport_stream_is_active(struct vsock_sock *vsk) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active); + +bool virtio_transport_stream_allow(u32 cid, u32 port) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_allow); + +int virtio_transport_dgram_bind(struct vsock_sock *vsk, + struct sockaddr_vm *addr) +{ + return vsock_bind_dgram_generic(vsk, addr); +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind); + +bool virtio_transport_dgram_allow(u32 cid, u32 port) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow); + +int virtio_transport_connect(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_REQUEST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + }; + + pr_debug("%s: vsk=%p send_request\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_connect); + +int virtio_transport_shutdown(struct vsock_sock *vsk, int mode) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_SHUTDOWN, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .flags = (mode & RCV_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_RCV : 0) | + (mode & SEND_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_SEND : 0), + }; + + pr_debug("%s: vsk=%p: send_shutdown\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_shutdown); + +void virtio_transport_release(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + struct sock *sk = &vsk->sk; + struct dgram_skb *dgram_skb; + struct dgram_skb *dgram_skb_tmp; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + + /* Tell other side to terminate connection */ + if (sk->sk_type == SOCK_STREAM && sk->sk_state == SS_CONNECTED) { + virtio_transport_shutdown(vsk, SHUTDOWN_MASK); + } + + /* Free incomplete dgrams */ + lock_sock(sk); + list_for_each_entry_safe(dgram_skb, dgram_skb_tmp, + &trans->incomplete_dgrams, list) { + list_del(&dgram_skb->list); + kfree_skb(dgram_skb->skb); + kfree(dgram_skb); + sock_put(sk); /* held in virtio_transport_recv_dgram() */ + } + release_sock(sk); +} +EXPORT_SYMBOL_GPL(virtio_transport_release); + +int +virtio_transport_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote_addr, + struct msghdr *msg, + size_t dgram_len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RW, + .type = VIRTIO_VSOCK_TYPE_DGRAM, + .msg = msg, + }; + size_t total_written = 0, pkt_off = 0, written; + u16 dgram_id; + + /* The max size of a single dgram we support is 64KB */ + if (dgram_len > VIRTIO_VSOCK_MAX_DGRAM_SIZE) + return -EMSGSIZE; + + info.dgram_len = dgram_len; + vsk->remote_addr = *remote_addr; + + dgram_id = trans->dgram_id++; + + /* TODO: To optimize, if we have enough credit to send the pkt already, + * do not ask the peer to send credit to use */ + virtio_transport_send_credit_request(vsk, VIRTIO_VSOCK_TYPE_DGRAM); + + while (total_written < dgram_len) { + info.pkt_len = dgram_len - total_written; + info.flags = dgram_id << 16 | pkt_off; + written = trans->ops->send_pkt(vsk, &info); + if (written < 0) + return -ENOMEM; + if (written == 0) { + /* TODO: if written = 0, we need a sleep & wakeup + * instead of sleep */ + pr_debug("%s: SHOULD WAIT written==0", __func__); + msleep(10); + } + total_written += written; + pkt_off += written; + pr_debug("%s:id=%d, dgram_len=%zu, off=%zu, total_written=%zu, written=%zu\n", + __func__, dgram_id, dgram_len, pkt_off, total_written, written); + } + + return dgram_len; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue); + +ssize_t +virtio_transport_stream_enqueue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RW, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .msg = msg, + .pkt_len = len, + }; + + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue); + +void virtio_transport_destruct(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + kfree(trans); +} +EXPORT_SYMBOL_GPL(virtio_transport_destruct); + +static int virtio_transport_send_ack(struct vsock_sock *vsk, u32 cookie) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_ACK, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .flags = cpu_to_le32(cookie), + }; + + pr_debug("%s: sk=%p send_offer\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static int virtio_transport_send_reset(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + }; + + pr_debug("%s\n", __func__); + + /* Send RST only if the original pkt is not a RST pkt */ + if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) + return 0; + + return trans->ops->send_pkt(vsk, &info); +} + +static int +virtio_transport_recv_connecting(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + int err; + int skerr; + u32 cookie; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RESPONSE: + cookie = le32_to_cpu(pkt->hdr.flags); + pr_debug("%s: got RESPONSE and send ACK, cookie=%x\n", __func__, cookie); + err = virtio_transport_send_ack(vsk, cookie); + if (err < 0) { + skerr = -err; + goto destroy; + } + sk->sk_state = SS_CONNECTED; + sk->sk_socket->state = SS_CONNECTED; + vsock_insert_connected(vsk); + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_INVALID: + pr_debug("%s: got invalid\n", __func__); + break; + case VIRTIO_VSOCK_OP_RST: + pr_debug("%s: got rst\n", __func__); + skerr = ECONNRESET; + err = 0; + goto destroy; + default: + pr_debug("%s: got def\n", __func__); + skerr = EPROTO; + err = -EINVAL; + goto destroy; + } + return 0; + +destroy: + virtio_transport_send_reset(vsk, pkt); + sk->sk_state = SS_UNCONNECTED; + sk->sk_err = skerr; + sk->sk_error_report(sk); + return err; +} + +static int +virtio_transport_recv_connected(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_transport *trans = vsk->trans; + int err = 0; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RW: + pkt->len = le32_to_cpu(pkt->hdr.len); + pkt->off = 0; + pkt->trans = trans; + + mutex_lock(&trans->rx_lock); + virtio_transport_inc_rx_pkt(pkt); + list_add_tail(&pkt->list, &trans->rx_queue); + mutex_unlock(&trans->rx_lock); + + sk->sk_data_ready(sk); + return err; + case VIRTIO_VSOCK_OP_CREDIT_UPDATE: + sk->sk_write_space(sk); + break; + case VIRTIO_VSOCK_OP_SHUTDOWN: + pr_debug("%s: got shutdown\n", __func__); + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) + vsk->peer_shutdown |= RCV_SHUTDOWN; + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) + vsk->peer_shutdown |= SEND_SHUTDOWN; + if (le32_to_cpu(pkt->hdr.flags)) + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_RST: + pr_debug("%s: got rst\n", __func__); + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown = SHUTDOWN_MASK; + if (vsock_stream_has_data(vsk) <= 0) + sk->sk_state = SS_DISCONNECTING; + sk->sk_state_change(sk); + break; + default: + err = -EINVAL; + break; + } + + virtio_transport_free_pkt(pkt); + return err; +} + +static int +virtio_transport_send_response(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RESPONSE, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .remote_cid = le32_to_cpu(pkt->hdr.src_cid), + .remote_port = le32_to_cpu(pkt->hdr.src_port), + }; + u32 cookie; + + cookie = virtio_vsock_secure_cookie(le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_port), + jiffies / (HZ * 60)); + info.flags = cpu_to_le32(cookie); + + pr_debug("%s: send_response, cookie=%x\n", __func__, le32_to_cpu(cookie)); + + return trans->ops->send_pkt(vsk, &info); +} + +/* Handle server socket */ +static int +virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct vsock_sock *vpending; + struct sock *pending; + int err; + u32 cookie; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_REQUEST: + err = virtio_transport_send_response(vsk, pkt); + if (err < 0) { + // FIXME vsk should be vpending + virtio_transport_send_reset(vsk, pkt); + return err; + } + break; + case VIRTIO_VSOCK_OP_ACK: + cookie = le32_to_cpu(pkt->hdr.flags); + err = virtio_vsock_check_cookie(le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_port), + jiffies / (HZ * 60), + le32_to_cpu(pkt->hdr.flags), + VSOCK_TIMEOUT_INIT); + pr_debug("%s: cookie=%x, err=%d\n", __func__, cookie, err); + if (err) + return err; + + /* So no pending socket are responsible for this pkt, create one */ + pr_debug("%s: create pending\n", __func__); + pending = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL, + sk->sk_type, 0); + if (!pending) { + virtio_transport_send_reset(vsk, pkt); + return -ENOMEM; + } + sk->sk_ack_backlog++; + pending->sk_state = SS_CONNECTING; + + vpending = vsock_sk(pending); + vsock_addr_init(&vpending->local_addr, le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port)); + vsock_addr_init(&vpending->remote_addr, le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port)); + vsock_add_pending(sk, pending); + + pr_debug("%s: get pending\n", __func__); + pending = virtio_transport_get_pending(sk, pkt); + vpending = vsock_sk(pending); + lock_sock(pending); + switch (pending->sk_state) { + case SS_CONNECTING: + if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_ACK) { + pr_debug("%s: op=%d != OP_ACK\n", __func__, + le16_to_cpu(pkt->hdr.op)); + virtio_transport_send_reset(vpending, pkt); + pending->sk_err = EPROTO; + pending->sk_state = SS_UNCONNECTED; + sock_put(pending); + } else { + pending->sk_state = SS_CONNECTED; + vsock_insert_connected(vpending); + + vsock_remove_pending(sk, pending); + vsock_enqueue_accept(sk, pending); + + sk->sk_data_ready(sk); + } + err = 0; + break; + default: + pr_debug("%s: sk->sk_ack_backlog=%d\n", __func__, + sk->sk_ack_backlog); + virtio_transport_send_reset(vpending, pkt); + err = -EINVAL; + break; + } + if (err < 0) + vsock_remove_pending(sk, pending); + release_sock(pending); + + /* Release refcnt obtained in virtio_transport_get_pending */ + sock_put(pending); + break; + default: + break; + } + + return 0; +} + +static void virtio_transport_space_update(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_transport *trans = vsk->trans; + bool space_available; + + /* buf_alloc and fwd_cnt is always included in the hdr */ + mutex_lock(&trans->tx_lock); + trans->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc); + trans->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt); + space_available = virtio_transport_has_space(vsk); + mutex_unlock(&trans->tx_lock); + + if (space_available) + sk->sk_write_space(sk); +} + +/* We are under the virtio-vsock's vsock->rx_lock or + * vhost-vsock's vq->mutex lock */ +void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans; + struct sockaddr_vm src, dst; + struct vsock_sock *vsk; + struct sock *sk; + + vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); + vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); + + virtio_vsock_dumppkt(__func__, pkt); + + if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_DGRAM) { + sk = vsock_find_unbound_socket(&dst); + if (!sk) + goto free_pkt; + + vsk = vsock_sk(sk); + trans = vsk->trans; + BUG_ON(!trans); + + virtio_transport_space_update(sk, pkt); + + lock_sock(sk); + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_CREDIT_UPDATE: + virtio_transport_free_pkt(pkt); + break; + case VIRTIO_VSOCK_OP_CREDIT_REQUEST: + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, + &pkt->hdr); + virtio_transport_free_pkt(pkt); + break; + case VIRTIO_VSOCK_OP_RW: + virtio_transport_recv_dgram(sk, pkt); + break; + default: + virtio_transport_free_pkt(pkt); + break; + } + release_sock(sk); + + /* Release refcnt obtained when we fetched this socket out of + * the unbound list. + */ + sock_put(sk); + return; + } else if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM) { + /* The socket must be in connected or bound table + * otherwise send reset back + */ + sk = vsock_find_connected_socket(&src, &dst); + if (!sk) { + sk = vsock_find_bound_socket(&dst); + if (!sk) { + pr_debug("%s: can not find bound_socket\n", __func__); + virtio_vsock_dumppkt(__func__, pkt); + /* Ignore this pkt instead of sending reset back */ + /* TODO send a RST unless this packet is a RST (to avoid infinite loops) */ + goto free_pkt; + } + } + + vsk = vsock_sk(sk); + trans = vsk->trans; + BUG_ON(!trans); + + virtio_transport_space_update(sk, pkt); + + lock_sock(sk); + switch (sk->sk_state) { + case VSOCK_SS_LISTEN: + virtio_transport_recv_listen(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTING: + virtio_transport_recv_connecting(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTED: + virtio_transport_recv_connected(sk, pkt); + break; + default: + virtio_transport_free_pkt(pkt); + break; + } + release_sock(sk); + + /* Release refcnt obtained when we fetched this socket out of the + * bound or connected list. + */ + sock_put(sk); + } + return; + +free_pkt: + virtio_transport_free_pkt(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt); + +void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt) +{ + kfree(pkt->buf); + kfree(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_free_pkt); + +static int __init virtio_vsock_common_init(void) +{ + get_random_bytes(vsockcookie_secret, sizeof(vsockcookie_secret)); + return 0; +} + +static void __exit virtio_vsock_common_exit(void) +{ +} + +module_init(virtio_vsock_common_init); +module_exit(virtio_vsock_common_exit); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Asias He"); +MODULE_DESCRIPTION("common code for virtio vsock"); -- cgit v1.2.3-59-g8ed1b From 3110489117581a980537b6d999a3724214ba772c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 22 Oct 2015 17:35:19 +0200 Subject: mac80211: allow driver to prevent two stations w/ same address Some devices or drivers cannot deal with having the same station address for different virtual interfaces, say as a client to two virtual AP interfaces. Rather than requiring each driver with a limitation like that to enforce it, add a hardware flag for it. Signed-off-by: Johannes Berg --- include/net/mac80211.h | 6 ++++++ net/mac80211/debugfs.c | 1 + net/mac80211/sta_info.c | 18 ++++++++++++++++-- 3 files changed, 23 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 760bc4d5a2cf..8628118214cc 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1901,6 +1901,11 @@ struct ieee80211_txq { * @IEEE80211_HW_BEACON_TX_STATUS: The device/driver provides TX status * for sent beacons. * + * @IEEE80211_HW_NEEDS_UNIQUE_STA_ADDR: Hardware (or driver) requires that each + * station has a unique address, i.e. each station entry can be identified + * by just its MAC address; this prevents, for example, the same station + * from connecting to two virtual AP interfaces at the same time. + * * @NUM_IEEE80211_HW_FLAGS: number of hardware flags, used for sizing arrays */ enum ieee80211_hw_flags { @@ -1936,6 +1941,7 @@ enum ieee80211_hw_flags { IEEE80211_HW_TDLS_WIDER_BW, IEEE80211_HW_SUPPORTS_AMSDU_IN_AMPDU, IEEE80211_HW_BEACON_TX_STATUS, + IEEE80211_HW_NEEDS_UNIQUE_STA_ADDR, /* keep last, obviously */ NUM_IEEE80211_HW_FLAGS diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index 4d2aaebd4f97..abbdff03ce92 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -125,6 +125,7 @@ static const char *hw_flag_names[NUM_IEEE80211_HW_FLAGS + 1] = { FLAG(TDLS_WIDER_BW), FLAG(SUPPORTS_AMSDU_IN_AMPDU), FLAG(BEACON_TX_STATUS), + FLAG(NEEDS_UNIQUE_STA_ADDR), /* keep last for the build bug below */ (void *)0x1 diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index f91d1873218c..8f630f51d9bd 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -435,6 +435,19 @@ static int sta_info_insert_check(struct sta_info *sta) is_multicast_ether_addr(sta->sta.addr))) return -EINVAL; + /* Strictly speaking this isn't necessary as we hold the mutex, but + * the rhashtable code can't really deal with that distinction. We + * do require the mutex for correctness though. + */ + rcu_read_lock(); + lockdep_assert_held(&sdata->local->sta_mtx); + if (ieee80211_hw_check(&sdata->local->hw, NEEDS_UNIQUE_STA_ADDR) && + ieee80211_find_sta_by_ifaddr(&sdata->local->hw, sta->addr, NULL)) { + rcu_read_unlock(); + return -ENOTUNIQ; + } + rcu_read_unlock(); + return 0; } @@ -554,14 +567,15 @@ int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU) might_sleep(); + mutex_lock(&local->sta_mtx); + err = sta_info_insert_check(sta); if (err) { + mutex_unlock(&local->sta_mtx); rcu_read_lock(); goto out_free; } - mutex_lock(&local->sta_mtx); - err = sta_info_insert_finish(sta); if (err) goto out_free; -- cgit v1.2.3-59-g8ed1b From 0483eeac59876ac37d4edbabd48727a468416d5b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 23 Oct 2015 09:50:03 +0200 Subject: cfg80211: replace ieee80211_ie_split() with an inline The function is a very simple wrapper around another one, just adds a few default parameters, so replace it with a static inline instead of using EXPORT_SYMBOL, reducing the module size slightly. Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 7 +++++-- net/wireless/util.c | 7 ------- 2 files changed, 5 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 2c7bdb81d30c..e568872203a5 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5173,8 +5173,11 @@ size_t ieee80211_ie_split_ric(const u8 *ies, size_t ielen, * buffer starts, which may be @ielen if the entire (remainder) * of the buffer should be used. */ -size_t ieee80211_ie_split(const u8 *ies, size_t ielen, - const u8 *ids, int n_ids, size_t offset); +static inline size_t ieee80211_ie_split(const u8 *ies, size_t ielen, + const u8 *ids, int n_ids, size_t offset) +{ + return ieee80211_ie_split_ric(ies, ielen, ids, n_ids, NULL, 0, offset); +} /** * cfg80211_report_wowlan_wakeup - report wakeup from WoWLAN diff --git a/net/wireless/util.c b/net/wireless/util.c index baf7218cec15..010a3c75a677 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1325,13 +1325,6 @@ size_t ieee80211_ie_split_ric(const u8 *ies, size_t ielen, } EXPORT_SYMBOL(ieee80211_ie_split_ric); -size_t ieee80211_ie_split(const u8 *ies, size_t ielen, - const u8 *ids, int n_ids, size_t offset) -{ - return ieee80211_ie_split_ric(ies, ielen, ids, n_ids, NULL, 0, offset); -} -EXPORT_SYMBOL(ieee80211_ie_split); - bool ieee80211_operating_class_to_band(u8 operating_class, enum ieee80211_band *band) { -- cgit v1.2.3-59-g8ed1b From cf595922b9a3b61e308224fd29909f54ecb557d4 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Mon, 9 Nov 2015 12:17:37 +0200 Subject: nl80211: clarify NL80211_ATTR_SCHED_SCAN_DELAY usage with net-detect In this attribute's documentation, it was not clear whether the delay started counting when WoWLAN net-detect was enabled or when the system was suspended. The correct answer is that it starts when the system suspends (which is when, in practice, the scan is scheduled). Clarify that in the nl80211.h documentation. Suggested-by: Samuel Tan Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- include/uapi/linux/nl80211.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 1f0b4cf5dd03..07099cb14778 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1764,8 +1764,9 @@ enum nl80211_commands { * over all channels. * * @NL80211_ATTR_SCHED_SCAN_DELAY: delay before the first cycle of a - * scheduled scan (or a WoWLAN net-detect scan) is started, u32 - * in seconds. + * scheduled scan is started. Or the delay before a WoWLAN + * net-detect scan is started, counting from the moment the + * system is suspended. This value is a u32, in seconds. * @NL80211_ATTR_REG_INDOOR: flag attribute, if set indicates that the device * is operating in an indoor environment. -- cgit v1.2.3-59-g8ed1b From 0ead2510f8cec11ce96308d79a1b4ee272fb5238 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Tue, 17 Nov 2015 10:24:36 +0200 Subject: mac80211: allow the driver to send EOSP when needed This can happen when the driver needs to send less frames than expected and then needs to close the SP. Mac80211 still needs to set the more_data properly based on its buffer state (ps_tx_buffer and buffered frames on other TIDs). To that end, refactor the code that delivers frames upon uAPSD trigger frames to be able to get only the more_data bit without actually delivering those frames in case the driver is just asking to set a NDP with EOSP and MORE_DATA bit properly set. Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- include/net/mac80211.h | 22 ++++++++ net/mac80211/sta_info.c | 144 +++++++++++++++++++++++++++++++++--------------- net/mac80211/trace.h | 25 +++++++++ 3 files changed, 148 insertions(+), 43 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 8628118214cc..18ac733afc91 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -4868,6 +4868,28 @@ void ieee80211_sta_block_awake(struct ieee80211_hw *hw, */ void ieee80211_sta_eosp(struct ieee80211_sta *pubsta); +/** + * ieee80211_send_eosp_nullfunc - ask mac80211 to send NDP with EOSP + * @pubsta: the station + * @tid: the tid of the NDP + * + * Sometimes the device understands that it needs to close + * the Service Period unexpectedly. This can happen when + * sending frames that are filling holes in the BA window. + * In this case, the device can ask mac80211 to send a + * Nullfunc frame with EOSP set. When that happens, the + * driver must have called ieee80211_sta_set_buffered() to + * let mac80211 know that there are no buffered frames any + * more, otherwise mac80211 will get the more_data bit wrong. + * The low level driver must have made sure that the frame + * will be sent despite the station being in power-save. + * Mac80211 won't call allow_buffered_frames(). + * Note that calling this function, doesn't exempt the driver + * from closing the EOSP properly, it will still have to call + * ieee80211_sta_eosp when the NDP is sent. + */ +void ieee80211_send_eosp_nullfunc(struct ieee80211_sta *pubsta, int tid); + /** * ieee80211_iter_keys - iterate keys programmed into the device * @hw: pointer obtained from ieee80211_alloc_hw() diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 8f630f51d9bd..723fa30aafc5 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -2,6 +2,7 @@ * Copyright 2002-2005, Instant802 Networks, Inc. * Copyright 2006-2007 Jiri Benc * Copyright 2013-2014 Intel Mobile Communications GmbH + * Copyright (C) 2015 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 @@ -1244,11 +1245,11 @@ void ieee80211_sta_ps_deliver_wakeup(struct sta_info *sta) ieee80211_check_fast_xmit(sta); } -static void ieee80211_send_null_response(struct ieee80211_sub_if_data *sdata, - struct sta_info *sta, int tid, +static void ieee80211_send_null_response(struct sta_info *sta, int tid, enum ieee80211_frame_release_type reason, - bool call_driver) + bool call_driver, bool more_data) { + struct ieee80211_sub_if_data *sdata = sta->sdata; struct ieee80211_local *local = sdata->local; struct ieee80211_qos_hdr *nullfunc; struct sk_buff *skb; @@ -1288,9 +1289,13 @@ static void ieee80211_send_null_response(struct ieee80211_sub_if_data *sdata, if (qos) { nullfunc->qos_ctrl = cpu_to_le16(tid); - if (reason == IEEE80211_FRAME_RELEASE_UAPSD) + if (reason == IEEE80211_FRAME_RELEASE_UAPSD) { nullfunc->qos_ctrl |= cpu_to_le16(IEEE80211_QOS_CTL_EOSP); + if (more_data) + nullfunc->frame_control |= + cpu_to_le16(IEEE80211_FCTL_MOREDATA); + } } info = IEEE80211_SKB_CB(skb); @@ -1337,22 +1342,48 @@ static int find_highest_prio_tid(unsigned long tids) return fls(tids) - 1; } +/* Indicates if the MORE_DATA bit should be set in the last + * frame obtained by ieee80211_sta_ps_get_frames. + * Note that driver_release_tids is relevant only if + * reason = IEEE80211_FRAME_RELEASE_PSPOLL + */ +static bool +ieee80211_sta_ps_more_data(struct sta_info *sta, u8 ignored_acs, + enum ieee80211_frame_release_type reason, + unsigned long driver_release_tids) +{ + int ac; + + /* If the driver has data on more than one TID then + * certainly there's more data if we release just a + * single frame now (from a single TID). This will + * only happen for PS-Poll. + */ + if (reason == IEEE80211_FRAME_RELEASE_PSPOLL && + hweight16(driver_release_tids) > 1) + return true; + + for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) { + if (ignored_acs & BIT(ac)) + continue; + + if (!skb_queue_empty(&sta->tx_filtered[ac]) || + !skb_queue_empty(&sta->ps_tx_buf[ac])) + return true; + } + + return false; +} + static void -ieee80211_sta_ps_deliver_response(struct sta_info *sta, - int n_frames, u8 ignored_acs, - enum ieee80211_frame_release_type reason) +ieee80211_sta_ps_get_frames(struct sta_info *sta, int n_frames, u8 ignored_acs, + enum ieee80211_frame_release_type reason, + struct sk_buff_head *frames, + unsigned long *driver_release_tids) { struct ieee80211_sub_if_data *sdata = sta->sdata; struct ieee80211_local *local = sdata->local; - bool more_data = false; int ac; - unsigned long driver_release_tids = 0; - struct sk_buff_head frames; - - /* Service or PS-Poll period starts */ - set_sta_flag(sta, WLAN_STA_SP); - - __skb_queue_head_init(&frames); /* Get response frame(s) and more data bit for the last one. */ for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) { @@ -1366,26 +1397,13 @@ ieee80211_sta_ps_deliver_response(struct sta_info *sta, /* if we already have frames from software, then we can't also * release from hardware queues */ - if (skb_queue_empty(&frames)) { - driver_release_tids |= sta->driver_buffered_tids & tids; - driver_release_tids |= sta->txq_buffered_tids & tids; + if (skb_queue_empty(frames)) { + *driver_release_tids |= + sta->driver_buffered_tids & tids; + *driver_release_tids |= sta->txq_buffered_tids & tids; } - if (driver_release_tids) { - /* If the driver has data on more than one TID then - * certainly there's more data if we release just a - * single frame now (from a single TID). This will - * only happen for PS-Poll. - */ - if (reason == IEEE80211_FRAME_RELEASE_PSPOLL && - hweight16(driver_release_tids) > 1) { - more_data = true; - driver_release_tids = - BIT(find_highest_prio_tid( - driver_release_tids)); - break; - } - } else { + if (!*driver_release_tids) { struct sk_buff *skb; while (n_frames > 0) { @@ -1399,20 +1417,44 @@ ieee80211_sta_ps_deliver_response(struct sta_info *sta, if (!skb) break; n_frames--; - __skb_queue_tail(&frames, skb); + __skb_queue_tail(frames, skb); } } - /* If we have more frames buffered on this AC, then set the - * more-data bit and abort the loop since we can't send more - * data from other ACs before the buffered frames from this. + /* If we have more frames buffered on this AC, then abort the + * loop since we can't send more data from other ACs before + * the buffered frames from this. */ if (!skb_queue_empty(&sta->tx_filtered[ac]) || - !skb_queue_empty(&sta->ps_tx_buf[ac])) { - more_data = true; + !skb_queue_empty(&sta->ps_tx_buf[ac])) break; - } } +} + +static void +ieee80211_sta_ps_deliver_response(struct sta_info *sta, + int n_frames, u8 ignored_acs, + enum ieee80211_frame_release_type reason) +{ + struct ieee80211_sub_if_data *sdata = sta->sdata; + struct ieee80211_local *local = sdata->local; + unsigned long driver_release_tids = 0; + struct sk_buff_head frames; + bool more_data; + + /* Service or PS-Poll period starts */ + set_sta_flag(sta, WLAN_STA_SP); + + __skb_queue_head_init(&frames); + + ieee80211_sta_ps_get_frames(sta, n_frames, ignored_acs, reason, + &frames, &driver_release_tids); + + more_data = ieee80211_sta_ps_more_data(sta, ignored_acs, reason, driver_release_tids); + + if (reason == IEEE80211_FRAME_RELEASE_PSPOLL) + driver_release_tids = + BIT(find_highest_prio_tid(driver_release_tids)); if (skb_queue_empty(&frames) && !driver_release_tids) { int tid; @@ -1435,7 +1477,7 @@ ieee80211_sta_ps_deliver_response(struct sta_info *sta, /* This will evaluate to 1, 3, 5 or 7. */ tid = 7 - ((ffs(~ignored_acs) - 1) << 1); - ieee80211_send_null_response(sdata, sta, tid, reason, true); + ieee80211_send_null_response(sta, tid, reason, true, false); } else if (!driver_release_tids) { struct sk_buff_head pending; struct sk_buff *skb; @@ -1535,8 +1577,8 @@ ieee80211_sta_ps_deliver_response(struct sta_info *sta, if (need_null) ieee80211_send_null_response( - sdata, sta, find_highest_prio_tid(tids), - reason, false); + sta, find_highest_prio_tid(tids), + reason, false, false); sta_info_recalc_tim(sta); } else { @@ -1674,6 +1716,22 @@ void ieee80211_sta_eosp(struct ieee80211_sta *pubsta) } EXPORT_SYMBOL(ieee80211_sta_eosp); +void ieee80211_send_eosp_nullfunc(struct ieee80211_sta *pubsta, int tid) +{ + struct sta_info *sta = container_of(pubsta, struct sta_info, sta); + enum ieee80211_frame_release_type reason; + bool more_data; + + trace_api_send_eosp_nullfunc(sta->local, pubsta, tid); + + reason = IEEE80211_FRAME_RELEASE_UAPSD; + more_data = ieee80211_sta_ps_more_data(sta, ~sta->sta.uapsd_queues, + reason, 0); + + ieee80211_send_null_response(sta, tid, reason, false, more_data); +} +EXPORT_SYMBOL(ieee80211_send_eosp_nullfunc); + void ieee80211_sta_set_buffered(struct ieee80211_sta *pubsta, u8 tid, bool buffered) { diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index 56c6d6cfa5a1..a6b4442776a0 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -2027,6 +2027,31 @@ TRACE_EVENT(api_eosp, ) ); +TRACE_EVENT(api_send_eosp_nullfunc, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sta *sta, + u8 tid), + + TP_ARGS(local, sta, tid), + + TP_STRUCT__entry( + LOCAL_ENTRY + STA_ENTRY + __field(u8, tid) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + STA_ASSIGN; + __entry->tid = tid; + ), + + TP_printk( + LOCAL_PR_FMT STA_PR_FMT " tid:%d", + LOCAL_PR_ARG, STA_PR_ARG, __entry->tid + ) +); + TRACE_EVENT(api_sta_set_buffered, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sta *sta, -- cgit v1.2.3-59-g8ed1b From ef044763a3ca6b9e0bb65a9ce0cb38c0eca62756 Mon Sep 17 00:00:00 2001 From: Eliad Peller Date: Tue, 17 Nov 2015 10:24:37 +0200 Subject: mac80211: add atomic uploaded keys iterator add ieee80211_iter_keys_rcu() to iterate over uploaded keys in atomic context (when rcu is locked) The station removal code removes the keys only after calling synchronize_net(), so it's not safe to iterate the keys at this point (and postponing the actual key deletion with call_rcu() might result in some badly-ordered ops calls). Add a flag to indicate a station is being removed, and skip the configured keys if it's set. Signed-off-by: Eliad Peller Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- include/net/mac80211.h | 24 +++++++++++++++++++++ net/mac80211/key.c | 56 +++++++++++++++++++++++++++++++++++++++++++++---- net/mac80211/sta_info.c | 1 + net/mac80211/sta_info.h | 2 ++ 4 files changed, 79 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 18ac733afc91..a68051c41ac3 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -4917,6 +4917,30 @@ void ieee80211_iter_keys(struct ieee80211_hw *hw, void *data), void *iter_data); +/** + * ieee80211_iter_keys_rcu - iterate keys programmed into the device + * @hw: pointer obtained from ieee80211_alloc_hw() + * @vif: virtual interface to iterate, may be %NULL for all + * @iter: iterator function that will be called for each key + * @iter_data: custom data to pass to the iterator function + * + * This function can be used to iterate all the keys known to + * mac80211, even those that weren't previously programmed into + * the device. Note that due to locking reasons, keys of station + * in removal process will be skipped. + * + * This function requires being called in an RCU critical section, + * and thus iter must be atomic. + */ +void ieee80211_iter_keys_rcu(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + void (*iter)(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta, + struct ieee80211_key_conf *key, + void *data), + void *iter_data); + /** * ieee80211_iter_chan_contexts_atomic - iterate channel contexts * @hw: pointre obtained from ieee80211_alloc_hw(). diff --git a/net/mac80211/key.c b/net/mac80211/key.c index 44388d6a1d8e..5e5bc599da4c 100644 --- a/net/mac80211/key.c +++ b/net/mac80211/key.c @@ -4,6 +4,7 @@ * Copyright 2006-2007 Jiri Benc * Copyright 2007-2008 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH + * Copyright 2015 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 @@ -320,7 +321,7 @@ static void ieee80211_key_replace(struct ieee80211_sub_if_data *sdata, return; if (new) - list_add_tail(&new->list, &sdata->key_list); + list_add_tail_rcu(&new->list, &sdata->key_list); WARN_ON(new && old && new->conf.keyidx != old->conf.keyidx); @@ -368,7 +369,7 @@ static void ieee80211_key_replace(struct ieee80211_sub_if_data *sdata, } if (old) - list_del(&old->list); + list_del_rcu(&old->list); } struct ieee80211_key * @@ -592,8 +593,8 @@ static void ieee80211_key_destroy(struct ieee80211_key *key, return; /* - * Synchronize so the TX path can no longer be using - * this key before we free/remove it. + * Synchronize so the TX path and rcu key iterators + * can no longer be using this key before we free/remove it. */ synchronize_net(); @@ -744,6 +745,53 @@ void ieee80211_iter_keys(struct ieee80211_hw *hw, } EXPORT_SYMBOL(ieee80211_iter_keys); +static void +_ieee80211_iter_keys_rcu(struct ieee80211_hw *hw, + struct ieee80211_sub_if_data *sdata, + void (*iter)(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta, + struct ieee80211_key_conf *key, + void *data), + void *iter_data) +{ + struct ieee80211_key *key; + + list_for_each_entry_rcu(key, &sdata->key_list, list) { + /* skip keys of station in removal process */ + if (key->sta && key->sta->removed) + continue; + if (!(key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE)) + continue; + + iter(hw, &sdata->vif, + key->sta ? &key->sta->sta : NULL, + &key->conf, iter_data); + } +} + +void ieee80211_iter_keys_rcu(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + void (*iter)(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_sta *sta, + struct ieee80211_key_conf *key, + void *data), + void *iter_data) +{ + struct ieee80211_local *local = hw_to_local(hw); + struct ieee80211_sub_if_data *sdata; + + if (vif) { + sdata = vif_to_sdata(vif); + _ieee80211_iter_keys_rcu(hw, sdata, iter, iter_data); + } else { + list_for_each_entry_rcu(sdata, &local->interfaces, list) + _ieee80211_iter_keys_rcu(hw, sdata, iter, iter_data); + } +} +EXPORT_SYMBOL(ieee80211_iter_keys_rcu); + static void ieee80211_free_keys_iface(struct ieee80211_sub_if_data *sdata, struct list_head *keys) { diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 723fa30aafc5..4402ad5b27d1 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -883,6 +883,7 @@ static int __must_check __sta_info_destroy_part1(struct sta_info *sta) } list_del_rcu(&sta->list); + sta->removed = true; drv_sta_pre_rcu_remove(local, sta->sdata, sta); diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 2cafb21b422f..d6051629ed15 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -367,6 +367,7 @@ DECLARE_EWMA(signal, 1024, 8) * @mesh: mesh STA information * @debugfs: debug filesystem info * @dead: set to true when sta is unlinked + * @removed: set to true when sta is being removed from sta_list * @uploaded: set to true when sta is uploaded to the driver * @sta: station information we share with the driver * @sta_state: duplicates information about station state (for debug) @@ -412,6 +413,7 @@ struct sta_info { u16 listen_interval; bool dead; + bool removed; bool uploaded; -- cgit v1.2.3-59-g8ed1b From b115b972997428b9134aba377721fea6486adbd0 Mon Sep 17 00:00:00 2001 From: "Janusz.Dziedzic@tieto.com" Date: Tue, 27 Oct 2015 08:38:40 +0100 Subject: mac80211: add new IEEE80211_VIF_GET_NOA_UPDATE flag Add new VIF flag, that will allow get NOA update notification when driver will request this, even this is not pure P2P vif (eg. STA vif). Signed-off-by: Janusz Dziedzic Signed-off-by: Johannes Berg --- include/net/mac80211.h | 4 ++++ net/mac80211/mlme.c | 6 ++++-- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a68051c41ac3..7c30faff245f 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1321,11 +1321,15 @@ struct ieee80211_channel_switch { * interface. This flag should be set during interface addition, * but may be set/cleared as late as authentication to an AP. It is * only valid for managed/station mode interfaces. + * @IEEE80211_VIF_GET_NOA_UPDATE: request to handle NOA attributes + * and send P2P_PS notification to the driver if NOA changed, even + * this is not pure P2P vif. */ enum ieee80211_vif_flags { IEEE80211_VIF_BEACON_FILTER = BIT(0), IEEE80211_VIF_SUPPORTS_CQM_RSSI = BIT(1), IEEE80211_VIF_SUPPORTS_UAPSD = BIT(2), + IEEE80211_VIF_GET_NOA_UPDATE = BIT(3), }; /** diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index b140cc6651f4..123b26d177e8 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -1930,7 +1930,8 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, sdata->u.mgd.flags |= IEEE80211_STA_RESET_SIGNAL_AVE; - if (sdata->vif.p2p) { + if (sdata->vif.p2p || + sdata->vif.driver_flags & IEEE80211_VIF_GET_NOA_UPDATE) { const struct cfg80211_bss_ies *ies; rcu_read_lock(); @@ -3458,7 +3459,8 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, } } - if (sdata->vif.p2p) { + if (sdata->vif.p2p || + sdata->vif.driver_flags & IEEE80211_VIF_GET_NOA_UPDATE) { struct ieee80211_p2p_noa_attr noa = {}; int ret; -- cgit v1.2.3-59-g8ed1b From 91d3ab46730379e89e1e908c6f62fbcadb3d8f08 Mon Sep 17 00:00:00 2001 From: Vidyullatha Kanchanapally Date: Fri, 30 Oct 2015 19:14:49 +0530 Subject: cfg80211: Add support for aborting an ongoing scan Implement new functionality for aborting an ongoing scan. Add NL80211_CMD_ABORT_SCAN to the nl80211 interface. After aborting the scan, driver shall provide the scan status by calling cfg80211_scan_done(). Reviewed-by: Jouni Malinen Signed-off-by: Vidyullatha Kanchanapally Signed-off-by: Sunil Dutt [change command to take wdev instead of netdev so that it can be used on p2p-device scans] Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 3 +++ include/uapi/linux/nl80211.h | 6 ++++++ net/wireless/nl80211.c | 26 ++++++++++++++++++++++++++ net/wireless/rdev-ops.h | 8 ++++++++ net/wireless/trace.h | 4 ++++ 5 files changed, 47 insertions(+) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index e568872203a5..9bcaaf7cd15a 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2321,6 +2321,8 @@ struct cfg80211_qos_map { * the driver, and will be valid until passed to cfg80211_scan_done(). * For scan results, call cfg80211_inform_bss(); you can call this outside * the scan/scan_done bracket too. + * @abort_scan: Tell the driver to abort an ongoing scan. The driver shall + * indicate the status of the scan through cfg80211_scan_done(). * * @auth: Request to authenticate with the specified peer * (invoked with the wireless_dev mutex held) @@ -2593,6 +2595,7 @@ struct cfg80211_ops { int (*scan)(struct wiphy *wiphy, struct cfg80211_scan_request *request); + void (*abort_scan)(struct wiphy *wiphy, struct wireless_dev *wdev); int (*auth)(struct wiphy *wiphy, struct net_device *dev, struct cfg80211_auth_request *req); diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 07099cb14778..5b7b5ebe7ca8 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -820,6 +820,10 @@ * as an event to indicate changes for devices with wiphy-specific regdom * management. * + * @NL80211_CMD_ABORT_SCAN: Stop an ongoing scan. Returns -ENOENT if a scan is + * not running. The driver indicates the status of the scan through + * cfg80211_scan_done(). + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -1006,6 +1010,8 @@ enum nl80211_commands { NL80211_CMD_WIPHY_REG_CHANGE, + NL80211_CMD_ABORT_SCAN, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 41e57d0c4d43..67e7b531db79 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -5997,6 +5997,24 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) return err; } +static int nl80211_abort_scan(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct wireless_dev *wdev = info->user_ptr[1]; + + if (!rdev->ops->abort_scan) + return -EOPNOTSUPP; + + if (rdev->scan_msg) + return 0; + + if (!rdev->scan_req) + return -ENOENT; + + rdev_abort_scan(rdev, wdev); + return 0; +} + static int nl80211_parse_sched_scan_plans(struct wiphy *wiphy, int n_plans, struct cfg80211_sched_scan_request *request, @@ -10944,6 +10962,14 @@ static const struct genl_ops nl80211_ops[] = { .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, + { + .cmd = NL80211_CMD_ABORT_SCAN, + .doit = nl80211_abort_scan, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_WDEV_UP | + NL80211_FLAG_NEED_RTNL, + }, { .cmd = NL80211_CMD_GET_SCAN, .policy = nl80211_policy, diff --git a/net/wireless/rdev-ops.h b/net/wireless/rdev-ops.h index b8cc594d409d..8ae0c04f9fc7 100644 --- a/net/wireless/rdev-ops.h +++ b/net/wireless/rdev-ops.h @@ -427,6 +427,14 @@ static inline int rdev_scan(struct cfg80211_registered_device *rdev, return ret; } +static inline void rdev_abort_scan(struct cfg80211_registered_device *rdev, + struct wireless_dev *wdev) +{ + trace_rdev_abort_scan(&rdev->wiphy, wdev); + rdev->ops->abort_scan(&rdev->wiphy, wdev); + trace_rdev_return_void(&rdev->wiphy); +} + static inline int rdev_auth(struct cfg80211_registered_device *rdev, struct net_device *dev, struct cfg80211_auth_request *req) diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 5b9139e53199..09b242b09bed 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2917,6 +2917,10 @@ TRACE_EVENT(rdev_set_coalesce, WIPHY_PR_ARG, __entry->n_rules) ); +DEFINE_EVENT(wiphy_wdev_evt, rdev_abort_scan, + TP_PROTO(struct wiphy *wiphy, struct wireless_dev *wdev), + TP_ARGS(wiphy, wdev) +); #endif /* !__RDEV_OPS_TRACE || TRACE_HEADER_MULTI_READ */ #undef TRACE_INCLUDE_PATH -- cgit v1.2.3-59-g8ed1b From 2f8364a291e8adde25c93f97a76abbcaf4b1ed3f Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Thu, 3 Dec 2015 21:12:31 +0100 Subject: WAN: HDLC: Call notifiers before and after changing device type An HDLC device can change type when the protocol driver is changed. Calling the notifier change allows potential users of the interface know about this planned change, and even block it. After the change has occurred, send a second notification to users can evaluate the new device type etc. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/wan/hdlc.c | 19 +++++++++++++++++-- drivers/net/wan/hdlc_cisco.c | 1 + drivers/net/wan/hdlc_fr.c | 1 + drivers/net/wan/hdlc_ppp.c | 1 + drivers/net/wan/hdlc_raw.c | 1 + drivers/net/wan/hdlc_raw_eth.c | 1 + drivers/net/wan/hdlc_x25.c | 1 + include/linux/hdlc.h | 2 +- 8 files changed, 24 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c index 2a6595b4ae15..9bd4aa8083ce 100644 --- a/drivers/net/wan/hdlc.c +++ b/drivers/net/wan/hdlc.c @@ -276,7 +276,11 @@ void unregister_hdlc_device(struct net_device *dev) int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto, size_t size) { - detach_hdlc_protocol(dev); + int err; + + err = detach_hdlc_protocol(dev); + if (err) + return err; if (!try_module_get(proto->module)) return -ENOSYS; @@ -289,15 +293,24 @@ int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto, } } dev_to_hdlc(dev)->proto = proto; + return 0; } -void detach_hdlc_protocol(struct net_device *dev) +int detach_hdlc_protocol(struct net_device *dev) { hdlc_device *hdlc = dev_to_hdlc(dev); + int err; if (hdlc->proto) { + err = call_netdevice_notifiers(NETDEV_PRE_TYPE_CHANGE, dev); + err = notifier_to_errno(err); + if (err) { + netdev_err(dev, "Refused to change device type\n"); + return err; + } + if (hdlc->proto->detach) hdlc->proto->detach(dev); module_put(hdlc->proto->module); @@ -306,6 +319,8 @@ void detach_hdlc_protocol(struct net_device *dev) kfree(hdlc->state); hdlc->state = NULL; hdlc_setup_dev(dev); + + return 0; } diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c index 3f20808b5ff8..a408abc25512 100644 --- a/drivers/net/wan/hdlc_cisco.c +++ b/drivers/net/wan/hdlc_cisco.c @@ -378,6 +378,7 @@ static int cisco_ioctl(struct net_device *dev, struct ifreq *ifr) spin_lock_init(&state(hdlc)->lock); dev->header_ops = &cisco_header_ops; dev->type = ARPHRD_CISCO; + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); netif_dormant_on(dev); return 0; } diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c index 89541cc90e87..b6e0cfb095d3 100644 --- a/drivers/net/wan/hdlc_fr.c +++ b/drivers/net/wan/hdlc_fr.c @@ -1240,6 +1240,7 @@ static int fr_ioctl(struct net_device *dev, struct ifreq *ifr) } memcpy(&state(hdlc)->settings, &new_settings, size); dev->type = ARPHRD_FRAD; + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); return 0; case IF_PROTO_FR_ADD_PVC: diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c index 0d7645581f91..47fdb87d3567 100644 --- a/drivers/net/wan/hdlc_ppp.c +++ b/drivers/net/wan/hdlc_ppp.c @@ -687,6 +687,7 @@ static int ppp_ioctl(struct net_device *dev, struct ifreq *ifr) dev->hard_header_len = sizeof(struct hdlc_header); dev->header_ops = &ppp_header_ops; dev->type = ARPHRD_PPP; + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); netif_dormant_on(dev); return 0; } diff --git a/drivers/net/wan/hdlc_raw.c b/drivers/net/wan/hdlc_raw.c index 5dc153e8a29d..4feb45001aac 100644 --- a/drivers/net/wan/hdlc_raw.c +++ b/drivers/net/wan/hdlc_raw.c @@ -84,6 +84,7 @@ static int raw_ioctl(struct net_device *dev, struct ifreq *ifr) return result; memcpy(hdlc->state, &new_settings, size); dev->type = ARPHRD_RAWHDLC; + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); netif_dormant_off(dev); return 0; } diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c index 3ab72b3082de..2f11836078ab 100644 --- a/drivers/net/wan/hdlc_raw_eth.c +++ b/drivers/net/wan/hdlc_raw_eth.c @@ -102,6 +102,7 @@ static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr) ether_setup(dev); dev->tx_queue_len = old_qlen; eth_hw_addr_random(dev); + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); netif_dormant_off(dev); return 0; } diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c index a49aec5efd20..e867638067a6 100644 --- a/drivers/net/wan/hdlc_x25.c +++ b/drivers/net/wan/hdlc_x25.c @@ -213,6 +213,7 @@ static int x25_ioctl(struct net_device *dev, struct ifreq *ifr) if ((result = attach_hdlc_protocol(dev, &proto, 0))) return result; dev->type = ARPHRD_X25; + call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev); netif_dormant_off(dev); return 0; } diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h index 1acb1445e05f..e31bcd4c7859 100644 --- a/include/linux/hdlc.h +++ b/include/linux/hdlc.h @@ -101,7 +101,7 @@ netdev_tx_t hdlc_start_xmit(struct sk_buff *skb, struct net_device *dev); int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto, size_t size); /* May be used by hardware driver to gain control over HDLC device */ -void detach_hdlc_protocol(struct net_device *dev); +int detach_hdlc_protocol(struct net_device *dev); static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb, struct net_device *dev) -- cgit v1.2.3-59-g8ed1b From b618aaa91b5870e7bd139987ac4b7bf0851142d0 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 4 Dec 2015 15:01:31 +0100 Subject: net: constify netif_is_* helpers net_device param As suggested by Eric, these helpers should have const dev param. Suggested-by: Eric Dumazet Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 2 +- include/linux/netdevice.h | 22 +++++++++++----------- net/core/dev.c | 2 +- 3 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 67ce5bd3b56a..05f5879821b8 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -73,7 +73,7 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb) /* found in socket.c */ extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *)); -static inline bool is_vlan_dev(struct net_device *dev) +static inline bool is_vlan_dev(const struct net_device *dev) { return dev->priv_flags & IFF_802_1Q_VLAN; } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3efe017fe419..1bb21ff0fa64 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3661,7 +3661,7 @@ extern u8 netdev_rss_key[NETDEV_RSS_KEY_LEN]; void netdev_rss_key_fill(void *buffer, size_t len); int dev_get_nest_level(struct net_device *dev, - bool (*type_check)(struct net_device *dev)); + bool (*type_check)(const struct net_device *dev)); int skb_checksum_help(struct sk_buff *skb); struct sk_buff *__skb_gso_segment(struct sk_buff *skb, netdev_features_t features, bool tx_path); @@ -3858,32 +3858,32 @@ static inline void skb_gso_error_unwind(struct sk_buff *skb, __be16 protocol, skb->mac_len = mac_len; } -static inline bool netif_is_macvlan(struct net_device *dev) +static inline bool netif_is_macvlan(const struct net_device *dev) { return dev->priv_flags & IFF_MACVLAN; } -static inline bool netif_is_macvlan_port(struct net_device *dev) +static inline bool netif_is_macvlan_port(const struct net_device *dev) { return dev->priv_flags & IFF_MACVLAN_PORT; } -static inline bool netif_is_ipvlan(struct net_device *dev) +static inline bool netif_is_ipvlan(const struct net_device *dev) { return dev->priv_flags & IFF_IPVLAN_SLAVE; } -static inline bool netif_is_ipvlan_port(struct net_device *dev) +static inline bool netif_is_ipvlan_port(const struct net_device *dev) { return dev->priv_flags & IFF_IPVLAN_MASTER; } -static inline bool netif_is_bond_master(struct net_device *dev) +static inline bool netif_is_bond_master(const struct net_device *dev) { return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING; } -static inline bool netif_is_bond_slave(struct net_device *dev) +static inline bool netif_is_bond_slave(const struct net_device *dev) { return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING; } @@ -3918,22 +3918,22 @@ static inline bool netif_is_ovs_master(const struct net_device *dev) return dev->priv_flags & IFF_OPENVSWITCH; } -static inline bool netif_is_team_master(struct net_device *dev) +static inline bool netif_is_team_master(const struct net_device *dev) { return dev->priv_flags & IFF_TEAM; } -static inline bool netif_is_team_port(struct net_device *dev) +static inline bool netif_is_team_port(const struct net_device *dev) { return dev->priv_flags & IFF_TEAM_PORT; } -static inline bool netif_is_lag_master(struct net_device *dev) +static inline bool netif_is_lag_master(const struct net_device *dev) { return netif_is_bond_master(dev) || netif_is_team_master(dev); } -static inline bool netif_is_lag_port(struct net_device *dev) +static inline bool netif_is_lag_port(const struct net_device *dev) { return netif_is_bond_slave(dev) || netif_is_team_port(dev); } diff --git a/net/core/dev.c b/net/core/dev.c index d1706e88fbeb..e5c395473eba 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5734,7 +5734,7 @@ EXPORT_SYMBOL(netdev_lower_dev_get_private); int dev_get_nest_level(struct net_device *dev, - bool (*type_check)(struct net_device *dev)) + bool (*type_check)(const struct net_device *dev)) { struct net_device *lower = NULL; struct list_head *iter; -- cgit v1.2.3-59-g8ed1b From 5f61385d2ebc2bd62bc389c7da0d8d2f263be1eb Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Sun, 6 Dec 2015 18:07:41 +0200 Subject: net/mlx4_core: Keep VLAN/MAC tables mirrored in multifunc HA mode Due to HW limitations, indexes to MAC and VLAN tables are always taken from the table of the actual port. So, if a resource holds an index to a table, it may refer to different values during the lifetime of the resource, unless the tables are mirrored. Also, even when driver is not in HA mode the policy of allocating an index to these tables is such to make sure, as much as possible, that when the time comes the mirroring will be successful. This means that in multifunction mode the allocation of a free index in a port's table tries to make sure that the same index in the other's port table is also free. Signed-off-by: Moni Shoua Reviewed-by: Jack Morgenstein Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/mlx4.h | 6 + drivers/net/ethernet/mellanox/mlx4/port.c | 598 ++++++++++++++++++++++++++++-- include/linux/mlx4/driver.h | 5 + 3 files changed, 586 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 33c4c6f2c4bb..2404c22ad2b2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -736,6 +736,7 @@ struct mlx4_catas_err { struct mlx4_mac_table { __be64 entries[MLX4_MAX_MAC_NUM]; int refs[MLX4_MAX_MAC_NUM]; + bool is_dup[MLX4_MAX_MAC_NUM]; struct mutex mutex; int total; int max; @@ -758,6 +759,7 @@ struct mlx4_roce_gid_table { struct mlx4_vlan_table { __be32 entries[MLX4_MAX_VLAN_NUM]; int refs[MLX4_MAX_VLAN_NUM]; + int is_dup[MLX4_MAX_VLAN_NUM]; struct mutex mutex; int total; int max; @@ -1225,6 +1227,10 @@ void mlx4_init_roce_gid_table(struct mlx4_dev *dev, struct mlx4_roce_gid_table *table); void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan); int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index); +int mlx4_bond_vlan_table(struct mlx4_dev *dev); +int mlx4_unbond_vlan_table(struct mlx4_dev *dev); +int mlx4_bond_mac_table(struct mlx4_dev *dev); +int mlx4_unbond_mac_table(struct mlx4_dev *dev); int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz); /* resource tracker functions*/ diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index c2b21313dba7..f2550425c251 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -61,6 +61,7 @@ void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { table->entries[i] = 0; table->refs[i] = 0; + table->is_dup[i] = false; } table->max = 1 << dev->caps.log_num_macs; table->total = 0; @@ -74,6 +75,7 @@ void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table) for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) { table->entries[i] = 0; table->refs[i] = 0; + table->is_dup[i] = false; } table->max = (1 << dev->caps.log_num_vlans) - MLX4_VLAN_REGULAR; table->total = 0; @@ -159,21 +161,94 @@ int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx) } EXPORT_SYMBOL_GPL(mlx4_find_cached_mac); +static bool mlx4_need_mf_bond(struct mlx4_dev *dev) +{ + int i, num_eth_ports = 0; + + if (!mlx4_is_mfunc(dev)) + return false; + mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) + ++num_eth_ports; + + return (num_eth_ports == 2) ? true : false; +} + int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac) { struct mlx4_port_info *info = &mlx4_priv(dev)->port[port]; struct mlx4_mac_table *table = &info->mac_table; int i, err = 0; int free = -1; + int free_for_dup = -1; + bool dup = mlx4_is_mf_bonded(dev); + u8 dup_port = (port == 1) ? 2 : 1; + struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table; + bool need_mf_bond = mlx4_need_mf_bond(dev); + bool can_mf_bond = true; + + mlx4_dbg(dev, "Registering MAC: 0x%llx for port %d %s duplicate\n", + (unsigned long long)mac, port, + dup ? "with" : "without"); + + if (need_mf_bond) { + if (port == 1) { + mutex_lock(&table->mutex); + mutex_lock(&dup_table->mutex); + } else { + mutex_lock(&dup_table->mutex); + mutex_lock(&table->mutex); + } + } else { + mutex_lock(&table->mutex); + } + + if (need_mf_bond) { + int index_at_port = -1; + int index_at_dup_port = -1; - mlx4_dbg(dev, "Registering MAC: 0x%llx for port %d\n", - (unsigned long long) mac, port); + for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { + if (((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i])))) + index_at_port = i; + if (((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(dup_table->entries[i])))) + index_at_dup_port = i; + } + + /* check that same mac is not in the tables at different indices */ + if ((index_at_port != index_at_dup_port) && + (index_at_port >= 0) && + (index_at_dup_port >= 0)) + can_mf_bond = false; + + /* If the mac is already in the primary table, the slot must be + * available in the duplicate table as well. + */ + if (index_at_port >= 0 && index_at_dup_port < 0 && + dup_table->refs[index_at_port]) { + can_mf_bond = false; + } + /* If the mac is already in the duplicate table, check that the + * corresponding index is not occupied in the primary table, or + * the primary table already contains the mac at the same index. + * Otherwise, you cannot bond (primary contains a different mac + * at that index). + */ + if (index_at_dup_port >= 0) { + if (!table->refs[index_at_dup_port] || + ((MLX4_MAC_MASK & mac) == (MLX4_MAC_MASK & be64_to_cpu(table->entries[index_at_dup_port])))) + free_for_dup = index_at_dup_port; + else + can_mf_bond = false; + } + } - mutex_lock(&table->mutex); for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { if (!table->refs[i]) { if (free < 0) free = i; + if (free_for_dup < 0 && need_mf_bond && can_mf_bond) { + if (!dup_table->refs[i]) + free_for_dup = i; + } continue; } @@ -182,10 +257,30 @@ int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac) /* MAC already registered, increment ref count */ err = i; ++table->refs[i]; + if (dup) { + u64 dup_mac = MLX4_MAC_MASK & be64_to_cpu(dup_table->entries[i]); + + if (dup_mac != mac || !dup_table->is_dup[i]) { + mlx4_warn(dev, "register mac: expect duplicate mac 0x%llx on port %d index %d\n", + mac, dup_port, i); + } + } goto out; } } + if (need_mf_bond && (free_for_dup < 0)) { + if (dup) { + mlx4_warn(dev, "Fail to allocate duplicate MAC table entry\n"); + mlx4_warn(dev, "High Availability for virtual functions may not work as expected\n"); + dup = false; + } + can_mf_bond = false; + } + + if (need_mf_bond && can_mf_bond) + free = free_for_dup; + mlx4_dbg(dev, "Free MAC index is %d\n", free); if (table->total == table->max) { @@ -205,10 +300,35 @@ int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac) goto out; } table->refs[free] = 1; - err = free; + table->is_dup[free] = false; ++table->total; + if (dup) { + dup_table->refs[free] = 0; + dup_table->is_dup[free] = true; + dup_table->entries[free] = cpu_to_be64(mac | MLX4_MAC_VALID); + + err = mlx4_set_port_mac_table(dev, dup_port, dup_table->entries); + if (unlikely(err)) { + mlx4_warn(dev, "Failed adding duplicate mac: 0x%llx\n", mac); + dup_table->is_dup[free] = false; + dup_table->entries[free] = 0; + goto out; + } + ++dup_table->total; + } + err = free; out: - mutex_unlock(&table->mutex); + if (need_mf_bond) { + if (port == 2) { + mutex_unlock(&table->mutex); + mutex_unlock(&dup_table->mutex); + } else { + mutex_unlock(&dup_table->mutex); + mutex_unlock(&table->mutex); + } + } else { + mutex_unlock(&table->mutex); + } return err; } EXPORT_SYMBOL_GPL(__mlx4_register_mac); @@ -255,6 +375,9 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac) struct mlx4_port_info *info; struct mlx4_mac_table *table; int index; + bool dup = mlx4_is_mf_bonded(dev); + u8 dup_port = (port == 1) ? 2 : 1; + struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table; if (port < 1 || port > dev->caps.num_ports) { mlx4_warn(dev, "invalid port number (%d), aborting...\n", port); @@ -262,22 +385,59 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac) } info = &mlx4_priv(dev)->port[port]; table = &info->mac_table; - mutex_lock(&table->mutex); + + if (dup) { + if (port == 1) { + mutex_lock(&table->mutex); + mutex_lock(&dup_table->mutex); + } else { + mutex_lock(&dup_table->mutex); + mutex_lock(&table->mutex); + } + } else { + mutex_lock(&table->mutex); + } + index = find_index(dev, table, mac); if (validate_index(dev, table, index)) goto out; - if (--table->refs[index]) { + + if (--table->refs[index] || table->is_dup[index]) { mlx4_dbg(dev, "Have more references for index %d, no need to modify mac table\n", index); + if (!table->refs[index]) + dup_table->is_dup[index] = false; goto out; } table->entries[index] = 0; - mlx4_set_port_mac_table(dev, port, table->entries); + if (mlx4_set_port_mac_table(dev, port, table->entries)) + mlx4_warn(dev, "Fail to set mac in port %d during unregister\n", port); --table->total; + + if (dup) { + dup_table->is_dup[index] = false; + if (dup_table->refs[index]) + goto out; + dup_table->entries[index] = 0; + if (mlx4_set_port_mac_table(dev, dup_port, dup_table->entries)) + mlx4_warn(dev, "Fail to set mac in duplicate port %d during unregister\n", dup_port); + + --table->total; + } out: - mutex_unlock(&table->mutex); + if (dup) { + if (port == 2) { + mutex_unlock(&table->mutex); + mutex_unlock(&dup_table->mutex); + } else { + mutex_unlock(&dup_table->mutex); + mutex_unlock(&table->mutex); + } + } else { + mutex_unlock(&table->mutex); + } } EXPORT_SYMBOL_GPL(__mlx4_unregister_mac); @@ -311,9 +471,22 @@ int __mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac) struct mlx4_mac_table *table = &info->mac_table; int index = qpn - info->base_qpn; int err = 0; + bool dup = mlx4_is_mf_bonded(dev); + u8 dup_port = (port == 1) ? 2 : 1; + struct mlx4_mac_table *dup_table = &mlx4_priv(dev)->port[dup_port].mac_table; /* CX1 doesn't support multi-functions */ - mutex_lock(&table->mutex); + if (dup) { + if (port == 1) { + mutex_lock(&table->mutex); + mutex_lock(&dup_table->mutex); + } else { + mutex_lock(&dup_table->mutex); + mutex_lock(&table->mutex); + } + } else { + mutex_lock(&table->mutex); + } err = validate_index(dev, table, index); if (err) @@ -326,9 +499,30 @@ int __mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac) mlx4_err(dev, "Failed adding MAC: 0x%llx\n", (unsigned long long) new_mac); table->entries[index] = 0; + } else { + if (dup) { + dup_table->entries[index] = cpu_to_be64(new_mac | MLX4_MAC_VALID); + + err = mlx4_set_port_mac_table(dev, dup_port, dup_table->entries); + if (unlikely(err)) { + mlx4_err(dev, "Failed adding duplicate MAC: 0x%llx\n", + (unsigned long long)new_mac); + dup_table->entries[index] = 0; + } + } } out: - mutex_unlock(&table->mutex); + if (dup) { + if (port == 2) { + mutex_unlock(&table->mutex); + mutex_unlock(&dup_table->mutex); + } else { + mutex_unlock(&dup_table->mutex); + mutex_unlock(&table->mutex); + } + } else { + mutex_unlock(&table->mutex); + } return err; } EXPORT_SYMBOL_GPL(__mlx4_replace_mac); @@ -380,8 +574,28 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table; int i, err = 0; int free = -1; - - mutex_lock(&table->mutex); + int free_for_dup = -1; + bool dup = mlx4_is_mf_bonded(dev); + u8 dup_port = (port == 1) ? 2 : 1; + struct mlx4_vlan_table *dup_table = &mlx4_priv(dev)->port[dup_port].vlan_table; + bool need_mf_bond = mlx4_need_mf_bond(dev); + bool can_mf_bond = true; + + mlx4_dbg(dev, "Registering VLAN: %d for port %d %s duplicate\n", + vlan, port, + dup ? "with" : "without"); + + if (need_mf_bond) { + if (port == 1) { + mutex_lock(&table->mutex); + mutex_lock(&dup_table->mutex); + } else { + mutex_lock(&dup_table->mutex); + mutex_lock(&table->mutex); + } + } else { + mutex_lock(&table->mutex); + } if (table->total == table->max) { /* No free vlan entries */ @@ -389,22 +603,85 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, goto out; } + if (need_mf_bond) { + int index_at_port = -1; + int index_at_dup_port = -1; + + for (i = MLX4_VLAN_REGULAR; i < MLX4_MAX_VLAN_NUM; i++) { + if ((vlan == (MLX4_VLAN_MASK & be32_to_cpu(table->entries[i])))) + index_at_port = i; + if ((vlan == (MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[i])))) + index_at_dup_port = i; + } + /* check that same vlan is not in the tables at different indices */ + if ((index_at_port != index_at_dup_port) && + (index_at_port >= 0) && + (index_at_dup_port >= 0)) + can_mf_bond = false; + + /* If the vlan is already in the primary table, the slot must be + * available in the duplicate table as well. + */ + if (index_at_port >= 0 && index_at_dup_port < 0 && + dup_table->refs[index_at_port]) { + can_mf_bond = false; + } + /* If the vlan is already in the duplicate table, check that the + * corresponding index is not occupied in the primary table, or + * the primary table already contains the vlan at the same index. + * Otherwise, you cannot bond (primary contains a different vlan + * at that index). + */ + if (index_at_dup_port >= 0) { + if (!table->refs[index_at_dup_port] || + (vlan == (MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[index_at_dup_port])))) + free_for_dup = index_at_dup_port; + else + can_mf_bond = false; + } + } + for (i = MLX4_VLAN_REGULAR; i < MLX4_MAX_VLAN_NUM; i++) { - if (free < 0 && (table->refs[i] == 0)) { - free = i; - continue; + if (!table->refs[i]) { + if (free < 0) + free = i; + if (free_for_dup < 0 && need_mf_bond && can_mf_bond) { + if (!dup_table->refs[i]) + free_for_dup = i; + } } - if (table->refs[i] && + if ((table->refs[i] || table->is_dup[i]) && (vlan == (MLX4_VLAN_MASK & be32_to_cpu(table->entries[i])))) { /* Vlan already registered, increase references count */ + mlx4_dbg(dev, "vlan %u is already registered.\n", vlan); *index = i; ++table->refs[i]; + if (dup) { + u16 dup_vlan = MLX4_VLAN_MASK & be32_to_cpu(dup_table->entries[i]); + + if (dup_vlan != vlan || !dup_table->is_dup[i]) { + mlx4_warn(dev, "register vlan: expected duplicate vlan %u on port %d index %d\n", + vlan, dup_port, i); + } + } goto out; } } + if (need_mf_bond && (free_for_dup < 0)) { + if (dup) { + mlx4_warn(dev, "Fail to allocate duplicate VLAN table entry\n"); + mlx4_warn(dev, "High Availability for virtual functions may not work as expected\n"); + dup = false; + } + can_mf_bond = false; + } + + if (need_mf_bond && can_mf_bond) + free = free_for_dup; + if (free < 0) { err = -ENOMEM; goto out; @@ -412,6 +689,7 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, /* Register new VLAN */ table->refs[free] = 1; + table->is_dup[free] = false; table->entries[free] = cpu_to_be32(vlan | MLX4_VLAN_VALID); err = mlx4_set_port_vlan_table(dev, port, table->entries); @@ -421,11 +699,35 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, table->entries[free] = 0; goto out; } + ++table->total; + if (dup) { + dup_table->refs[free] = 0; + dup_table->is_dup[free] = true; + dup_table->entries[free] = cpu_to_be32(vlan | MLX4_VLAN_VALID); + + err = mlx4_set_port_vlan_table(dev, dup_port, dup_table->entries); + if (unlikely(err)) { + mlx4_warn(dev, "Failed adding duplicate vlan: %u\n", vlan); + dup_table->is_dup[free] = false; + dup_table->entries[free] = 0; + goto out; + } + ++dup_table->total; + } *index = free; - ++table->total; out: - mutex_unlock(&table->mutex); + if (need_mf_bond) { + if (port == 2) { + mutex_unlock(&table->mutex); + mutex_unlock(&dup_table->mutex); + } else { + mutex_unlock(&dup_table->mutex); + mutex_unlock(&table->mutex); + } + } else { + mutex_unlock(&table->mutex); + } return err; } @@ -455,8 +757,22 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan) { struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table; int index; + bool dup = mlx4_is_mf_bonded(dev); + u8 dup_port = (port == 1) ? 2 : 1; + struct mlx4_vlan_table *dup_table = &mlx4_priv(dev)->port[dup_port].vlan_table; + + if (dup) { + if (port == 1) { + mutex_lock(&table->mutex); + mutex_lock(&dup_table->mutex); + } else { + mutex_lock(&dup_table->mutex); + mutex_lock(&table->mutex); + } + } else { + mutex_lock(&table->mutex); + } - mutex_lock(&table->mutex); if (mlx4_find_cached_vlan(dev, port, vlan, &index)) { mlx4_warn(dev, "vlan 0x%x is not in the vlan table\n", vlan); goto out; @@ -467,16 +783,38 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan) goto out; } - if (--table->refs[index]) { + if (--table->refs[index] || table->is_dup[index]) { mlx4_dbg(dev, "Have %d more references for index %d, no need to modify vlan table\n", table->refs[index], index); + if (!table->refs[index]) + dup_table->is_dup[index] = false; goto out; } table->entries[index] = 0; - mlx4_set_port_vlan_table(dev, port, table->entries); + if (mlx4_set_port_vlan_table(dev, port, table->entries)) + mlx4_warn(dev, "Fail to set vlan in port %d during unregister\n", port); --table->total; + if (dup) { + dup_table->is_dup[index] = false; + if (dup_table->refs[index]) + goto out; + dup_table->entries[index] = 0; + if (mlx4_set_port_vlan_table(dev, dup_port, dup_table->entries)) + mlx4_warn(dev, "Fail to set vlan in duplicate port %d during unregister\n", dup_port); + --dup_table->total; + } out: - mutex_unlock(&table->mutex); + if (dup) { + if (port == 2) { + mutex_unlock(&table->mutex); + mutex_unlock(&dup_table->mutex); + } else { + mutex_unlock(&dup_table->mutex); + mutex_unlock(&table->mutex); + } + } else { + mutex_unlock(&table->mutex); + } } void mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan) @@ -495,6 +833,220 @@ void mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan) } EXPORT_SYMBOL_GPL(mlx4_unregister_vlan); +int mlx4_bond_mac_table(struct mlx4_dev *dev) +{ + struct mlx4_mac_table *t1 = &mlx4_priv(dev)->port[1].mac_table; + struct mlx4_mac_table *t2 = &mlx4_priv(dev)->port[2].mac_table; + int ret = 0; + int i; + bool update1 = false; + bool update2 = false; + + mutex_lock(&t1->mutex); + mutex_lock(&t2->mutex); + for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { + if ((t1->entries[i] != t2->entries[i]) && + t1->entries[i] && t2->entries[i]) { + mlx4_warn(dev, "can't duplicate entry %d in mac table\n", i); + ret = -EINVAL; + goto unlock; + } + } + + for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { + if (t1->entries[i] && !t2->entries[i]) { + t2->entries[i] = t1->entries[i]; + t2->is_dup[i] = true; + update2 = true; + } else if (!t1->entries[i] && t2->entries[i]) { + t1->entries[i] = t2->entries[i]; + t1->is_dup[i] = true; + update1 = true; + } else if (t1->entries[i] && t2->entries[i]) { + t1->is_dup[i] = true; + t2->is_dup[i] = true; + } + } + + if (update1) { + ret = mlx4_set_port_mac_table(dev, 1, t1->entries); + if (ret) + mlx4_warn(dev, "failed to set MAC table for port 1 (%d)\n", ret); + } + if (!ret && update2) { + ret = mlx4_set_port_mac_table(dev, 2, t2->entries); + if (ret) + mlx4_warn(dev, "failed to set MAC table for port 2 (%d)\n", ret); + } + + if (ret) + mlx4_warn(dev, "failed to create mirror MAC tables\n"); +unlock: + mutex_unlock(&t2->mutex); + mutex_unlock(&t1->mutex); + return ret; +} + +int mlx4_unbond_mac_table(struct mlx4_dev *dev) +{ + struct mlx4_mac_table *t1 = &mlx4_priv(dev)->port[1].mac_table; + struct mlx4_mac_table *t2 = &mlx4_priv(dev)->port[2].mac_table; + int ret = 0; + int ret1; + int i; + bool update1 = false; + bool update2 = false; + + mutex_lock(&t1->mutex); + mutex_lock(&t2->mutex); + for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { + if (t1->entries[i] != t2->entries[i]) { + mlx4_warn(dev, "mac table is in an unexpected state when trying to unbond\n"); + ret = -EINVAL; + goto unlock; + } + } + + for (i = 0; i < MLX4_MAX_MAC_NUM; i++) { + if (!t1->entries[i]) + continue; + t1->is_dup[i] = false; + if (!t1->refs[i]) { + t1->entries[i] = 0; + update1 = true; + } + t2->is_dup[i] = false; + if (!t2->refs[i]) { + t2->entries[i] = 0; + update2 = true; + } + } + + if (update1) { + ret = mlx4_set_port_mac_table(dev, 1, t1->entries); + if (ret) + mlx4_warn(dev, "failed to unmirror MAC tables for port 1(%d)\n", ret); + } + if (update2) { + ret1 = mlx4_set_port_mac_table(dev, 2, t2->entries); + if (ret1) { + mlx4_warn(dev, "failed to unmirror MAC tables for port 2(%d)\n", ret1); + ret = ret1; + } + } +unlock: + mutex_unlock(&t2->mutex); + mutex_unlock(&t1->mutex); + return ret; +} + +int mlx4_bond_vlan_table(struct mlx4_dev *dev) +{ + struct mlx4_vlan_table *t1 = &mlx4_priv(dev)->port[1].vlan_table; + struct mlx4_vlan_table *t2 = &mlx4_priv(dev)->port[2].vlan_table; + int ret = 0; + int i; + bool update1 = false; + bool update2 = false; + + mutex_lock(&t1->mutex); + mutex_lock(&t2->mutex); + for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) { + if ((t1->entries[i] != t2->entries[i]) && + t1->entries[i] && t2->entries[i]) { + mlx4_warn(dev, "can't duplicate entry %d in vlan table\n", i); + ret = -EINVAL; + goto unlock; + } + } + + for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) { + if (t1->entries[i] && !t2->entries[i]) { + t2->entries[i] = t1->entries[i]; + t2->is_dup[i] = true; + update2 = true; + } else if (!t1->entries[i] && t2->entries[i]) { + t1->entries[i] = t2->entries[i]; + t1->is_dup[i] = true; + update1 = true; + } else if (t1->entries[i] && t2->entries[i]) { + t1->is_dup[i] = true; + t2->is_dup[i] = true; + } + } + + if (update1) { + ret = mlx4_set_port_vlan_table(dev, 1, t1->entries); + if (ret) + mlx4_warn(dev, "failed to set VLAN table for port 1 (%d)\n", ret); + } + if (!ret && update2) { + ret = mlx4_set_port_vlan_table(dev, 2, t2->entries); + if (ret) + mlx4_warn(dev, "failed to set VLAN table for port 2 (%d)\n", ret); + } + + if (ret) + mlx4_warn(dev, "failed to create mirror VLAN tables\n"); +unlock: + mutex_unlock(&t2->mutex); + mutex_unlock(&t1->mutex); + return ret; +} + +int mlx4_unbond_vlan_table(struct mlx4_dev *dev) +{ + struct mlx4_vlan_table *t1 = &mlx4_priv(dev)->port[1].vlan_table; + struct mlx4_vlan_table *t2 = &mlx4_priv(dev)->port[2].vlan_table; + int ret = 0; + int ret1; + int i; + bool update1 = false; + bool update2 = false; + + mutex_lock(&t1->mutex); + mutex_lock(&t2->mutex); + for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) { + if (t1->entries[i] != t2->entries[i]) { + mlx4_warn(dev, "vlan table is in an unexpected state when trying to unbond\n"); + ret = -EINVAL; + goto unlock; + } + } + + for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) { + if (!t1->entries[i]) + continue; + t1->is_dup[i] = false; + if (!t1->refs[i]) { + t1->entries[i] = 0; + update1 = true; + } + t2->is_dup[i] = false; + if (!t2->refs[i]) { + t2->entries[i] = 0; + update2 = true; + } + } + + if (update1) { + ret = mlx4_set_port_vlan_table(dev, 1, t1->entries); + if (ret) + mlx4_warn(dev, "failed to unmirror VLAN tables for port 1(%d)\n", ret); + } + if (update2) { + ret1 = mlx4_set_port_vlan_table(dev, 2, t2->entries); + if (ret1) { + mlx4_warn(dev, "failed to unmirror VLAN tables for port 2(%d)\n", ret1); + ret = ret1; + } + } +unlock: + mutex_unlock(&t2->mutex); + mutex_unlock(&t1->mutex); + return ret; +} + int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps) { struct mlx4_cmd_mailbox *inmailbox, *outmailbox; diff --git a/include/linux/mlx4/driver.h b/include/linux/mlx4/driver.h index 5a06d969338e..2e8af001c5da 100644 --- a/include/linux/mlx4/driver.h +++ b/include/linux/mlx4/driver.h @@ -75,6 +75,11 @@ static inline int mlx4_is_bonded(struct mlx4_dev *dev) return !!(dev->flags & MLX4_FLAG_BONDED); } +static inline int mlx4_is_mf_bonded(struct mlx4_dev *dev) +{ + return (mlx4_is_bonded(dev) && mlx4_is_mfunc(dev)); +} + struct mlx4_port_map { u8 port1; u8 port2; -- cgit v1.2.3-59-g8ed1b From ea3793ee29d3621faf857fa8ef5425e9ff9a756d Mon Sep 17 00:00:00 2001 From: Rainer Weikusat Date: Sun, 6 Dec 2015 21:11:34 +0000 Subject: core: enable more fine-grained datagram reception control The __skb_recv_datagram routine in core/ datagram.c provides a general skb reception factility supposed to be utilized by protocol modules providing datagram sockets. It encompasses both the actual recvmsg code and a surrounding 'sleep until data is available' loop. This is inconvenient if a protocol module has to use additional locking in order to maintain some per-socket state the generic datagram socket code is unaware of (as the af_unix code does). The patch below moves the recvmsg proper code into a new __skb_try_recv_datagram routine which doesn't sleep and renames wait_for_more_packets to __skb_wait_for_more_packets, both routines being exported interfaces. The original __skb_recv_datagram routine is reimplemented on top of these two functions such that its user-visible behaviour remains unchanged. Signed-off-by: Rainer Weikusat Signed-off-by: David S. Miller --- include/linux/skbuff.h | 6 ++++ net/core/datagram.c | 77 +++++++++++++++++++++++++++++++------------------- 2 files changed, 54 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c9c394bf0771..9b9b9ead7bb3 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2785,6 +2785,12 @@ static inline void skb_frag_list_init(struct sk_buff *skb) #define skb_walk_frags(skb, iter) \ for (iter = skb_shinfo(skb)->frag_list; iter; iter = iter->next) + +int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p, + const struct sk_buff *skb); +struct sk_buff *__skb_try_recv_datagram(struct sock *sk, unsigned flags, + int *peeked, int *off, int *err, + struct sk_buff **last); struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags, int *peeked, int *off, int *err); struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned flags, int noblock, diff --git a/net/core/datagram.c b/net/core/datagram.c index d62af69ad844..7daff66d3d0b 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -83,8 +83,8 @@ static int receiver_wake_function(wait_queue_t *wait, unsigned int mode, int syn /* * Wait for the last received packet to be different from skb */ -static int wait_for_more_packets(struct sock *sk, int *err, long *timeo_p, - const struct sk_buff *skb) +int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p, + const struct sk_buff *skb) { int error; DEFINE_WAIT_FUNC(wait, receiver_wake_function); @@ -130,6 +130,7 @@ out_noerr: error = 1; goto out; } +EXPORT_SYMBOL(__skb_wait_for_more_packets); static struct sk_buff *skb_set_peeked(struct sk_buff *skb) { @@ -161,13 +162,15 @@ done: } /** - * __skb_recv_datagram - Receive a datagram skbuff + * __skb_try_recv_datagram - Receive a datagram skbuff * @sk: socket * @flags: MSG_ flags * @peeked: returns non-zero if this packet has been seen before * @off: an offset in bytes to peek skb from. Returns an offset * within an skb where data actually starts * @err: error code returned + * @last: set to last peeked message to inform the wait function + * what to look for when peeking * * Get a datagram skbuff, understands the peeking, nonblocking wakeups * and possible races. This replaces identical code in packet, raw and @@ -175,9 +178,11 @@ done: * the long standing peek and read race for datagram sockets. If you * alter this routine remember it must be re-entrant. * - * This function will lock the socket if a skb is returned, so the caller - * needs to unlock the socket in that case (usually by calling - * skb_free_datagram) + * This function will lock the socket if a skb is returned, so + * the caller needs to unlock the socket in that case (usually by + * calling skb_free_datagram). Returns NULL with *err set to + * -EAGAIN if no data was available or to some other value if an + * error was detected. * * * It does not lock socket since today. This function is * * free of race conditions. This measure should/can improve @@ -191,13 +196,13 @@ done: * quite explicitly by POSIX 1003.1g, don't change them without having * the standard around please. */ -struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, - int *peeked, int *off, int *err) +struct sk_buff *__skb_try_recv_datagram(struct sock *sk, unsigned int flags, + int *peeked, int *off, int *err, + struct sk_buff **last) { struct sk_buff_head *queue = &sk->sk_receive_queue; - struct sk_buff *skb, *last; + struct sk_buff *skb; unsigned long cpu_flags; - long timeo; /* * Caller is allowed not to check sk->sk_err before skb_recv_datagram() */ @@ -206,8 +211,6 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, if (error) goto no_packet; - timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); - do { /* Again only user level code calls this function, so nothing * interrupt level will suddenly eat the receive_queue. @@ -217,10 +220,10 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, */ int _off = *off; - last = (struct sk_buff *)queue; + *last = (struct sk_buff *)queue; spin_lock_irqsave(&queue->lock, cpu_flags); skb_queue_walk(queue, skb) { - last = skb; + *last = skb; *peeked = skb->peeked; if (flags & MSG_PEEK) { if (_off >= skb->len && (skb->len || _off || @@ -231,8 +234,11 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, skb = skb_set_peeked(skb); error = PTR_ERR(skb); - if (IS_ERR(skb)) - goto unlock_err; + if (IS_ERR(skb)) { + spin_unlock_irqrestore(&queue->lock, + cpu_flags); + goto no_packet; + } atomic_inc(&skb->users); } else @@ -242,25 +248,38 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, *off = _off; return skb; } + spin_unlock_irqrestore(&queue->lock, cpu_flags); + } while (sk_can_busy_loop(sk) && + sk_busy_loop(sk, flags & MSG_DONTWAIT)); - if (sk_can_busy_loop(sk) && - sk_busy_loop(sk, flags & MSG_DONTWAIT)) - continue; + error = -EAGAIN; - /* User doesn't want to wait */ - error = -EAGAIN; - if (!timeo) - goto no_packet; +no_packet: + *err = error; + return NULL; +} +EXPORT_SYMBOL(__skb_try_recv_datagram); - } while (!wait_for_more_packets(sk, err, &timeo, last)); +struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, + int *peeked, int *off, int *err) +{ + struct sk_buff *skb, *last; + long timeo; - return NULL; + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + + do { + skb = __skb_try_recv_datagram(sk, flags, peeked, off, err, + &last); + if (skb) + return skb; + + if (*err != EAGAIN) + break; + } while (timeo && + !__skb_wait_for_more_packets(sk, err, &timeo, last)); -unlock_err: - spin_unlock_irqrestore(&queue->lock, cpu_flags); -no_packet: - *err = error; return NULL; } EXPORT_SYMBOL(__skb_recv_datagram); -- cgit v1.2.3-59-g8ed1b From 4baee937b8d551c89f61542a575378e407b63415 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 7 Dec 2015 13:57:32 +0100 Subject: net: dsa: remove DSA link polling Since no more DSA driver uses the polling callback, and since the phylib handles the link detection, remove the link polling work and timer code. Signed-off-by: Neil Armstrong Signed-off-by: David S. Miller --- include/net/dsa.h | 12 ------------ net/dsa/dsa.c | 43 ------------------------------------------- 2 files changed, 55 deletions(-) (limited to 'include') diff --git a/include/net/dsa.h b/include/net/dsa.h index 3f23dd9d6a69..26a0e86e611e 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -116,13 +116,6 @@ struct dsa_switch_tree { s8 cpu_switch; s8 cpu_port; - /* - * Link state polling. - */ - int link_poll_needed; - struct work_struct link_poll_work; - struct timer_list link_poll_timer; - /* * Data for the individual switch chips. */ @@ -231,11 +224,6 @@ struct dsa_switch_driver { int (*phy_write)(struct dsa_switch *ds, int port, int regnum, u16 val); - /* - * Link state polling and IRQ handling. - */ - void (*poll_link)(struct dsa_switch *ds); - /* * Link state adjustment (called from libphy) */ diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index b7448c8490ac..0f41f71efac1 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -508,33 +508,6 @@ static int dsa_switch_resume(struct dsa_switch *ds) } #endif - -/* link polling *************************************************************/ -static void dsa_link_poll_work(struct work_struct *ugly) -{ - struct dsa_switch_tree *dst; - int i; - - dst = container_of(ugly, struct dsa_switch_tree, link_poll_work); - - for (i = 0; i < dst->pd->nr_chips; i++) { - struct dsa_switch *ds = dst->ds[i]; - - if (ds != NULL && ds->drv->poll_link != NULL) - ds->drv->poll_link(ds); - } - - mod_timer(&dst->link_poll_timer, round_jiffies(jiffies + HZ)); -} - -static void dsa_link_poll_timer(unsigned long _dst) -{ - struct dsa_switch_tree *dst = (void *)_dst; - - schedule_work(&dst->link_poll_work); -} - - /* platform driver init and cleanup *****************************************/ static int dev_is_class(struct device *dev, void *class) { @@ -877,8 +850,6 @@ static int dsa_setup_dst(struct dsa_switch_tree *dst, struct net_device *dev, } dst->ds[i] = ds; - if (ds->drv->poll_link != NULL) - dst->link_poll_needed = 1; ++configured; } @@ -897,15 +868,6 @@ static int dsa_setup_dst(struct dsa_switch_tree *dst, struct net_device *dev, wmb(); dev->dsa_ptr = (void *)dst; - if (dst->link_poll_needed) { - INIT_WORK(&dst->link_poll_work, dsa_link_poll_work); - init_timer(&dst->link_poll_timer); - dst->link_poll_timer.data = (unsigned long)dst; - dst->link_poll_timer.function = dsa_link_poll_timer; - dst->link_poll_timer.expires = round_jiffies(jiffies + HZ); - add_timer(&dst->link_poll_timer); - } - return 0; } @@ -972,11 +934,6 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst) { int i; - if (dst->link_poll_needed) - del_timer_sync(&dst->link_poll_timer); - - flush_work(&dst->link_poll_work); - for (i = 0; i < dst->pd->nr_chips; i++) { struct dsa_switch *ds = dst->ds[i]; -- cgit v1.2.3-59-g8ed1b From 8ac2837c89c8c0fcad557e4380aeef80580390f9 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Wed, 9 Dec 2015 10:51:12 +0800 Subject: Revert "Merge branch 'vsock-virtio'" This reverts commit 0d76d6e8b2507983a2cae4c09880798079007421 and merge commit c402293bd76fbc93e52ef8c0947ab81eea3ae019, reversing changes made to c89359a42e2a49656451569c382eed63e781153c. The virtio-vsock device specification is not finalized yet. Michael Tsirkin voiced concerned about merging this code when the hardware interface (and possibly the userspace interface) could still change. Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- drivers/vhost/Kconfig | 4 - drivers/vhost/Kconfig.vsock | 7 - drivers/vhost/Makefile | 4 - drivers/vhost/vsock.c | 630 --------------- drivers/vhost/vsock.h | 4 - include/linux/virtio_vsock.h | 209 ----- include/net/af_vsock.h | 2 - include/uapi/linux/virtio_ids.h | 1 - include/uapi/linux/virtio_vsock.h | 89 --- net/vmw_vsock/Kconfig | 18 - net/vmw_vsock/Makefile | 2 - net/vmw_vsock/af_vsock.c | 70 -- net/vmw_vsock/virtio_transport.c | 466 ----------- net/vmw_vsock/virtio_transport_common.c | 1272 ------------------------------- 14 files changed, 2778 deletions(-) delete mode 100644 drivers/vhost/Kconfig.vsock delete mode 100644 drivers/vhost/vsock.c delete mode 100644 drivers/vhost/vsock.h delete mode 100644 include/linux/virtio_vsock.h delete mode 100644 include/uapi/linux/virtio_vsock.h delete mode 100644 net/vmw_vsock/virtio_transport.c delete mode 100644 net/vmw_vsock/virtio_transport_common.c (limited to 'include') diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig index 81449bfc8d3b..533eaf04f12f 100644 --- a/drivers/vhost/Kconfig +++ b/drivers/vhost/Kconfig @@ -47,7 +47,3 @@ config VHOST_CROSS_ENDIAN_LEGACY adds some overhead, it is disabled by default. If unsure, say "N". - -if STAGING -source "drivers/vhost/Kconfig.vsock" -endif diff --git a/drivers/vhost/Kconfig.vsock b/drivers/vhost/Kconfig.vsock deleted file mode 100644 index 3491865d3eb9..000000000000 --- a/drivers/vhost/Kconfig.vsock +++ /dev/null @@ -1,7 +0,0 @@ -config VHOST_VSOCK - tristate "vhost virtio-vsock driver" - depends on VSOCKETS && EVENTFD - select VIRTIO_VSOCKETS_COMMON - default n - ---help--- - Say M here to enable the vhost-vsock for virtio-vsock guests diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile index 6b012b986b57..e0441c34db1c 100644 --- a/drivers/vhost/Makefile +++ b/drivers/vhost/Makefile @@ -4,9 +4,5 @@ vhost_net-y := net.o obj-$(CONFIG_VHOST_SCSI) += vhost_scsi.o vhost_scsi-y := scsi.o -obj-$(CONFIG_VHOST_VSOCK) += vhost_vsock.o -vhost_vsock-y := vsock.o - obj-$(CONFIG_VHOST_RING) += vringh.o - obj-$(CONFIG_VHOST) += vhost.o diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c deleted file mode 100644 index 64bcb10bb901..000000000000 --- a/drivers/vhost/vsock.c +++ /dev/null @@ -1,630 +0,0 @@ -/* - * vhost transport for vsock - * - * Copyright (C) 2013-2015 Red Hat, Inc. - * Author: Asias He - * Stefan Hajnoczi - * - * This work is licensed under the terms of the GNU GPL, version 2. - */ -#include -#include -#include -#include -#include -#include - -#include -#include "vhost.h" -#include "vsock.h" - -#define VHOST_VSOCK_DEFAULT_HOST_CID 2 - -static int vhost_transport_socket_init(struct vsock_sock *vsk, - struct vsock_sock *psk); - -enum { - VHOST_VSOCK_FEATURES = VHOST_FEATURES, -}; - -/* Used to track all the vhost_vsock instances on the system. */ -static LIST_HEAD(vhost_vsock_list); -static DEFINE_MUTEX(vhost_vsock_mutex); - -struct vhost_vsock_virtqueue { - struct vhost_virtqueue vq; -}; - -struct vhost_vsock { - /* Vhost device */ - struct vhost_dev dev; - /* Vhost vsock virtqueue*/ - struct vhost_vsock_virtqueue vqs[VSOCK_VQ_MAX]; - /* Link to global vhost_vsock_list*/ - struct list_head list; - /* Head for pkt from host to guest */ - struct list_head send_pkt_list; - /* Work item to send pkt */ - struct vhost_work send_pkt_work; - /* Wait queue for send pkt */ - wait_queue_head_t queue_wait; - /* Used for global tx buf limitation */ - u32 total_tx_buf; - /* Guest contex id this vhost_vsock instance handles */ - u32 guest_cid; -}; - -static u32 vhost_transport_get_local_cid(void) -{ - return VHOST_VSOCK_DEFAULT_HOST_CID; -} - -static struct vhost_vsock *vhost_vsock_get(u32 guest_cid) -{ - struct vhost_vsock *vsock; - - mutex_lock(&vhost_vsock_mutex); - list_for_each_entry(vsock, &vhost_vsock_list, list) { - if (vsock->guest_cid == guest_cid) { - mutex_unlock(&vhost_vsock_mutex); - return vsock; - } - } - mutex_unlock(&vhost_vsock_mutex); - - return NULL; -} - -static void -vhost_transport_do_send_pkt(struct vhost_vsock *vsock, - struct vhost_virtqueue *vq) -{ - bool added = false; - - mutex_lock(&vq->mutex); - vhost_disable_notify(&vsock->dev, vq); - for (;;) { - struct virtio_vsock_pkt *pkt; - struct iov_iter iov_iter; - unsigned out, in; - struct sock *sk; - size_t nbytes; - size_t len; - int head; - - if (list_empty(&vsock->send_pkt_list)) { - vhost_enable_notify(&vsock->dev, vq); - break; - } - - head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), - &out, &in, NULL, NULL); - pr_debug("%s: head = %d\n", __func__, head); - if (head < 0) - break; - - if (head == vq->num) { - if (unlikely(vhost_enable_notify(&vsock->dev, vq))) { - vhost_disable_notify(&vsock->dev, vq); - continue; - } - break; - } - - pkt = list_first_entry(&vsock->send_pkt_list, - struct virtio_vsock_pkt, list); - list_del_init(&pkt->list); - - if (out) { - virtio_transport_free_pkt(pkt); - vq_err(vq, "Expected 0 output buffers, got %u\n", out); - break; - } - - len = iov_length(&vq->iov[out], in); - iov_iter_init(&iov_iter, READ, &vq->iov[out], in, len); - - nbytes = copy_to_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter); - if (nbytes != sizeof(pkt->hdr)) { - virtio_transport_free_pkt(pkt); - vq_err(vq, "Faulted on copying pkt hdr\n"); - break; - } - - nbytes = copy_to_iter(pkt->buf, pkt->len, &iov_iter); - if (nbytes != pkt->len) { - virtio_transport_free_pkt(pkt); - vq_err(vq, "Faulted on copying pkt buf\n"); - break; - } - - vhost_add_used(vq, head, pkt->len); /* TODO should this be sizeof(pkt->hdr) + pkt->len? */ - added = true; - - virtio_transport_dec_tx_pkt(pkt); - vsock->total_tx_buf -= pkt->len; - - sk = sk_vsock(pkt->trans->vsk); - /* Release refcnt taken in vhost_transport_send_pkt */ - sock_put(sk); - - virtio_transport_free_pkt(pkt); - } - if (added) - vhost_signal(&vsock->dev, vq); - mutex_unlock(&vq->mutex); - - if (added) - wake_up(&vsock->queue_wait); -} - -static void vhost_transport_send_pkt_work(struct vhost_work *work) -{ - struct vhost_virtqueue *vq; - struct vhost_vsock *vsock; - - vsock = container_of(work, struct vhost_vsock, send_pkt_work); - vq = &vsock->vqs[VSOCK_VQ_RX].vq; - - vhost_transport_do_send_pkt(vsock, vq); -} - -static int -vhost_transport_send_pkt(struct vsock_sock *vsk, - struct virtio_vsock_pkt_info *info) -{ - u32 src_cid, src_port, dst_cid, dst_port; - struct virtio_transport *trans; - struct virtio_vsock_pkt *pkt; - struct vhost_virtqueue *vq; - struct vhost_vsock *vsock; - u32 pkt_len = info->pkt_len; - DEFINE_WAIT(wait); - - src_cid = vhost_transport_get_local_cid(); - src_port = vsk->local_addr.svm_port; - if (!info->remote_cid) { - dst_cid = vsk->remote_addr.svm_cid; - dst_port = vsk->remote_addr.svm_port; - } else { - dst_cid = info->remote_cid; - dst_port = info->remote_port; - } - - /* Find the vhost_vsock according to guest context id */ - vsock = vhost_vsock_get(dst_cid); - if (!vsock) - return -ENODEV; - - trans = vsk->trans; - vq = &vsock->vqs[VSOCK_VQ_RX].vq; - - /* we can send less than pkt_len bytes */ - if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) - pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; - - /* virtio_transport_get_credit might return less than pkt_len credit */ - pkt_len = virtio_transport_get_credit(trans, pkt_len); - - /* Do not send zero length OP_RW pkt*/ - if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW) - return pkt_len; - - /* Respect global tx buf limitation */ - mutex_lock(&vq->mutex); - while (pkt_len + vsock->total_tx_buf > VIRTIO_VSOCK_MAX_TX_BUF_SIZE) { - prepare_to_wait_exclusive(&vsock->queue_wait, &wait, - TASK_UNINTERRUPTIBLE); - mutex_unlock(&vq->mutex); - schedule(); - mutex_lock(&vq->mutex); - finish_wait(&vsock->queue_wait, &wait); - } - vsock->total_tx_buf += pkt_len; - mutex_unlock(&vq->mutex); - - pkt = virtio_transport_alloc_pkt(vsk, info, pkt_len, - src_cid, src_port, - dst_cid, dst_port); - if (!pkt) { - mutex_lock(&vq->mutex); - vsock->total_tx_buf -= pkt_len; - mutex_unlock(&vq->mutex); - virtio_transport_put_credit(trans, pkt_len); - return -ENOMEM; - } - - pr_debug("%s:info->pkt_len= %d\n", __func__, pkt_len); - /* Released in vhost_transport_do_send_pkt */ - sock_hold(&trans->vsk->sk); - virtio_transport_inc_tx_pkt(pkt); - - /* Queue it up in vhost work */ - mutex_lock(&vq->mutex); - list_add_tail(&pkt->list, &vsock->send_pkt_list); - vhost_work_queue(&vsock->dev, &vsock->send_pkt_work); - mutex_unlock(&vq->mutex); - - return pkt_len; -} - -static struct virtio_transport_pkt_ops vhost_ops = { - .send_pkt = vhost_transport_send_pkt, -}; - -static struct virtio_vsock_pkt * -vhost_vsock_alloc_pkt(struct vhost_virtqueue *vq, - unsigned int out, unsigned int in) -{ - struct virtio_vsock_pkt *pkt; - struct iov_iter iov_iter; - size_t nbytes; - size_t len; - - if (in != 0) { - vq_err(vq, "Expected 0 input buffers, got %u\n", in); - return NULL; - } - - pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); - if (!pkt) - return NULL; - - len = iov_length(vq->iov, out); - iov_iter_init(&iov_iter, WRITE, vq->iov, out, len); - - nbytes = copy_from_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter); - if (nbytes != sizeof(pkt->hdr)) { - vq_err(vq, "Expected %zu bytes for pkt->hdr, got %zu bytes\n", - sizeof(pkt->hdr), nbytes); - kfree(pkt); - return NULL; - } - - if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_DGRAM) - pkt->len = le32_to_cpu(pkt->hdr.len) & 0XFFFF; - else if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM) - pkt->len = le32_to_cpu(pkt->hdr.len); - - /* No payload */ - if (!pkt->len) - return pkt; - - /* The pkt is too big */ - if (pkt->len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) { - kfree(pkt); - return NULL; - } - - pkt->buf = kmalloc(pkt->len, GFP_KERNEL); - if (!pkt->buf) { - kfree(pkt); - return NULL; - } - - nbytes = copy_from_iter(pkt->buf, pkt->len, &iov_iter); - if (nbytes != pkt->len) { - vq_err(vq, "Expected %u byte payload, got %zu bytes\n", - pkt->len, nbytes); - virtio_transport_free_pkt(pkt); - return NULL; - } - - return pkt; -} - -static void vhost_vsock_handle_ctl_kick(struct vhost_work *work) -{ - struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue, - poll.work); - struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock, - dev); - - pr_debug("%s vq=%p, vsock=%p\n", __func__, vq, vsock); -} - -static void vhost_vsock_handle_tx_kick(struct vhost_work *work) -{ - struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue, - poll.work); - struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock, - dev); - struct virtio_vsock_pkt *pkt; - int head; - unsigned int out, in; - bool added = false; - u32 len; - - mutex_lock(&vq->mutex); - vhost_disable_notify(&vsock->dev, vq); - for (;;) { - head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), - &out, &in, NULL, NULL); - if (head < 0) - break; - - if (head == vq->num) { - if (unlikely(vhost_enable_notify(&vsock->dev, vq))) { - vhost_disable_notify(&vsock->dev, vq); - continue; - } - break; - } - - pkt = vhost_vsock_alloc_pkt(vq, out, in); - if (!pkt) { - vq_err(vq, "Faulted on pkt\n"); - continue; - } - - len = pkt->len; - - /* Only accept correctly addressed packets */ - if (le32_to_cpu(pkt->hdr.src_cid) == vsock->guest_cid && - le32_to_cpu(pkt->hdr.dst_cid) == vhost_transport_get_local_cid()) - virtio_transport_recv_pkt(pkt); - else - virtio_transport_free_pkt(pkt); - - vhost_add_used(vq, head, len); - added = true; - } - if (added) - vhost_signal(&vsock->dev, vq); - mutex_unlock(&vq->mutex); -} - -static void vhost_vsock_handle_rx_kick(struct vhost_work *work) -{ - struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue, - poll.work); - struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock, - dev); - - vhost_transport_do_send_pkt(vsock, vq); -} - -static int vhost_vsock_dev_open(struct inode *inode, struct file *file) -{ - struct vhost_virtqueue **vqs; - struct vhost_vsock *vsock; - int ret; - - vsock = kzalloc(sizeof(*vsock), GFP_KERNEL); - if (!vsock) - return -ENOMEM; - - pr_debug("%s:vsock=%p\n", __func__, vsock); - - vqs = kmalloc(VSOCK_VQ_MAX * sizeof(*vqs), GFP_KERNEL); - if (!vqs) { - ret = -ENOMEM; - goto out; - } - - vqs[VSOCK_VQ_CTRL] = &vsock->vqs[VSOCK_VQ_CTRL].vq; - vqs[VSOCK_VQ_TX] = &vsock->vqs[VSOCK_VQ_TX].vq; - vqs[VSOCK_VQ_RX] = &vsock->vqs[VSOCK_VQ_RX].vq; - vsock->vqs[VSOCK_VQ_CTRL].vq.handle_kick = vhost_vsock_handle_ctl_kick; - vsock->vqs[VSOCK_VQ_TX].vq.handle_kick = vhost_vsock_handle_tx_kick; - vsock->vqs[VSOCK_VQ_RX].vq.handle_kick = vhost_vsock_handle_rx_kick; - - vhost_dev_init(&vsock->dev, vqs, VSOCK_VQ_MAX); - - file->private_data = vsock; - init_waitqueue_head(&vsock->queue_wait); - INIT_LIST_HEAD(&vsock->send_pkt_list); - vhost_work_init(&vsock->send_pkt_work, vhost_transport_send_pkt_work); - - mutex_lock(&vhost_vsock_mutex); - list_add_tail(&vsock->list, &vhost_vsock_list); - mutex_unlock(&vhost_vsock_mutex); - return 0; - -out: - kfree(vsock); - return ret; -} - -static void vhost_vsock_flush(struct vhost_vsock *vsock) -{ - int i; - - for (i = 0; i < VSOCK_VQ_MAX; i++) - vhost_poll_flush(&vsock->vqs[i].vq.poll); - vhost_work_flush(&vsock->dev, &vsock->send_pkt_work); -} - -static int vhost_vsock_dev_release(struct inode *inode, struct file *file) -{ - struct vhost_vsock *vsock = file->private_data; - - mutex_lock(&vhost_vsock_mutex); - list_del(&vsock->list); - mutex_unlock(&vhost_vsock_mutex); - - vhost_dev_stop(&vsock->dev); - vhost_vsock_flush(vsock); - vhost_dev_cleanup(&vsock->dev, false); - kfree(vsock->dev.vqs); - kfree(vsock); - return 0; -} - -static int vhost_vsock_set_cid(struct vhost_vsock *vsock, u32 guest_cid) -{ - struct vhost_vsock *other; - - /* Refuse reserved CIDs */ - if (guest_cid <= VMADDR_CID_HOST) { - return -EINVAL; - } - - /* Refuse if CID is already in use */ - other = vhost_vsock_get(guest_cid); - if (other && other != vsock) { - return -EADDRINUSE; - } - - mutex_lock(&vhost_vsock_mutex); - vsock->guest_cid = guest_cid; - pr_debug("%s:guest_cid=%d\n", __func__, guest_cid); - mutex_unlock(&vhost_vsock_mutex); - - return 0; -} - -static int vhost_vsock_set_features(struct vhost_vsock *vsock, u64 features) -{ - struct vhost_virtqueue *vq; - int i; - - if (features & ~VHOST_VSOCK_FEATURES) - return -EOPNOTSUPP; - - mutex_lock(&vsock->dev.mutex); - if ((features & (1 << VHOST_F_LOG_ALL)) && - !vhost_log_access_ok(&vsock->dev)) { - mutex_unlock(&vsock->dev.mutex); - return -EFAULT; - } - - for (i = 0; i < VSOCK_VQ_MAX; i++) { - vq = &vsock->vqs[i].vq; - mutex_lock(&vq->mutex); - vq->acked_features = features; - mutex_unlock(&vq->mutex); - } - mutex_unlock(&vsock->dev.mutex); - return 0; -} - -static long vhost_vsock_dev_ioctl(struct file *f, unsigned int ioctl, - unsigned long arg) -{ - struct vhost_vsock *vsock = f->private_data; - void __user *argp = (void __user *)arg; - u64 __user *featurep = argp; - u32 __user *cidp = argp; - u32 guest_cid; - u64 features; - int r; - - switch (ioctl) { - case VHOST_VSOCK_SET_GUEST_CID: - if (get_user(guest_cid, cidp)) - return -EFAULT; - return vhost_vsock_set_cid(vsock, guest_cid); - case VHOST_GET_FEATURES: - features = VHOST_VSOCK_FEATURES; - if (copy_to_user(featurep, &features, sizeof(features))) - return -EFAULT; - return 0; - case VHOST_SET_FEATURES: - if (copy_from_user(&features, featurep, sizeof(features))) - return -EFAULT; - return vhost_vsock_set_features(vsock, features); - default: - mutex_lock(&vsock->dev.mutex); - r = vhost_dev_ioctl(&vsock->dev, ioctl, argp); - if (r == -ENOIOCTLCMD) - r = vhost_vring_ioctl(&vsock->dev, ioctl, argp); - else - vhost_vsock_flush(vsock); - mutex_unlock(&vsock->dev.mutex); - return r; - } -} - -static const struct file_operations vhost_vsock_fops = { - .owner = THIS_MODULE, - .open = vhost_vsock_dev_open, - .release = vhost_vsock_dev_release, - .llseek = noop_llseek, - .unlocked_ioctl = vhost_vsock_dev_ioctl, -}; - -static struct miscdevice vhost_vsock_misc = { - .minor = MISC_DYNAMIC_MINOR, - .name = "vhost-vsock", - .fops = &vhost_vsock_fops, -}; - -static int -vhost_transport_socket_init(struct vsock_sock *vsk, struct vsock_sock *psk) -{ - struct virtio_transport *trans; - int ret; - - ret = virtio_transport_do_socket_init(vsk, psk); - if (ret) - return ret; - - trans = vsk->trans; - trans->ops = &vhost_ops; - - return ret; -} - -static struct vsock_transport vhost_transport = { - .get_local_cid = vhost_transport_get_local_cid, - - .init = vhost_transport_socket_init, - .destruct = virtio_transport_destruct, - .release = virtio_transport_release, - .connect = virtio_transport_connect, - .shutdown = virtio_transport_shutdown, - - .dgram_enqueue = virtio_transport_dgram_enqueue, - .dgram_dequeue = virtio_transport_dgram_dequeue, - .dgram_bind = virtio_transport_dgram_bind, - .dgram_allow = virtio_transport_dgram_allow, - - .stream_enqueue = virtio_transport_stream_enqueue, - .stream_dequeue = virtio_transport_stream_dequeue, - .stream_has_data = virtio_transport_stream_has_data, - .stream_has_space = virtio_transport_stream_has_space, - .stream_rcvhiwat = virtio_transport_stream_rcvhiwat, - .stream_is_active = virtio_transport_stream_is_active, - .stream_allow = virtio_transport_stream_allow, - - .notify_poll_in = virtio_transport_notify_poll_in, - .notify_poll_out = virtio_transport_notify_poll_out, - .notify_recv_init = virtio_transport_notify_recv_init, - .notify_recv_pre_block = virtio_transport_notify_recv_pre_block, - .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue, - .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue, - .notify_send_init = virtio_transport_notify_send_init, - .notify_send_pre_block = virtio_transport_notify_send_pre_block, - .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, - .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, - - .set_buffer_size = virtio_transport_set_buffer_size, - .set_min_buffer_size = virtio_transport_set_min_buffer_size, - .set_max_buffer_size = virtio_transport_set_max_buffer_size, - .get_buffer_size = virtio_transport_get_buffer_size, - .get_min_buffer_size = virtio_transport_get_min_buffer_size, - .get_max_buffer_size = virtio_transport_get_max_buffer_size, -}; - -static int __init vhost_vsock_init(void) -{ - int ret; - - ret = vsock_core_init(&vhost_transport); - if (ret < 0) - return ret; - return misc_register(&vhost_vsock_misc); -}; - -static void __exit vhost_vsock_exit(void) -{ - misc_deregister(&vhost_vsock_misc); - vsock_core_exit(); -}; - -module_init(vhost_vsock_init); -module_exit(vhost_vsock_exit); -MODULE_LICENSE("GPL v2"); -MODULE_AUTHOR("Asias He"); -MODULE_DESCRIPTION("vhost transport for vsock "); diff --git a/drivers/vhost/vsock.h b/drivers/vhost/vsock.h deleted file mode 100644 index 0ddb107b86ca..000000000000 --- a/drivers/vhost/vsock.h +++ /dev/null @@ -1,4 +0,0 @@ -#ifndef VHOST_VSOCK_H -#define VHOST_VSOCK_H -#define VHOST_VSOCK_SET_GUEST_CID _IOW(VHOST_VIRTIO, 0x60, __u32) -#endif diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h deleted file mode 100644 index a5f3ecc038f7..000000000000 --- a/include/linux/virtio_vsock.h +++ /dev/null @@ -1,209 +0,0 @@ -/* - * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so - * anyone can use the definitions to implement compatible drivers/servers: - * - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of IBM nor the names of its contributors - * may be used to endorse or promote products derived from this software - * without specific prior written permission. - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - * - * Copyright (C) Red Hat, Inc., 2013-2015 - * Copyright (C) Asias He , 2013 - * Copyright (C) Stefan Hajnoczi , 2015 - */ - -#ifndef _LINUX_VIRTIO_VSOCK_H -#define _LINUX_VIRTIO_VSOCK_H - -#include -#include -#include - -#define VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE 128 -#define VIRTIO_VSOCK_DEFAULT_BUF_SIZE (1024 * 256) -#define VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE (1024 * 256) -#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE (1024 * 4) -#define VIRTIO_VSOCK_MAX_BUF_SIZE 0xFFFFFFFFUL -#define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE (1024 * 64) -#define VIRTIO_VSOCK_MAX_TX_BUF_SIZE (1024 * 1024 * 16) -#define VIRTIO_VSOCK_MAX_DGRAM_SIZE (1024 * 64) - -struct vsock_transport_recv_notify_data; -struct vsock_transport_send_notify_data; -struct sockaddr_vm; -struct vsock_sock; - -enum { - VSOCK_VQ_CTRL = 0, - VSOCK_VQ_RX = 1, /* for host to guest data */ - VSOCK_VQ_TX = 2, /* for guest to host data */ - VSOCK_VQ_MAX = 3, -}; - -/* virtio transport socket state */ -struct virtio_transport { - struct virtio_transport_pkt_ops *ops; - struct vsock_sock *vsk; - - u32 buf_size; - u32 buf_size_min; - u32 buf_size_max; - - struct mutex tx_lock; - struct mutex rx_lock; - - struct list_head rx_queue; - u32 rx_bytes; - - /* Protected by trans->tx_lock */ - u32 tx_cnt; - u32 buf_alloc; - u32 peer_fwd_cnt; - u32 peer_buf_alloc; - /* Protected by trans->rx_lock */ - u32 fwd_cnt; - - /* Protected by sk_lock */ - u16 dgram_id; - struct list_head incomplete_dgrams; /* dgram fragments */ -}; - -struct virtio_vsock_pkt { - struct virtio_vsock_hdr hdr; - struct virtio_transport *trans; - struct work_struct work; - struct list_head list; - void *buf; - u32 len; - u32 off; -}; - -struct virtio_vsock_pkt_info { - u32 remote_cid, remote_port; - struct msghdr *msg; - u32 pkt_len; - u16 type; - u16 op; - u32 flags; - u16 dgram_id; - u16 dgram_len; -}; - -struct virtio_transport_pkt_ops { - int (*send_pkt)(struct vsock_sock *vsk, - struct virtio_vsock_pkt_info *info); -}; - -void virtio_vsock_dumppkt(const char *func, - const struct virtio_vsock_pkt *pkt); - -struct sock * -virtio_transport_get_pending(struct sock *listener, - struct virtio_vsock_pkt *pkt); -struct virtio_vsock_pkt * -virtio_transport_alloc_pkt(struct vsock_sock *vsk, - struct virtio_vsock_pkt_info *info, - size_t len, - u32 src_cid, - u32 src_port, - u32 dst_cid, - u32 dst_port); -ssize_t -virtio_transport_stream_dequeue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len, - int type); -int -virtio_transport_dgram_dequeue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len, int flags); - -s64 virtio_transport_stream_has_data(struct vsock_sock *vsk); -s64 virtio_transport_stream_has_space(struct vsock_sock *vsk); - -int virtio_transport_do_socket_init(struct vsock_sock *vsk, - struct vsock_sock *psk); -u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk); -u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk); -u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk); -void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val); -void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val); -void virtio_transport_set_max_buffer_size(struct vsock_sock *vs, u64 val); -int -virtio_transport_notify_poll_in(struct vsock_sock *vsk, - size_t target, - bool *data_ready_now); -int -virtio_transport_notify_poll_out(struct vsock_sock *vsk, - size_t target, - bool *space_available_now); - -int virtio_transport_notify_recv_init(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data); -int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data); -int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data); -int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, - size_t target, ssize_t copied, bool data_read, - struct vsock_transport_recv_notify_data *data); -int virtio_transport_notify_send_init(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data); -int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data); -int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data); -int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, - ssize_t written, struct vsock_transport_send_notify_data *data); - -u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk); -bool virtio_transport_stream_is_active(struct vsock_sock *vsk); -bool virtio_transport_stream_allow(u32 cid, u32 port); -int virtio_transport_dgram_bind(struct vsock_sock *vsk, - struct sockaddr_vm *addr); -bool virtio_transport_dgram_allow(u32 cid, u32 port); - -int virtio_transport_connect(struct vsock_sock *vsk); - -int virtio_transport_shutdown(struct vsock_sock *vsk, int mode); - -void virtio_transport_release(struct vsock_sock *vsk); - -ssize_t -virtio_transport_stream_enqueue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len); -int -virtio_transport_dgram_enqueue(struct vsock_sock *vsk, - struct sockaddr_vm *remote_addr, - struct msghdr *msg, - size_t len); - -void virtio_transport_destruct(struct vsock_sock *vsk); - -void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt); -void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt); -void virtio_transport_inc_tx_pkt(struct virtio_vsock_pkt *pkt); -void virtio_transport_dec_tx_pkt(struct virtio_vsock_pkt *pkt); -u32 virtio_transport_get_credit(struct virtio_transport *trans, u32 wanted); -void virtio_transport_put_credit(struct virtio_transport *trans, u32 credit); -#endif /* _LINUX_VIRTIO_VSOCK_H */ diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index a0c8fa2ababf..e9eb2d6791b3 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -175,10 +175,8 @@ void vsock_insert_connected(struct vsock_sock *vsk); void vsock_remove_bound(struct vsock_sock *vsk); void vsock_remove_connected(struct vsock_sock *vsk); struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr); -struct sock *vsock_find_unbound_socket(struct sockaddr_vm *addr); struct sock *vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst); void vsock_for_each_connected_socket(void (*fn)(struct sock *sk)); -int vsock_bind_dgram_generic(struct vsock_sock *vsk, struct sockaddr_vm *addr); #endif /* __AF_VSOCK_H__ */ diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h index 16dcf5d06cd7..77925f587b15 100644 --- a/include/uapi/linux/virtio_ids.h +++ b/include/uapi/linux/virtio_ids.h @@ -39,7 +39,6 @@ #define VIRTIO_ID_9P 9 /* 9p virtio console */ #define VIRTIO_ID_RPROC_SERIAL 11 /* virtio remoteproc serial link */ #define VIRTIO_ID_CAIF 12 /* Virtio caif */ -#define VIRTIO_ID_VSOCK 13 /* virtio vsock transport */ #define VIRTIO_ID_GPU 16 /* virtio GPU */ #define VIRTIO_ID_INPUT 18 /* virtio input */ diff --git a/include/uapi/linux/virtio_vsock.h b/include/uapi/linux/virtio_vsock.h deleted file mode 100644 index 8cf9b5682628..000000000000 --- a/include/uapi/linux/virtio_vsock.h +++ /dev/null @@ -1,89 +0,0 @@ -/* - * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so - * anyone can use the definitions to implement compatible drivers/servers: - * - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of IBM nor the names of its contributors - * may be used to endorse or promote products derived from this software - * without specific prior written permission. - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - * - * Copyright (C) Red Hat, Inc., 2013-2015 - * Copyright (C) Asias He , 2013 - * Copyright (C) Stefan Hajnoczi , 2015 - */ - -#ifndef _UAPI_LINUX_VIRTIO_VSOCK_H -#define _UAPI_LINUX_VIRTIO_VOSCK_H - -#include -#include -#include - -struct virtio_vsock_config { - __le32 guest_cid; - __le32 max_virtqueue_pairs; -}; - -struct virtio_vsock_hdr { - __le32 src_cid; - __le32 src_port; - __le32 dst_cid; - __le32 dst_port; - __le32 len; - __le16 type; /* enum virtio_vsock_type */ - __le16 op; /* enum virtio_vsock_op */ - __le32 flags; - __le32 buf_alloc; - __le32 fwd_cnt; -}; - -enum virtio_vsock_type { - VIRTIO_VSOCK_TYPE_STREAM = 1, - VIRTIO_VSOCK_TYPE_DGRAM = 2, -}; - -enum virtio_vsock_op { - VIRTIO_VSOCK_OP_INVALID = 0, - - /* Connect operations */ - VIRTIO_VSOCK_OP_REQUEST = 1, - VIRTIO_VSOCK_OP_RESPONSE = 2, - VIRTIO_VSOCK_OP_ACK = 3, - VIRTIO_VSOCK_OP_RST = 4, - VIRTIO_VSOCK_OP_SHUTDOWN = 5, - - /* To send payload */ - VIRTIO_VSOCK_OP_RW = 6, - - /* Tell the peer our credit info */ - VIRTIO_VSOCK_OP_CREDIT_UPDATE = 7, - /* Request the peer to send the credit info to us */ - VIRTIO_VSOCK_OP_CREDIT_REQUEST = 8, -}; - -/* VIRTIO_VSOCK_OP_SHUTDOWN flags values */ -enum virtio_vsock_shutdown { - VIRTIO_VSOCK_SHUTDOWN_RCV = 1, - VIRTIO_VSOCK_SHUTDOWN_SEND = 2, -}; - -#endif /* _UAPI_LINUX_VIRTIO_VSOCK_H */ diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig index 74e0bc887a33..14810abedc2e 100644 --- a/net/vmw_vsock/Kconfig +++ b/net/vmw_vsock/Kconfig @@ -26,21 +26,3 @@ config VMWARE_VMCI_VSOCKETS To compile this driver as a module, choose M here: the module will be called vmw_vsock_vmci_transport. If unsure, say N. - -config VIRTIO_VSOCKETS - tristate "virtio transport for Virtual Sockets" - depends on VSOCKETS && VIRTIO - select VIRTIO_VSOCKETS_COMMON - help - This module implements a virtio transport for Virtual Sockets. - - Enable this transport if your Virtual Machine runs on Qemu/KVM. - - To compile this driver as a module, choose M here: the module - will be called virtio_vsock_transport. If unsure, say N. - -config VIRTIO_VSOCKETS_COMMON - tristate - ---help--- - This option is selected by any driver which needs to access - the virtio_vsock. diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile index cf4c29439081..2ce52d70f224 100644 --- a/net/vmw_vsock/Makefile +++ b/net/vmw_vsock/Makefile @@ -1,7 +1,5 @@ obj-$(CONFIG_VSOCKETS) += vsock.o obj-$(CONFIG_VMWARE_VMCI_VSOCKETS) += vmw_vsock_vmci_transport.o -obj-$(CONFIG_VIRTIO_VSOCKETS) += virtio_transport.o -obj-$(CONFIG_VIRTIO_VSOCKETS_COMMON) += virtio_transport_common.o vsock-y += af_vsock.o vsock_addr.o diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 77247a2b670b..7fd1220fbfa0 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -223,17 +223,6 @@ static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) return NULL; } -static struct sock *__vsock_find_unbound_socket(struct sockaddr_vm *addr) -{ - struct vsock_sock *vsk; - - list_for_each_entry(vsk, vsock_unbound_sockets, bound_table) - if (addr->svm_port == vsk->local_addr.svm_port) - return sk_vsock(vsk); - - return NULL; -} - static struct sock *__vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -309,21 +298,6 @@ struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr) } EXPORT_SYMBOL_GPL(vsock_find_bound_socket); -struct sock *vsock_find_unbound_socket(struct sockaddr_vm *addr) -{ - struct sock *sk; - - spin_lock_bh(&vsock_table_lock); - sk = __vsock_find_unbound_socket(addr); - if (sk) - sock_hold(sk); - - spin_unlock_bh(&vsock_table_lock); - - return sk; -} -EXPORT_SYMBOL_GPL(vsock_find_unbound_socket); - struct sock *vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -558,50 +532,6 @@ static int __vsock_bind_stream(struct vsock_sock *vsk, return 0; } -int vsock_bind_dgram_generic(struct vsock_sock *vsk, struct sockaddr_vm *addr) -{ - static u32 port = LAST_RESERVED_PORT + 1; - struct sockaddr_vm new_addr; - - vsock_addr_init(&new_addr, addr->svm_cid, addr->svm_port); - - if (addr->svm_port == VMADDR_PORT_ANY) { - bool found = false; - unsigned int i; - - for (i = 0; i < MAX_PORT_RETRIES; i++) { - if (port <= LAST_RESERVED_PORT) - port = LAST_RESERVED_PORT + 1; - - new_addr.svm_port = port++; - - if (!__vsock_find_unbound_socket(&new_addr)) { - found = true; - break; - } - } - - if (!found) - return -EADDRNOTAVAIL; - } else { - /* If port is in reserved range, ensure caller - * has necessary privileges. - */ - if (addr->svm_port <= LAST_RESERVED_PORT && - !capable(CAP_NET_BIND_SERVICE)) { - return -EACCES; - } - - if (__vsock_find_unbound_socket(&new_addr)) - return -EADDRINUSE; - } - - vsock_addr_init(&vsk->local_addr, new_addr.svm_cid, new_addr.svm_port); - - return 0; -} -EXPORT_SYMBOL_GPL(vsock_bind_dgram_generic); - static int __vsock_bind_dgram(struct vsock_sock *vsk, struct sockaddr_vm *addr) { diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c deleted file mode 100644 index df65dca55fa1..000000000000 --- a/net/vmw_vsock/virtio_transport.c +++ /dev/null @@ -1,466 +0,0 @@ -/* - * virtio transport for vsock - * - * Copyright (C) 2013-2015 Red Hat, Inc. - * Author: Asias He - * Stefan Hajnoczi - * - * Some of the code is take from Gerd Hoffmann 's - * early virtio-vsock proof-of-concept bits. - * - * This work is licensed under the terms of the GNU GPL, version 2. - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -static struct workqueue_struct *virtio_vsock_workqueue; -static struct virtio_vsock *the_virtio_vsock; -static DEFINE_MUTEX(the_virtio_vsock_mutex); /* protects the_virtio_vsock */ -static void virtio_vsock_rx_fill(struct virtio_vsock *vsock); - -struct virtio_vsock { - /* Virtio device */ - struct virtio_device *vdev; - /* Virtio virtqueue */ - struct virtqueue *vqs[VSOCK_VQ_MAX]; - /* Wait queue for send pkt */ - wait_queue_head_t queue_wait; - /* Work item to send pkt */ - struct work_struct tx_work; - /* Work item to recv pkt */ - struct work_struct rx_work; - /* Mutex to protect send pkt*/ - struct mutex tx_lock; - /* Mutex to protect recv pkt*/ - struct mutex rx_lock; - /* Number of recv buffers */ - int rx_buf_nr; - /* Number of max recv buffers */ - int rx_buf_max_nr; - /* Used for global tx buf limitation */ - u32 total_tx_buf; - /* Guest context id, just like guest ip address */ - u32 guest_cid; -}; - -static struct virtio_vsock *virtio_vsock_get(void) -{ - return the_virtio_vsock; -} - -static u32 virtio_transport_get_local_cid(void) -{ - struct virtio_vsock *vsock = virtio_vsock_get(); - - return vsock->guest_cid; -} - -static int -virtio_transport_send_pkt(struct vsock_sock *vsk, - struct virtio_vsock_pkt_info *info) -{ - u32 src_cid, src_port, dst_cid, dst_port; - int ret, in_sg = 0, out_sg = 0; - struct virtio_transport *trans; - struct virtio_vsock_pkt *pkt; - struct virtio_vsock *vsock; - struct scatterlist hdr, buf, *sgs[2]; - struct virtqueue *vq; - u32 pkt_len = info->pkt_len; - DEFINE_WAIT(wait); - - vsock = virtio_vsock_get(); - if (!vsock) - return -ENODEV; - - src_cid = virtio_transport_get_local_cid(); - src_port = vsk->local_addr.svm_port; - if (!info->remote_cid) { - dst_cid = vsk->remote_addr.svm_cid; - dst_port = vsk->remote_addr.svm_port; - } else { - dst_cid = info->remote_cid; - dst_port = info->remote_port; - } - - trans = vsk->trans; - vq = vsock->vqs[VSOCK_VQ_TX]; - - if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) - pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; - pkt_len = virtio_transport_get_credit(trans, pkt_len); - /* Do not send zero length OP_RW pkt*/ - if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW) - return pkt_len; - - /* Respect global tx buf limitation */ - mutex_lock(&vsock->tx_lock); - while (pkt_len + vsock->total_tx_buf > VIRTIO_VSOCK_MAX_TX_BUF_SIZE) { - prepare_to_wait_exclusive(&vsock->queue_wait, &wait, - TASK_UNINTERRUPTIBLE); - mutex_unlock(&vsock->tx_lock); - schedule(); - mutex_lock(&vsock->tx_lock); - finish_wait(&vsock->queue_wait, &wait); - } - vsock->total_tx_buf += pkt_len; - mutex_unlock(&vsock->tx_lock); - - pkt = virtio_transport_alloc_pkt(vsk, info, pkt_len, - src_cid, src_port, - dst_cid, dst_port); - if (!pkt) { - mutex_lock(&vsock->tx_lock); - vsock->total_tx_buf -= pkt_len; - mutex_unlock(&vsock->tx_lock); - virtio_transport_put_credit(trans, pkt_len); - return -ENOMEM; - } - - pr_debug("%s:info->pkt_len= %d\n", __func__, info->pkt_len); - - /* Will be released in virtio_transport_send_pkt_work */ - sock_hold(&trans->vsk->sk); - virtio_transport_inc_tx_pkt(pkt); - - /* Put pkt in the virtqueue */ - sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr)); - sgs[out_sg++] = &hdr; - if (info->msg && info->pkt_len > 0) { - sg_init_one(&buf, pkt->buf, pkt->len); - sgs[out_sg++] = &buf; - } - - mutex_lock(&vsock->tx_lock); - while ((ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, pkt, - GFP_KERNEL)) < 0) { - prepare_to_wait_exclusive(&vsock->queue_wait, &wait, - TASK_UNINTERRUPTIBLE); - mutex_unlock(&vsock->tx_lock); - schedule(); - mutex_lock(&vsock->tx_lock); - finish_wait(&vsock->queue_wait, &wait); - } - virtqueue_kick(vq); - mutex_unlock(&vsock->tx_lock); - - return pkt_len; -} - -static struct virtio_transport_pkt_ops virtio_ops = { - .send_pkt = virtio_transport_send_pkt, -}; - -static void virtio_vsock_rx_fill(struct virtio_vsock *vsock) -{ - int buf_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; - struct virtio_vsock_pkt *pkt; - struct scatterlist hdr, buf, *sgs[2]; - struct virtqueue *vq; - int ret; - - vq = vsock->vqs[VSOCK_VQ_RX]; - - do { - pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); - if (!pkt) { - pr_debug("%s: fail to allocate pkt\n", __func__); - goto out; - } - - /* TODO: use mergeable rx buffer */ - pkt->buf = kmalloc(buf_len, GFP_KERNEL); - if (!pkt->buf) { - pr_debug("%s: fail to allocate pkt->buf\n", __func__); - goto err; - } - - sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr)); - sgs[0] = &hdr; - - sg_init_one(&buf, pkt->buf, buf_len); - sgs[1] = &buf; - ret = virtqueue_add_sgs(vq, sgs, 0, 2, pkt, GFP_KERNEL); - if (ret) - goto err; - vsock->rx_buf_nr++; - } while (vq->num_free); - if (vsock->rx_buf_nr > vsock->rx_buf_max_nr) - vsock->rx_buf_max_nr = vsock->rx_buf_nr; -out: - virtqueue_kick(vq); - return; -err: - virtqueue_kick(vq); - virtio_transport_free_pkt(pkt); - return; -} - -static void virtio_transport_send_pkt_work(struct work_struct *work) -{ - struct virtio_vsock *vsock = - container_of(work, struct virtio_vsock, tx_work); - struct virtio_vsock_pkt *pkt; - bool added = false; - struct virtqueue *vq; - unsigned int len; - struct sock *sk; - - vq = vsock->vqs[VSOCK_VQ_TX]; - mutex_lock(&vsock->tx_lock); - do { - virtqueue_disable_cb(vq); - while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) { - sk = &pkt->trans->vsk->sk; - virtio_transport_dec_tx_pkt(pkt); - /* Release refcnt taken in virtio_transport_send_pkt */ - sock_put(sk); - vsock->total_tx_buf -= pkt->len; - virtio_transport_free_pkt(pkt); - added = true; - } - } while (!virtqueue_enable_cb(vq)); - mutex_unlock(&vsock->tx_lock); - - if (added) - wake_up(&vsock->queue_wait); -} - -static void virtio_transport_recv_pkt_work(struct work_struct *work) -{ - struct virtio_vsock *vsock = - container_of(work, struct virtio_vsock, rx_work); - struct virtio_vsock_pkt *pkt; - struct virtqueue *vq; - unsigned int len; - - vq = vsock->vqs[VSOCK_VQ_RX]; - mutex_lock(&vsock->rx_lock); - do { - virtqueue_disable_cb(vq); - while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) { - pkt->len = len; - virtio_transport_recv_pkt(pkt); - vsock->rx_buf_nr--; - } - } while (!virtqueue_enable_cb(vq)); - - if (vsock->rx_buf_nr < vsock->rx_buf_max_nr / 2) - virtio_vsock_rx_fill(vsock); - mutex_unlock(&vsock->rx_lock); -} - -static void virtio_vsock_ctrl_done(struct virtqueue *vq) -{ -} - -static void virtio_vsock_tx_done(struct virtqueue *vq) -{ - struct virtio_vsock *vsock = vq->vdev->priv; - - if (!vsock) - return; - queue_work(virtio_vsock_workqueue, &vsock->tx_work); -} - -static void virtio_vsock_rx_done(struct virtqueue *vq) -{ - struct virtio_vsock *vsock = vq->vdev->priv; - - if (!vsock) - return; - queue_work(virtio_vsock_workqueue, &vsock->rx_work); -} - -static int -virtio_transport_socket_init(struct vsock_sock *vsk, struct vsock_sock *psk) -{ - struct virtio_transport *trans; - int ret; - - ret = virtio_transport_do_socket_init(vsk, psk); - if (ret) - return ret; - - trans = vsk->trans; - trans->ops = &virtio_ops; - return ret; -} - -static struct vsock_transport virtio_transport = { - .get_local_cid = virtio_transport_get_local_cid, - - .init = virtio_transport_socket_init, - .destruct = virtio_transport_destruct, - .release = virtio_transport_release, - .connect = virtio_transport_connect, - .shutdown = virtio_transport_shutdown, - - .dgram_bind = virtio_transport_dgram_bind, - .dgram_dequeue = virtio_transport_dgram_dequeue, - .dgram_enqueue = virtio_transport_dgram_enqueue, - .dgram_allow = virtio_transport_dgram_allow, - - .stream_dequeue = virtio_transport_stream_dequeue, - .stream_enqueue = virtio_transport_stream_enqueue, - .stream_has_data = virtio_transport_stream_has_data, - .stream_has_space = virtio_transport_stream_has_space, - .stream_rcvhiwat = virtio_transport_stream_rcvhiwat, - .stream_is_active = virtio_transport_stream_is_active, - .stream_allow = virtio_transport_stream_allow, - - .notify_poll_in = virtio_transport_notify_poll_in, - .notify_poll_out = virtio_transport_notify_poll_out, - .notify_recv_init = virtio_transport_notify_recv_init, - .notify_recv_pre_block = virtio_transport_notify_recv_pre_block, - .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue, - .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue, - .notify_send_init = virtio_transport_notify_send_init, - .notify_send_pre_block = virtio_transport_notify_send_pre_block, - .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, - .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, - - .set_buffer_size = virtio_transport_set_buffer_size, - .set_min_buffer_size = virtio_transport_set_min_buffer_size, - .set_max_buffer_size = virtio_transport_set_max_buffer_size, - .get_buffer_size = virtio_transport_get_buffer_size, - .get_min_buffer_size = virtio_transport_get_min_buffer_size, - .get_max_buffer_size = virtio_transport_get_max_buffer_size, -}; - -static int virtio_vsock_probe(struct virtio_device *vdev) -{ - vq_callback_t *callbacks[] = { - virtio_vsock_ctrl_done, - virtio_vsock_rx_done, - virtio_vsock_tx_done, - }; - const char *names[] = { - "ctrl", - "rx", - "tx", - }; - struct virtio_vsock *vsock = NULL; - u32 guest_cid; - int ret; - - ret = mutex_lock_interruptible(&the_virtio_vsock_mutex); - if (ret) - return ret; - - /* Only one virtio-vsock device per guest is supported */ - if (the_virtio_vsock) { - ret = -EBUSY; - goto out; - } - - vsock = kzalloc(sizeof(*vsock), GFP_KERNEL); - if (!vsock) { - ret = -ENOMEM; - goto out; - } - - vsock->vdev = vdev; - - ret = vsock->vdev->config->find_vqs(vsock->vdev, VSOCK_VQ_MAX, - vsock->vqs, callbacks, names); - if (ret < 0) - goto out; - - vdev->config->get(vdev, offsetof(struct virtio_vsock_config, guest_cid), - &guest_cid, sizeof(guest_cid)); - vsock->guest_cid = le32_to_cpu(guest_cid); - pr_debug("%s:guest_cid=%d\n", __func__, vsock->guest_cid); - - ret = vsock_core_init(&virtio_transport); - if (ret < 0) - goto out_vqs; - - vsock->rx_buf_nr = 0; - vsock->rx_buf_max_nr = 0; - - vdev->priv = the_virtio_vsock = vsock; - init_waitqueue_head(&vsock->queue_wait); - mutex_init(&vsock->tx_lock); - mutex_init(&vsock->rx_lock); - INIT_WORK(&vsock->rx_work, virtio_transport_recv_pkt_work); - INIT_WORK(&vsock->tx_work, virtio_transport_send_pkt_work); - - mutex_lock(&vsock->rx_lock); - virtio_vsock_rx_fill(vsock); - mutex_unlock(&vsock->rx_lock); - - mutex_unlock(&the_virtio_vsock_mutex); - return 0; - -out_vqs: - vsock->vdev->config->del_vqs(vsock->vdev); -out: - kfree(vsock); - mutex_unlock(&the_virtio_vsock_mutex); - return ret; -} - -static void virtio_vsock_remove(struct virtio_device *vdev) -{ - struct virtio_vsock *vsock = vdev->priv; - - mutex_lock(&the_virtio_vsock_mutex); - the_virtio_vsock = NULL; - vsock_core_exit(); - mutex_unlock(&the_virtio_vsock_mutex); - - kfree(vsock); -} - -static struct virtio_device_id id_table[] = { - { VIRTIO_ID_VSOCK, VIRTIO_DEV_ANY_ID }, - { 0 }, -}; - -static unsigned int features[] = { -}; - -static struct virtio_driver virtio_vsock_driver = { - .feature_table = features, - .feature_table_size = ARRAY_SIZE(features), - .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, - .id_table = id_table, - .probe = virtio_vsock_probe, - .remove = virtio_vsock_remove, -}; - -static int __init virtio_vsock_init(void) -{ - int ret; - - virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", 0, 0); - if (!virtio_vsock_workqueue) - return -ENOMEM; - ret = register_virtio_driver(&virtio_vsock_driver); - if (ret) - destroy_workqueue(virtio_vsock_workqueue); - return ret; -} - -static void __exit virtio_vsock_exit(void) -{ - unregister_virtio_driver(&virtio_vsock_driver); - destroy_workqueue(virtio_vsock_workqueue); -} - -module_init(virtio_vsock_init); -module_exit(virtio_vsock_exit); -MODULE_LICENSE("GPL v2"); -MODULE_AUTHOR("Asias He"); -MODULE_DESCRIPTION("virtio transport for vsock"); -MODULE_DEVICE_TABLE(virtio, id_table); diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c deleted file mode 100644 index 28f790da6f15..000000000000 --- a/net/vmw_vsock/virtio_transport_common.c +++ /dev/null @@ -1,1272 +0,0 @@ -/* - * common code for virtio vsock - * - * Copyright (C) 2013-2015 Red Hat, Inc. - * Author: Asias He - * Stefan Hajnoczi - * - * This work is licensed under the terms of the GNU GPL, version 2. - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -#define COOKIEBITS 24 -#define COOKIEMASK (((u32)1 << COOKIEBITS) - 1) -#define VSOCK_TIMEOUT_INIT 4 - -#define SHA_MESSAGE_WORDS 16 -#define SHA_VSOCK_WORDS 5 - -static u32 vsockcookie_secret[2][SHA_MESSAGE_WORDS - SHA_VSOCK_WORDS + - SHA_DIGEST_WORDS]; - -static DEFINE_PER_CPU(__u32[SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS + - SHA_WORKSPACE_WORDS], vsock_cookie_scratch); - -static u32 cookie_hash(u32 saddr, u32 daddr, u16 sport, u16 dport, - u32 count, int c) -{ - __u32 *tmp = this_cpu_ptr(vsock_cookie_scratch); - - memcpy(tmp + SHA_VSOCK_WORDS, vsockcookie_secret[c], - sizeof(vsockcookie_secret[c])); - tmp[0] = saddr; - tmp[1] = daddr; - tmp[2] = sport; - tmp[3] = dport; - tmp[4] = count; - sha_transform(tmp + SHA_MESSAGE_WORDS, (__u8 *)tmp, - tmp + SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS); - - return tmp[17]; -} - -static u32 -virtio_vsock_secure_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, - u32 count) -{ - u32 h1, h2; - - h1 = cookie_hash(saddr, daddr, sport, dport, 0, 0); - h2 = cookie_hash(saddr, daddr, sport, dport, count, 1); - - return h1 + (count << COOKIEBITS) + (h2 & COOKIEMASK); -} - -static u32 -virtio_vsock_check_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, - u32 count, u32 cookie, u32 maxdiff) -{ - u32 diff; - u32 ret; - - cookie -= cookie_hash(saddr, daddr, sport, dport, 0, 0); - - diff = (count - (cookie >> COOKIEBITS)) & ((u32)-1 >> COOKIEBITS); - pr_debug("%s: diff=%x\n", __func__, diff); - if (diff >= maxdiff) - return (u32)-1; - - ret = (cookie - - cookie_hash(saddr, daddr, sport, dport, count - diff, 1)) - & COOKIEMASK; - pr_debug("%s: ret=%x\n", __func__, diff); - - return ret; -} - -void virtio_vsock_dumppkt(const char *func, const struct virtio_vsock_pkt *pkt) -{ - pr_debug("%s: pkt=%p, op=%d, len=%d, %d:%d---%d:%d, len=%d\n", - func, pkt, - le16_to_cpu(pkt->hdr.op), - le32_to_cpu(pkt->hdr.len), - le32_to_cpu(pkt->hdr.src_cid), - le32_to_cpu(pkt->hdr.src_port), - le32_to_cpu(pkt->hdr.dst_cid), - le32_to_cpu(pkt->hdr.dst_port), - pkt->len); -} -EXPORT_SYMBOL_GPL(virtio_vsock_dumppkt); - -struct virtio_vsock_pkt * -virtio_transport_alloc_pkt(struct vsock_sock *vsk, - struct virtio_vsock_pkt_info *info, - size_t len, - u32 src_cid, - u32 src_port, - u32 dst_cid, - u32 dst_port) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt *pkt; - int err; - - BUG_ON(!trans); - - pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); - if (!pkt) - return NULL; - - pkt->hdr.type = cpu_to_le16(info->type); - pkt->hdr.op = cpu_to_le16(info->op); - pkt->hdr.src_cid = cpu_to_le32(src_cid); - pkt->hdr.src_port = cpu_to_le32(src_port); - pkt->hdr.dst_cid = cpu_to_le32(dst_cid); - pkt->hdr.dst_port = cpu_to_le32(dst_port); - pkt->hdr.flags = cpu_to_le32(info->flags); - pkt->len = len; - pkt->trans = trans; - if (info->type == VIRTIO_VSOCK_TYPE_DGRAM) - pkt->hdr.len = cpu_to_le32(len + (info->dgram_len << 16)); - else if (info->type == VIRTIO_VSOCK_TYPE_STREAM) - pkt->hdr.len = cpu_to_le32(len); - - if (info->msg && len > 0) { - pkt->buf = kmalloc(len, GFP_KERNEL); - if (!pkt->buf) - goto out_pkt; - err = memcpy_from_msg(pkt->buf, info->msg, len); - if (err) - goto out; - } - - return pkt; - -out: - kfree(pkt->buf); -out_pkt: - kfree(pkt); - return NULL; -} -EXPORT_SYMBOL_GPL(virtio_transport_alloc_pkt); - -struct sock * -virtio_transport_get_pending(struct sock *listener, - struct virtio_vsock_pkt *pkt) -{ - struct vsock_sock *vlistener; - struct vsock_sock *vpending; - struct sockaddr_vm src; - struct sockaddr_vm dst; - struct sock *pending; - - vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); - vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); - - vlistener = vsock_sk(listener); - list_for_each_entry(vpending, &vlistener->pending_links, - pending_links) { - if (vsock_addr_equals_addr(&src, &vpending->remote_addr) && - vsock_addr_equals_addr(&dst, &vpending->local_addr)) { - pending = sk_vsock(vpending); - sock_hold(pending); - return pending; - } - } - - return NULL; -} -EXPORT_SYMBOL_GPL(virtio_transport_get_pending); - -static void virtio_transport_inc_rx_pkt(struct virtio_vsock_pkt *pkt) -{ - pkt->trans->rx_bytes += pkt->len; -} - -static void virtio_transport_dec_rx_pkt(struct virtio_vsock_pkt *pkt) -{ - pkt->trans->rx_bytes -= pkt->len; - pkt->trans->fwd_cnt += pkt->len; -} - -void virtio_transport_inc_tx_pkt(struct virtio_vsock_pkt *pkt) -{ - mutex_lock(&pkt->trans->tx_lock); - pkt->hdr.fwd_cnt = cpu_to_le32(pkt->trans->fwd_cnt); - pkt->hdr.buf_alloc = cpu_to_le32(pkt->trans->buf_alloc); - mutex_unlock(&pkt->trans->tx_lock); -} -EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt); - -void virtio_transport_dec_tx_pkt(struct virtio_vsock_pkt *pkt) -{ -} -EXPORT_SYMBOL_GPL(virtio_transport_dec_tx_pkt); - -u32 virtio_transport_get_credit(struct virtio_transport *trans, u32 credit) -{ - u32 ret; - - mutex_lock(&trans->tx_lock); - ret = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); - if (ret > credit) - ret = credit; - trans->tx_cnt += ret; - mutex_unlock(&trans->tx_lock); - - pr_debug("%s: ret=%d, buf_alloc=%d, peer_buf_alloc=%d," - "tx_cnt=%d, fwd_cnt=%d, peer_fwd_cnt=%d\n", __func__, - ret, trans->buf_alloc, trans->peer_buf_alloc, - trans->tx_cnt, trans->fwd_cnt, trans->peer_fwd_cnt); - - return ret; -} -EXPORT_SYMBOL_GPL(virtio_transport_get_credit); - -void virtio_transport_put_credit(struct virtio_transport *trans, u32 credit) -{ - mutex_lock(&trans->tx_lock); - trans->tx_cnt -= credit; - mutex_unlock(&trans->tx_lock); -} -EXPORT_SYMBOL_GPL(virtio_transport_put_credit); - -static int virtio_transport_send_credit_update(struct vsock_sock *vsk, int type, struct virtio_vsock_hdr *hdr) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_CREDIT_UPDATE, - .type = type, - }; - - if (hdr && type == VIRTIO_VSOCK_TYPE_DGRAM) { - info.remote_cid = le32_to_cpu(hdr->src_cid); - info.remote_port = le32_to_cpu(hdr->src_port); - } - - pr_debug("%s: sk=%p send_credit_update\n", __func__, vsk); - return trans->ops->send_pkt(vsk, &info); -} - -static int virtio_transport_send_credit_request(struct vsock_sock *vsk, int type) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_CREDIT_REQUEST, - .type = type, - }; - - pr_debug("%s: sk=%p send_credit_request\n", __func__, vsk); - return trans->ops->send_pkt(vsk, &info); -} - -static ssize_t -virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt *pkt; - size_t bytes, total = 0; - int err = -EFAULT; - - mutex_lock(&trans->rx_lock); - while (total < len && trans->rx_bytes > 0 && - !list_empty(&trans->rx_queue)) { - pkt = list_first_entry(&trans->rx_queue, - struct virtio_vsock_pkt, list); - - bytes = len - total; - if (bytes > pkt->len - pkt->off) - bytes = pkt->len - pkt->off; - - err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes); - if (err) - goto out; - total += bytes; - pkt->off += bytes; - if (pkt->off == pkt->len) { - virtio_transport_dec_rx_pkt(pkt); - list_del(&pkt->list); - virtio_transport_free_pkt(pkt); - } - } - mutex_unlock(&trans->rx_lock); - - /* Send a credit pkt to peer */ - virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM, - NULL); - - return total; - -out: - mutex_unlock(&trans->rx_lock); - if (total) - err = total; - return err; -} - -ssize_t -virtio_transport_stream_dequeue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len, int flags) -{ - if (flags & MSG_PEEK) - return -EOPNOTSUPP; - - return virtio_transport_stream_do_dequeue(vsk, msg, len); -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue); - -struct dgram_skb { - struct list_head list; - struct sk_buff *skb; - u16 id; -}; - -static struct dgram_skb *dgram_id_to_skb(struct virtio_transport *trans, - u16 id) -{ - struct dgram_skb *dgram_skb; - - list_for_each_entry(dgram_skb, &trans->incomplete_dgrams, list) { - if (dgram_skb->id == id) - return dgram_skb; - } - - return NULL; -} - -static void -virtio_transport_recv_dgram(struct sock *sk, - struct virtio_vsock_pkt *pkt) -{ - struct sk_buff *skb = NULL; - struct vsock_sock *vsk; - struct virtio_transport *trans; - size_t size; - u16 dgram_id, pkt_off, dgram_len, pkt_len; - u32 flags, len; - struct dgram_skb *dgram_skb; - - vsk = vsock_sk(sk); - trans = vsk->trans; - - /* len: dgram_len | pkt_len */ - len = le32_to_cpu(pkt->hdr.len); - dgram_len = len >> 16; - pkt_len = len & 0xFFFF; - - /* flags: dgram_id | pkt_off */ - flags = le32_to_cpu(pkt->hdr.flags); - dgram_id = flags >> 16; - pkt_off = flags & 0xFFFF; - - pr_debug("%s: dgram_len=%d, pkt_len=%d, id=%d, off=%d\n", __func__, - dgram_len, pkt_len, dgram_id, pkt_off); - - dgram_skb = dgram_id_to_skb(trans, dgram_id); - if (dgram_skb) { - /* This pkt is for a existing dgram */ - skb = dgram_skb->skb; - pr_debug("%s:found skb\n", __func__); - } - - /* Packet payload must be within datagram bounds */ - if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) - goto drop; - if (pkt_len > dgram_len) - goto drop; - if (pkt_off > dgram_len) - goto drop; - if (dgram_len - pkt_off < pkt_len) - goto drop; - - if (!skb) { - /* This pkt is for a new dgram */ - pr_debug("%s:create skb\n", __func__); - - size = sizeof(pkt->hdr) + dgram_len; - /* Attach the packet to the socket's receive queue as an sk_buff. */ - dgram_skb = kzalloc(sizeof(struct dgram_skb), GFP_ATOMIC); - if (!dgram_skb) - goto drop; - - skb = alloc_skb(size, GFP_ATOMIC); - if (!skb) { - kfree(dgram_skb); - dgram_skb = NULL; - goto drop; - } - dgram_skb->id = dgram_id; - dgram_skb->skb = skb; - list_add_tail(&dgram_skb->list, &trans->incomplete_dgrams); - - /* sk_receive_skb() will do a sock_put(), so hold here. */ - sock_hold(sk); - skb_put(skb, size); - memcpy(skb->data, &pkt->hdr, sizeof(pkt->hdr)); - } - - memcpy(skb->data + sizeof(pkt->hdr) + pkt_off, pkt->buf, pkt_len); - - pr_debug("%s:C, off=%d, pkt_len=%d, dgram_len=%d\n", __func__, - pkt_off, pkt_len, dgram_len); - - /* We are done with this dgram */ - if (pkt_off + pkt_len == dgram_len) { - pr_debug("%s:dgram_id=%d is done\n", __func__, dgram_id); - list_del(&dgram_skb->list); - kfree(dgram_skb); - sk_receive_skb(sk, skb, 0); - } - virtio_transport_free_pkt(pkt); - return; - -drop: - if (dgram_skb) { - list_del(&dgram_skb->list); - kfree(dgram_skb); - kfree_skb(skb); - sock_put(sk); - } - virtio_transport_free_pkt(pkt); -} - -int -virtio_transport_dgram_dequeue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len, int flags) -{ - struct virtio_vsock_hdr *hdr; - struct sk_buff *skb; - int noblock; - int err; - int dgram_len; - - noblock = flags & MSG_DONTWAIT; - - if (flags & MSG_OOB || flags & MSG_ERRQUEUE) - return -EOPNOTSUPP; - - /* Retrieve the head sk_buff from the socket's receive queue. */ - err = 0; - skb = skb_recv_datagram(&vsk->sk, flags, noblock, &err); - if (err) - return err; - if (!skb) - return -EAGAIN; - - hdr = (struct virtio_vsock_hdr *)skb->data; - if (!hdr) - goto out; - - dgram_len = le32_to_cpu(hdr->len) >> 16; - /* Place the datagram payload in the user's iovec. */ - err = skb_copy_datagram_msg(skb, sizeof(*hdr), msg, dgram_len); - if (err) - goto out; - - if (msg->msg_name) { - /* Provide the address of the sender. */ - DECLARE_SOCKADDR(struct sockaddr_vm *, vm_addr, msg->msg_name); - vsock_addr_init(vm_addr, le32_to_cpu(hdr->src_cid), le32_to_cpu(hdr->src_port)); - msg->msg_namelen = sizeof(*vm_addr); - } - err = dgram_len; - - /* Send a credit pkt to peer */ - virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, hdr); - - pr_debug("%s:done, recved =%d\n", __func__, dgram_len); -out: - skb_free_datagram(&vsk->sk, skb); - return err; -} -EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue); - -s64 virtio_transport_stream_has_data(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - s64 bytes; - - mutex_lock(&trans->rx_lock); - bytes = trans->rx_bytes; - mutex_unlock(&trans->rx_lock); - - return bytes; -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data); - -static s64 virtio_transport_has_space(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - s64 bytes; - - bytes = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); - if (bytes < 0) - bytes = 0; - - return bytes; -} - -s64 virtio_transport_stream_has_space(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - s64 bytes; - - mutex_lock(&trans->tx_lock); - bytes = virtio_transport_has_space(vsk); - mutex_unlock(&trans->tx_lock); - - pr_debug("%s: bytes=%lld\n", __func__, bytes); - - return bytes; -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space); - -int virtio_transport_do_socket_init(struct vsock_sock *vsk, - struct vsock_sock *psk) -{ - struct virtio_transport *trans; - - trans = kzalloc(sizeof(*trans), GFP_KERNEL); - if (!trans) - return -ENOMEM; - - vsk->trans = trans; - trans->vsk = vsk; - if (psk) { - struct virtio_transport *ptrans = psk->trans; - trans->buf_size = ptrans->buf_size; - trans->buf_size_min = ptrans->buf_size_min; - trans->buf_size_max = ptrans->buf_size_max; - trans->peer_buf_alloc = ptrans->peer_buf_alloc; - } else { - trans->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE; - trans->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE; - trans->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE; - } - - trans->buf_alloc = trans->buf_size; - - pr_debug("%s: trans->buf_alloc=%d\n", __func__, trans->buf_alloc); - - mutex_init(&trans->rx_lock); - mutex_init(&trans->tx_lock); - INIT_LIST_HEAD(&trans->rx_queue); - INIT_LIST_HEAD(&trans->incomplete_dgrams); - - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init); - -u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - - return trans->buf_size; -} -EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size); - -u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - - return trans->buf_size_min; -} -EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size); - -u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - - return trans->buf_size_max; -} -EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size); - -void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val) -{ - struct virtio_transport *trans = vsk->trans; - - if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) - val = VIRTIO_VSOCK_MAX_BUF_SIZE; - if (val < trans->buf_size_min) - trans->buf_size_min = val; - if (val > trans->buf_size_max) - trans->buf_size_max = val; - trans->buf_size = val; - trans->buf_alloc = val; -} -EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size); - -void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val) -{ - struct virtio_transport *trans = vsk->trans; - - if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) - val = VIRTIO_VSOCK_MAX_BUF_SIZE; - if (val > trans->buf_size) - trans->buf_size = val; - trans->buf_size_min = val; -} -EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size); - -void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val) -{ - struct virtio_transport *trans = vsk->trans; - - if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) - val = VIRTIO_VSOCK_MAX_BUF_SIZE; - if (val < trans->buf_size) - trans->buf_size = val; - trans->buf_size_max = val; -} -EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size); - -int -virtio_transport_notify_poll_in(struct vsock_sock *vsk, - size_t target, - bool *data_ready_now) -{ - if (vsock_stream_has_data(vsk)) - *data_ready_now = true; - else - *data_ready_now = false; - - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in); - -int -virtio_transport_notify_poll_out(struct vsock_sock *vsk, - size_t target, - bool *space_avail_now) -{ - s64 free_space; - - free_space = vsock_stream_has_space(vsk); - if (free_space > 0) - *space_avail_now = true; - else if (free_space == 0) - *space_avail_now = false; - - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out); - -int virtio_transport_notify_recv_init(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init); - -int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block); - -int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, - size_t target, struct vsock_transport_recv_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue); - -int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, - size_t target, ssize_t copied, bool data_read, - struct vsock_transport_recv_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue); - -int virtio_transport_notify_send_init(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init); - -int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block); - -int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, - struct vsock_transport_send_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue); - -int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, - ssize_t written, struct vsock_transport_send_notify_data *data) -{ - return 0; -} -EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue); - -u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - - return trans->buf_size; -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat); - -bool virtio_transport_stream_is_active(struct vsock_sock *vsk) -{ - return true; -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active); - -bool virtio_transport_stream_allow(u32 cid, u32 port) -{ - return true; -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_allow); - -int virtio_transport_dgram_bind(struct vsock_sock *vsk, - struct sockaddr_vm *addr) -{ - return vsock_bind_dgram_generic(vsk, addr); -} -EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind); - -bool virtio_transport_dgram_allow(u32 cid, u32 port) -{ - return true; -} -EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow); - -int virtio_transport_connect(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_REQUEST, - .type = VIRTIO_VSOCK_TYPE_STREAM, - }; - - pr_debug("%s: vsk=%p send_request\n", __func__, vsk); - return trans->ops->send_pkt(vsk, &info); -} -EXPORT_SYMBOL_GPL(virtio_transport_connect); - -int virtio_transport_shutdown(struct vsock_sock *vsk, int mode) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_SHUTDOWN, - .type = VIRTIO_VSOCK_TYPE_STREAM, - .flags = (mode & RCV_SHUTDOWN ? - VIRTIO_VSOCK_SHUTDOWN_RCV : 0) | - (mode & SEND_SHUTDOWN ? - VIRTIO_VSOCK_SHUTDOWN_SEND : 0), - }; - - pr_debug("%s: vsk=%p: send_shutdown\n", __func__, vsk); - return trans->ops->send_pkt(vsk, &info); -} -EXPORT_SYMBOL_GPL(virtio_transport_shutdown); - -void virtio_transport_release(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - struct sock *sk = &vsk->sk; - struct dgram_skb *dgram_skb; - struct dgram_skb *dgram_skb_tmp; - - pr_debug("%s: vsk=%p\n", __func__, vsk); - - /* Tell other side to terminate connection */ - if (sk->sk_type == SOCK_STREAM && sk->sk_state == SS_CONNECTED) { - virtio_transport_shutdown(vsk, SHUTDOWN_MASK); - } - - /* Free incomplete dgrams */ - lock_sock(sk); - list_for_each_entry_safe(dgram_skb, dgram_skb_tmp, - &trans->incomplete_dgrams, list) { - list_del(&dgram_skb->list); - kfree_skb(dgram_skb->skb); - kfree(dgram_skb); - sock_put(sk); /* held in virtio_transport_recv_dgram() */ - } - release_sock(sk); -} -EXPORT_SYMBOL_GPL(virtio_transport_release); - -int -virtio_transport_dgram_enqueue(struct vsock_sock *vsk, - struct sockaddr_vm *remote_addr, - struct msghdr *msg, - size_t dgram_len) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_RW, - .type = VIRTIO_VSOCK_TYPE_DGRAM, - .msg = msg, - }; - size_t total_written = 0, pkt_off = 0, written; - u16 dgram_id; - - /* The max size of a single dgram we support is 64KB */ - if (dgram_len > VIRTIO_VSOCK_MAX_DGRAM_SIZE) - return -EMSGSIZE; - - info.dgram_len = dgram_len; - vsk->remote_addr = *remote_addr; - - dgram_id = trans->dgram_id++; - - /* TODO: To optimize, if we have enough credit to send the pkt already, - * do not ask the peer to send credit to use */ - virtio_transport_send_credit_request(vsk, VIRTIO_VSOCK_TYPE_DGRAM); - - while (total_written < dgram_len) { - info.pkt_len = dgram_len - total_written; - info.flags = dgram_id << 16 | pkt_off; - written = trans->ops->send_pkt(vsk, &info); - if (written < 0) - return -ENOMEM; - if (written == 0) { - /* TODO: if written = 0, we need a sleep & wakeup - * instead of sleep */ - pr_debug("%s: SHOULD WAIT written==0", __func__); - msleep(10); - } - total_written += written; - pkt_off += written; - pr_debug("%s:id=%d, dgram_len=%zu, off=%zu, total_written=%zu, written=%zu\n", - __func__, dgram_id, dgram_len, pkt_off, total_written, written); - } - - return dgram_len; -} -EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue); - -ssize_t -virtio_transport_stream_enqueue(struct vsock_sock *vsk, - struct msghdr *msg, - size_t len) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_RW, - .type = VIRTIO_VSOCK_TYPE_STREAM, - .msg = msg, - .pkt_len = len, - }; - - return trans->ops->send_pkt(vsk, &info); -} -EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue); - -void virtio_transport_destruct(struct vsock_sock *vsk) -{ - struct virtio_transport *trans = vsk->trans; - - pr_debug("%s: vsk=%p\n", __func__, vsk); - kfree(trans); -} -EXPORT_SYMBOL_GPL(virtio_transport_destruct); - -static int virtio_transport_send_ack(struct vsock_sock *vsk, u32 cookie) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_ACK, - .type = VIRTIO_VSOCK_TYPE_STREAM, - .flags = cpu_to_le32(cookie), - }; - - pr_debug("%s: sk=%p send_offer\n", __func__, vsk); - return trans->ops->send_pkt(vsk, &info); -} - -static int virtio_transport_send_reset(struct vsock_sock *vsk, - struct virtio_vsock_pkt *pkt) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_RST, - .type = VIRTIO_VSOCK_TYPE_STREAM, - }; - - pr_debug("%s\n", __func__); - - /* Send RST only if the original pkt is not a RST pkt */ - if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) - return 0; - - return trans->ops->send_pkt(vsk, &info); -} - -static int -virtio_transport_recv_connecting(struct sock *sk, - struct virtio_vsock_pkt *pkt) -{ - struct vsock_sock *vsk = vsock_sk(sk); - int err; - int skerr; - u32 cookie; - - pr_debug("%s: vsk=%p\n", __func__, vsk); - switch (le16_to_cpu(pkt->hdr.op)) { - case VIRTIO_VSOCK_OP_RESPONSE: - cookie = le32_to_cpu(pkt->hdr.flags); - pr_debug("%s: got RESPONSE and send ACK, cookie=%x\n", __func__, cookie); - err = virtio_transport_send_ack(vsk, cookie); - if (err < 0) { - skerr = -err; - goto destroy; - } - sk->sk_state = SS_CONNECTED; - sk->sk_socket->state = SS_CONNECTED; - vsock_insert_connected(vsk); - sk->sk_state_change(sk); - break; - case VIRTIO_VSOCK_OP_INVALID: - pr_debug("%s: got invalid\n", __func__); - break; - case VIRTIO_VSOCK_OP_RST: - pr_debug("%s: got rst\n", __func__); - skerr = ECONNRESET; - err = 0; - goto destroy; - default: - pr_debug("%s: got def\n", __func__); - skerr = EPROTO; - err = -EINVAL; - goto destroy; - } - return 0; - -destroy: - virtio_transport_send_reset(vsk, pkt); - sk->sk_state = SS_UNCONNECTED; - sk->sk_err = skerr; - sk->sk_error_report(sk); - return err; -} - -static int -virtio_transport_recv_connected(struct sock *sk, - struct virtio_vsock_pkt *pkt) -{ - struct vsock_sock *vsk = vsock_sk(sk); - struct virtio_transport *trans = vsk->trans; - int err = 0; - - switch (le16_to_cpu(pkt->hdr.op)) { - case VIRTIO_VSOCK_OP_RW: - pkt->len = le32_to_cpu(pkt->hdr.len); - pkt->off = 0; - pkt->trans = trans; - - mutex_lock(&trans->rx_lock); - virtio_transport_inc_rx_pkt(pkt); - list_add_tail(&pkt->list, &trans->rx_queue); - mutex_unlock(&trans->rx_lock); - - sk->sk_data_ready(sk); - return err; - case VIRTIO_VSOCK_OP_CREDIT_UPDATE: - sk->sk_write_space(sk); - break; - case VIRTIO_VSOCK_OP_SHUTDOWN: - pr_debug("%s: got shutdown\n", __func__); - if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) - vsk->peer_shutdown |= RCV_SHUTDOWN; - if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) - vsk->peer_shutdown |= SEND_SHUTDOWN; - if (le32_to_cpu(pkt->hdr.flags)) - sk->sk_state_change(sk); - break; - case VIRTIO_VSOCK_OP_RST: - pr_debug("%s: got rst\n", __func__); - sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; - if (vsock_stream_has_data(vsk) <= 0) - sk->sk_state = SS_DISCONNECTING; - sk->sk_state_change(sk); - break; - default: - err = -EINVAL; - break; - } - - virtio_transport_free_pkt(pkt); - return err; -} - -static int -virtio_transport_send_response(struct vsock_sock *vsk, - struct virtio_vsock_pkt *pkt) -{ - struct virtio_transport *trans = vsk->trans; - struct virtio_vsock_pkt_info info = { - .op = VIRTIO_VSOCK_OP_RESPONSE, - .type = VIRTIO_VSOCK_TYPE_STREAM, - .remote_cid = le32_to_cpu(pkt->hdr.src_cid), - .remote_port = le32_to_cpu(pkt->hdr.src_port), - }; - u32 cookie; - - cookie = virtio_vsock_secure_cookie(le32_to_cpu(pkt->hdr.src_cid), - le32_to_cpu(pkt->hdr.dst_cid), - le32_to_cpu(pkt->hdr.src_port), - le32_to_cpu(pkt->hdr.dst_port), - jiffies / (HZ * 60)); - info.flags = cpu_to_le32(cookie); - - pr_debug("%s: send_response, cookie=%x\n", __func__, le32_to_cpu(cookie)); - - return trans->ops->send_pkt(vsk, &info); -} - -/* Handle server socket */ -static int -virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt) -{ - struct vsock_sock *vsk = vsock_sk(sk); - struct vsock_sock *vpending; - struct sock *pending; - int err; - u32 cookie; - - switch (le16_to_cpu(pkt->hdr.op)) { - case VIRTIO_VSOCK_OP_REQUEST: - err = virtio_transport_send_response(vsk, pkt); - if (err < 0) { - // FIXME vsk should be vpending - virtio_transport_send_reset(vsk, pkt); - return err; - } - break; - case VIRTIO_VSOCK_OP_ACK: - cookie = le32_to_cpu(pkt->hdr.flags); - err = virtio_vsock_check_cookie(le32_to_cpu(pkt->hdr.src_cid), - le32_to_cpu(pkt->hdr.dst_cid), - le32_to_cpu(pkt->hdr.src_port), - le32_to_cpu(pkt->hdr.dst_port), - jiffies / (HZ * 60), - le32_to_cpu(pkt->hdr.flags), - VSOCK_TIMEOUT_INIT); - pr_debug("%s: cookie=%x, err=%d\n", __func__, cookie, err); - if (err) - return err; - - /* So no pending socket are responsible for this pkt, create one */ - pr_debug("%s: create pending\n", __func__); - pending = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL, - sk->sk_type, 0); - if (!pending) { - virtio_transport_send_reset(vsk, pkt); - return -ENOMEM; - } - sk->sk_ack_backlog++; - pending->sk_state = SS_CONNECTING; - - vpending = vsock_sk(pending); - vsock_addr_init(&vpending->local_addr, le32_to_cpu(pkt->hdr.dst_cid), - le32_to_cpu(pkt->hdr.dst_port)); - vsock_addr_init(&vpending->remote_addr, le32_to_cpu(pkt->hdr.src_cid), - le32_to_cpu(pkt->hdr.src_port)); - vsock_add_pending(sk, pending); - - pr_debug("%s: get pending\n", __func__); - pending = virtio_transport_get_pending(sk, pkt); - vpending = vsock_sk(pending); - lock_sock(pending); - switch (pending->sk_state) { - case SS_CONNECTING: - if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_ACK) { - pr_debug("%s: op=%d != OP_ACK\n", __func__, - le16_to_cpu(pkt->hdr.op)); - virtio_transport_send_reset(vpending, pkt); - pending->sk_err = EPROTO; - pending->sk_state = SS_UNCONNECTED; - sock_put(pending); - } else { - pending->sk_state = SS_CONNECTED; - vsock_insert_connected(vpending); - - vsock_remove_pending(sk, pending); - vsock_enqueue_accept(sk, pending); - - sk->sk_data_ready(sk); - } - err = 0; - break; - default: - pr_debug("%s: sk->sk_ack_backlog=%d\n", __func__, - sk->sk_ack_backlog); - virtio_transport_send_reset(vpending, pkt); - err = -EINVAL; - break; - } - if (err < 0) - vsock_remove_pending(sk, pending); - release_sock(pending); - - /* Release refcnt obtained in virtio_transport_get_pending */ - sock_put(pending); - break; - default: - break; - } - - return 0; -} - -static void virtio_transport_space_update(struct sock *sk, - struct virtio_vsock_pkt *pkt) -{ - struct vsock_sock *vsk = vsock_sk(sk); - struct virtio_transport *trans = vsk->trans; - bool space_available; - - /* buf_alloc and fwd_cnt is always included in the hdr */ - mutex_lock(&trans->tx_lock); - trans->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc); - trans->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt); - space_available = virtio_transport_has_space(vsk); - mutex_unlock(&trans->tx_lock); - - if (space_available) - sk->sk_write_space(sk); -} - -/* We are under the virtio-vsock's vsock->rx_lock or - * vhost-vsock's vq->mutex lock */ -void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt) -{ - struct virtio_transport *trans; - struct sockaddr_vm src, dst; - struct vsock_sock *vsk; - struct sock *sk; - - vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); - vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); - - virtio_vsock_dumppkt(__func__, pkt); - - if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_DGRAM) { - sk = vsock_find_unbound_socket(&dst); - if (!sk) - goto free_pkt; - - vsk = vsock_sk(sk); - trans = vsk->trans; - BUG_ON(!trans); - - virtio_transport_space_update(sk, pkt); - - lock_sock(sk); - switch (le16_to_cpu(pkt->hdr.op)) { - case VIRTIO_VSOCK_OP_CREDIT_UPDATE: - virtio_transport_free_pkt(pkt); - break; - case VIRTIO_VSOCK_OP_CREDIT_REQUEST: - virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, - &pkt->hdr); - virtio_transport_free_pkt(pkt); - break; - case VIRTIO_VSOCK_OP_RW: - virtio_transport_recv_dgram(sk, pkt); - break; - default: - virtio_transport_free_pkt(pkt); - break; - } - release_sock(sk); - - /* Release refcnt obtained when we fetched this socket out of - * the unbound list. - */ - sock_put(sk); - return; - } else if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM) { - /* The socket must be in connected or bound table - * otherwise send reset back - */ - sk = vsock_find_connected_socket(&src, &dst); - if (!sk) { - sk = vsock_find_bound_socket(&dst); - if (!sk) { - pr_debug("%s: can not find bound_socket\n", __func__); - virtio_vsock_dumppkt(__func__, pkt); - /* Ignore this pkt instead of sending reset back */ - /* TODO send a RST unless this packet is a RST (to avoid infinite loops) */ - goto free_pkt; - } - } - - vsk = vsock_sk(sk); - trans = vsk->trans; - BUG_ON(!trans); - - virtio_transport_space_update(sk, pkt); - - lock_sock(sk); - switch (sk->sk_state) { - case VSOCK_SS_LISTEN: - virtio_transport_recv_listen(sk, pkt); - virtio_transport_free_pkt(pkt); - break; - case SS_CONNECTING: - virtio_transport_recv_connecting(sk, pkt); - virtio_transport_free_pkt(pkt); - break; - case SS_CONNECTED: - virtio_transport_recv_connected(sk, pkt); - break; - default: - virtio_transport_free_pkt(pkt); - break; - } - release_sock(sk); - - /* Release refcnt obtained when we fetched this socket out of the - * bound or connected list. - */ - sock_put(sk); - } - return; - -free_pkt: - virtio_transport_free_pkt(pkt); -} -EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt); - -void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt) -{ - kfree(pkt->buf); - kfree(pkt); -} -EXPORT_SYMBOL_GPL(virtio_transport_free_pkt); - -static int __init virtio_vsock_common_init(void) -{ - get_random_bytes(vsockcookie_secret, sizeof(vsockcookie_secret)); - return 0; -} - -static void __exit virtio_vsock_common_exit(void) -{ -} - -module_init(virtio_vsock_common_init); -module_exit(virtio_vsock_common_exit); -MODULE_LICENSE("GPL v2"); -MODULE_AUTHOR("Asias He"); -MODULE_DESCRIPTION("common code for virtio vsock"); -- cgit v1.2.3-59-g8ed1b From 297dbde19cf6a0ccb6fd4396c6220a5912ed61e8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 7 Dec 2015 17:38:51 -0500 Subject: netprio_cgroup: limit the maximum css->id to USHRT_MAX netprio builds per-netdev contiguous priomap array which is indexed by css->id. The array is allocated using kzalloc() effectively limiting the maximum ID supported to some thousand range. This patch caps the maximum supported css->id to USHRT_MAX which should be way above what is actually useable. This allows reducing sock->sk_cgrp_prioidx to u16 from u32. The freed up part will be used to overload the cgroup related fields. sock->sk_cgrp_prioidx's position is swapped with sk_mark so that the two cgroup related fields are adjacent. Signed-off-by: Tejun Heo Acked-by: Daniel Wagner Cc: Daniel Borkmann CC: Neil Horman Signed-off-by: David S. Miller --- include/net/sock.h | 10 +++++----- net/core/netprio_cgroup.c | 9 +++++++++ 2 files changed, 14 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 6f58b84fc742..a95bcf7d6efa 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -288,7 +288,6 @@ struct cg_proto; * @sk_ack_backlog: current listen backlog * @sk_max_ack_backlog: listen backlog set in listen() * @sk_priority: %SO_PRIORITY setting - * @sk_cgrp_prioidx: socket group's priority map index * @sk_type: socket type (%SOCK_STREAM, etc) * @sk_protocol: which protocol this socket belongs in this network family * @sk_peer_pid: &struct pid for this socket's peer @@ -309,6 +308,7 @@ struct cg_proto; * @sk_send_head: front of stuff to transmit * @sk_security: used by security modules * @sk_mark: generic packet mark + * @sk_cgrp_prioidx: socket group's priority map index * @sk_classid: this socket's cgroup classid * @sk_cgrp: this socket's cgroup-specific proto data * @sk_write_pending: a write to stream socket waits to start @@ -425,9 +425,7 @@ struct sock { u32 sk_ack_backlog; u32 sk_max_ack_backlog; __u32 sk_priority; -#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) - __u32 sk_cgrp_prioidx; -#endif + __u32 sk_mark; struct pid *sk_peer_pid; const struct cred *sk_peer_cred; long sk_rcvtimeo; @@ -445,7 +443,9 @@ struct sock { #ifdef CONFIG_SECURITY void *sk_security; #endif - __u32 sk_mark; +#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) + u16 sk_cgrp_prioidx; +#endif #ifdef CONFIG_CGROUP_NET_CLASSID u32 sk_classid; #endif diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index cbd0a199bf52..2b9159b7a28a 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -27,6 +27,12 @@ #include +/* + * netprio allocates per-net_device priomap array which is indexed by + * css->id. Limiting css ID to 16bits doesn't lose anything. + */ +#define NETPRIO_ID_MAX USHRT_MAX + #define PRIOMAP_MIN_SZ 128 /* @@ -144,6 +150,9 @@ static int cgrp_css_online(struct cgroup_subsys_state *css) struct net_device *dev; int ret = 0; + if (css->id > NETPRIO_ID_MAX) + return -ENOSPC; + if (!parent_css) return 0; -- cgit v1.2.3-59-g8ed1b From 2a56a1fec290bf0bc4676bbf4efdb3744953a3e7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 7 Dec 2015 17:38:52 -0500 Subject: net: wrap sock->sk_cgrp_prioidx and ->sk_classid inside a struct Introduce sock->sk_cgrp_data which is a struct sock_cgroup_data. ->sk_cgroup_prioidx and ->sk_classid are moved into it. The struct and its accessors are defined in cgroup-defs.h. This is to prepare for overloading the fields with a cgroup pointer. This patch mostly performs equivalent conversions but the followings are noteworthy. * Equality test before updating classid is removed from sock_update_classid(). This shouldn't make any noticeable difference and a similar test will be implemented on the helper side later. * sock_update_netprioidx() now takes struct sock_cgroup_data and can be moved to netprio_cgroup.h without causing include dependency loop. Moved. * The dummy version of sock_update_netprioidx() converted to a static inline function while at it. Signed-off-by: Tejun Heo Signed-off-by: David S. Miller --- include/linux/cgroup-defs.h | 36 ++++++++++++++++++++++++++++++++++++ include/net/cls_cgroup.h | 11 +++++------ include/net/netprio_cgroup.h | 16 +++++++++++++--- include/net/sock.h | 11 +++-------- net/Kconfig | 6 ++++++ net/core/dev.c | 3 ++- net/core/netclassid_cgroup.c | 4 ++-- net/core/netprio_cgroup.c | 3 ++- net/core/scm.c | 4 ++-- net/core/sock.c | 15 ++------------- net/netfilter/nft_meta.c | 2 +- net/netfilter/xt_cgroup.c | 3 ++- 12 files changed, 76 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 504d8591b6d3..ed128fed0335 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -542,4 +542,40 @@ static inline void cgroup_threadgroup_change_end(struct task_struct *tsk) {} #endif /* CONFIG_CGROUPS */ +#ifdef CONFIG_SOCK_CGROUP_DATA + +struct sock_cgroup_data { + u16 prioidx; + u32 classid; +}; + +static inline u16 sock_cgroup_prioidx(struct sock_cgroup_data *skcd) +{ + return skcd->prioidx; +} + +static inline u32 sock_cgroup_classid(struct sock_cgroup_data *skcd) +{ + return skcd->classid; +} + +static inline void sock_cgroup_set_prioidx(struct sock_cgroup_data *skcd, + u16 prioidx) +{ + skcd->prioidx = prioidx; +} + +static inline void sock_cgroup_set_classid(struct sock_cgroup_data *skcd, + u32 classid) +{ + skcd->classid = classid; +} + +#else /* CONFIG_SOCK_CGROUP_DATA */ + +struct sock_cgroup_data { +}; + +#endif /* CONFIG_SOCK_CGROUP_DATA */ + #endif /* _LINUX_CGROUP_DEFS_H */ diff --git a/include/net/cls_cgroup.h b/include/net/cls_cgroup.h index ccd6d8bffa4d..c0a92e2c286d 100644 --- a/include/net/cls_cgroup.h +++ b/include/net/cls_cgroup.h @@ -41,13 +41,12 @@ static inline u32 task_cls_classid(struct task_struct *p) return classid; } -static inline void sock_update_classid(struct sock *sk) +static inline void sock_update_classid(struct sock_cgroup_data *skcd) { u32 classid; classid = task_cls_classid(current); - if (classid != sk->sk_classid) - sk->sk_classid = classid; + sock_cgroup_set_classid(skcd, classid); } static inline u32 task_get_classid(const struct sk_buff *skb) @@ -64,17 +63,17 @@ static inline u32 task_get_classid(const struct sk_buff *skb) * softirqs always disables bh. */ if (in_serving_softirq()) { - /* If there is an sk_classid we'll use that. */ + /* If there is an sock_cgroup_classid we'll use that. */ if (!skb->sk) return 0; - classid = skb->sk->sk_classid; + classid = sock_cgroup_classid(&skb->sk->sk_cgrp_data); } return classid; } #else /* !CONFIG_CGROUP_NET_CLASSID */ -static inline void sock_update_classid(struct sock *sk) +static inline void sock_update_classid(struct sock_cgroup_data *skcd) { } diff --git a/include/net/netprio_cgroup.h b/include/net/netprio_cgroup.h index f2a9597ff53c..604190596cde 100644 --- a/include/net/netprio_cgroup.h +++ b/include/net/netprio_cgroup.h @@ -25,8 +25,6 @@ struct netprio_map { u32 priomap[]; }; -void sock_update_netprioidx(struct sock *sk); - static inline u32 task_netprioidx(struct task_struct *p) { struct cgroup_subsys_state *css; @@ -38,13 +36,25 @@ static inline u32 task_netprioidx(struct task_struct *p) rcu_read_unlock(); return idx; } + +static inline void sock_update_netprioidx(struct sock_cgroup_data *skcd) +{ + if (in_interrupt()) + return; + + sock_cgroup_set_prioidx(skcd, task_netprioidx(current)); +} + #else /* !CONFIG_CGROUP_NET_PRIO */ + static inline u32 task_netprioidx(struct task_struct *p) { return 0; } -#define sock_update_netprioidx(sk) +static inline void sock_update_netprioidx(struct sock_cgroup_data *skcd) +{ +} #endif /* CONFIG_CGROUP_NET_PRIO */ #endif /* _NET_CLS_CGROUP_H */ diff --git a/include/net/sock.h b/include/net/sock.h index a95bcf7d6efa..0ca22b014de1 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -59,6 +59,7 @@ #include #include #include +#include #include #include @@ -308,8 +309,7 @@ struct cg_proto; * @sk_send_head: front of stuff to transmit * @sk_security: used by security modules * @sk_mark: generic packet mark - * @sk_cgrp_prioidx: socket group's priority map index - * @sk_classid: this socket's cgroup classid + * @sk_cgrp_data: cgroup data for this cgroup * @sk_cgrp: this socket's cgroup-specific proto data * @sk_write_pending: a write to stream socket waits to start * @sk_state_change: callback to indicate change in the state of the sock @@ -443,12 +443,7 @@ struct sock { #ifdef CONFIG_SECURITY void *sk_security; #endif -#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) - u16 sk_cgrp_prioidx; -#endif -#ifdef CONFIG_CGROUP_NET_CLASSID - u32 sk_classid; -#endif + struct sock_cgroup_data sk_cgrp_data; struct cg_proto *sk_cgrp; void (*sk_state_change)(struct sock *sk); void (*sk_data_ready)(struct sock *sk); diff --git a/net/Kconfig b/net/Kconfig index 127da94ae25e..11f8c22af34d 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -250,9 +250,14 @@ config XPS depends on SMP default y +config SOCK_CGROUP_DATA + bool + default n + config CGROUP_NET_PRIO bool "Network priority cgroup" depends on CGROUPS + select SOCK_CGROUP_DATA ---help--- Cgroup subsystem for use in assigning processes to network priorities on a per-interface basis. @@ -260,6 +265,7 @@ config CGROUP_NET_PRIO config CGROUP_NET_CLASSID bool "Network classid cgroup" depends on CGROUPS + select SOCK_CGROUP_DATA ---help--- Cgroup subsystem for use as general purpose socket classid marker that is being used in cls_cgroup and for netfilter matching. diff --git a/net/core/dev.c b/net/core/dev.c index e5c395473eba..8f705fcedb94 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2929,7 +2929,8 @@ static void skb_update_prio(struct sk_buff *skb) struct netprio_map *map = rcu_dereference_bh(skb->dev->priomap); if (!skb->priority && skb->sk && map) { - unsigned int prioidx = skb->sk->sk_cgrp_prioidx; + unsigned int prioidx = + sock_cgroup_prioidx(&skb->sk->sk_cgrp_data); if (prioidx < map->priomap_len) skb->priority = map->priomap[prioidx]; diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c index 2e4df84c34a1..e60ded46b3ac 100644 --- a/net/core/netclassid_cgroup.c +++ b/net/core/netclassid_cgroup.c @@ -62,8 +62,8 @@ static int update_classid_sock(const void *v, struct file *file, unsigned n) struct socket *sock = sock_from_file(file, &err); if (sock) - sock->sk->sk_classid = (u32)(unsigned long)v; - + sock_cgroup_set_classid(&sock->sk->sk_cgrp_data, + (unsigned long)v); return 0; } diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index 2b9159b7a28a..de42aa7f6c77 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -223,7 +223,8 @@ static int update_netprio(const void *v, struct file *file, unsigned n) int err; struct socket *sock = sock_from_file(file, &err); if (sock) - sock->sk->sk_cgrp_prioidx = (u32)(unsigned long)v; + sock_cgroup_set_prioidx(&sock->sk->sk_cgrp_data, + (unsigned long)v); return 0; } diff --git a/net/core/scm.c b/net/core/scm.c index 8a1741b14302..14596fb37172 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -289,8 +289,8 @@ void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm) /* Bump the usage count and install the file. */ sock = sock_from_file(fp[i], &err); if (sock) { - sock_update_netprioidx(sock->sk); - sock_update_classid(sock->sk); + sock_update_netprioidx(&sock->sk->sk_cgrp_data); + sock_update_classid(&sock->sk->sk_cgrp_data); } fd_install(new_fd, get_file(fp[i])); } diff --git a/net/core/sock.c b/net/core/sock.c index 7965ef487375..947741dc43fa 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1393,17 +1393,6 @@ static void sk_prot_free(struct proto *prot, struct sock *sk) module_put(owner); } -#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) -void sock_update_netprioidx(struct sock *sk) -{ - if (in_interrupt()) - return; - - sk->sk_cgrp_prioidx = task_netprioidx(current); -} -EXPORT_SYMBOL_GPL(sock_update_netprioidx); -#endif - /** * sk_alloc - All socket objects are allocated here * @net: the applicable net namespace @@ -1432,8 +1421,8 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, sock_net_set(sk, net); atomic_set(&sk->sk_wmem_alloc, 1); - sock_update_classid(sk); - sock_update_netprioidx(sk); + sock_update_classid(&sk->sk_cgrp_data); + sock_update_netprioidx(&sk->sk_cgrp_data); } return sk; diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 9dfaf4d55ee0..1915cab7f32d 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -174,7 +174,7 @@ void nft_meta_get_eval(const struct nft_expr *expr, sk = skb_to_full_sk(skb); if (!sk || !sk_fullsock(sk)) goto err; - *dest = sk->sk_classid; + *dest = sock_cgroup_classid(&sk->sk_cgrp_data); break; #endif default: diff --git a/net/netfilter/xt_cgroup.c b/net/netfilter/xt_cgroup.c index a1d126f29463..54eaeb45ce99 100644 --- a/net/netfilter/xt_cgroup.c +++ b/net/netfilter/xt_cgroup.c @@ -42,7 +42,8 @@ cgroup_mt(const struct sk_buff *skb, struct xt_action_param *par) if (skb->sk == NULL || !sk_fullsock(skb->sk)) return false; - return (info->id == skb->sk->sk_classid) ^ info->invert; + return (info->id == sock_cgroup_classid(&skb->sk->sk_cgrp_data)) ^ + info->invert; } static struct xt_match cgroup_mt_reg __read_mostly = { -- cgit v1.2.3-59-g8ed1b From bd1060a1d67128bb8fbe2e1384c518912cbe54e7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 7 Dec 2015 17:38:53 -0500 Subject: sock, cgroup: add sock->sk_cgroup In cgroup v1, dealing with cgroup membership was difficult because the number of membership associations was unbound. As a result, cgroup v1 grew several controllers whose primary purpose is either tagging membership or pull in configuration knobs from other subsystems so that cgroup membership test can be avoided. net_cls and net_prio controllers are examples of the latter. They allow configuring network-specific attributes from cgroup side so that network subsystem can avoid testing cgroup membership; unfortunately, these are not only cumbersome but also problematic. Both net_cls and net_prio aren't properly hierarchical. Both inherit configuration from the parent on creation but there's no interaction afterwards. An ancestor doesn't restrict the behavior in its subtree in anyway and configuration changes aren't propagated downwards. Especially when combined with cgroup delegation, this is problematic because delegatees can mess up whatever network configuration implemented at the system level. net_prio would allow the delegatees to set whatever priority value regardless of CAP_NET_ADMIN and net_cls the same for classid. While it is possible to solve these issues from controller side by implementing hierarchical allowable ranges in both controllers, it would involve quite a bit of complexity in the controllers and further obfuscate network configuration as it becomes even more difficult to tell what's actually being configured looking from the network side. While not much can be done for v1 at this point, as membership handling is sane on cgroup v2, it'd be better to make cgroup matching behave like other network matches and classifiers than introducing further complications. In preparation, this patch updates sock->sk_cgrp_data handling so that it points to the v2 cgroup that sock was created in until either net_prio or net_cls is used. Once either of the two is used, sock->sk_cgrp_data reverts to its previous role of carrying prioidx and classid. This is to avoid adding yet another cgroup related field to struct sock. As the mode switching can happen at most once per boot, the switching mechanism is aimed at lowering hot path overhead. It may leak a finite, likely small, number of cgroup refs and report spurious prioidx or classid on switching; however, dynamic updates of prioidx and classid have always been racy and lossy - socks between creation and fd installation are never updated, config changes don't update existing sockets at all, and prioidx may index with dead and recycled cgroup IDs. Non-critical inaccuracies from small race windows won't make any noticeable difference. This patch doesn't make use of the pointer yet. The following patch will implement netfilter match for cgroup2 membership. v2: Use sock_cgroup_data to avoid inflating struct sock w/ another cgroup specific field. v3: Add comments explaining why sock_data_prioidx() and sock_data_classid() use different fallback values. Signed-off-by: Tejun Heo Cc: Daniel Borkmann Cc: Daniel Wagner CC: Neil Horman Signed-off-by: David S. Miller --- include/linux/cgroup-defs.h | 88 +++++++++++++++++++++++++++++++++++++++++--- include/linux/cgroup.h | 41 +++++++++++++++++++++ kernel/cgroup.c | 55 ++++++++++++++++++++++++++- net/core/netclassid_cgroup.c | 7 +++- net/core/netprio_cgroup.c | 7 +++- net/core/sock.c | 2 + 6 files changed, 191 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index ed128fed0335..9dc226345e4e 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -544,31 +544,107 @@ static inline void cgroup_threadgroup_change_end(struct task_struct *tsk) {} #ifdef CONFIG_SOCK_CGROUP_DATA +/* + * sock_cgroup_data is embedded at sock->sk_cgrp_data and contains + * per-socket cgroup information except for memcg association. + * + * On legacy hierarchies, net_prio and net_cls controllers directly set + * attributes on each sock which can then be tested by the network layer. + * On the default hierarchy, each sock is associated with the cgroup it was + * created in and the networking layer can match the cgroup directly. + * + * To avoid carrying all three cgroup related fields separately in sock, + * sock_cgroup_data overloads (prioidx, classid) and the cgroup pointer. + * On boot, sock_cgroup_data records the cgroup that the sock was created + * in so that cgroup2 matches can be made; however, once either net_prio or + * net_cls starts being used, the area is overriden to carry prioidx and/or + * classid. The two modes are distinguished by whether the lowest bit is + * set. Clear bit indicates cgroup pointer while set bit prioidx and + * classid. + * + * While userland may start using net_prio or net_cls at any time, once + * either is used, cgroup2 matching no longer works. There is no reason to + * mix the two and this is in line with how legacy and v2 compatibility is + * handled. On mode switch, cgroup references which are already being + * pointed to by socks may be leaked. While this can be remedied by adding + * synchronization around sock_cgroup_data, given that the number of leaked + * cgroups is bound and highly unlikely to be high, this seems to be the + * better trade-off. + */ struct sock_cgroup_data { - u16 prioidx; - u32 classid; + union { +#ifdef __LITTLE_ENDIAN + struct { + u8 is_data; + u8 padding; + u16 prioidx; + u32 classid; + } __packed; +#else + struct { + u32 classid; + u16 prioidx; + u8 padding; + u8 is_data; + } __packed; +#endif + u64 val; + }; }; +/* + * There's a theoretical window where the following accessors race with + * updaters and return part of the previous pointer as the prioidx or + * classid. Such races are short-lived and the result isn't critical. + */ static inline u16 sock_cgroup_prioidx(struct sock_cgroup_data *skcd) { - return skcd->prioidx; + /* fallback to 1 which is always the ID of the root cgroup */ + return (skcd->is_data & 1) ? skcd->prioidx : 1; } static inline u32 sock_cgroup_classid(struct sock_cgroup_data *skcd) { - return skcd->classid; + /* fallback to 0 which is the unconfigured default classid */ + return (skcd->is_data & 1) ? skcd->classid : 0; } +/* + * If invoked concurrently, the updaters may clobber each other. The + * caller is responsible for synchronization. + */ static inline void sock_cgroup_set_prioidx(struct sock_cgroup_data *skcd, u16 prioidx) { - skcd->prioidx = prioidx; + struct sock_cgroup_data skcd_buf = { .val = READ_ONCE(skcd->val) }; + + if (sock_cgroup_prioidx(&skcd_buf) == prioidx) + return; + + if (!(skcd_buf.is_data & 1)) { + skcd_buf.val = 0; + skcd_buf.is_data = 1; + } + + skcd_buf.prioidx = prioidx; + WRITE_ONCE(skcd->val, skcd_buf.val); /* see sock_cgroup_ptr() */ } static inline void sock_cgroup_set_classid(struct sock_cgroup_data *skcd, u32 classid) { - skcd->classid = classid; + struct sock_cgroup_data skcd_buf = { .val = READ_ONCE(skcd->val) }; + + if (sock_cgroup_classid(&skcd_buf) == classid) + return; + + if (!(skcd_buf.is_data & 1)) { + skcd_buf.val = 0; + skcd_buf.is_data = 1; + } + + skcd_buf.classid = classid; + WRITE_ONCE(skcd->val, skcd_buf.val); /* see sock_cgroup_ptr() */ } #else /* CONFIG_SOCK_CGROUP_DATA */ diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 4c3ffab81ba7..a8ba1ea0ea5a 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -578,4 +578,45 @@ static inline int cgroup_init(void) { return 0; } #endif /* !CONFIG_CGROUPS */ +/* + * sock->sk_cgrp_data handling. For more info, see sock_cgroup_data + * definition in cgroup-defs.h. + */ +#ifdef CONFIG_SOCK_CGROUP_DATA + +#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID) +extern spinlock_t cgroup_sk_update_lock; +#endif + +void cgroup_sk_alloc_disable(void); +void cgroup_sk_alloc(struct sock_cgroup_data *skcd); +void cgroup_sk_free(struct sock_cgroup_data *skcd); + +static inline struct cgroup *sock_cgroup_ptr(struct sock_cgroup_data *skcd) +{ +#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID) + unsigned long v; + + /* + * @skcd->val is 64bit but the following is safe on 32bit too as we + * just need the lower ulong to be written and read atomically. + */ + v = READ_ONCE(skcd->val); + + if (v & 1) + return &cgrp_dfl_root.cgrp; + + return (struct cgroup *)(unsigned long)v ?: &cgrp_dfl_root.cgrp; +#else + return (struct cgroup *)(unsigned long)skcd->val; +#endif +} + +#else /* CONFIG_CGROUP_DATA */ + +static inline void cgroup_sk_alloc(struct sock_cgroup_data *skcd) {} +static inline void cgroup_sk_free(struct sock_cgroup_data *skcd) {} + +#endif /* CONFIG_CGROUP_DATA */ + #endif /* _LINUX_CGROUP_H */ diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 3db5e8f5b702..4f8f7927b422 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -57,8 +57,8 @@ #include /* TODO: replace with more sophisticated array */ #include #include - #include +#include /* * pidlists linger the following amount before being destroyed. The goal @@ -5782,6 +5782,59 @@ struct cgroup *cgroup_get_from_path(const char *path) } EXPORT_SYMBOL_GPL(cgroup_get_from_path); +/* + * sock->sk_cgrp_data handling. For more info, see sock_cgroup_data + * definition in cgroup-defs.h. + */ +#ifdef CONFIG_SOCK_CGROUP_DATA + +#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID) + +spinlock_t cgroup_sk_update_lock; +static bool cgroup_sk_alloc_disabled __read_mostly; + +void cgroup_sk_alloc_disable(void) +{ + if (cgroup_sk_alloc_disabled) + return; + pr_info("cgroup: disabling cgroup2 socket matching due to net_prio or net_cls activation\n"); + cgroup_sk_alloc_disabled = true; +} + +#else + +#define cgroup_sk_alloc_disabled false + +#endif + +void cgroup_sk_alloc(struct sock_cgroup_data *skcd) +{ + if (cgroup_sk_alloc_disabled) + return; + + rcu_read_lock(); + + while (true) { + struct css_set *cset; + + cset = task_css_set(current); + if (likely(cgroup_tryget(cset->dfl_cgrp))) { + skcd->val = (unsigned long)cset->dfl_cgrp; + break; + } + cpu_relax(); + } + + rcu_read_unlock(); +} + +void cgroup_sk_free(struct sock_cgroup_data *skcd) +{ + cgroup_put(sock_cgroup_ptr(skcd)); +} + +#endif /* CONFIG_SOCK_CGROUP_DATA */ + #ifdef CONFIG_CGROUP_DEBUG static struct cgroup_subsys_state * debug_css_alloc(struct cgroup_subsys_state *parent_css) diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c index e60ded46b3ac..04257a0e3534 100644 --- a/net/core/netclassid_cgroup.c +++ b/net/core/netclassid_cgroup.c @@ -61,9 +61,12 @@ static int update_classid_sock(const void *v, struct file *file, unsigned n) int err; struct socket *sock = sock_from_file(file, &err); - if (sock) + if (sock) { + spin_lock(&cgroup_sk_update_lock); sock_cgroup_set_classid(&sock->sk->sk_cgrp_data, (unsigned long)v); + spin_unlock(&cgroup_sk_update_lock); + } return 0; } @@ -98,6 +101,8 @@ static int write_classid(struct cgroup_subsys_state *css, struct cftype *cft, { struct cgroup_cls_state *cs = css_cls_state(css); + cgroup_sk_alloc_disable(); + cs->classid = (u32)value; update_classid(css, (void *)(unsigned long)cs->classid); diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index de42aa7f6c77..053d60c33395 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -209,6 +209,8 @@ static ssize_t write_priomap(struct kernfs_open_file *of, if (!dev) return -ENODEV; + cgroup_sk_alloc_disable(); + rtnl_lock(); ret = netprio_set_prio(of_css(of), dev, prio); @@ -222,9 +224,12 @@ static int update_netprio(const void *v, struct file *file, unsigned n) { int err; struct socket *sock = sock_from_file(file, &err); - if (sock) + if (sock) { + spin_lock(&cgroup_sk_update_lock); sock_cgroup_set_prioidx(&sock->sk->sk_cgrp_data, (unsigned long)v); + spin_unlock(&cgroup_sk_update_lock); + } return 0; } diff --git a/net/core/sock.c b/net/core/sock.c index 947741dc43fa..1278d7b7bd9a 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1363,6 +1363,7 @@ static struct sock *sk_prot_alloc(struct proto *prot, gfp_t priority, if (!try_module_get(prot->owner)) goto out_free_sec; sk_tx_queue_clear(sk); + cgroup_sk_alloc(&sk->sk_cgrp_data); } return sk; @@ -1385,6 +1386,7 @@ static void sk_prot_free(struct proto *prot, struct sock *sk) owner = prot->owner; slab = prot->slab; + cgroup_sk_free(&sk->sk_cgrp_data); security_sk_free(sk); if (slab != NULL) kmem_cache_free(slab, sk); -- cgit v1.2.3-59-g8ed1b From 33d5a7b14bfd02e60af9d223db8dfff0cbcabe6b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 28 Nov 2015 21:53:04 +0100 Subject: netfilter: nf_tables: extend tracing infrastructure nft monitor mode can then decode and display this trace data. Parts of LL/Network/Transport headers are provided as separate attributes. Otherwise, printing IP address data becomes virtually impossible for userspace since in the case of the netdev family we really don't want userspace to have to know all the possible link layer types and/or sizes just to display/print an ip address. We also don't want userspace to have to follow ipv6 header chains to get the s/dport info, the kernel already did this work for us. To avoid bloating nft_do_chain all data required for tracing is encapsulated in nft_traceinfo. The structure is initialized unconditionally(!) for each nft_do_chain invocation. This unconditionall call will be moved under a static key in a followup patch. With lots of help from Patrick McHardy and Pablo Neira. Signed-off-by: Florian Westphal Acked-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 32 ++++ include/uapi/linux/netfilter/nf_tables.h | 52 ++++++ include/uapi/linux/netfilter/nfnetlink.h | 2 + net/netfilter/Makefile | 2 +- net/netfilter/nf_tables_api.c | 12 +- net/netfilter/nf_tables_core.c | 45 +++-- net/netfilter/nf_tables_trace.c | 271 +++++++++++++++++++++++++++++++ net/netfilter/nfnetlink.c | 1 + 8 files changed, 398 insertions(+), 19 deletions(-) create mode 100644 net/netfilter/nf_tables_trace.c (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 101d7d7ec243..b313cda49194 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -888,6 +888,38 @@ void nft_unregister_chain_type(const struct nf_chain_type *); int nft_register_expr(struct nft_expr_type *); void nft_unregister_expr(struct nft_expr_type *); +int nft_verdict_dump(struct sk_buff *skb, int type, + const struct nft_verdict *v); + +/** + * struct nft_traceinfo - nft tracing information and state + * + * @pkt: pktinfo currently processed + * @basechain: base chain currently processed + * @chain: chain currently processed + * @rule: rule that was evaluated + * @verdict: verdict given by rule + * @type: event type (enum nft_trace_types) + * @packet_dumped: packet headers sent in a previous traceinfo message + * @trace: other struct members are initialised + */ +struct nft_traceinfo { + const struct nft_pktinfo *pkt; + const struct nft_base_chain *basechain; + const struct nft_chain *chain; + const struct nft_rule *rule; + const struct nft_verdict *verdict; + enum nft_trace_types type; + bool packet_dumped; + bool trace; +}; + +void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, + const struct nft_verdict *verdict, + const struct nft_chain *basechain); + +void nft_trace_notify(struct nft_traceinfo *info); + #define nft_dereference(p) \ nfnl_dereference(p, NFNL_SUBSYS_NFTABLES) diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 5f3ececf84b3..b48a3ab761f8 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -83,6 +83,7 @@ enum nft_verdicts { * @NFT_MSG_DELSETELEM: delete a set element (enum nft_set_elem_attributes) * @NFT_MSG_NEWGEN: announce a new generation, only for events (enum nft_gen_attributes) * @NFT_MSG_GETGEN: get the rule-set generation (enum nft_gen_attributes) + * @NFT_MSG_TRACE: trace event (enum nft_trace_attributes) */ enum nf_tables_msg_types { NFT_MSG_NEWTABLE, @@ -102,6 +103,7 @@ enum nf_tables_msg_types { NFT_MSG_DELSETELEM, NFT_MSG_NEWGEN, NFT_MSG_GETGEN, + NFT_MSG_TRACE, NFT_MSG_MAX, }; @@ -987,4 +989,54 @@ enum nft_gen_attributes { }; #define NFTA_GEN_MAX (__NFTA_GEN_MAX - 1) +/** + * enum nft_trace_attributes - nf_tables trace netlink attributes + * + * @NFTA_TRACE_TABLE: name of the table (NLA_STRING) + * @NFTA_TRACE_CHAIN: name of the chain (NLA_STRING) + * @NFTA_TRACE_RULE_HANDLE: numeric handle of the rule (NLA_U64) + * @NFTA_TRACE_TYPE: type of the event (NLA_U32: nft_trace_types) + * @NFTA_TRACE_VERDICT: verdict returned by hook (NLA_NESTED: nft_verdicts) + * @NFTA_TRACE_ID: pseudo-id, same for each skb traced (NLA_U32) + * @NFTA_TRACE_LL_HEADER: linklayer header (NLA_BINARY) + * @NFTA_TRACE_NETWORK_HEADER: network header (NLA_BINARY) + * @NFTA_TRACE_TRANSPORT_HEADER: transport header (NLA_BINARY) + * @NFTA_TRACE_IIF: indev ifindex (NLA_U32) + * @NFTA_TRACE_IIFTYPE: netdev->type of indev (NLA_U16) + * @NFTA_TRACE_OIF: outdev ifindex (NLA_U32) + * @NFTA_TRACE_OIFTYPE: netdev->type of outdev (NLA_U16) + * @NFTA_TRACE_MARK: nfmark (NLA_U32) + * @NFTA_TRACE_NFPROTO: nf protocol processed (NLA_U32) + * @NFTA_TRACE_POLICY: policy that decided fate of packet (NLA_U32) + */ +enum nft_trace_attibutes { + NFTA_TRACE_UNSPEC, + NFTA_TRACE_TABLE, + NFTA_TRACE_CHAIN, + NFTA_TRACE_RULE_HANDLE, + NFTA_TRACE_TYPE, + NFTA_TRACE_VERDICT, + NFTA_TRACE_ID, + NFTA_TRACE_LL_HEADER, + NFTA_TRACE_NETWORK_HEADER, + NFTA_TRACE_TRANSPORT_HEADER, + NFTA_TRACE_IIF, + NFTA_TRACE_IIFTYPE, + NFTA_TRACE_OIF, + NFTA_TRACE_OIFTYPE, + NFTA_TRACE_MARK, + NFTA_TRACE_NFPROTO, + NFTA_TRACE_POLICY, + __NFTA_TRACE_MAX +}; +#define NFTA_TRACE_MAX (__NFTA_TRACE_MAX - 1) + +enum nft_trace_types { + NFT_TRACETYPE_UNSPEC, + NFT_TRACETYPE_POLICY, + NFT_TRACETYPE_RETURN, + NFT_TRACETYPE_RULE, + __NFT_TRACETYPE_MAX +}; +#define NFT_TRACETYPE_MAX (__NFT_TRACETYPE_MAX - 1) #endif /* _LINUX_NF_TABLES_H */ diff --git a/include/uapi/linux/netfilter/nfnetlink.h b/include/uapi/linux/netfilter/nfnetlink.h index 354a7e5e50f2..4bb8cb7730e7 100644 --- a/include/uapi/linux/netfilter/nfnetlink.h +++ b/include/uapi/linux/netfilter/nfnetlink.h @@ -22,6 +22,8 @@ enum nfnetlink_groups { #define NFNLGRP_NFTABLES NFNLGRP_NFTABLES NFNLGRP_ACCT_QUOTA, #define NFNLGRP_ACCT_QUOTA NFNLGRP_ACCT_QUOTA + NFNLGRP_NFTRACE, +#define NFNLGRP_NFTRACE NFNLGRP_NFTRACE __NFNLGRP_MAX, }; #define NFNLGRP_MAX (__NFNLGRP_MAX - 1) diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 7638c36b498c..22934846b5d1 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -67,7 +67,7 @@ obj-$(CONFIG_NF_NAT_TFTP) += nf_nat_tftp.o obj-$(CONFIG_NETFILTER_SYNPROXY) += nf_synproxy_core.o # nf_tables -nf_tables-objs += nf_tables_core.o nf_tables_api.o +nf_tables-objs += nf_tables_core.o nf_tables_api.o nf_tables_trace.o nf_tables-objs += nft_immediate.o nft_cmp.o nft_lookup.o nft_dynset.o nf_tables-objs += nft_bitwise.o nft_byteorder.o nft_payload.o diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 93cc4737018f..c4969a0d54ba 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4446,22 +4446,22 @@ static void nft_verdict_uninit(const struct nft_data *data) } } -static int nft_verdict_dump(struct sk_buff *skb, const struct nft_data *data) +int nft_verdict_dump(struct sk_buff *skb, int type, const struct nft_verdict *v) { struct nlattr *nest; - nest = nla_nest_start(skb, NFTA_DATA_VERDICT); + nest = nla_nest_start(skb, type); if (!nest) goto nla_put_failure; - if (nla_put_be32(skb, NFTA_VERDICT_CODE, htonl(data->verdict.code))) + if (nla_put_be32(skb, NFTA_VERDICT_CODE, htonl(v->code))) goto nla_put_failure; - switch (data->verdict.code) { + switch (v->code) { case NFT_JUMP: case NFT_GOTO: if (nla_put_string(skb, NFTA_VERDICT_CHAIN, - data->verdict.chain->name)) + v->chain->name)) goto nla_put_failure; } nla_nest_end(skb, nest); @@ -4572,7 +4572,7 @@ int nft_data_dump(struct sk_buff *skb, int attr, const struct nft_data *data, err = nft_value_dump(skb, data, len); break; case NFT_DATA_VERDICT: - err = nft_verdict_dump(skb, data); + err = nft_verdict_dump(skb, NFTA_DATA_VERDICT, &data->verdict); break; default: err = -EINVAL; diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index f3695a497408..2395de7c8ab2 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -44,22 +44,36 @@ static struct nf_loginfo trace_loginfo = { }, }; -static void __nft_trace_packet(const struct nft_pktinfo *pkt, - const struct nft_chain *chain, - int rulenum, enum nft_trace type) +static noinline void __nft_trace_packet(struct nft_traceinfo *info, + const struct nft_chain *chain, + int rulenum, enum nft_trace type) { + const struct nft_pktinfo *pkt = info->pkt; + + if (!pkt->skb->nf_trace) + return; + + info->chain = chain; + info->type = type; + + nft_trace_notify(info); + nf_log_trace(pkt->net, pkt->pf, pkt->hook, pkt->skb, pkt->in, pkt->out, &trace_loginfo, "TRACE: %s:%s:%s:%u ", chain->table->name, chain->name, comments[type], rulenum); } -static inline void nft_trace_packet(const struct nft_pktinfo *pkt, +static inline void nft_trace_packet(struct nft_traceinfo *info, const struct nft_chain *chain, - int rulenum, enum nft_trace type) + const struct nft_rule *rule, + int rulenum, + enum nft_trace_types type) { - if (unlikely(pkt->skb->nf_trace)) - __nft_trace_packet(pkt, chain, rulenum, type); + if (unlikely(info->trace)) { + info->rule = rule; + __nft_trace_packet(info, chain, rulenum, type); + } } static void nft_cmp_fast_eval(const struct nft_expr *expr, @@ -121,7 +135,9 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv) struct nft_stats *stats; int rulenum; unsigned int gencursor = nft_genmask_cur(net); + struct nft_traceinfo info; + nft_trace_init(&info, pkt, ®s.verdict, basechain); do_chain: rulenum = 0; rule = list_entry(&chain->rules, struct nft_rule, list); @@ -151,7 +167,8 @@ next_rule: regs.verdict.code = NFT_CONTINUE; continue; case NFT_CONTINUE: - nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); + nft_trace_packet(&info, chain, rule, + rulenum, NFT_TRACETYPE_RULE); continue; } break; @@ -161,7 +178,8 @@ next_rule: case NF_ACCEPT: case NF_DROP: case NF_QUEUE: - nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); + nft_trace_packet(&info, chain, rule, + rulenum, NFT_TRACETYPE_RULE); return regs.verdict.code; } @@ -174,7 +192,8 @@ next_rule: stackptr++; /* fall through */ case NFT_GOTO: - nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); + nft_trace_packet(&info, chain, rule, + rulenum, NFT_TRACETYPE_RULE); chain = regs.verdict.chain; goto do_chain; @@ -182,7 +201,8 @@ next_rule: rulenum++; /* fall through */ case NFT_RETURN: - nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RETURN); + nft_trace_packet(&info, chain, rule, + rulenum, NFT_TRACETYPE_RETURN); break; default: WARN_ON(1); @@ -196,7 +216,8 @@ next_rule: goto next_rule; } - nft_trace_packet(pkt, basechain, -1, NFT_TRACE_POLICY); + 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)); diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c new file mode 100644 index 000000000000..36fd7ad6729a --- /dev/null +++ b/net/netfilter/nf_tables_trace.c @@ -0,0 +1,271 @@ +/* + * (C) 2015 Red Hat GmbH + * Author: Florian Westphal + * + * 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 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define NFT_TRACETYPE_LL_HSIZE 20 +#define NFT_TRACETYPE_NETWORK_HSIZE 40 +#define NFT_TRACETYPE_TRANSPORT_HSIZE 20 + +static int trace_fill_id(struct sk_buff *nlskb, struct sk_buff *skb) +{ + __be32 id; + + /* using skb address as ID results in a limited number of + * values (and quick reuse). + * + * So we attempt to use as many skb members that will not + * change while skb is with netfilter. + */ + id = (__be32)jhash_2words(hash32_ptr(skb), skb_get_hash(skb), + skb->skb_iif); + + return nla_put_be32(nlskb, NFTA_TRACE_ID, id); +} + +static int trace_fill_header(struct sk_buff *nlskb, u16 type, + const struct sk_buff *skb, + int off, unsigned int len) +{ + struct nlattr *nla; + + if (len == 0) + return 0; + + nla = nla_reserve(nlskb, type, len); + if (!nla || skb_copy_bits(skb, off, nla_data(nla), len)) + return -1; + + return 0; +} + +static int nf_trace_fill_ll_header(struct sk_buff *nlskb, + const struct sk_buff *skb) +{ + struct vlan_ethhdr veth; + int off; + + BUILD_BUG_ON(sizeof(veth) > NFT_TRACETYPE_LL_HSIZE); + + off = skb_mac_header(skb) - skb->data; + if (off != -ETH_HLEN) + return -1; + + if (skb_copy_bits(skb, off, &veth, ETH_HLEN)) + return -1; + + veth.h_vlan_proto = skb->vlan_proto; + veth.h_vlan_TCI = htons(skb_vlan_tag_get(skb)); + veth.h_vlan_encapsulated_proto = skb->protocol; + + return nla_put(nlskb, NFTA_TRACE_LL_HEADER, sizeof(veth), &veth); +} + +static int nf_trace_fill_dev_info(struct sk_buff *nlskb, + const struct net_device *indev, + const struct net_device *outdev) +{ + if (indev) { + if (nla_put_be32(nlskb, NFTA_TRACE_IIF, + htonl(indev->ifindex))) + return -1; + + if (nla_put_be16(nlskb, NFTA_TRACE_IIFTYPE, + htons(indev->type))) + return -1; + } + + if (outdev) { + if (nla_put_be32(nlskb, NFTA_TRACE_OIF, + htonl(outdev->ifindex))) + return -1; + + if (nla_put_be16(nlskb, NFTA_TRACE_OIFTYPE, + htons(outdev->type))) + return -1; + } + + return 0; +} + +static int nf_trace_fill_pkt_info(struct sk_buff *nlskb, + const struct nft_pktinfo *pkt) +{ + const struct sk_buff *skb = pkt->skb; + unsigned int len = min_t(unsigned int, + pkt->xt.thoff - skb_network_offset(skb), + NFT_TRACETYPE_NETWORK_HSIZE); + int off = skb_network_offset(skb); + + if (trace_fill_header(nlskb, NFTA_TRACE_NETWORK_HEADER, skb, off, len)) + return -1; + + len = min_t(unsigned int, skb->len - pkt->xt.thoff, + NFT_TRACETYPE_TRANSPORT_HSIZE); + + if (trace_fill_header(nlskb, NFTA_TRACE_TRANSPORT_HEADER, skb, + pkt->xt.thoff, len)) + return -1; + + if (!skb_mac_header_was_set(skb)) + return 0; + + if (skb_vlan_tag_get(skb)) + return nf_trace_fill_ll_header(nlskb, skb); + + off = skb_mac_header(skb) - skb->data; + len = min_t(unsigned int, -off, NFT_TRACETYPE_LL_HSIZE); + return trace_fill_header(nlskb, NFTA_TRACE_LL_HEADER, + skb, off, len); +} + +static int nf_trace_fill_rule_info(struct sk_buff *nlskb, + const struct nft_traceinfo *info) +{ + if (!info->rule) + return 0; + + /* a continue verdict with ->type == RETURN means that this is + * an implicit return (end of chain reached). + * + * Since no rule matched, the ->rule pointer is invalid. + */ + if (info->type == NFT_TRACETYPE_RETURN && + info->verdict->code == NFT_CONTINUE) + return 0; + + return nla_put_be64(nlskb, NFTA_TRACE_RULE_HANDLE, + cpu_to_be64(info->rule->handle)); +} + +void nft_trace_notify(struct nft_traceinfo *info) +{ + const struct nft_pktinfo *pkt = info->pkt; + struct nfgenmsg *nfmsg; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned int size; + int event = (NFNL_SUBSYS_NFTABLES << 8) | NFT_MSG_TRACE; + + if (!nfnetlink_has_listeners(pkt->net, NFNLGRP_NFTRACE)) + return; + + size = nlmsg_total_size(sizeof(struct nfgenmsg)) + + nla_total_size(NFT_TABLE_MAXNAMELEN) + + nla_total_size(NFT_CHAIN_MAXNAMELEN) + + nla_total_size(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) + + nla_total_size(NFT_TRACETYPE_TRANSPORT_HSIZE) + + nla_total_size(sizeof(u32)) + /* iif */ + nla_total_size(sizeof(__be16)) + /* iiftype */ + nla_total_size(sizeof(u32)) + /* oif */ + nla_total_size(sizeof(__be16)) + /* oiftype */ + nla_total_size(sizeof(u32)) + /* mark */ + nla_total_size(sizeof(u32)) + /* nfproto */ + nla_total_size(sizeof(u32)); /* policy */ + + skb = nlmsg_new(size, GFP_ATOMIC); + if (!skb) + return; + + nlh = nlmsg_put(skb, 0, 0, event, sizeof(struct nfgenmsg), 0); + if (!nlh) + goto nla_put_failure; + + nfmsg = nlmsg_data(nlh); + nfmsg->nfgen_family = info->basechain->type->family; + nfmsg->version = NFNETLINK_V0; + nfmsg->res_id = 0; + + if (nla_put_be32(skb, NFTA_TRACE_NFPROTO, htonl(pkt->pf))) + goto nla_put_failure; + + if (nla_put_be32(skb, NFTA_TRACE_TYPE, htonl(info->type))) + goto nla_put_failure; + + 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 (nf_trace_fill_rule_info(skb, info)) + goto nla_put_failure; + + switch (info->type) { + case NFT_TRACETYPE_UNSPEC: + case __NFT_TRACETYPE_MAX: + break; + case NFT_TRACETYPE_RETURN: + case NFT_TRACETYPE_RULE: + if (nft_verdict_dump(skb, NFTA_TRACE_VERDICT, info->verdict)) + goto nla_put_failure; + break; + case NFT_TRACETYPE_POLICY: + if (nla_put_be32(skb, NFTA_TRACE_POLICY, + info->basechain->policy)) + goto nla_put_failure; + break; + } + + if (pkt->skb->mark && + nla_put_be32(skb, NFTA_TRACE_MARK, htonl(pkt->skb->mark))) + goto nla_put_failure; + + if (!info->packet_dumped) { + if (nf_trace_fill_dev_info(skb, pkt->in, pkt->out)) + goto nla_put_failure; + + if (nf_trace_fill_pkt_info(skb, pkt)) + goto nla_put_failure; + info->packet_dumped = true; + } + + nlmsg_end(skb, nlh); + nfnetlink_send(skb, pkt->net, 0, NFNLGRP_NFTRACE, 0, GFP_ATOMIC); + return; + + nla_put_failure: + WARN_ON_ONCE(1); + kfree_skb(skb); +} + +void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, + const struct nft_verdict *verdict, + const struct nft_chain *chain) +{ + info->basechain = nft_base_chain(chain); + info->trace = true; + info->packet_dumped = false; + info->pkt = pkt; + info->verdict = verdict; +} diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 46453ab318db..28591fa94ba5 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -49,6 +49,7 @@ static const int nfnl_group2type[NFNLGRP_MAX+1] = { [NFNLGRP_CONNTRACK_EXP_DESTROY] = NFNL_SUBSYS_CTNETLINK_EXP, [NFNLGRP_NFTABLES] = NFNL_SUBSYS_NFTABLES, [NFNLGRP_ACCT_QUOTA] = NFNL_SUBSYS_ACCT, + [NFNLGRP_NFTRACE] = NFNL_SUBSYS_NFTABLES, }; void nfnl_lock(__u8 subsys_id) -- cgit v1.2.3-59-g8ed1b From e639f7ab079b5256660018511d87aa34b54f1a9d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 28 Nov 2015 21:53:05 +0100 Subject: netfilter: nf_tables: wrap tracing with a static key Only needed when meta nftrace rule(s) were added. The assumption is that no such rules are active, so the call to nft_trace_init is "never" needed. When nftrace rules are active, we always call the nft_trace_* functions, but will only send netlink messages when all of the following are true: - traceinfo structure was initialised - skb->nf_trace == 1 - at least one subscriber to trace group. Adding an extra conditional (static_branch ... && skb->nf_trace) nft_trace_init( ..) Is possible but results in a larger nft_do_chain footprint. Signed-off-by: Florian Westphal Acked-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables_core.h | 1 + include/net/netfilter/nft_meta.h | 3 +++ net/bridge/netfilter/nft_meta_bridge.c | 1 + net/netfilter/nf_tables_core.c | 9 ++++++--- net/netfilter/nf_tables_trace.c | 4 ++++ net/netfilter/nft_meta.c | 16 ++++++++++++++++ 6 files changed, 31 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h index 4ff5424909aa..a9060dd99db7 100644 --- a/include/net/netfilter/nf_tables_core.h +++ b/include/net/netfilter/nf_tables_core.h @@ -57,6 +57,7 @@ struct nft_payload_set { }; extern const struct nft_expr_ops nft_payload_fast_ops; +extern struct static_key_false nft_trace_enabled; int nft_payload_module_init(void); void nft_payload_module_exit(void); diff --git a/include/net/netfilter/nft_meta.h b/include/net/netfilter/nft_meta.h index 711887a09e91..d27588c8dbd9 100644 --- a/include/net/netfilter/nft_meta.h +++ b/include/net/netfilter/nft_meta.h @@ -33,4 +33,7 @@ void nft_meta_set_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt); +void nft_meta_set_destroy(const struct nft_ctx *ctx, + const struct nft_expr *expr); + #endif diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c index a21269b83f16..4b901d9f2e7c 100644 --- a/net/bridge/netfilter/nft_meta_bridge.c +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -84,6 +84,7 @@ static const struct nft_expr_ops nft_meta_bridge_set_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_set_eval, .init = nft_meta_set_init, + .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, }; diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 2395de7c8ab2..67fa41d317f6 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -50,7 +51,7 @@ static noinline void __nft_trace_packet(struct nft_traceinfo *info, { const struct nft_pktinfo *pkt = info->pkt; - if (!pkt->skb->nf_trace) + if (!info->trace || !pkt->skb->nf_trace) return; info->chain = chain; @@ -70,7 +71,7 @@ static inline void nft_trace_packet(struct nft_traceinfo *info, int rulenum, enum nft_trace_types type) { - if (unlikely(info->trace)) { + if (static_branch_unlikely(&nft_trace_enabled)) { info->rule = rule; __nft_trace_packet(info, chain, rulenum, type); } @@ -137,7 +138,9 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv) unsigned int gencursor = nft_genmask_cur(net); struct nft_traceinfo info; - nft_trace_init(&info, pkt, ®s.verdict, basechain); + info.trace = false; + if (static_branch_unlikely(&nft_trace_enabled)) + nft_trace_init(&info, pkt, ®s.verdict, basechain); do_chain: rulenum = 0; rule = list_entry(&chain->rules, struct nft_rule, list); diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index 36fd7ad6729a..e9e959f65d91 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -8,6 +8,7 @@ */ #include +#include #include #include #include @@ -24,6 +25,9 @@ #define NFT_TRACETYPE_NETWORK_HSIZE 40 #define NFT_TRACETYPE_TRANSPORT_HSIZE 20 +DEFINE_STATIC_KEY_FALSE(nft_trace_enabled); +EXPORT_SYMBOL_GPL(nft_trace_enabled); + static int trace_fill_id(struct sk_buff *nlskb, struct sk_buff *skb) { __be32 id; diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 9dfaf4d55ee0..85a465b773e5 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -18,10 +18,12 @@ #include #include #include +#include #include #include #include /* for TCP_TIME_WAIT */ #include +#include #include void nft_meta_get_eval(const struct nft_expr *expr, @@ -297,6 +299,9 @@ int nft_meta_set_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (priv->key == NFT_META_NFTRACE) + static_branch_inc(&nft_trace_enabled); + return 0; } EXPORT_SYMBOL_GPL(nft_meta_set_init); @@ -334,6 +339,16 @@ nla_put_failure: } EXPORT_SYMBOL_GPL(nft_meta_set_dump); +void nft_meta_set_destroy(const struct nft_ctx *ctx, + const struct nft_expr *expr) +{ + const struct nft_meta *priv = nft_expr_priv(expr); + + if (priv->key == NFT_META_NFTRACE) + static_branch_dec(&nft_trace_enabled); +} +EXPORT_SYMBOL_GPL(nft_meta_set_destroy); + static struct nft_expr_type nft_meta_type; static const struct nft_expr_ops nft_meta_get_ops = { .type = &nft_meta_type, @@ -348,6 +363,7 @@ static const struct nft_expr_ops nft_meta_set_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_set_eval, .init = nft_meta_set_init, + .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, }; -- cgit v1.2.3-59-g8ed1b From ad2c8c73d29702c3193f739390f6661f9a4ecad9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 9 Dec 2015 12:30:46 -0500 Subject: cgroup: fix sock_cgroup_data initialization on earlier compilers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit sock_cgroup_data is a struct containing an anonymous union. sock_cgroup_set_prioidx() and sock_cgroup_set_classid() were initializing a field inside the anonymous union as follows. struct sock_ccgroup_data skcd_buf = { .val = VAL }; While this is fine on more recent compilers, gcc-4.4.7 triggers the following errors. include/linux/cgroup-defs.h: In function ‘sock_cgroup_set_prioidx’: include/linux/cgroup-defs.h:619: error: unknown field ‘val’ specified in initializer include/linux/cgroup-defs.h:619: warning: missing braces around initializer include/linux/cgroup-defs.h:619: warning: (near initialization for ‘skcd_buf.’) This is because .val belongs to the anonymous union nested inside the struct but the initializer is missing the nesting. Fix it by adding an extra pair of braces. Signed-off-by: Tejun Heo Reported-by: Alaa Hleihel Fixes: bd1060a1d671 ("sock, cgroup: add sock->sk_cgroup") Signed-off-by: David S. Miller --- include/linux/cgroup-defs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 9dc226345e4e..097901a68671 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -616,7 +616,7 @@ static inline u32 sock_cgroup_classid(struct sock_cgroup_data *skcd) static inline void sock_cgroup_set_prioidx(struct sock_cgroup_data *skcd, u16 prioidx) { - struct sock_cgroup_data skcd_buf = { .val = READ_ONCE(skcd->val) }; + struct sock_cgroup_data skcd_buf = {{ .val = READ_ONCE(skcd->val) }}; if (sock_cgroup_prioidx(&skcd_buf) == prioidx) return; @@ -633,7 +633,7 @@ static inline void sock_cgroup_set_prioidx(struct sock_cgroup_data *skcd, static inline void sock_cgroup_set_classid(struct sock_cgroup_data *skcd, u32 classid) { - struct sock_cgroup_data skcd_buf = { .val = READ_ONCE(skcd->val) }; + struct sock_cgroup_data skcd_buf = {{ .val = READ_ONCE(skcd->val) }}; if (sock_cgroup_classid(&skcd_buf) == classid) return; -- cgit v1.2.3-59-g8ed1b From 01b1cb87d37fb19cdaa5e7002416fdde156873d0 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Mon, 16 Nov 2015 12:52:21 +0200 Subject: Bluetooth: Run page scan updates through hdev->req_workqueue Since Add/Remove Device perform the page scan updates independently from the HCI command completion we've introduced a potential race when multiple mgmt commands are queued. Doing the page scan updates through the req_workqueue ensures that the state changes are performed in a race-free manner. At the same time, to make the request helper more widely usable, extend it to also cover Inquiry Scan changes since those are behind the same HCI command. This is also reflected in the new name of the API as well as the work struct name. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_event.c | 4 ++-- net/bluetooth/hci_request.c | 27 ++++++++++++++++++--------- net/bluetooth/hci_request.h | 8 ++++++-- net/bluetooth/mgmt.c | 16 ++++++++-------- 5 files changed, 35 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 55ce209157b1..eda809a5c3df 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -329,6 +329,7 @@ struct hci_dev { struct work_struct discov_update; struct work_struct bg_scan_update; + struct work_struct scan_update; struct delayed_work le_scan_disable; struct delayed_work le_scan_restart; diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d57c11c1c6b5..703e37f1a955 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2176,7 +2176,7 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) hci_send_cmd(hdev, HCI_OP_READ_REMOTE_FEATURES, sizeof(cp), &cp); - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); } /* Set packet type for incoming connection */ @@ -2362,7 +2362,7 @@ static void hci_disconn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) if (test_bit(HCI_CONN_FLUSH_KEY, &conn->flags)) hci_remove_link_key(hdev, &conn->dst); - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); } params = hci_conn_params_lookup(hdev, &conn->dst, conn->dst_type); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index e639671f54bd..78c026b4ffa1 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -637,7 +637,7 @@ static bool disconnected_whitelist_entries(struct hci_dev *hdev) return false; } -void __hci_update_page_scan(struct hci_request *req) +void __hci_req_update_scan(struct hci_request *req) { struct hci_dev *hdev = req->hdev; u8 scan; @@ -657,22 +657,29 @@ void __hci_update_page_scan(struct hci_request *req) else scan = SCAN_DISABLED; - if (test_bit(HCI_PSCAN, &hdev->flags) == !!(scan & SCAN_PAGE)) - return; - if (hci_dev_test_flag(hdev, HCI_DISCOVERABLE)) scan |= SCAN_INQUIRY; + if (test_bit(HCI_PSCAN, &hdev->flags) == !!(scan & SCAN_PAGE) && + test_bit(HCI_ISCAN, &hdev->flags) == !!(scan & SCAN_INQUIRY)) + return; + hci_req_add(req, HCI_OP_WRITE_SCAN_ENABLE, 1, &scan); } -void hci_update_page_scan(struct hci_dev *hdev) +static int update_scan(struct hci_request *req, unsigned long opt) { - struct hci_request req; + hci_dev_lock(req->hdev); + __hci_req_update_scan(req); + hci_dev_unlock(req->hdev); + return 0; +} - hci_req_init(&req, hdev); - __hci_update_page_scan(&req); - hci_req_run(&req, NULL); +static void scan_update_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, scan_update); + + hci_req_sync(hdev, update_scan, 0, HCI_CMD_TIMEOUT, NULL); } /* This function controls the background scanning based on hdev->pend_le_conns @@ -1270,6 +1277,7 @@ void hci_request_setup(struct hci_dev *hdev) { INIT_WORK(&hdev->discov_update, discov_update); INIT_WORK(&hdev->bg_scan_update, bg_scan_update); + INIT_WORK(&hdev->scan_update, scan_update_work); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); } @@ -1280,6 +1288,7 @@ void hci_request_cancel_all(struct hci_dev *hdev) cancel_work_sync(&hdev->discov_update); cancel_work_sync(&hdev->bg_scan_update); + cancel_work_sync(&hdev->scan_update); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); } diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 6b9e59f7f7a9..cc8275520fb2 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -61,8 +61,12 @@ void hci_req_add_le_passive_scan(struct hci_request *req); /* Returns true if HCI commands were queued */ bool hci_req_stop_discovery(struct hci_request *req); -void hci_update_page_scan(struct hci_dev *hdev); -void __hci_update_page_scan(struct hci_request *req); +static inline void hci_req_update_scan(struct hci_dev *hdev) +{ + queue_work(hdev->req_workqueue, &hdev->scan_update); +} + +void __hci_req_update_scan(struct hci_request *req); int hci_update_random_address(struct hci_request *req, bool require_privacy, u8 *own_addr_type); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 3d9d2e4839c5..0d20e1328528 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1810,7 +1810,7 @@ static void set_discoverable_complete(struct hci_dev *hdev, u8 status, * entries. */ hci_req_init(&req, hdev); - __hci_update_page_scan(&req); + __hci_req_update_scan(&req); update_class(&req); hci_req_run(&req, NULL); @@ -2058,7 +2058,7 @@ static void set_connectable_complete(struct hci_dev *hdev, u8 status, if (conn_changed || discov_changed) { new_settings(hdev, cmd->sk); - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); if (discov_changed) mgmt_update_adv_data(hdev); hci_update_background_scan(hdev); @@ -2092,7 +2092,7 @@ static int set_connectable_update_settings(struct hci_dev *hdev, return err; if (changed) { - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); hci_update_background_scan(hdev); return new_settings(hdev, sk); } @@ -5041,7 +5041,7 @@ static int set_bredr(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) hci_req_init(&req, hdev); write_fast_connectable(&req, false); - __hci_update_page_scan(&req); + __hci_req_update_scan(&req); /* Since only the advertising data flags will change, there * is no need to update the scan response data. @@ -5927,7 +5927,7 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, if (err) goto unlock; - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); goto added; } @@ -6024,7 +6024,7 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, goto unlock; } - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); device_removed(sk, hdev, &cp->addr.bdaddr, cp->addr.type); @@ -6089,7 +6089,7 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, kfree(b); } - hci_update_page_scan(hdev); + hci_req_update_scan(hdev); list_for_each_entry_safe(p, tmp, &hdev->le_conn_params, list) { if (p->auto_connect == HCI_AUTO_CONN_DISABLED) @@ -7397,7 +7397,7 @@ static int powered_update_hci(struct hci_dev *hdev) write_fast_connectable(&req, true); else write_fast_connectable(&req, false); - __hci_update_page_scan(&req); + __hci_req_update_scan(&req); update_class(&req); update_name(&req); update_eir(&req); -- cgit v1.2.3-59-g8ed1b From f22525700b2ae34eb97a29a91e2eee902062b484 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 18 Nov 2015 12:49:20 +0200 Subject: Bluetooth: Move advertising instance management to hci_request.c This paves the way for eventually performing advertising changes through the hdev->req_workqueue. Some new APIs need to be exposed from mgmt.c to hci_request.c and vice-versa, but many of them will go away once hdev->req_workqueue gets used. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 9 +- net/bluetooth/hci_conn.c | 2 +- net/bluetooth/hci_core.c | 19 +- net/bluetooth/hci_event.c | 4 +- net/bluetooth/hci_request.c | 533 +++++++++++++++++++++++++++++++++++- net/bluetooth/hci_request.h | 14 + net/bluetooth/mgmt.c | 574 +++------------------------------------ 7 files changed, 589 insertions(+), 566 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index eda809a5c3df..b56085b6ecce 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1434,9 +1434,7 @@ void mgmt_index_added(struct hci_dev *hdev); void mgmt_index_removed(struct hci_dev *hdev); void mgmt_set_powered_failed(struct hci_dev *hdev, int err); int mgmt_powered(struct hci_dev *hdev, u8 powered); -int mgmt_update_adv_data(struct hci_dev *hdev); void mgmt_discoverable_timeout(struct hci_dev *hdev); -void mgmt_adv_timeout_expired(struct hci_dev *hdev); void mgmt_new_link_key(struct hci_dev *hdev, struct link_key *key, bool persistent); void mgmt_device_connected(struct hci_dev *hdev, struct hci_conn *conn, @@ -1491,8 +1489,13 @@ void mgmt_new_csrk(struct hci_dev *hdev, struct smp_csrk *csrk, void mgmt_new_conn_param(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 bdaddr_type, u8 store_hint, u16 min_interval, u16 max_interval, u16 latency, u16 timeout); -void mgmt_reenable_advertising(struct hci_dev *hdev); void mgmt_smp_complete(struct hci_conn *conn, bool complete); +bool mgmt_get_connectable(struct hci_dev *hdev); +u8 mgmt_get_adv_discov_flags(struct hci_dev *hdev); +void mgmt_advertising_added(struct sock *sk, struct hci_dev *hdev, + u8 instance); +void mgmt_advertising_removed(struct sock *sk, struct hci_dev *hdev, + u8 instance); u8 hci_le_conn_update(struct hci_conn *conn, u16 min, u16 max, u16 latency, u16 to_multiplier); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 2d334e07fd77..e2600213cd50 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -683,7 +683,7 @@ void hci_le_conn_failed(struct hci_conn *conn, u8 status) /* Re-enable advertising in case this was a failed connection * attempt as a peripheral. */ - mgmt_reenable_advertising(hdev); + hci_req_reenable_advertising(hdev); } static void create_le_conn_complete(struct hci_dev *hdev, u8 status, u16 opcode) diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 89af7e4fac02..bab8958bf46e 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1549,11 +1549,6 @@ int hci_dev_do_close(struct hci_dev *hdev) if (hci_dev_test_flag(hdev, HCI_MGMT)) cancel_delayed_work_sync(&hdev->rpa_expired); - if (hdev->adv_instance_timeout) { - cancel_delayed_work_sync(&hdev->adv_instance_expire); - hdev->adv_instance_timeout = 0; - } - /* Avoid potential lockdep warnings from the *_flush() calls by * ensuring the workqueue is empty up front. */ @@ -1774,7 +1769,7 @@ static void hci_update_scan_state(struct hci_dev *hdev, u8 scan) hci_dev_set_flag(hdev, HCI_BREDR_ENABLED); if (hci_dev_test_flag(hdev, HCI_LE_ENABLED)) - mgmt_update_adv_data(hdev); + hci_req_update_adv_data(hdev, HCI_ADV_CURRENT); mgmt_new_settings(hdev); } @@ -2112,17 +2107,6 @@ static void hci_discov_off(struct work_struct *work) mgmt_discoverable_timeout(hdev); } -static void hci_adv_timeout_expire(struct work_struct *work) -{ - struct hci_dev *hdev; - - hdev = container_of(work, struct hci_dev, adv_instance_expire.work); - - BT_DBG("%s", hdev->name); - - mgmt_adv_timeout_expired(hdev); -} - void hci_uuids_clear(struct hci_dev *hdev) { struct bt_uuid *uuid, *tmp; @@ -3003,7 +2987,6 @@ struct hci_dev *hci_alloc_dev(void) INIT_DELAYED_WORK(&hdev->power_off, hci_power_off); INIT_DELAYED_WORK(&hdev->discov_off, hci_discov_off); - INIT_DELAYED_WORK(&hdev->adv_instance_expire, hci_adv_timeout_expire); skb_queue_head_init(&hdev->rx_q); skb_queue_head_init(&hdev->cmd_q); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 703e37f1a955..7554da5b7a8f 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1183,7 +1183,7 @@ static void hci_cc_le_set_scan_enable(struct hci_dev *hdev, hci_discovery_set_state(hdev, DISCOVERY_STOPPED); else if (!hci_dev_test_flag(hdev, HCI_LE_ADV) && hdev->discovery.state == DISCOVERY_FINDING) - mgmt_reenable_advertising(hdev); + hci_req_reenable_advertising(hdev); break; @@ -2401,7 +2401,7 @@ static void hci_disconn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) * is timed out due to Directed Advertising." */ if (type == LE_LINK) - mgmt_reenable_advertising(hdev); + hci_req_reenable_advertising(hdev); unlock: hci_dev_unlock(hdev); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 7c85435b8982..e6622bd1926d 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -23,6 +23,7 @@ #include #include +#include #include "smp.h" #include "hci_request.h" @@ -580,6 +581,524 @@ void hci_req_add_le_passive_scan(struct hci_request *req) &enable_cp); } +static u8 get_current_adv_instance(struct hci_dev *hdev) +{ + /* The "Set Advertising" setting supersedes the "Add Advertising" + * setting. Here we set the advertising data based on which + * setting was set. When neither apply, default to the global settings, + * represented by instance "0". + */ + if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && + !hci_dev_test_flag(hdev, HCI_ADVERTISING)) + return hdev->cur_adv_instance; + + return 0x00; +} + +static u8 get_cur_adv_instance_scan_rsp_len(struct hci_dev *hdev) +{ + u8 instance = get_current_adv_instance(hdev); + struct adv_info *adv_instance; + + /* Ignore instance 0 */ + if (instance == 0x00) + return 0; + + adv_instance = hci_find_adv_instance(hdev, instance); + if (!adv_instance) + return 0; + + /* TODO: Take into account the "appearance" and "local-name" flags here. + * These are currently being ignored as they are not supported. + */ + return adv_instance->scan_rsp_len; +} + +void __hci_req_disable_advertising(struct hci_request *req) +{ + u8 enable = 0x00; + + hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); +} + +static u32 get_adv_instance_flags(struct hci_dev *hdev, u8 instance) +{ + u32 flags; + struct adv_info *adv_instance; + + if (instance == 0x00) { + /* Instance 0 always manages the "Tx Power" and "Flags" + * fields + */ + flags = MGMT_ADV_FLAG_TX_POWER | MGMT_ADV_FLAG_MANAGED_FLAGS; + + /* For instance 0, the HCI_ADVERTISING_CONNECTABLE setting + * corresponds to the "connectable" instance flag. + */ + if (hci_dev_test_flag(hdev, HCI_ADVERTISING_CONNECTABLE)) + flags |= MGMT_ADV_FLAG_CONNECTABLE; + + return flags; + } + + adv_instance = hci_find_adv_instance(hdev, instance); + + /* Return 0 when we got an invalid instance identifier. */ + if (!adv_instance) + return 0; + + return adv_instance->flags; +} + +void __hci_req_enable_advertising(struct hci_request *req) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_adv_param cp; + u8 own_addr_type, enable = 0x01; + bool connectable; + u8 instance; + u32 flags; + + if (hci_conn_num(hdev, LE_LINK) > 0) + return; + + if (hci_dev_test_flag(hdev, HCI_LE_ADV)) + __hci_req_disable_advertising(req); + + /* Clear the HCI_LE_ADV bit temporarily so that the + * hci_update_random_address knows that it's safe to go ahead + * and write a new random address. The flag will be set back on + * as soon as the SET_ADV_ENABLE HCI command completes. + */ + hci_dev_clear_flag(hdev, HCI_LE_ADV); + + instance = get_current_adv_instance(hdev); + flags = get_adv_instance_flags(hdev, instance); + + /* If the "connectable" instance flag was not set, then choose between + * ADV_IND and ADV_NONCONN_IND based on the global connectable setting. + */ + connectable = (flags & MGMT_ADV_FLAG_CONNECTABLE) || + mgmt_get_connectable(hdev); + + /* Set require_privacy to true only when non-connectable + * advertising is used. In that case it is fine to use a + * non-resolvable private address. + */ + if (hci_update_random_address(req, !connectable, &own_addr_type) < 0) + return; + + memset(&cp, 0, sizeof(cp)); + cp.min_interval = cpu_to_le16(hdev->le_adv_min_interval); + cp.max_interval = cpu_to_le16(hdev->le_adv_max_interval); + + if (connectable) + cp.type = LE_ADV_IND; + else if (get_cur_adv_instance_scan_rsp_len(hdev)) + cp.type = LE_ADV_SCAN_IND; + else + cp.type = LE_ADV_NONCONN_IND; + + cp.own_address_type = own_addr_type; + cp.channel_map = hdev->le_adv_channel_map; + + hci_req_add(req, HCI_OP_LE_SET_ADV_PARAM, sizeof(cp), &cp); + + hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); +} + +static u8 create_default_scan_rsp_data(struct hci_dev *hdev, u8 *ptr) +{ + u8 ad_len = 0; + size_t name_len; + + name_len = strlen(hdev->dev_name); + if (name_len > 0) { + size_t max_len = HCI_MAX_AD_LENGTH - ad_len - 2; + + if (name_len > max_len) { + name_len = max_len; + ptr[1] = EIR_NAME_SHORT; + } else + ptr[1] = EIR_NAME_COMPLETE; + + ptr[0] = name_len + 1; + + memcpy(ptr + 2, hdev->dev_name, name_len); + + ad_len += (name_len + 2); + ptr += (name_len + 2); + } + + return ad_len; +} + +static u8 create_instance_scan_rsp_data(struct hci_dev *hdev, u8 instance, + u8 *ptr) +{ + struct adv_info *adv_instance; + + adv_instance = hci_find_adv_instance(hdev, instance); + if (!adv_instance) + return 0; + + /* TODO: Set the appropriate entries based on advertising instance flags + * here once flags other than 0 are supported. + */ + memcpy(ptr, adv_instance->scan_rsp_data, + adv_instance->scan_rsp_len); + + return adv_instance->scan_rsp_len; +} + +static void update_inst_scan_rsp_data(struct hci_request *req, u8 instance) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_scan_rsp_data cp; + u8 len; + + if (!hci_dev_test_flag(hdev, HCI_LE_ENABLED)) + return; + + memset(&cp, 0, sizeof(cp)); + + if (instance) + len = create_instance_scan_rsp_data(hdev, instance, cp.data); + else + len = create_default_scan_rsp_data(hdev, cp.data); + + if (hdev->scan_rsp_data_len == len && + !memcmp(cp.data, hdev->scan_rsp_data, len)) + return; + + memcpy(hdev->scan_rsp_data, cp.data, sizeof(cp.data)); + hdev->scan_rsp_data_len = len; + + cp.length = len; + + hci_req_add(req, HCI_OP_LE_SET_SCAN_RSP_DATA, sizeof(cp), &cp); +} + +void __hci_req_update_scan_rsp_data(struct hci_request *req, int instance) +{ + if (instance == HCI_ADV_CURRENT) + instance = get_current_adv_instance(req->hdev); + + update_inst_scan_rsp_data(req, get_current_adv_instance(req->hdev)); +} + +static u8 create_instance_adv_data(struct hci_dev *hdev, u8 instance, u8 *ptr) +{ + struct adv_info *adv_instance = NULL; + u8 ad_len = 0, flags = 0; + u32 instance_flags; + + /* Return 0 when the current instance identifier is invalid. */ + if (instance) { + adv_instance = hci_find_adv_instance(hdev, instance); + if (!adv_instance) + return 0; + } + + instance_flags = get_adv_instance_flags(hdev, instance); + + /* The Add Advertising command allows userspace to set both the general + * and limited discoverable flags. + */ + if (instance_flags & MGMT_ADV_FLAG_DISCOV) + flags |= LE_AD_GENERAL; + + if (instance_flags & MGMT_ADV_FLAG_LIMITED_DISCOV) + flags |= LE_AD_LIMITED; + + if (flags || (instance_flags & MGMT_ADV_FLAG_MANAGED_FLAGS)) { + /* If a discovery flag wasn't provided, simply use the global + * settings. + */ + if (!flags) + flags |= mgmt_get_adv_discov_flags(hdev); + + if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) + flags |= LE_AD_NO_BREDR; + + /* If flags would still be empty, then there is no need to + * include the "Flags" AD field". + */ + if (flags) { + ptr[0] = 0x02; + ptr[1] = EIR_FLAGS; + ptr[2] = flags; + + ad_len += 3; + ptr += 3; + } + } + + if (adv_instance) { + memcpy(ptr, adv_instance->adv_data, + adv_instance->adv_data_len); + ad_len += adv_instance->adv_data_len; + ptr += adv_instance->adv_data_len; + } + + /* Provide Tx Power only if we can provide a valid value for it */ + if (hdev->adv_tx_power != HCI_TX_POWER_INVALID && + (instance_flags & MGMT_ADV_FLAG_TX_POWER)) { + ptr[0] = 0x02; + ptr[1] = EIR_TX_POWER; + ptr[2] = (u8)hdev->adv_tx_power; + + ad_len += 3; + ptr += 3; + } + + return ad_len; +} + +static void update_inst_adv_data(struct hci_request *req, u8 instance) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_adv_data cp; + u8 len; + + if (!hci_dev_test_flag(hdev, HCI_LE_ENABLED)) + return; + + memset(&cp, 0, sizeof(cp)); + + len = create_instance_adv_data(hdev, instance, cp.data); + + /* There's nothing to do if the data hasn't changed */ + if (hdev->adv_data_len == len && + memcmp(cp.data, hdev->adv_data, len) == 0) + return; + + memcpy(hdev->adv_data, cp.data, sizeof(cp.data)); + hdev->adv_data_len = len; + + cp.length = len; + + hci_req_add(req, HCI_OP_LE_SET_ADV_DATA, sizeof(cp), &cp); +} + +void __hci_req_update_adv_data(struct hci_request *req, int instance) +{ + if (instance == HCI_ADV_CURRENT) + instance = get_current_adv_instance(req->hdev); + + update_inst_adv_data(req, instance); +} + +int hci_req_update_adv_data(struct hci_dev *hdev, int instance) +{ + struct hci_request req; + + hci_req_init(&req, hdev); + __hci_req_update_adv_data(&req, instance); + + return hci_req_run(&req, NULL); +} + +static void adv_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode) +{ + BT_DBG("%s status %u", hdev->name, status); +} + +void hci_req_reenable_advertising(struct hci_dev *hdev) +{ + struct hci_request req; + u8 instance; + + if (!hci_dev_test_flag(hdev, HCI_ADVERTISING) && + !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + return; + + instance = get_current_adv_instance(hdev); + + hci_req_init(&req, hdev); + + if (instance) { + __hci_req_schedule_adv_instance(&req, instance, true); + } else { + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); + __hci_req_update_scan_rsp_data(&req, HCI_ADV_CURRENT); + __hci_req_enable_advertising(&req); + } + + hci_req_run(&req, adv_enable_complete); +} + +static void adv_timeout_expire(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + adv_instance_expire.work); + + struct hci_request req; + u8 instance; + + BT_DBG("%s", hdev->name); + + hci_dev_lock(hdev); + + hdev->adv_instance_timeout = 0; + + instance = get_current_adv_instance(hdev); + if (instance == 0x00) + goto unlock; + + hci_req_init(&req, hdev); + + hci_req_clear_adv_instance(hdev, &req, instance, false); + + if (list_empty(&hdev->adv_instances)) + __hci_req_disable_advertising(&req); + + if (!skb_queue_empty(&req.cmd_q)) + hci_req_run(&req, NULL); + +unlock: + hci_dev_unlock(hdev); +} + +int __hci_req_schedule_adv_instance(struct hci_request *req, u8 instance, + bool force) +{ + struct hci_dev *hdev = req->hdev; + struct adv_info *adv_instance = NULL; + u16 timeout; + + if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || + !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + return -EPERM; + + if (hdev->adv_instance_timeout) + return -EBUSY; + + adv_instance = hci_find_adv_instance(hdev, instance); + if (!adv_instance) + return -ENOENT; + + /* A zero timeout means unlimited advertising. As long as there is + * only one instance, duration should be ignored. We still set a timeout + * in case further instances are being added later on. + * + * If the remaining lifetime of the instance is more than the duration + * then the timeout corresponds to the duration, otherwise it will be + * reduced to the remaining instance lifetime. + */ + if (adv_instance->timeout == 0 || + adv_instance->duration <= adv_instance->remaining_time) + timeout = adv_instance->duration; + else + timeout = adv_instance->remaining_time; + + /* The remaining time is being reduced unless the instance is being + * advertised without time limit. + */ + if (adv_instance->timeout) + adv_instance->remaining_time = + adv_instance->remaining_time - timeout; + + hdev->adv_instance_timeout = timeout; + queue_delayed_work(hdev->req_workqueue, + &hdev->adv_instance_expire, + msecs_to_jiffies(timeout * 1000)); + + /* If we're just re-scheduling the same instance again then do not + * execute any HCI commands. This happens when a single instance is + * being advertised. + */ + if (!force && hdev->cur_adv_instance == instance && + hci_dev_test_flag(hdev, HCI_LE_ADV)) + return 0; + + hdev->cur_adv_instance = instance; + __hci_req_update_adv_data(req, HCI_ADV_CURRENT); + __hci_req_update_scan_rsp_data(req, HCI_ADV_CURRENT); + __hci_req_enable_advertising(req); + + return 0; +} + +static void cancel_adv_timeout(struct hci_dev *hdev) +{ + if (hdev->adv_instance_timeout) { + hdev->adv_instance_timeout = 0; + cancel_delayed_work(&hdev->adv_instance_expire); + } +} + +/* For a single instance: + * - force == true: The instance will be removed even when its remaining + * lifetime is not zero. + * - force == false: the instance will be deactivated but kept stored unless + * the remaining lifetime is zero. + * + * For instance == 0x00: + * - force == true: All instances will be removed regardless of their timeout + * setting. + * - force == false: Only instances that have a timeout will be removed. + */ +void hci_req_clear_adv_instance(struct hci_dev *hdev, struct hci_request *req, + u8 instance, bool force) +{ + struct adv_info *adv_instance, *n, *next_instance = NULL; + int err; + u8 rem_inst; + + /* Cancel any timeout concerning the removed instance(s). */ + if (!instance || hdev->cur_adv_instance == instance) + cancel_adv_timeout(hdev); + + /* Get the next instance to advertise BEFORE we remove + * the current one. This can be the same instance again + * if there is only one instance. + */ + if (instance && hdev->cur_adv_instance == instance) + next_instance = hci_get_next_instance(hdev, instance); + + if (instance == 0x00) { + list_for_each_entry_safe(adv_instance, n, &hdev->adv_instances, + list) { + if (!(force || adv_instance->timeout)) + continue; + + rem_inst = adv_instance->instance; + err = hci_remove_adv_instance(hdev, rem_inst); + if (!err) + mgmt_advertising_removed(NULL, hdev, rem_inst); + } + hdev->cur_adv_instance = 0x00; + } else { + adv_instance = hci_find_adv_instance(hdev, instance); + + if (force || (adv_instance && adv_instance->timeout && + !adv_instance->remaining_time)) { + /* Don't advertise a removed instance. */ + if (next_instance && + next_instance->instance == instance) + next_instance = NULL; + + err = hci_remove_adv_instance(hdev, instance); + if (!err) + mgmt_advertising_removed(NULL, hdev, instance); + } + } + + if (list_empty(&hdev->adv_instances)) { + hdev->cur_adv_instance = 0x00; + hci_dev_clear_flag(hdev, HCI_ADVERTISING_INSTANCE); + } + + if (!req || !hdev_is_powered(hdev) || + hci_dev_test_flag(hdev, HCI_ADVERTISING)) + return; + + if (next_instance) + __hci_req_schedule_adv_instance(req, next_instance->instance, + false); +} + static void set_random_addr(struct hci_request *req, bdaddr_t *rpa) { struct hci_dev *hdev = req->hdev; @@ -1031,14 +1550,6 @@ unlock: hci_dev_unlock(hdev); } -static void cancel_adv_timeout(struct hci_dev *hdev) -{ - if (hdev->adv_instance_timeout) { - hdev->adv_instance_timeout = 0; - cancel_delayed_work(&hdev->adv_instance_expire); - } -} - static void disable_advertising(struct hci_request *req) { u8 enable = 0x00; @@ -1280,6 +1791,7 @@ void hci_request_setup(struct hci_dev *hdev) INIT_WORK(&hdev->scan_update, scan_update_work); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); + INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire); } void hci_request_cancel_all(struct hci_dev *hdev) @@ -1291,4 +1803,9 @@ void hci_request_cancel_all(struct hci_dev *hdev) cancel_work_sync(&hdev->scan_update); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); + + if (hdev->adv_instance_timeout) { + cancel_delayed_work_sync(&hdev->adv_instance_expire); + hdev->adv_instance_timeout = 0; + } } diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index cc8275520fb2..5358b1b12ca0 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -58,6 +58,20 @@ struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, void hci_req_add_le_scan_disable(struct hci_request *req); void hci_req_add_le_passive_scan(struct hci_request *req); +#define HCI_ADV_CURRENT (-1) + +void hci_req_reenable_advertising(struct hci_dev *hdev); +void __hci_req_enable_advertising(struct hci_request *req); +void __hci_req_disable_advertising(struct hci_request *req); +void __hci_req_update_adv_data(struct hci_request *req, int instance); +int hci_req_update_adv_data(struct hci_dev *hdev, int instance); +void __hci_req_update_scan_rsp_data(struct hci_request *req, int instance); + +int __hci_req_schedule_adv_instance(struct hci_request *req, u8 instance, + bool force); +void hci_req_clear_adv_instance(struct hci_dev *hdev, struct hci_request *req, + u8 instance, bool force); + /* Returns true if HCI commands were queued */ bool hci_req_stop_discovery(struct hci_request *req); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 0d20e1328528..6d0f0025052f 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -841,98 +841,7 @@ static struct mgmt_pending_cmd *pending_find_data(u16 opcode, return mgmt_pending_find_data(HCI_CHANNEL_CONTROL, opcode, hdev, data); } -static u8 get_current_adv_instance(struct hci_dev *hdev) -{ - /* The "Set Advertising" setting supersedes the "Add Advertising" - * setting. Here we set the advertising data based on which - * setting was set. When neither apply, default to the global settings, - * represented by instance "0". - */ - if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && - !hci_dev_test_flag(hdev, HCI_ADVERTISING)) - return hdev->cur_adv_instance; - - return 0x00; -} - -static u8 create_default_scan_rsp_data(struct hci_dev *hdev, u8 *ptr) -{ - u8 ad_len = 0; - size_t name_len; - - name_len = strlen(hdev->dev_name); - if (name_len > 0) { - size_t max_len = HCI_MAX_AD_LENGTH - ad_len - 2; - - if (name_len > max_len) { - name_len = max_len; - ptr[1] = EIR_NAME_SHORT; - } else - ptr[1] = EIR_NAME_COMPLETE; - - ptr[0] = name_len + 1; - - memcpy(ptr + 2, hdev->dev_name, name_len); - - ad_len += (name_len + 2); - ptr += (name_len + 2); - } - - return ad_len; -} - -static u8 create_instance_scan_rsp_data(struct hci_dev *hdev, u8 instance, - u8 *ptr) -{ - struct adv_info *adv_instance; - - adv_instance = hci_find_adv_instance(hdev, instance); - if (!adv_instance) - return 0; - - /* TODO: Set the appropriate entries based on advertising instance flags - * here once flags other than 0 are supported. - */ - memcpy(ptr, adv_instance->scan_rsp_data, - adv_instance->scan_rsp_len); - - return adv_instance->scan_rsp_len; -} - -static void update_inst_scan_rsp_data(struct hci_request *req, u8 instance) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_le_set_scan_rsp_data cp; - u8 len; - - if (!hci_dev_test_flag(hdev, HCI_LE_ENABLED)) - return; - - memset(&cp, 0, sizeof(cp)); - - if (instance) - len = create_instance_scan_rsp_data(hdev, instance, cp.data); - else - len = create_default_scan_rsp_data(hdev, cp.data); - - if (hdev->scan_rsp_data_len == len && - !memcmp(cp.data, hdev->scan_rsp_data, len)) - return; - - memcpy(hdev->scan_rsp_data, cp.data, sizeof(cp.data)); - hdev->scan_rsp_data_len = len; - - cp.length = len; - - hci_req_add(req, HCI_OP_LE_SET_SCAN_RSP_DATA, sizeof(cp), &cp); -} - -static void update_scan_rsp_data(struct hci_request *req) -{ - update_inst_scan_rsp_data(req, get_current_adv_instance(req->hdev)); -} - -static u8 get_adv_discov_flags(struct hci_dev *hdev) +u8 mgmt_get_adv_discov_flags(struct hci_dev *hdev) { struct mgmt_pending_cmd *cmd; @@ -956,7 +865,7 @@ static u8 get_adv_discov_flags(struct hci_dev *hdev) return 0; } -static bool get_connectable(struct hci_dev *hdev) +bool mgmt_get_connectable(struct hci_dev *hdev) { struct mgmt_pending_cmd *cmd; @@ -973,163 +882,6 @@ static bool get_connectable(struct hci_dev *hdev) return hci_dev_test_flag(hdev, HCI_CONNECTABLE); } -static u32 get_adv_instance_flags(struct hci_dev *hdev, u8 instance) -{ - u32 flags; - struct adv_info *adv_instance; - - if (instance == 0x00) { - /* Instance 0 always manages the "Tx Power" and "Flags" - * fields - */ - flags = MGMT_ADV_FLAG_TX_POWER | MGMT_ADV_FLAG_MANAGED_FLAGS; - - /* For instance 0, the HCI_ADVERTISING_CONNECTABLE setting - * corresponds to the "connectable" instance flag. - */ - if (hci_dev_test_flag(hdev, HCI_ADVERTISING_CONNECTABLE)) - flags |= MGMT_ADV_FLAG_CONNECTABLE; - - return flags; - } - - adv_instance = hci_find_adv_instance(hdev, instance); - - /* Return 0 when we got an invalid instance identifier. */ - if (!adv_instance) - return 0; - - return adv_instance->flags; -} - -static u8 get_cur_adv_instance_scan_rsp_len(struct hci_dev *hdev) -{ - u8 instance = get_current_adv_instance(hdev); - struct adv_info *adv_instance; - - /* Ignore instance 0 */ - if (instance == 0x00) - return 0; - - adv_instance = hci_find_adv_instance(hdev, instance); - if (!adv_instance) - return 0; - - /* TODO: Take into account the "appearance" and "local-name" flags here. - * These are currently being ignored as they are not supported. - */ - return adv_instance->scan_rsp_len; -} - -static u8 create_instance_adv_data(struct hci_dev *hdev, u8 instance, u8 *ptr) -{ - struct adv_info *adv_instance = NULL; - u8 ad_len = 0, flags = 0; - u32 instance_flags; - - /* Return 0 when the current instance identifier is invalid. */ - if (instance) { - adv_instance = hci_find_adv_instance(hdev, instance); - if (!adv_instance) - return 0; - } - - instance_flags = get_adv_instance_flags(hdev, instance); - - /* The Add Advertising command allows userspace to set both the general - * and limited discoverable flags. - */ - if (instance_flags & MGMT_ADV_FLAG_DISCOV) - flags |= LE_AD_GENERAL; - - if (instance_flags & MGMT_ADV_FLAG_LIMITED_DISCOV) - flags |= LE_AD_LIMITED; - - if (flags || (instance_flags & MGMT_ADV_FLAG_MANAGED_FLAGS)) { - /* If a discovery flag wasn't provided, simply use the global - * settings. - */ - if (!flags) - flags |= get_adv_discov_flags(hdev); - - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) - flags |= LE_AD_NO_BREDR; - - /* If flags would still be empty, then there is no need to - * include the "Flags" AD field". - */ - if (flags) { - ptr[0] = 0x02; - ptr[1] = EIR_FLAGS; - ptr[2] = flags; - - ad_len += 3; - ptr += 3; - } - } - - if (adv_instance) { - memcpy(ptr, adv_instance->adv_data, - adv_instance->adv_data_len); - ad_len += adv_instance->adv_data_len; - ptr += adv_instance->adv_data_len; - } - - /* Provide Tx Power only if we can provide a valid value for it */ - if (hdev->adv_tx_power != HCI_TX_POWER_INVALID && - (instance_flags & MGMT_ADV_FLAG_TX_POWER)) { - ptr[0] = 0x02; - ptr[1] = EIR_TX_POWER; - ptr[2] = (u8)hdev->adv_tx_power; - - ad_len += 3; - ptr += 3; - } - - return ad_len; -} - -static void update_inst_adv_data(struct hci_request *req, u8 instance) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_le_set_adv_data cp; - u8 len; - - if (!hci_dev_test_flag(hdev, HCI_LE_ENABLED)) - return; - - memset(&cp, 0, sizeof(cp)); - - len = create_instance_adv_data(hdev, instance, cp.data); - - /* There's nothing to do if the data hasn't changed */ - if (hdev->adv_data_len == len && - memcmp(cp.data, hdev->adv_data, len) == 0) - return; - - memcpy(hdev->adv_data, cp.data, sizeof(cp.data)); - hdev->adv_data_len = len; - - cp.length = len; - - hci_req_add(req, HCI_OP_LE_SET_ADV_DATA, sizeof(cp), &cp); -} - -static void update_adv_data(struct hci_request *req) -{ - update_inst_adv_data(req, get_current_adv_instance(req->hdev)); -} - -int mgmt_update_adv_data(struct hci_dev *hdev) -{ - struct hci_request req; - - hci_req_init(&req, hdev); - update_adv_data(&req); - - return hci_req_run(&req, NULL); -} - static void create_eir(struct hci_dev *hdev, u8 *data) { u8 *ptr = data; @@ -1247,70 +999,6 @@ static void update_class(struct hci_request *req) hci_req_add(req, HCI_OP_WRITE_CLASS_OF_DEV, sizeof(cod), cod); } -static void disable_advertising(struct hci_request *req) -{ - u8 enable = 0x00; - - hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); -} - -static void enable_advertising(struct hci_request *req) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_le_set_adv_param cp; - u8 own_addr_type, enable = 0x01; - bool connectable; - u8 instance; - u32 flags; - - if (hci_conn_num(hdev, LE_LINK) > 0) - return; - - if (hci_dev_test_flag(hdev, HCI_LE_ADV)) - disable_advertising(req); - - /* Clear the HCI_LE_ADV bit temporarily so that the - * hci_update_random_address knows that it's safe to go ahead - * and write a new random address. The flag will be set back on - * as soon as the SET_ADV_ENABLE HCI command completes. - */ - hci_dev_clear_flag(hdev, HCI_LE_ADV); - - instance = get_current_adv_instance(hdev); - flags = get_adv_instance_flags(hdev, instance); - - /* If the "connectable" instance flag was not set, then choose between - * ADV_IND and ADV_NONCONN_IND based on the global connectable setting. - */ - connectable = (flags & MGMT_ADV_FLAG_CONNECTABLE) || - get_connectable(hdev); - - /* Set require_privacy to true only when non-connectable - * advertising is used. In that case it is fine to use a - * non-resolvable private address. - */ - if (hci_update_random_address(req, !connectable, &own_addr_type) < 0) - return; - - memset(&cp, 0, sizeof(cp)); - cp.min_interval = cpu_to_le16(hdev->le_adv_min_interval); - cp.max_interval = cpu_to_le16(hdev->le_adv_max_interval); - - if (connectable) - cp.type = LE_ADV_IND; - else if (get_cur_adv_instance_scan_rsp_len(hdev)) - cp.type = LE_ADV_SCAN_IND; - else - cp.type = LE_ADV_NONCONN_IND; - - cp.own_address_type = own_addr_type; - cp.channel_map = hdev->le_adv_channel_map; - - hci_req_add(req, HCI_OP_LE_SET_ADV_PARAM, sizeof(cp), &cp); - - hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); -} - static void service_cache_off(struct work_struct *work) { struct hci_dev *hdev = container_of(work, struct hci_dev, @@ -1346,10 +1034,11 @@ static void rpa_expired(struct work_struct *work) return; /* The generation of a new RPA and programming it into the - * controller happens in the enable_advertising() function. + * controller happens in the hci_req_enable_advertising() + * function. */ hci_req_init(&req, hdev); - enable_advertising(&req); + __hci_req_enable_advertising(&req); hci_req_run(&req, NULL); } @@ -1417,8 +1106,7 @@ static void clean_up_hci_complete(struct hci_dev *hdev, u8 status, u16 opcode) } } -static void advertising_added(struct sock *sk, struct hci_dev *hdev, - u8 instance) +void mgmt_advertising_added(struct sock *sk, struct hci_dev *hdev, u8 instance) { struct mgmt_ev_advertising_added ev; @@ -1427,8 +1115,8 @@ static void advertising_added(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_ADVERTISING_ADDED, hdev, &ev, sizeof(ev), sk); } -static void advertising_removed(struct sock *sk, struct hci_dev *hdev, - u8 instance) +void mgmt_advertising_removed(struct sock *sk, struct hci_dev *hdev, + u8 instance) { struct mgmt_ev_advertising_removed ev; @@ -1437,65 +1125,6 @@ static void advertising_removed(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_ADVERTISING_REMOVED, hdev, &ev, sizeof(ev), sk); } -static int schedule_adv_instance(struct hci_request *req, u8 instance, - bool force) { - struct hci_dev *hdev = req->hdev; - struct adv_info *adv_instance = NULL; - u16 timeout; - - if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) - return -EPERM; - - if (hdev->adv_instance_timeout) - return -EBUSY; - - adv_instance = hci_find_adv_instance(hdev, instance); - if (!adv_instance) - return -ENOENT; - - /* A zero timeout means unlimited advertising. As long as there is - * only one instance, duration should be ignored. We still set a timeout - * in case further instances are being added later on. - * - * If the remaining lifetime of the instance is more than the duration - * then the timeout corresponds to the duration, otherwise it will be - * reduced to the remaining instance lifetime. - */ - if (adv_instance->timeout == 0 || - adv_instance->duration <= adv_instance->remaining_time) - timeout = adv_instance->duration; - else - timeout = adv_instance->remaining_time; - - /* The remaining time is being reduced unless the instance is being - * advertised without time limit. - */ - if (adv_instance->timeout) - adv_instance->remaining_time = - adv_instance->remaining_time - timeout; - - hdev->adv_instance_timeout = timeout; - queue_delayed_work(hdev->workqueue, - &hdev->adv_instance_expire, - msecs_to_jiffies(timeout * 1000)); - - /* If we're just re-scheduling the same instance again then do not - * execute any HCI commands. This happens when a single instance is - * being advertised. - */ - if (!force && hdev->cur_adv_instance == instance && - hci_dev_test_flag(hdev, HCI_LE_ADV)) - return 0; - - hdev->cur_adv_instance = instance; - update_adv_data(req); - update_scan_rsp_data(req); - enable_advertising(req); - - return 0; -} - static void cancel_adv_timeout(struct hci_dev *hdev) { if (hdev->adv_instance_timeout) { @@ -1504,76 +1133,6 @@ static void cancel_adv_timeout(struct hci_dev *hdev) } } -/* For a single instance: - * - force == true: The instance will be removed even when its remaining - * lifetime is not zero. - * - force == false: the instance will be deactivated but kept stored unless - * the remaining lifetime is zero. - * - * For instance == 0x00: - * - force == true: All instances will be removed regardless of their timeout - * setting. - * - force == false: Only instances that have a timeout will be removed. - */ -static void clear_adv_instance(struct hci_dev *hdev, struct hci_request *req, - u8 instance, bool force) -{ - struct adv_info *adv_instance, *n, *next_instance = NULL; - int err; - u8 rem_inst; - - /* Cancel any timeout concerning the removed instance(s). */ - if (!instance || hdev->cur_adv_instance == instance) - cancel_adv_timeout(hdev); - - /* Get the next instance to advertise BEFORE we remove - * the current one. This can be the same instance again - * if there is only one instance. - */ - if (instance && hdev->cur_adv_instance == instance) - next_instance = hci_get_next_instance(hdev, instance); - - if (instance == 0x00) { - list_for_each_entry_safe(adv_instance, n, &hdev->adv_instances, - list) { - if (!(force || adv_instance->timeout)) - continue; - - rem_inst = adv_instance->instance; - err = hci_remove_adv_instance(hdev, rem_inst); - if (!err) - advertising_removed(NULL, hdev, rem_inst); - } - hdev->cur_adv_instance = 0x00; - } else { - adv_instance = hci_find_adv_instance(hdev, instance); - - if (force || (adv_instance && adv_instance->timeout && - !adv_instance->remaining_time)) { - /* Don't advertise a removed instance. */ - if (next_instance && - next_instance->instance == instance) - next_instance = NULL; - - err = hci_remove_adv_instance(hdev, instance); - if (!err) - advertising_removed(NULL, hdev, instance); - } - } - - if (list_empty(&hdev->adv_instances)) { - hdev->cur_adv_instance = 0x00; - hci_dev_clear_flag(hdev, HCI_ADVERTISING_INSTANCE); - } - - if (!req || !hdev_is_powered(hdev) || - hci_dev_test_flag(hdev, HCI_ADVERTISING)) - return; - - if (next_instance) - schedule_adv_instance(req, next_instance->instance, false); -} - static int clean_up_hci_state(struct hci_dev *hdev) { struct hci_request req; @@ -1589,10 +1148,10 @@ static int clean_up_hci_state(struct hci_dev *hdev) hci_req_add(&req, HCI_OP_WRITE_SCAN_ENABLE, 1, &scan); } - clear_adv_instance(hdev, NULL, 0x00, false); + hci_req_clear_adv_instance(hdev, NULL, 0x00, false); if (hci_dev_test_flag(hdev, HCI_LE_ADV)) - disable_advertising(&req); + __hci_req_disable_advertising(&req); discov_stopped = hci_req_stop_discovery(&req); @@ -1975,7 +1534,7 @@ static int set_discoverable(struct sock *sk, struct hci_dev *hdev, void *data, hci_req_add(&req, HCI_OP_WRITE_SCAN_ENABLE, sizeof(scan), &scan); update_ad: - update_adv_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); err = hci_req_run(&req, set_discoverable_complete); if (err < 0) @@ -2060,7 +1619,7 @@ static void set_connectable_complete(struct hci_dev *hdev, u8 status, new_settings(hdev, cmd->sk); hci_req_update_scan(hdev); if (discov_changed) - mgmt_update_adv_data(hdev); + hci_req_update_adv_data(hdev, HCI_ADV_CURRENT); hci_update_background_scan(hdev); } @@ -2151,7 +1710,7 @@ static int set_connectable(struct sock *sk, struct hci_dev *hdev, void *data, hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); } - update_adv_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); } else if (cp->val != test_bit(HCI_PSCAN, &hdev->flags)) { if (cp->val) { scan = SCAN_PAGE; @@ -2181,7 +1740,7 @@ no_scan_update: /* Update the advertising parameters if necessary */ if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) - enable_advertising(&req); + __hci_req_enable_advertising(&req); err = hci_req_run(&req, set_connectable_complete); if (err < 0) { @@ -2466,8 +2025,8 @@ static void le_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode) struct hci_request req; hci_req_init(&req, hdev); - update_adv_data(&req); - update_scan_rsp_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); + __hci_req_update_scan_rsp_data(&req, HCI_ADV_CURRENT); hci_req_run(&req, NULL); hci_update_background_scan(hdev); } @@ -2518,7 +2077,7 @@ static int set_le(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) enabled = lmp_host_le_capable(hdev); if (!val) - clear_adv_instance(hdev, NULL, 0x00, true); + hci_req_clear_adv_instance(hdev, NULL, 0x00, true); if (!hdev_is_powered(hdev) || val == enabled) { bool changed = false; @@ -2565,7 +2124,7 @@ static int set_le(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) hci_cp.simul = 0x00; } else { if (hci_dev_test_flag(hdev, HCI_LE_ADV)) - disable_advertising(&req); + __hci_req_disable_advertising(&req); } hci_req_add(&req, HCI_OP_WRITE_LE_HOST_SUPPORTED, sizeof(hci_cp), @@ -3856,7 +3415,7 @@ static int set_local_name(struct sock *sk, struct hci_dev *hdev, void *data, * no need to udpate the advertising data here. */ if (lmp_le_capable(hdev)) - update_scan_rsp_data(&req); + __hci_req_update_scan_rsp_data(&req, HCI_ADV_CURRENT); err = hci_req_run(&req, set_name_complete); if (err < 0) @@ -4600,7 +4159,7 @@ static void set_advertising_complete(struct hci_dev *hdev, u8 status, hci_req_init(&req, hdev); - err = schedule_adv_instance(&req, instance, true); + err = __hci_req_schedule_adv_instance(&req, instance, true); if (!err) err = hci_req_run(&req, enable_advertising_instance); @@ -4697,11 +4256,11 @@ static int set_advertising(struct sock *sk, struct hci_dev *hdev, void *data, * We cannot use update_[adv|scan_rsp]_data() here as the * HCI_ADVERTISING flag is not yet set. */ - update_inst_adv_data(&req, 0x00); - update_inst_scan_rsp_data(&req, 0x00); - enable_advertising(&req); + __hci_req_update_adv_data(&req, 0x00); + __hci_req_update_scan_rsp_data(&req, 0x00); + __hci_req_enable_advertising(&req); } else { - disable_advertising(&req); + __hci_req_disable_advertising(&req); } err = hci_req_run(&req, set_advertising_complete); @@ -5033,8 +4592,8 @@ static int set_bredr(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) goto unlock; } - /* We need to flip the bit already here so that update_adv_data - * generates the correct flags. + /* We need to flip the bit already here so that + * hci_req_update_adv_data generates the correct flags. */ hci_dev_set_flag(hdev, HCI_BREDR_ENABLED); @@ -5046,7 +4605,7 @@ static int set_bredr(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) /* Since only the advertising data flags will change, there * is no need to update the scan response data. */ - update_adv_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); err = hci_req_run(&req, set_bredr_complete); if (err < 0) @@ -6583,7 +6142,7 @@ static int read_local_oob_ext_data(struct sock *sk, struct hci_dev *hdev, rand, sizeof(rand)); } - flags = get_adv_discov_flags(hdev); + flags = mgmt_get_adv_discov_flags(hdev); if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) flags |= LE_AD_NO_BREDR; @@ -6772,7 +6331,7 @@ static void add_advertising_complete(struct hci_dev *hdev, u8 status, cancel_adv_timeout(hdev); hci_remove_adv_instance(hdev, instance); - advertising_removed(cmd ? cmd->sk : NULL, hdev, instance); + mgmt_advertising_removed(cmd ? cmd->sk : NULL, hdev, instance); } if (!cmd) @@ -6794,31 +6353,6 @@ unlock: hci_dev_unlock(hdev); } -void mgmt_adv_timeout_expired(struct hci_dev *hdev) -{ - u8 instance; - struct hci_request req; - - hdev->adv_instance_timeout = 0; - - instance = get_current_adv_instance(hdev); - if (instance == 0x00) - return; - - hci_dev_lock(hdev); - hci_req_init(&req, hdev); - - clear_adv_instance(hdev, &req, instance, false); - - if (list_empty(&hdev->adv_instances)) - disable_advertising(&req); - - if (!skb_queue_empty(&req.cmd_q)) - hci_req_run(&req, NULL); - - hci_dev_unlock(hdev); -} - static int add_advertising(struct sock *sk, struct hci_dev *hdev, void *data, u16 data_len) { @@ -6897,7 +6431,7 @@ static int add_advertising(struct sock *sk, struct hci_dev *hdev, * actually added. */ if (hdev->adv_instance_cnt > prev_instance_cnt) - advertising_added(sk, hdev, cp->instance); + mgmt_advertising_added(sk, hdev, cp->instance); hci_dev_set_flag(hdev, HCI_ADVERTISING_INSTANCE); @@ -6944,7 +6478,7 @@ static int add_advertising(struct sock *sk, struct hci_dev *hdev, hci_req_init(&req, hdev); - err = schedule_adv_instance(&req, schedule_instance, true); + err = __hci_req_schedule_adv_instance(&req, schedule_instance, true); if (!err) err = hci_req_run(&req, add_advertising_complete); @@ -7024,10 +6558,10 @@ static int remove_advertising(struct sock *sk, struct hci_dev *hdev, hci_req_init(&req, hdev); - clear_adv_instance(hdev, &req, cp->instance, true); + hci_req_clear_adv_instance(hdev, &req, cp->instance, true); if (list_empty(&hdev->adv_instances)) - disable_advertising(&req); + __hci_req_disable_advertising(&req); /* If no HCI commands have been collected so far or the HCI_ADVERTISING * flag is set or the device isn't powered then we have no HCI @@ -7367,8 +6901,8 @@ static int powered_update_hci(struct hci_dev *hdev) if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && (hci_dev_test_flag(hdev, HCI_ADVERTISING) || !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE))) { - update_adv_data(&req); - update_scan_rsp_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); + __hci_req_update_scan_rsp_data(&req, HCI_ADV_CURRENT); } if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && @@ -7380,11 +6914,12 @@ static int powered_update_hci(struct hci_dev *hdev) } if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - enable_advertising(&req); + __hci_req_enable_advertising(&req); else if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && hdev->cur_adv_instance) - schedule_adv_instance(&req, hdev->cur_adv_instance, - true); + __hci_req_schedule_adv_instance(&req, + hdev->cur_adv_instance, + true); } link_sec = hci_dev_test_flag(hdev, HCI_LINK_SECURITY); @@ -7505,7 +7040,7 @@ void mgmt_discoverable_timeout(struct hci_dev *hdev) * only update AD if advertising was enabled using Set Advertising. */ if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - update_adv_data(&req); + __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); hci_req_run(&req, NULL); @@ -8352,35 +7887,6 @@ void mgmt_discovering(struct hci_dev *hdev, u8 discovering) mgmt_event(MGMT_EV_DISCOVERING, hdev, &ev, sizeof(ev), NULL); } -static void adv_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode) -{ - BT_DBG("%s status %u", hdev->name, status); -} - -void mgmt_reenable_advertising(struct hci_dev *hdev) -{ - struct hci_request req; - u8 instance; - - if (!hci_dev_test_flag(hdev, HCI_ADVERTISING) && - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) - return; - - instance = get_current_adv_instance(hdev); - - hci_req_init(&req, hdev); - - if (instance) { - schedule_adv_instance(&req, instance, true); - } else { - update_adv_data(&req); - update_scan_rsp_data(&req); - enable_advertising(&req); - } - - hci_req_run(&req, adv_enable_complete); -} - static struct hci_mgmt_chan chan = { .channel = HCI_CHANNEL_CONTROL, .handler_count = ARRAY_SIZE(mgmt_handlers), -- cgit v1.2.3-59-g8ed1b From 53c0ba74510c1182786dcd1e3710215467777601 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Sun, 22 Nov 2015 16:43:43 +0300 Subject: Bluetooth: Move connectable changes to hdev->req_workqueue This way the connectable changes are synchronized against each other, which helps avoid potential races. The connectable mode is also linked together with LE advertising which makes is more convenient to have it behind the same workqueue. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 + net/bluetooth/hci_request.c | 39 ++++++++++++++++++ net/bluetooth/mgmt.c | 86 ++++++---------------------------------- 3 files changed, 53 insertions(+), 74 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index b56085b6ecce..a855e41df68c 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -330,6 +330,7 @@ struct hci_dev { struct work_struct discov_update; struct work_struct bg_scan_update; struct work_struct scan_update; + struct work_struct connectable_update; struct delayed_work le_scan_disable; struct delayed_work le_scan_restart; @@ -1491,6 +1492,7 @@ void mgmt_new_conn_param(struct hci_dev *hdev, bdaddr_t *bdaddr, u16 max_interval, u16 latency, u16 timeout); void mgmt_smp_complete(struct hci_conn *conn, bool complete); bool mgmt_get_connectable(struct hci_dev *hdev); +void mgmt_set_connectable_complete(struct hci_dev *hdev, u8 status); u8 mgmt_get_adv_discov_flags(struct hci_dev *hdev); void mgmt_advertising_added(struct sock *sk, struct hci_dev *hdev, u8 instance); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index e6622bd1926d..167c90644b4b 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1274,6 +1274,43 @@ static void scan_update_work(struct work_struct *work) hci_req_sync(hdev, update_scan, 0, HCI_CMD_TIMEOUT, NULL); } +static int connectable_update(struct hci_request *req, unsigned long opt) +{ + struct hci_dev *hdev = req->hdev; + + hci_dev_lock(hdev); + + __hci_req_update_scan(req); + + /* If BR/EDR is not enabled and we disable advertising as a + * by-product of disabling connectable, we need to update the + * advertising flags. + */ + if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) + __hci_req_update_adv_data(req, HCI_ADV_CURRENT); + + /* Update the advertising parameters if necessary */ + if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || + hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + __hci_req_enable_advertising(req); + + __hci_update_background_scan(req); + + hci_dev_unlock(hdev); + + return 0; +} + +static void connectable_update_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + connectable_update); + u8 status; + + hci_req_sync(hdev, connectable_update, 0, HCI_CMD_TIMEOUT, &status); + mgmt_set_connectable_complete(hdev, status); +} + void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason) { @@ -1789,6 +1826,7 @@ void hci_request_setup(struct hci_dev *hdev) INIT_WORK(&hdev->discov_update, discov_update); INIT_WORK(&hdev->bg_scan_update, bg_scan_update); INIT_WORK(&hdev->scan_update, scan_update_work); + INIT_WORK(&hdev->connectable_update, connectable_update_work); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire); @@ -1801,6 +1839,7 @@ void hci_request_cancel_all(struct hci_dev *hdev) cancel_work_sync(&hdev->discov_update); cancel_work_sync(&hdev->bg_scan_update); cancel_work_sync(&hdev->scan_update); + cancel_work_sync(&hdev->connectable_update); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 6d0f0025052f..d8b76ca5c820 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1580,12 +1580,9 @@ static void write_fast_connectable(struct hci_request *req, bool enable) hci_req_add(req, HCI_OP_WRITE_PAGE_SCAN_TYPE, 1, &type); } -static void set_connectable_complete(struct hci_dev *hdev, u8 status, - u16 opcode) +void mgmt_set_connectable_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; - struct mgmt_mode *cp; - bool conn_changed, discov_changed; BT_DBG("status 0x%02x", status); @@ -1601,27 +1598,8 @@ static void set_connectable_complete(struct hci_dev *hdev, u8 status, goto remove_cmd; } - cp = cmd->param; - if (cp->val) { - conn_changed = !hci_dev_test_and_set_flag(hdev, - HCI_CONNECTABLE); - discov_changed = false; - } else { - conn_changed = hci_dev_test_and_clear_flag(hdev, - HCI_CONNECTABLE); - discov_changed = hci_dev_test_and_clear_flag(hdev, - HCI_DISCOVERABLE); - } - send_settings_rsp(cmd->sk, MGMT_OP_SET_CONNECTABLE, hdev); - - if (conn_changed || discov_changed) { - new_settings(hdev, cmd->sk); - hci_req_update_scan(hdev); - if (discov_changed) - hci_req_update_adv_data(hdev, HCI_ADV_CURRENT); - hci_update_background_scan(hdev); - } + new_settings(hdev, cmd->sk); remove_cmd: mgmt_pending_remove(cmd); @@ -1664,8 +1642,6 @@ static int set_connectable(struct sock *sk, struct hci_dev *hdev, void *data, { struct mgmt_mode *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; - u8 scan; int err; BT_DBG("request for %s", hdev->name); @@ -1699,57 +1675,19 @@ static int set_connectable(struct sock *sk, struct hci_dev *hdev, void *data, goto failed; } - hci_req_init(&req, hdev); - - /* If BR/EDR is not enabled and we disable advertising as a - * by-product of disabling connectable, we need to update the - * advertising flags. - */ - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { - if (!cp->val) { - hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); - hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); - } - __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); - } else if (cp->val != test_bit(HCI_PSCAN, &hdev->flags)) { - if (cp->val) { - scan = SCAN_PAGE; - } else { - /* If we don't have any whitelist entries just - * disable all scanning. If there are entries - * and we had both page and inquiry scanning - * enabled then fall back to only page scanning. - * Otherwise no changes are needed. - */ - if (list_empty(&hdev->whitelist)) - scan = SCAN_DISABLED; - else if (test_bit(HCI_ISCAN, &hdev->flags)) - scan = SCAN_PAGE; - else - goto no_scan_update; - - if (test_bit(HCI_ISCAN, &hdev->flags) && - hdev->discov_timeout > 0) - cancel_delayed_work(&hdev->discov_off); - } + if (cp->val) { + hci_dev_set_flag(hdev, HCI_CONNECTABLE); + } else { + if (hdev->discov_timeout > 0) + cancel_delayed_work(&hdev->discov_off); - hci_req_add(&req, HCI_OP_WRITE_SCAN_ENABLE, 1, &scan); + hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); + hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); + hci_dev_clear_flag(hdev, HCI_CONNECTABLE); } -no_scan_update: - /* Update the advertising parameters if necessary */ - if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) - __hci_req_enable_advertising(&req); - - err = hci_req_run(&req, set_connectable_complete); - if (err < 0) { - mgmt_pending_remove(cmd); - if (err == -ENODATA) - err = set_connectable_update_settings(hdev, sk, - cp->val); - goto failed; - } + queue_work(hdev->req_workqueue, &hdev->connectable_update); + err = 0; failed: hci_dev_unlock(hdev); -- cgit v1.2.3-59-g8ed1b From aed1a8851db022c3bd22af41a343068b8c6e40c1 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Sun, 22 Nov 2015 17:24:44 +0300 Subject: Bluetooth: Move discoverable changes to hdev->req_workqueue The discoverable mode is intrinsically linked with the connectable mode e.g. through sharing the same HCI command (Write Scan Enable) for BR/EDR. It makes therefore sense to move it to hci_request.c and run the changes through the same hdev->req_workqueue. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 + net/bluetooth/hci_request.c | 64 ++++++++++++++++++++++++++++ net/bluetooth/mgmt.c | 90 ++++++---------------------------------- 3 files changed, 79 insertions(+), 77 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a855e41df68c..0a6966ed7ee1 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -331,6 +331,7 @@ struct hci_dev { struct work_struct bg_scan_update; struct work_struct scan_update; struct work_struct connectable_update; + struct work_struct discoverable_update; struct delayed_work le_scan_disable; struct delayed_work le_scan_restart; @@ -1493,6 +1494,7 @@ void mgmt_new_conn_param(struct hci_dev *hdev, bdaddr_t *bdaddr, void mgmt_smp_complete(struct hci_conn *conn, bool complete); bool mgmt_get_connectable(struct hci_dev *hdev); void mgmt_set_connectable_complete(struct hci_dev *hdev, u8 status); +void mgmt_set_discoverable_complete(struct hci_dev *hdev, u8 status); u8 mgmt_get_adv_discov_flags(struct hci_dev *hdev); void mgmt_advertising_added(struct sock *sk, struct hci_dev *hdev, u8 instance); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index e5e827b762b9..8f72218ed805 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1351,6 +1351,68 @@ void __hci_req_update_class(struct hci_request *req) hci_req_add(req, HCI_OP_WRITE_CLASS_OF_DEV, sizeof(cod), cod); } +static void write_iac(struct hci_request *req) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_write_current_iac_lap cp; + + if (!hci_dev_test_flag(hdev, HCI_DISCOVERABLE)) + return; + + if (hci_dev_test_flag(hdev, HCI_LIMITED_DISCOVERABLE)) { + /* Limited discoverable mode */ + cp.num_iac = min_t(u8, hdev->num_iac, 2); + cp.iac_lap[0] = 0x00; /* LIAC */ + cp.iac_lap[1] = 0x8b; + cp.iac_lap[2] = 0x9e; + cp.iac_lap[3] = 0x33; /* GIAC */ + cp.iac_lap[4] = 0x8b; + cp.iac_lap[5] = 0x9e; + } else { + /* General discoverable mode */ + cp.num_iac = 1; + cp.iac_lap[0] = 0x33; /* GIAC */ + cp.iac_lap[1] = 0x8b; + cp.iac_lap[2] = 0x9e; + } + + hci_req_add(req, HCI_OP_WRITE_CURRENT_IAC_LAP, + (cp.num_iac * 3) + 1, &cp); +} + +static int discoverable_update(struct hci_request *req, unsigned long opt) +{ + struct hci_dev *hdev = req->hdev; + + hci_dev_lock(hdev); + + if (hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { + write_iac(req); + __hci_req_update_scan(req); + __hci_req_update_class(req); + } + + /* Advertising instances don't use the global discoverable setting, so + * only update AD if advertising was enabled using Set Advertising. + */ + if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) + __hci_req_update_adv_data(req, HCI_ADV_CURRENT); + + hci_dev_unlock(hdev); + + return 0; +} + +static void discoverable_update_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + discoverable_update); + u8 status; + + hci_req_sync(hdev, discoverable_update, 0, HCI_CMD_TIMEOUT, &status); + mgmt_set_discoverable_complete(hdev, status); +} + void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason) { @@ -1867,6 +1929,7 @@ void hci_request_setup(struct hci_dev *hdev) INIT_WORK(&hdev->bg_scan_update, bg_scan_update); INIT_WORK(&hdev->scan_update, scan_update_work); INIT_WORK(&hdev->connectable_update, connectable_update_work); + INIT_WORK(&hdev->discoverable_update, discoverable_update_work); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire); @@ -1880,6 +1943,7 @@ void hci_request_cancel_all(struct hci_dev *hdev) cancel_work_sync(&hdev->bg_scan_update); cancel_work_sync(&hdev->scan_update); cancel_work_sync(&hdev->connectable_update); + cancel_work_sync(&hdev->discoverable_update); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index f5a4ee92f2bf..8846cb3b0aaa 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1282,13 +1282,9 @@ static u8 mgmt_le_support(struct hci_dev *hdev) return MGMT_STATUS_SUCCESS; } -static void set_discoverable_complete(struct hci_dev *hdev, u8 status, - u16 opcode) +void mgmt_set_discoverable_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; - struct mgmt_mode *cp; - struct hci_request req; - bool changed; BT_DBG("status 0x%02x", status); @@ -1305,33 +1301,14 @@ static void set_discoverable_complete(struct hci_dev *hdev, u8 status, goto remove_cmd; } - cp = cmd->param; - if (cp->val) { - changed = !hci_dev_test_and_set_flag(hdev, HCI_DISCOVERABLE); - - if (hdev->discov_timeout > 0) { - int to = msecs_to_jiffies(hdev->discov_timeout * 1000); - queue_delayed_work(hdev->workqueue, &hdev->discov_off, - to); - } - } else { - changed = hci_dev_test_and_clear_flag(hdev, HCI_DISCOVERABLE); + if (hci_dev_test_flag(hdev, HCI_DISCOVERABLE) && + hdev->discov_timeout > 0) { + int to = msecs_to_jiffies(hdev->discov_timeout * 1000); + queue_delayed_work(hdev->req_workqueue, &hdev->discov_off, to); } send_settings_rsp(cmd->sk, MGMT_OP_SET_DISCOVERABLE, hdev); - - if (changed) - new_settings(hdev, cmd->sk); - - /* When the discoverable mode gets changed, make sure - * that class of device has the limited discoverable - * bit correctly set. Also update page scan based on whitelist - * entries. - */ - hci_req_init(&req, hdev); - __hci_req_update_scan(&req); - __hci_req_update_class(&req); - hci_req_run(&req, NULL); + new_settings(hdev, cmd->sk); remove_cmd: mgmt_pending_remove(cmd); @@ -1345,9 +1322,7 @@ static int set_discoverable(struct sock *sk, struct hci_dev *hdev, void *data, { struct mgmt_cp_set_discoverable *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; u16 timeout; - u8 scan; int err; BT_DBG("request for %s", hdev->name); @@ -1447,58 +1422,19 @@ static int set_discoverable(struct sock *sk, struct hci_dev *hdev, void *data, cancel_delayed_work(&hdev->discov_off); hdev->discov_timeout = timeout; + if (cp->val) + hci_dev_set_flag(hdev, HCI_DISCOVERABLE); + else + hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); + /* Limited discoverable mode */ if (cp->val == 0x02) hci_dev_set_flag(hdev, HCI_LIMITED_DISCOVERABLE); else hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); - hci_req_init(&req, hdev); - - /* The procedure for LE-only controllers is much simpler - just - * update the advertising data. - */ - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) - goto update_ad; - - scan = SCAN_PAGE; - - if (cp->val) { - struct hci_cp_write_current_iac_lap hci_cp; - - if (cp->val == 0x02) { - /* Limited discoverable mode */ - hci_cp.num_iac = min_t(u8, hdev->num_iac, 2); - hci_cp.iac_lap[0] = 0x00; /* LIAC */ - hci_cp.iac_lap[1] = 0x8b; - hci_cp.iac_lap[2] = 0x9e; - hci_cp.iac_lap[3] = 0x33; /* GIAC */ - hci_cp.iac_lap[4] = 0x8b; - hci_cp.iac_lap[5] = 0x9e; - } else { - /* General discoverable mode */ - hci_cp.num_iac = 1; - hci_cp.iac_lap[0] = 0x33; /* GIAC */ - hci_cp.iac_lap[1] = 0x8b; - hci_cp.iac_lap[2] = 0x9e; - } - - hci_req_add(&req, HCI_OP_WRITE_CURRENT_IAC_LAP, - (hci_cp.num_iac * 3) + 1, &hci_cp); - - scan |= SCAN_INQUIRY; - } else { - hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); - } - - hci_req_add(&req, HCI_OP_WRITE_SCAN_ENABLE, sizeof(scan), &scan); - -update_ad: - __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); - - err = hci_req_run(&req, set_discoverable_complete); - if (err < 0) - mgmt_pending_remove(cmd); + queue_work(hdev->req_workqueue, &hdev->discoverable_update); + err = 0; failed: hci_dev_unlock(hdev); -- cgit v1.2.3-59-g8ed1b From c366f555b8df67633b849a5088bb897d6c63aaa5 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Mon, 23 Nov 2015 15:43:06 +0200 Subject: Bluetooth: Move discoverable timeout behind hdev->req_workqueue Since the other discoverable changes are behind req_workqueue now it only makes sense to move the discoverable timeout there as well. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 - net/bluetooth/hci_core.c | 13 ------------- net/bluetooth/hci_request.c | 26 +++++++++++++++++++++++++ net/bluetooth/mgmt.c | 41 ++-------------------------------------- 4 files changed, 28 insertions(+), 53 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 0a6966ed7ee1..319bf020cea6 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1436,7 +1436,6 @@ void mgmt_index_added(struct hci_dev *hdev); void mgmt_index_removed(struct hci_dev *hdev); void mgmt_set_powered_failed(struct hci_dev *hdev, int err); int mgmt_powered(struct hci_dev *hdev, u8 powered); -void mgmt_discoverable_timeout(struct hci_dev *hdev); void mgmt_new_link_key(struct hci_dev *hdev, struct link_key *key, bool persistent); void mgmt_device_connected(struct hci_dev *hdev, struct hci_conn *conn, diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index bab8958bf46e..484c75f3332c 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1537,7 +1537,6 @@ int hci_dev_do_close(struct hci_dev *hdev) flush_work(&hdev->rx_work); if (hdev->discov_timeout > 0) { - cancel_delayed_work(&hdev->discov_off); hdev->discov_timeout = 0; hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); @@ -2096,17 +2095,6 @@ static void hci_error_reset(struct work_struct *work) hci_dev_do_open(hdev); } -static void hci_discov_off(struct work_struct *work) -{ - struct hci_dev *hdev; - - hdev = container_of(work, struct hci_dev, discov_off.work); - - BT_DBG("%s", hdev->name); - - mgmt_discoverable_timeout(hdev); -} - void hci_uuids_clear(struct hci_dev *hdev) { struct bt_uuid *uuid, *tmp; @@ -2986,7 +2974,6 @@ struct hci_dev *hci_alloc_dev(void) INIT_WORK(&hdev->error_reset, hci_error_reset); INIT_DELAYED_WORK(&hdev->power_off, hci_power_off); - INIT_DELAYED_WORK(&hdev->discov_off, hci_discov_off); skb_queue_head_init(&hdev->rx_q); skb_queue_head_init(&hdev->cmd_q); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 8f72218ed805..fe14fd121d36 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1923,6 +1923,30 @@ static void discov_update(struct work_struct *work) } } +static void discov_off(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + discov_off.work); + + BT_DBG("%s", hdev->name); + + hci_dev_lock(hdev); + + /* When discoverable timeout triggers, then just make sure + * the limited discoverable flag is cleared. Even in the case + * of a timeout triggered from general discoverable, it is + * safe to unconditionally clear the flag. + */ + hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); + hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); + hdev->discov_timeout = 0; + + hci_dev_unlock(hdev); + + hci_req_sync(hdev, discoverable_update, 0, HCI_CMD_TIMEOUT, NULL); + mgmt_new_settings(hdev); +} + void hci_request_setup(struct hci_dev *hdev) { INIT_WORK(&hdev->discov_update, discov_update); @@ -1930,6 +1954,7 @@ void hci_request_setup(struct hci_dev *hdev) INIT_WORK(&hdev->scan_update, scan_update_work); INIT_WORK(&hdev->connectable_update, connectable_update_work); INIT_WORK(&hdev->discoverable_update, discoverable_update_work); + INIT_DELAYED_WORK(&hdev->discov_off, discov_off); INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire); @@ -1944,6 +1969,7 @@ void hci_request_cancel_all(struct hci_dev *hdev) cancel_work_sync(&hdev->scan_update); cancel_work_sync(&hdev->connectable_update); cancel_work_sync(&hdev->discoverable_update); + cancel_delayed_work_sync(&hdev->discov_off); cancel_delayed_work_sync(&hdev->le_scan_disable); cancel_delayed_work_sync(&hdev->le_scan_restart); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 8846cb3b0aaa..29b3bb70ae9f 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1401,8 +1401,8 @@ static int set_discoverable(struct sock *sk, struct hci_dev *hdev, void *data, if (cp->val && hdev->discov_timeout > 0) { int to = msecs_to_jiffies(hdev->discov_timeout * 1000); - queue_delayed_work(hdev->workqueue, &hdev->discov_off, - to); + queue_delayed_work(hdev->req_workqueue, + &hdev->discov_off, to); } err = send_settings_rsp(sk, MGMT_OP_SET_DISCOVERABLE, hdev); @@ -6848,43 +6848,6 @@ void mgmt_set_powered_failed(struct hci_dev *hdev, int err) mgmt_pending_remove(cmd); } -void mgmt_discoverable_timeout(struct hci_dev *hdev) -{ - struct hci_request req; - - hci_dev_lock(hdev); - - /* When discoverable timeout triggers, then just make sure - * the limited discoverable flag is cleared. Even in the case - * of a timeout triggered from general discoverable, it is - * safe to unconditionally clear the flag. - */ - hci_dev_clear_flag(hdev, HCI_LIMITED_DISCOVERABLE); - hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); - - hci_req_init(&req, hdev); - if (hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { - u8 scan = SCAN_PAGE; - hci_req_add(&req, HCI_OP_WRITE_SCAN_ENABLE, - sizeof(scan), &scan); - } - __hci_req_update_class(&req); - - /* Advertising instances don't use the global discoverable setting, so - * only update AD if advertising was enabled using Set Advertising. - */ - if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); - - hci_req_run(&req, NULL); - - hdev->discov_timeout = 0; - - new_settings(hdev, NULL); - - hci_dev_unlock(hdev); -} - void mgmt_new_link_key(struct hci_dev *hdev, struct link_key *key, bool persistent) { -- cgit v1.2.3-59-g8ed1b From 2ff13894cfb877cb3d02d96a8402202f0a6f3efd Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 25 Nov 2015 16:15:44 +0200 Subject: Bluetooth: Perform HCI update for power on synchronously The request to update HCI during power on is always coming either from hdev->req_workqueue or through an ioctl, so it's safe to use hci_req_sync for it. This way we also eliminate potential races with incoming mgmt commands or other actions while powering on. Part of this refactoring is the splitting of mgmt_powered() into mgmt_power_on() and __mgmt_power_off() functions. The main reason is the different requirements as far as hdev locking is concerned, as highlighted with the __ prefix of the power off API. Since the power on in the case of clearing the AUTO_OFF flag cannot be done synchronously in the set_powered mgmt handler, the hci_power_on work callback is extended to cover this (which also simplifies the set_powered helper a lot). Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 3 +- net/bluetooth/hci_core.c | 21 ++++-- net/bluetooth/hci_request.c | 100 ++++++++++++++++++++++++++++ net/bluetooth/hci_request.h | 2 + net/bluetooth/mgmt.c | 136 +++------------------------------------ 5 files changed, 128 insertions(+), 134 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 319bf020cea6..c95e0326c41a 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1435,7 +1435,8 @@ int mgmt_new_settings(struct hci_dev *hdev); void mgmt_index_added(struct hci_dev *hdev); void mgmt_index_removed(struct hci_dev *hdev); void mgmt_set_powered_failed(struct hci_dev *hdev, int err); -int mgmt_powered(struct hci_dev *hdev, u8 powered); +void mgmt_power_on(struct hci_dev *hdev, int err); +void __mgmt_power_off(struct hci_dev *hdev); void mgmt_new_link_key(struct hci_dev *hdev, struct link_key *key, bool persistent); void mgmt_device_connected(struct hci_dev *hdev, struct hci_conn *conn, diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 484c75f3332c..eac3f6fa1272 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1399,10 +1399,10 @@ static int hci_dev_do_open(struct hci_dev *hdev) !hci_dev_test_flag(hdev, HCI_CONFIG) && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED) && !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && + hci_dev_test_flag(hdev, HCI_MGMT) && hdev->dev_type == HCI_BREDR) { - hci_dev_lock(hdev); - mgmt_powered(hdev, 1); - hci_dev_unlock(hdev); + ret = __hci_req_hci_power_on(hdev); + mgmt_power_on(hdev, ret); } } else { /* Init failed, cleanup */ @@ -1559,8 +1559,9 @@ int hci_dev_do_close(struct hci_dev *hdev) auto_off = hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF); - if (!auto_off && hdev->dev_type == HCI_BREDR) - mgmt_powered(hdev, 0); + if (!auto_off && hdev->dev_type == HCI_BREDR && + hci_dev_test_flag(hdev, HCI_MGMT)) + __mgmt_power_off(hdev); hci_inquiry_cache_flush(hdev); hci_pend_le_actions_clear(hdev); @@ -2013,6 +2014,16 @@ static void hci_power_on(struct work_struct *work) BT_DBG("%s", hdev->name); + if (test_bit(HCI_UP, &hdev->flags) && + hci_dev_test_flag(hdev, HCI_MGMT) && + hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF)) { + hci_req_sync_lock(hdev); + err = __hci_req_hci_power_on(hdev); + hci_req_sync_unlock(hdev); + mgmt_power_on(hdev, err); + return; + } + err = hci_dev_do_open(hdev); if (err < 0) { hci_dev_lock(hdev); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 0abd83ddd4fb..7cc24f1448bd 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -2181,6 +2181,106 @@ static void discov_off(struct work_struct *work) mgmt_new_settings(hdev); } +static int powered_update_hci(struct hci_request *req, unsigned long opt) +{ + struct hci_dev *hdev = req->hdev; + struct adv_info *adv_instance; + u8 link_sec; + + hci_dev_lock(hdev); + + if (hci_dev_test_flag(hdev, HCI_SSP_ENABLED) && + !lmp_host_ssp_capable(hdev)) { + u8 mode = 0x01; + + hci_req_add(req, HCI_OP_WRITE_SSP_MODE, sizeof(mode), &mode); + + if (bredr_sc_enabled(hdev) && !lmp_host_sc_capable(hdev)) { + u8 support = 0x01; + + hci_req_add(req, HCI_OP_WRITE_SC_SUPPORT, + sizeof(support), &support); + } + } + + if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && + lmp_bredr_capable(hdev)) { + struct hci_cp_write_le_host_supported cp; + + cp.le = 0x01; + cp.simul = 0x00; + + /* Check first if we already have the right + * host state (host features set) + */ + if (cp.le != lmp_host_le_capable(hdev) || + cp.simul != lmp_host_le_br_capable(hdev)) + hci_req_add(req, HCI_OP_WRITE_LE_HOST_SUPPORTED, + sizeof(cp), &cp); + } + + if (lmp_le_capable(hdev)) { + /* Make sure the controller has a good default for + * advertising data. This also applies to the case + * where BR/EDR was toggled during the AUTO_OFF phase. + */ + if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && + (hci_dev_test_flag(hdev, HCI_ADVERTISING) || + !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE))) { + __hci_req_update_adv_data(req, HCI_ADV_CURRENT); + __hci_req_update_scan_rsp_data(req, HCI_ADV_CURRENT); + } + + if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && + hdev->cur_adv_instance == 0x00 && + !list_empty(&hdev->adv_instances)) { + adv_instance = list_first_entry(&hdev->adv_instances, + struct adv_info, list); + hdev->cur_adv_instance = adv_instance->instance; + } + + if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) + __hci_req_enable_advertising(req); + else if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && + hdev->cur_adv_instance) + __hci_req_schedule_adv_instance(req, + hdev->cur_adv_instance, + true); + } + + link_sec = hci_dev_test_flag(hdev, HCI_LINK_SECURITY); + if (link_sec != test_bit(HCI_AUTH, &hdev->flags)) + hci_req_add(req, HCI_OP_WRITE_AUTH_ENABLE, + sizeof(link_sec), &link_sec); + + if (lmp_bredr_capable(hdev)) { + if (hci_dev_test_flag(hdev, HCI_FAST_CONNECTABLE)) + __hci_req_write_fast_connectable(req, true); + else + __hci_req_write_fast_connectable(req, false); + __hci_req_update_scan(req); + __hci_req_update_class(req); + __hci_req_update_name(req); + __hci_req_update_eir(req); + } + + hci_dev_unlock(hdev); + return 0; +} + +int __hci_req_hci_power_on(struct hci_dev *hdev) +{ + /* Register the available SMP channels (BR/EDR and LE) only when + * successfully powering on the controller. This late + * registration is required so that LE SMP can clearly decide if + * the public address or static address is used. + */ + smp_register(hdev); + + return __hci_req_sync(hdev, powered_update_hci, 0, HCI_CMD_TIMEOUT, + NULL); +} + void hci_request_setup(struct hci_dev *hdev) { INIT_WORK(&hdev->discov_update, discov_update); diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index d3dd24deca74..a24d3b55094c 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -55,6 +55,8 @@ void hci_req_sync_cancel(struct hci_dev *hdev, int err); struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, const void *param); +int __hci_req_hci_power_on(struct hci_dev *hdev); + void __hci_req_write_fast_connectable(struct hci_request *req, bool enable); void __hci_req_update_name(struct hci_request *req); void __hci_req_update_eir(struct hci_request *req); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 0a7e6f4de383..468402ad933c 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -961,17 +961,6 @@ static int set_powered(struct sock *sk, struct hci_dev *hdev, void *data, goto failed; } - if (hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF)) { - cancel_delayed_work(&hdev->power_off); - - if (cp->val) { - mgmt_pending_add(sk, MGMT_OP_SET_POWERED, hdev, - data, len); - err = mgmt_powered(hdev, 1); - goto failed; - } - } - if (!!cp->val == hdev_is_powered(hdev)) { err = send_settings_rsp(sk, MGMT_OP_SET_POWERED, hdev); goto failed; @@ -6434,139 +6423,33 @@ static void restart_le_actions(struct hci_dev *hdev) } } -static void powered_complete(struct hci_dev *hdev, u8 status, u16 opcode) +void mgmt_power_on(struct hci_dev *hdev, int err) { struct cmd_lookup match = { NULL, hdev }; - BT_DBG("status 0x%02x", status); + BT_DBG("err %d", err); - if (!status) { + hci_dev_lock(hdev); + + if (!err) { restart_le_actions(hdev); hci_update_background_scan(hdev); } - hci_dev_lock(hdev); - mgmt_pending_foreach(MGMT_OP_SET_POWERED, hdev, settings_rsp, &match); new_settings(hdev, match.sk); - hci_dev_unlock(hdev); - if (match.sk) sock_put(match.sk); -} -static int powered_update_hci(struct hci_dev *hdev) -{ - struct hci_request req; - struct adv_info *adv_instance; - u8 link_sec; - - hci_req_init(&req, hdev); - - if (hci_dev_test_flag(hdev, HCI_SSP_ENABLED) && - !lmp_host_ssp_capable(hdev)) { - u8 mode = 0x01; - - hci_req_add(&req, HCI_OP_WRITE_SSP_MODE, sizeof(mode), &mode); - - if (bredr_sc_enabled(hdev) && !lmp_host_sc_capable(hdev)) { - u8 support = 0x01; - - hci_req_add(&req, HCI_OP_WRITE_SC_SUPPORT, - sizeof(support), &support); - } - } - - if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && - lmp_bredr_capable(hdev)) { - struct hci_cp_write_le_host_supported cp; - - cp.le = 0x01; - cp.simul = 0x00; - - /* Check first if we already have the right - * host state (host features set) - */ - if (cp.le != lmp_host_le_capable(hdev) || - cp.simul != lmp_host_le_br_capable(hdev)) - hci_req_add(&req, HCI_OP_WRITE_LE_HOST_SUPPORTED, - sizeof(cp), &cp); - } - - if (lmp_le_capable(hdev)) { - /* Make sure the controller has a good default for - * advertising data. This also applies to the case - * where BR/EDR was toggled during the AUTO_OFF phase. - */ - if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && - (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE))) { - __hci_req_update_adv_data(&req, HCI_ADV_CURRENT); - __hci_req_update_scan_rsp_data(&req, HCI_ADV_CURRENT); - } - - if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && - hdev->cur_adv_instance == 0x00 && - !list_empty(&hdev->adv_instances)) { - adv_instance = list_first_entry(&hdev->adv_instances, - struct adv_info, list); - hdev->cur_adv_instance = adv_instance->instance; - } - - if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - __hci_req_enable_advertising(&req); - else if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && - hdev->cur_adv_instance) - __hci_req_schedule_adv_instance(&req, - hdev->cur_adv_instance, - true); - } - - link_sec = hci_dev_test_flag(hdev, HCI_LINK_SECURITY); - if (link_sec != test_bit(HCI_AUTH, &hdev->flags)) - hci_req_add(&req, HCI_OP_WRITE_AUTH_ENABLE, - sizeof(link_sec), &link_sec); - - if (lmp_bredr_capable(hdev)) { - if (hci_dev_test_flag(hdev, HCI_FAST_CONNECTABLE)) - __hci_req_write_fast_connectable(&req, true); - else - __hci_req_write_fast_connectable(&req, false); - __hci_req_update_scan(&req); - __hci_req_update_class(&req); - __hci_req_update_name(&req); - __hci_req_update_eir(&req); - } - - return hci_req_run(&req, powered_complete); + hci_dev_unlock(hdev); } -int mgmt_powered(struct hci_dev *hdev, u8 powered) +void __mgmt_power_off(struct hci_dev *hdev) { struct cmd_lookup match = { NULL, hdev }; u8 status, zero_cod[] = { 0, 0, 0 }; - int err; - - if (!hci_dev_test_flag(hdev, HCI_MGMT)) - return 0; - - if (powered) { - /* Register the available SMP channels (BR/EDR and LE) only - * when successfully powering on the controller. This late - * registration is required so that LE SMP can clearly - * decide if the public address or static address is used. - */ - smp_register(hdev); - - if (powered_update_hci(hdev) == 0) - return 0; - - mgmt_pending_foreach(MGMT_OP_SET_POWERED, hdev, settings_rsp, - &match); - goto new_settings; - } mgmt_pending_foreach(MGMT_OP_SET_POWERED, hdev, settings_rsp, &match); @@ -6588,13 +6471,10 @@ int mgmt_powered(struct hci_dev *hdev, u8 powered) mgmt_generic_event(MGMT_EV_CLASS_OF_DEV_CHANGED, hdev, zero_cod, sizeof(zero_cod), NULL); -new_settings: - err = new_settings(hdev, match.sk); + new_settings(hdev, match.sk); if (match.sk) sock_put(match.sk); - - return err; } void mgmt_set_powered_failed(struct hci_dev *hdev, int err) -- cgit v1.2.3-59-g8ed1b From 17fd08ffb5981cff2c921eb479f46b872b02b2b9 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Thu, 26 Nov 2015 12:15:59 +0200 Subject: Bluetooth: Remove unnecessary HCI_ADVERTISING_INSTANCE flag This flag just tells us whether hdev->adv_instances is empty or not. We can equally well use the list_empty() function to get this information. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 1 - net/bluetooth/hci_request.c | 19 ++++++++----------- net/bluetooth/mgmt.c | 8 +------- 3 files changed, 9 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index cc2216727655..339ea57be423 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -239,7 +239,6 @@ enum { HCI_LE_ENABLED, HCI_ADVERTISING, HCI_ADVERTISING_CONNECTABLE, - HCI_ADVERTISING_INSTANCE, HCI_CONNECTABLE, HCI_DISCOVERABLE, HCI_LIMITED_DISCOVERABLE, diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 7cc24f1448bd..adfcd6f1d0de 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -822,7 +822,7 @@ static u8 get_current_adv_instance(struct hci_dev *hdev) * setting was set. When neither apply, default to the global settings, * represented by instance "0". */ - if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && + if (!list_empty(&hdev->adv_instances) && !hci_dev_test_flag(hdev, HCI_ADVERTISING)) return hdev->cur_adv_instance; @@ -1144,7 +1144,7 @@ void hci_req_reenable_advertising(struct hci_dev *hdev) u8 instance; if (!hci_dev_test_flag(hdev, HCI_ADVERTISING) && - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + list_empty(&hdev->adv_instances)) return; instance = get_current_adv_instance(hdev); @@ -1202,7 +1202,7 @@ int __hci_req_schedule_adv_instance(struct hci_request *req, u8 instance, u16 timeout; if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + list_empty(&hdev->adv_instances)) return -EPERM; if (hdev->adv_instance_timeout) @@ -1319,10 +1319,8 @@ void hci_req_clear_adv_instance(struct hci_dev *hdev, struct hci_request *req, } } - if (list_empty(&hdev->adv_instances)) { + if (list_empty(&hdev->adv_instances)) hdev->cur_adv_instance = 0x00; - hci_dev_clear_flag(hdev, HCI_ADVERTISING_INSTANCE); - } if (!req || !hdev_is_powered(hdev) || hci_dev_test_flag(hdev, HCI_ADVERTISING)) @@ -1525,7 +1523,7 @@ static int connectable_update(struct hci_request *req, unsigned long opt) /* Update the advertising parameters if necessary */ if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) + !list_empty(&hdev->adv_instances)) __hci_req_enable_advertising(req); __hci_update_background_scan(req); @@ -2226,13 +2224,12 @@ static int powered_update_hci(struct hci_request *req, unsigned long opt) */ if (hci_dev_test_flag(hdev, HCI_LE_ENABLED) && (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE))) { + list_empty(&hdev->adv_instances))) { __hci_req_update_adv_data(req, HCI_ADV_CURRENT); __hci_req_update_scan_rsp_data(req, HCI_ADV_CURRENT); } - if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && - hdev->cur_adv_instance == 0x00 && + if (hdev->cur_adv_instance == 0x00 && !list_empty(&hdev->adv_instances)) { adv_instance = list_first_entry(&hdev->adv_instances, struct adv_info, list); @@ -2241,7 +2238,7 @@ static int powered_update_hci(struct hci_request *req, unsigned long opt) if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) __hci_req_enable_advertising(req); - else if (hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) && + else if (!list_empty(&hdev->adv_instances) && hdev->cur_adv_instance) __hci_req_schedule_adv_instance(req, hdev->cur_adv_instance, diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 9ce2bb2fc977..03a65e89a7d7 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -3734,7 +3734,6 @@ static void set_advertising_complete(struct hci_dev *hdev, u8 status, * set up earlier, then re-enable multi-instance advertising. */ if (hci_dev_test_flag(hdev, HCI_ADVERTISING) || - !hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE) || list_empty(&hdev->adv_instances)) goto unlock; @@ -5892,9 +5891,6 @@ static void add_advertising_complete(struct hci_dev *hdev, u8 status, cmd = pending_find(MGMT_OP_ADD_ADVERTISING, hdev); - if (status) - hci_dev_clear_flag(hdev, HCI_ADVERTISING_INSTANCE); - list_for_each_entry_safe(adv_instance, n, &hdev->adv_instances, list) { if (!adv_instance->pending) continue; @@ -6012,8 +6008,6 @@ static int add_advertising(struct sock *sk, struct hci_dev *hdev, if (hdev->adv_instance_cnt > prev_instance_cnt) mgmt_advertising_added(sk, hdev, cp->instance); - hci_dev_set_flag(hdev, HCI_ADVERTISING_INSTANCE); - if (hdev->cur_adv_instance == cp->instance) { /* If the currently advertised instance is being changed then * cancel the current advertising and schedule the next @@ -6129,7 +6123,7 @@ static int remove_advertising(struct sock *sk, struct hci_dev *hdev, goto unlock; } - if (!hci_dev_test_flag(hdev, HCI_ADVERTISING_INSTANCE)) { + if (list_empty(&hdev->adv_instances)) { err = mgmt_cmd_status(sk, hdev->id, MGMT_OP_REMOVE_ADVERTISING, MGMT_STATUS_INVALID_PARAMS); goto unlock; -- cgit v1.2.3-59-g8ed1b From 00f59314111a6b18ee65b238b38c470dbdbf3be5 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 9 Dec 2015 22:46:29 +0100 Subject: 6lowpan: add lowpan dev register helpers This patch introduces register and unregister functionality for lowpan interfaces. While register a lowpan interface there are several things which need to be initialize by the 6lowpan subsystem. Upcoming functionality need to register/unregister per interface components e.g. debugfs entry. Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/net/6lowpan.h | 7 ++++++- net/6lowpan/core.c | 33 +++++++++++++++++++++++++++++++-- net/bluetooth/6lowpan.c | 8 +++----- net/ieee802154/6lowpan/core.c | 6 ++---- 4 files changed, 42 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h index cf3bc564ac03..730211fd8ed7 100644 --- a/include/net/6lowpan.h +++ b/include/net/6lowpan.h @@ -185,7 +185,12 @@ static inline void lowpan_push_hc_data(u8 **hc_ptr, const void *data, *hc_ptr += len; } -void lowpan_netdev_setup(struct net_device *dev, enum lowpan_lltypes lltype); +int lowpan_register_netdevice(struct net_device *dev, + enum lowpan_lltypes lltype); +int lowpan_register_netdev(struct net_device *dev, + enum lowpan_lltypes lltype); +void lowpan_unregister_netdevice(struct net_device *dev); +void lowpan_unregister_netdev(struct net_device *dev); /** * lowpan_header_decompress - replace 6LoWPAN header with IPv6 header diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c index 83b19e072224..80fc50987cf3 100644 --- a/net/6lowpan/core.c +++ b/net/6lowpan/core.c @@ -15,7 +15,8 @@ #include -void lowpan_netdev_setup(struct net_device *dev, enum lowpan_lltypes lltype) +int lowpan_register_netdevice(struct net_device *dev, + enum lowpan_lltypes lltype) { dev->addr_len = EUI64_ADDR_LEN; dev->type = ARPHRD_6LOWPAN; @@ -23,8 +24,36 @@ void lowpan_netdev_setup(struct net_device *dev, enum lowpan_lltypes lltype) dev->priv_flags |= IFF_NO_QUEUE; lowpan_priv(dev)->lltype = lltype; + + return register_netdevice(dev); +} +EXPORT_SYMBOL(lowpan_register_netdevice); + +int lowpan_register_netdev(struct net_device *dev, + enum lowpan_lltypes lltype) +{ + int ret; + + rtnl_lock(); + ret = lowpan_register_netdevice(dev, lltype); + rtnl_unlock(); + return ret; +} +EXPORT_SYMBOL(lowpan_register_netdev); + +void lowpan_unregister_netdevice(struct net_device *dev) +{ + unregister_netdevice(dev); +} +EXPORT_SYMBOL(lowpan_unregister_netdevice); + +void lowpan_unregister_netdev(struct net_device *dev) +{ + rtnl_lock(); + lowpan_unregister_netdevice(dev); + rtnl_unlock(); } -EXPORT_SYMBOL(lowpan_netdev_setup); +EXPORT_SYMBOL(lowpan_unregister_netdev); static int __init lowpan_module_init(void) { diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 9e9cca3689a0..d040365ba98e 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -825,9 +825,7 @@ static int setup_netdev(struct l2cap_chan *chan, struct lowpan_dev **dev) list_add_rcu(&(*dev)->list, &bt_6lowpan_devices); spin_unlock(&devices_lock); - lowpan_netdev_setup(netdev, LOWPAN_LLTYPE_BTLE); - - err = register_netdev(netdev); + err = lowpan_register_netdev(netdev, LOWPAN_LLTYPE_BTLE); if (err < 0) { BT_INFO("register_netdev failed %d", err); spin_lock(&devices_lock); @@ -890,7 +888,7 @@ static void delete_netdev(struct work_struct *work) struct lowpan_dev *entry = container_of(work, struct lowpan_dev, delete_netdev); - unregister_netdev(entry->netdev); + lowpan_unregister_netdev(entry->netdev); /* The entry pointer is deleted by the netdev destructor. */ } @@ -1348,7 +1346,7 @@ static void disconnect_devices(void) ifdown(entry->netdev); BT_DBG("Unregistering netdev %s %p", entry->netdev->name, entry->netdev); - unregister_netdev(entry->netdev); + lowpan_unregister_netdev(entry->netdev); kfree(entry); } } diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 20c49c724ba0..737c87a2a41e 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -161,9 +161,7 @@ static int lowpan_newlink(struct net *src_net, struct net_device *ldev, wdev->needed_headroom; ldev->needed_tailroom = wdev->needed_tailroom; - lowpan_netdev_setup(ldev, LOWPAN_LLTYPE_IEEE802154); - - ret = register_netdevice(ldev); + ret = lowpan_register_netdevice(ldev, LOWPAN_LLTYPE_IEEE802154); if (ret < 0) { dev_put(wdev); return ret; @@ -180,7 +178,7 @@ static void lowpan_dellink(struct net_device *ldev, struct list_head *head) ASSERT_RTNL(); wdev->ieee802154_ptr->lowpan_dev = NULL; - unregister_netdevice(ldev); + lowpan_unregister_netdevice(ldev); dev_put(wdev); } -- cgit v1.2.3-59-g8ed1b From b1815fd949e5bd06d118019acf68f87c9414f705 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 9 Dec 2015 22:46:30 +0100 Subject: 6lowpan: add debugfs support This patch will introduce a 6lowpan entry into the debugfs if enabled. Inside this 6lowpan directory we create a subdirectories of all 6lowpan interfaces to offer a per interface debugfs support. Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/net/6lowpan.h | 3 +++ net/6lowpan/6lowpan_i.h | 28 ++++++++++++++++++++++++++ net/6lowpan/Kconfig | 8 ++++++++ net/6lowpan/Makefile | 1 + net/6lowpan/core.c | 28 +++++++++++++++++++++++++- net/6lowpan/debugfs.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 120 insertions(+), 1 deletion(-) create mode 100644 net/6lowpan/6lowpan_i.h create mode 100644 net/6lowpan/debugfs.c (limited to 'include') diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h index 730211fd8ed7..2f6a3f2233ed 100644 --- a/include/net/6lowpan.h +++ b/include/net/6lowpan.h @@ -53,6 +53,8 @@ #ifndef __6LOWPAN_H__ #define __6LOWPAN_H__ +#include + #include #include @@ -98,6 +100,7 @@ enum lowpan_lltypes { struct lowpan_priv { enum lowpan_lltypes lltype; + struct dentry *iface_debugfs; /* must be last */ u8 priv[0] __aligned(sizeof(void *)); diff --git a/net/6lowpan/6lowpan_i.h b/net/6lowpan/6lowpan_i.h new file mode 100644 index 000000000000..d16bb4b14aa1 --- /dev/null +++ b/net/6lowpan/6lowpan_i.h @@ -0,0 +1,28 @@ +#ifndef __6LOWPAN_I_H +#define __6LOWPAN_I_H + +#include + +#ifdef CONFIG_6LOWPAN_DEBUGFS +int lowpan_dev_debugfs_init(struct net_device *dev); +void lowpan_dev_debugfs_exit(struct net_device *dev); + +int __init lowpan_debugfs_init(void); +void lowpan_debugfs_exit(void); +#else +static inline int lowpan_dev_debugfs_init(struct net_device *dev) +{ + return 0; +} + +static inline void lowpan_dev_debugfs_exit(struct net_device *dev) { } + +static inline int __init lowpan_debugfs_init(void) +{ + return 0; +} + +static inline void lowpan_debugfs_exit(void) { } +#endif /* CONFIG_6LOWPAN_DEBUGFS */ + +#endif /* __6LOWPAN_I_H */ diff --git a/net/6lowpan/Kconfig b/net/6lowpan/Kconfig index bcb9d8a21fc0..9c051512d14f 100644 --- a/net/6lowpan/Kconfig +++ b/net/6lowpan/Kconfig @@ -5,6 +5,14 @@ menuconfig 6LOWPAN This enables IPv6 over Low power Wireless Personal Area Network - "6LoWPAN" which is supported by IEEE 802.15.4 or Bluetooth stacks. +config 6LOWPAN_DEBUGFS + bool "6LoWPAN debugfs support" + depends on 6LOWPAN + depends on DEBUG_FS + ---help--- + This enables 6LoWPAN debugfs support. For example to manipulate + IPHC context information at runtime. + menuconfig 6LOWPAN_NHC tristate "Next Header and Generic Header Compression Support" depends on 6LOWPAN diff --git a/net/6lowpan/Makefile b/net/6lowpan/Makefile index 9e35a5d7b92d..e44f3bf2dd42 100644 --- a/net/6lowpan/Makefile +++ b/net/6lowpan/Makefile @@ -1,6 +1,7 @@ obj-$(CONFIG_6LOWPAN) += 6lowpan.o 6lowpan-y := core.o iphc.o nhc.o +6lowpan-$(CONFIG_6LOWPAN_DEBUGFS) += debugfs.o #rfc6282 nhcs obj-$(CONFIG_6LOWPAN_NHC_DEST) += nhc_dest.o diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c index 80fc50987cf3..c7f06f5c0121 100644 --- a/net/6lowpan/core.c +++ b/net/6lowpan/core.c @@ -15,9 +15,13 @@ #include +#include "6lowpan_i.h" + int lowpan_register_netdevice(struct net_device *dev, enum lowpan_lltypes lltype) { + int ret; + dev->addr_len = EUI64_ADDR_LEN; dev->type = ARPHRD_6LOWPAN; dev->mtu = IPV6_MIN_MTU; @@ -25,7 +29,15 @@ int lowpan_register_netdevice(struct net_device *dev, lowpan_priv(dev)->lltype = lltype; - return register_netdevice(dev); + ret = lowpan_dev_debugfs_init(dev); + if (ret < 0) + return ret; + + ret = register_netdevice(dev); + if (ret < 0) + lowpan_dev_debugfs_exit(dev); + + return ret; } EXPORT_SYMBOL(lowpan_register_netdevice); @@ -44,6 +56,7 @@ EXPORT_SYMBOL(lowpan_register_netdev); void lowpan_unregister_netdevice(struct net_device *dev) { unregister_netdevice(dev); + lowpan_dev_debugfs_exit(dev); } EXPORT_SYMBOL(lowpan_unregister_netdevice); @@ -57,6 +70,12 @@ EXPORT_SYMBOL(lowpan_unregister_netdev); static int __init lowpan_module_init(void) { + int ret; + + ret = lowpan_debugfs_init(); + if (ret < 0) + return ret; + request_module_nowait("ipv6"); request_module_nowait("nhc_dest"); @@ -69,6 +88,13 @@ static int __init lowpan_module_init(void) return 0; } + +static void __exit lowpan_module_exit(void) +{ + lowpan_debugfs_exit(); +} + module_init(lowpan_module_init); +module_exit(lowpan_module_exit); MODULE_LICENSE("GPL"); diff --git a/net/6lowpan/debugfs.c b/net/6lowpan/debugfs.c new file mode 100644 index 000000000000..88eef84df0fc --- /dev/null +++ b/net/6lowpan/debugfs.c @@ -0,0 +1,53 @@ +/* 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 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. + * + * Authors: + * (C) 2015 Pengutronix, Alexander Aring + * Copyright (c) 2015 Nordic Semiconductor. All Rights Reserved. + */ + +#include + +#include "6lowpan_i.h" + +static struct dentry *lowpan_debugfs; + +int lowpan_dev_debugfs_init(struct net_device *dev) +{ + struct lowpan_priv *lpriv = lowpan_priv(dev); + + /* creating the root */ + lpriv->iface_debugfs = debugfs_create_dir(dev->name, lowpan_debugfs); + if (!lpriv->iface_debugfs) + goto fail; + + return 0; + +fail: + return -EINVAL; +} + +void lowpan_dev_debugfs_exit(struct net_device *dev) +{ + debugfs_remove_recursive(lowpan_priv(dev)->iface_debugfs); +} + +int __init lowpan_debugfs_init(void) +{ + lowpan_debugfs = debugfs_create_dir("6lowpan", NULL); + if (!lowpan_debugfs) + return -EINVAL; + + return 0; +} + +void lowpan_debugfs_exit(void) +{ + debugfs_remove_recursive(lowpan_debugfs); +} -- cgit v1.2.3-59-g8ed1b From 818f1f3e70dd4c8e6f8d59c617857be0fa0fce7c Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 9 Dec 2015 22:46:31 +0100 Subject: ipv6: add ipv6_addr_prefix_copy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a static inline function ipv6_addr_prefix_copy which copies a ipv6 address prefix(argument pfx) into the ipv6 address prefix. The prefix len is given by plen as bits. This function mainly based on ipv6_addr_prefix which copies one address prefix from address into a new ipv6 address destination and zero all other address bits. The difference is that ipv6_addr_prefix_copy don't get a prefix from an ipv6 address, it sets a prefix to an ipv6 address with keeping other address bits. The use case is for context based address compression inside 6LoWPAN IPHC header which keeping ipv6 prefixes inside a context table to lookup address-bits without sending them. Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: Łukasz Duda Acked-by: Hannes Frederic Sowa Acked-by: YOSHIFUJI Hideaki Acked-by: David S. Miller Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/net/ipv6.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 9a5c9f013784..6570f379aba2 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -401,6 +401,21 @@ static inline void ipv6_addr_prefix(struct in6_addr *pfx, pfx->s6_addr[o] = addr->s6_addr[o] & (0xff00 >> b); } +static inline void ipv6_addr_prefix_copy(struct in6_addr *addr, + const struct in6_addr *pfx, + int plen) +{ + /* caller must guarantee 0 <= plen <= 128 */ + int o = plen >> 3, + b = plen & 0x7; + + memcpy(addr->s6_addr, pfx, o); + if (b != 0) { + addr->s6_addr[o] &= ~(0xff00 >> b); + addr->s6_addr[o] |= (pfx->s6_addr[o] & (0xff00 >> b)); + } +} + static inline void __ipv6_addr_set_half(__be32 *addr, __be32 wh, __be32 wl) { -- cgit v1.2.3-59-g8ed1b From 899077791403ff7a2d8cfaa87bd1a82d729463e2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 8 Dec 2015 16:32:27 +0100 Subject: netcp: try to reduce type confusion in descriptors The netcp driver produces tons of warnings when CONFIG_LPAE is enabled on ARM: drivers/net/ethernet/ti/netcp_core.c: In function 'netcp_tx_map_skb': drivers/net/ethernet/ti/netcp_core.c:1084:13: warning: passing argument 1 of 'set_words' from incompatible pointer type [-Wincompatible-pointer-types] This is the result of trying to pass a pointer to a dma_addr_t to a function that expects a u32 pointer to copy that into a DMA descriptor. Looking at that code in more detail to fix the warnings, I see multiple related problems: * The conversion functions are not endian-safe, as the DMA descriptors are almost certainly fixed-endian, but the CPU is not. * On 64-bit machines, passing a pointer through a u32 variable is a bug, accessing an indirect pointer as a u32 pointer even more so. * The handling of epib and psdata mixes native-endian and device-endian data. In this patch, I try to sort out the types for most accesses here, adding le32_to_cpu/cpu_to_le32 where appropriate, and passing pointers through two 32-bit words in the descriptor padding, to make it plausible that the driver does the right thing if compiled for big-endian or 64-bit systems. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/netcp_core.c | 123 ++++++++++++++++++++--------------- include/linux/soc/ti/knav_dma.h | 22 +++---- 2 files changed, 82 insertions(+), 63 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index e5e20e734f21..eb2585e777e1 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -109,69 +109,80 @@ module_param(netcp_debug_level, int, 0); MODULE_PARM_DESC(netcp_debug_level, "Netcp debug level (NETIF_MSG bits) (0=none,...,16=all)"); /* Helper functions - Get/Set */ -static void get_pkt_info(u32 *buff, u32 *buff_len, u32 *ndesc, +static void get_pkt_info(dma_addr_t *buff, u32 *buff_len, dma_addr_t *ndesc, struct knav_dma_desc *desc) { - *buff_len = desc->buff_len; - *buff = desc->buff; - *ndesc = desc->next_desc; + *buff_len = le32_to_cpu(desc->buff_len); + *buff = le32_to_cpu(desc->buff); + *ndesc = le32_to_cpu(desc->next_desc); } -static void get_pad_info(u32 *pad0, u32 *pad1, struct knav_dma_desc *desc) +static void get_pad_info(u32 *pad0, u32 *pad1, u32 *pad2, struct knav_dma_desc *desc) { - *pad0 = desc->pad[0]; - *pad1 = desc->pad[1]; + *pad0 = le32_to_cpu(desc->pad[0]); + *pad1 = le32_to_cpu(desc->pad[1]); + *pad2 = le32_to_cpu(desc->pad[2]); } -static void get_org_pkt_info(u32 *buff, u32 *buff_len, +static void get_pad_ptr(void **padptr, struct knav_dma_desc *desc) +{ + u64 pad64; + + pad64 = le32_to_cpu(desc->pad[0]) + + ((u64)le32_to_cpu(desc->pad[1]) << 32); + *padptr = (void *)(uintptr_t)pad64; +} + +static void get_org_pkt_info(dma_addr_t *buff, u32 *buff_len, struct knav_dma_desc *desc) { - *buff = desc->orig_buff; - *buff_len = desc->orig_len; + *buff = le32_to_cpu(desc->orig_buff); + *buff_len = le32_to_cpu(desc->orig_len); } -static void get_words(u32 *words, int num_words, u32 *desc) +static void get_words(dma_addr_t *words, int num_words, __le32 *desc) { int i; for (i = 0; i < num_words; i++) - words[i] = desc[i]; + words[i] = le32_to_cpu(desc[i]); } -static void set_pkt_info(u32 buff, u32 buff_len, u32 ndesc, +static void set_pkt_info(dma_addr_t buff, u32 buff_len, u32 ndesc, struct knav_dma_desc *desc) { - desc->buff_len = buff_len; - desc->buff = buff; - desc->next_desc = ndesc; + desc->buff_len = cpu_to_le32(buff_len); + desc->buff = cpu_to_le32(buff); + desc->next_desc = cpu_to_le32(ndesc); } static void set_desc_info(u32 desc_info, u32 pkt_info, struct knav_dma_desc *desc) { - desc->desc_info = desc_info; - desc->packet_info = pkt_info; + desc->desc_info = cpu_to_le32(desc_info); + desc->packet_info = cpu_to_le32(pkt_info); } -static void set_pad_info(u32 pad0, u32 pad1, struct knav_dma_desc *desc) +static void set_pad_info(u32 pad0, u32 pad1, u32 pad2, struct knav_dma_desc *desc) { - desc->pad[0] = pad0; - desc->pad[1] = pad1; + desc->pad[0] = cpu_to_le32(pad0); + desc->pad[1] = cpu_to_le32(pad1); + desc->pad[2] = cpu_to_le32(pad1); } -static void set_org_pkt_info(u32 buff, u32 buff_len, +static void set_org_pkt_info(dma_addr_t buff, u32 buff_len, struct knav_dma_desc *desc) { - desc->orig_buff = buff; - desc->orig_len = buff_len; + desc->orig_buff = cpu_to_le32(buff); + desc->orig_len = cpu_to_le32(buff_len); } -static void set_words(u32 *words, int num_words, u32 *desc) +static void set_words(u32 *words, int num_words, __le32 *desc) { int i; for (i = 0; i < num_words; i++) - desc[i] = words[i]; + desc[i] = cpu_to_le32(words[i]); } /* Read the e-fuse value as 32 bit values to be endian independent */ @@ -570,7 +581,7 @@ static void netcp_free_rx_desc_chain(struct netcp_intf *netcp, dma_addr_t dma_desc, dma_buf; unsigned int buf_len, dma_sz = sizeof(*ndesc); void *buf_ptr; - u32 tmp; + u32 pad[2]; get_words(&dma_desc, 1, &desc->next_desc); @@ -580,14 +591,15 @@ static void netcp_free_rx_desc_chain(struct netcp_intf *netcp, dev_err(netcp->ndev_dev, "failed to unmap Rx desc\n"); break; } - get_pkt_info(&dma_buf, &tmp, &dma_desc, ndesc); - get_pad_info((u32 *)&buf_ptr, &tmp, ndesc); + get_pad_ptr(&buf_ptr, ndesc); dma_unmap_page(netcp->dev, dma_buf, PAGE_SIZE, DMA_FROM_DEVICE); __free_page(buf_ptr); knav_pool_desc_put(netcp->rx_pool, desc); } - get_pad_info((u32 *)&buf_ptr, &buf_len, desc); + get_pad_info(&pad[0], &pad[1], &buf_len, desc); + buf_ptr = (void *)(uintptr_t)(pad[0] + ((u64)pad[1] << 32)); + if (buf_ptr) netcp_frag_free(buf_len <= PAGE_SIZE, buf_ptr); knav_pool_desc_put(netcp->rx_pool, desc); @@ -626,7 +638,6 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp) struct netcp_packet p_info; struct sk_buff *skb; void *org_buf_ptr; - u32 tmp; dma_desc = knav_queue_pop(netcp->rx_queue, &dma_sz); if (!dma_desc) @@ -639,7 +650,7 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp) } get_pkt_info(&dma_buff, &buf_len, &dma_desc, desc); - get_pad_info((u32 *)&org_buf_ptr, &org_buf_len, desc); + get_pad_ptr(&org_buf_ptr, desc); if (unlikely(!org_buf_ptr)) { dev_err(netcp->ndev_dev, "NULL bufptr in desc\n"); @@ -664,6 +675,7 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp) /* Fill in the page fragment list */ while (dma_desc) { struct page *page; + void *ptr; ndesc = knav_pool_desc_unmap(netcp->rx_pool, dma_desc, dma_sz); if (unlikely(!ndesc)) { @@ -672,14 +684,15 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp) } get_pkt_info(&dma_buff, &buf_len, &dma_desc, ndesc); - get_pad_info((u32 *)&page, &tmp, ndesc); + get_pad_ptr(ptr, ndesc); + page = ptr; if (likely(dma_buff && buf_len && page)) { dma_unmap_page(netcp->dev, dma_buff, PAGE_SIZE, DMA_FROM_DEVICE); } else { - dev_err(netcp->ndev_dev, "Bad Rx desc dma_buff(%p), len(%d), page(%p)\n", - (void *)dma_buff, buf_len, page); + dev_err(netcp->ndev_dev, "Bad Rx desc dma_buff(%pad), len(%d), page(%p)\n", + &dma_buff, buf_len, page); goto free_desc; } @@ -750,7 +763,6 @@ static void netcp_free_rx_buf(struct netcp_intf *netcp, int fdq) unsigned int buf_len, dma_sz; dma_addr_t dma; void *buf_ptr; - u32 tmp; /* Allocate descriptor */ while ((dma = knav_queue_pop(netcp->rx_fdq[fdq], &dma_sz))) { @@ -761,7 +773,7 @@ static void netcp_free_rx_buf(struct netcp_intf *netcp, int fdq) } get_org_pkt_info(&dma, &buf_len, desc); - get_pad_info((u32 *)&buf_ptr, &tmp, desc); + get_pad_ptr(buf_ptr, desc); if (unlikely(!dma)) { dev_err(netcp->ndev_dev, "NULL orig_buff in desc\n"); @@ -813,7 +825,7 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) struct page *page; dma_addr_t dma; void *bufptr; - u32 pad[2]; + u32 pad[3]; /* Allocate descriptor */ hwdesc = knav_pool_desc_get(netcp->rx_pool); @@ -830,7 +842,7 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); bufptr = netdev_alloc_frag(primary_buf_len); - pad[1] = primary_buf_len; + pad[2] = primary_buf_len; if (unlikely(!bufptr)) { dev_warn_ratelimited(netcp->ndev_dev, @@ -842,7 +854,8 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) if (unlikely(dma_mapping_error(netcp->dev, dma))) goto fail; - pad[0] = (u32)bufptr; + pad[0] = lower_32_bits((uintptr_t)bufptr); + pad[1] = upper_32_bits((uintptr_t)bufptr); } else { /* Allocate a secondary receive queue entry */ @@ -853,8 +866,9 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) } buf_len = PAGE_SIZE; dma = dma_map_page(netcp->dev, page, 0, buf_len, DMA_TO_DEVICE); - pad[0] = (u32)page; - pad[1] = 0; + pad[0] = lower_32_bits(dma); + pad[1] = upper_32_bits(dma); + pad[2] = 0; } desc_info = KNAV_DMA_DESC_PS_INFO_IN_DESC; @@ -864,7 +878,7 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq) pkt_info |= (netcp->rx_queue_id & KNAV_DMA_DESC_RETQ_MASK) << KNAV_DMA_DESC_RETQ_SHIFT; set_org_pkt_info(dma, buf_len, hwdesc); - set_pad_info(pad[0], pad[1], hwdesc); + set_pad_info(pad[0], pad[1], pad[2], hwdesc); set_desc_info(desc_info, pkt_info, hwdesc); /* Push to FDQs */ @@ -935,8 +949,8 @@ static void netcp_free_tx_desc_chain(struct netcp_intf *netcp, dma_unmap_single(netcp->dev, dma_buf, buf_len, DMA_TO_DEVICE); else - dev_warn(netcp->ndev_dev, "bad Tx desc buf(%p), len(%d)\n", - (void *)dma_buf, buf_len); + dev_warn(netcp->ndev_dev, "bad Tx desc buf(%pad), len(%d)\n", + &dma_buf, buf_len); knav_pool_desc_put(netcp->tx_pool, ndesc); ndesc = NULL; @@ -953,11 +967,11 @@ static int netcp_process_tx_compl_packets(struct netcp_intf *netcp, unsigned int budget) { struct knav_dma_desc *desc; + void *ptr; struct sk_buff *skb; unsigned int dma_sz; dma_addr_t dma; int pkts = 0; - u32 tmp; while (budget--) { dma = knav_queue_pop(netcp->tx_compl_q, &dma_sz); @@ -970,7 +984,8 @@ static int netcp_process_tx_compl_packets(struct netcp_intf *netcp, continue; } - get_pad_info((u32 *)&skb, &tmp, desc); + get_pad_ptr(&ptr, desc); + skb = ptr; netcp_free_tx_desc_chain(netcp, desc, dma_sz); if (!skb) { dev_err(netcp->ndev_dev, "No skb in Tx desc\n"); @@ -1059,6 +1074,7 @@ netcp_tx_map_skb(struct sk_buff *skb, struct netcp_intf *netcp) u32 page_offset = frag->page_offset; u32 buf_len = skb_frag_size(frag); dma_addr_t desc_dma; + u32 desc_dma_32; u32 pkt_info; dma_addr = dma_map_page(dev, page, page_offset, buf_len, @@ -1075,13 +1091,13 @@ netcp_tx_map_skb(struct sk_buff *skb, struct netcp_intf *netcp) goto free_descs; } - desc_dma = knav_pool_desc_virt_to_dma(netcp->tx_pool, - (void *)ndesc); + desc_dma = knav_pool_desc_virt_to_dma(netcp->tx_pool, ndesc); pkt_info = (netcp->tx_compl_qid & KNAV_DMA_DESC_RETQ_MASK) << KNAV_DMA_DESC_RETQ_SHIFT; set_pkt_info(dma_addr, buf_len, 0, ndesc); - set_words(&desc_dma, 1, &pdesc->next_desc); + desc_dma_32 = (u32)desc_dma; + set_words(&desc_dma_32, 1, &pdesc->next_desc); pkt_len += buf_len; if (pdesc != desc) knav_pool_desc_map(netcp->tx_pool, pdesc, @@ -1173,11 +1189,14 @@ static int netcp_tx_submit_skb(struct netcp_intf *netcp, } set_words(&tmp, 1, &desc->packet_info); - set_words((u32 *)&skb, 1, &desc->pad[0]); + tmp = lower_32_bits((uintptr_t)&skb); + set_words(&tmp, 1, &desc->pad[0]); + tmp = upper_32_bits((uintptr_t)&skb); + set_words(&tmp, 1, &desc->pad[1]); if (tx_pipe->flags & SWITCH_TO_PORT_IN_TAGINFO) { tmp = tx_pipe->switch_to_port; - set_words((u32 *)&tmp, 1, &desc->tag_info); + set_words(&tmp, 1, &desc->tag_info); } /* submit packet descriptor */ diff --git a/include/linux/soc/ti/knav_dma.h b/include/linux/soc/ti/knav_dma.h index dad035c16d94..343c13ac4f71 100644 --- a/include/linux/soc/ti/knav_dma.h +++ b/include/linux/soc/ti/knav_dma.h @@ -144,17 +144,17 @@ struct knav_dma_cfg { * @psdata: Protocol specific */ struct knav_dma_desc { - u32 desc_info; - u32 tag_info; - u32 packet_info; - u32 buff_len; - u32 buff; - u32 next_desc; - u32 orig_len; - u32 orig_buff; - u32 epib[KNAV_DMA_NUM_EPIB_WORDS]; - u32 psdata[KNAV_DMA_NUM_PS_WORDS]; - u32 pad[4]; + __le32 desc_info; + __le32 tag_info; + __le32 packet_info; + __le32 buff_len; + __le32 buff; + __le32 next_desc; + __le32 orig_len; + __le32 orig_buff; + __le32 epib[KNAV_DMA_NUM_EPIB_WORDS]; + __le32 psdata[KNAV_DMA_NUM_PS_WORDS]; + __le32 pad[4]; } ____cacheline_aligned; #if IS_ENABLED(CONFIG_KEYSTONE_NAVIGATOR_DMA) -- cgit v1.2.3-59-g8ed1b From 26a8145390b36cbe97a5bd0b9e97249f21af6aea Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Thu, 10 Dec 2015 17:12:39 +0200 Subject: net/mlx5_core: Introduce flow steering firmware commands Introduce new Flow Steering (FS) firmware commands, in-order to support the new flow steering infrastructure. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +- drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 239 ++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h | 65 ++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 81 ++++++++ include/linux/mlx5/fs.h | 47 +++++ include/linux/mlx5/mlx5_ifc.h | 32 ++- 6 files changed, 455 insertions(+), 11 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/fs_core.h create mode 100644 include/linux/mlx5/fs.h (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index a0755919ccaf..be10592e0518 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -2,7 +2,7 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ - mad.o transobj.o vport.o sriov.o + mad.o transobj.o vport.o sriov.o fs_cmd.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o eswitch.o \ en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \ en_txrx.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c new file mode 100644 index 000000000000..5096f4f336bd --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -0,0 +1,239 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include + +#include "fs_core.h" +#include "fs_cmd.h" +#include "mlx5_core.h" + +int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, + enum fs_flow_table_type type, unsigned int level, + unsigned int log_size, unsigned int *table_id) +{ + u32 out[MLX5_ST_SZ_DW(create_flow_table_out)]; + u32 in[MLX5_ST_SZ_DW(create_flow_table_in)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(create_flow_table_in, in, opcode, + MLX5_CMD_OP_CREATE_FLOW_TABLE); + + MLX5_SET(create_flow_table_in, in, table_type, type); + MLX5_SET(create_flow_table_in, in, level, level); + MLX5_SET(create_flow_table_in, in, log_size, log_size); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, + sizeof(out)); + + if (!err) + *table_id = MLX5_GET(create_flow_table_out, out, + table_id); + return err; +} + +int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft) +{ + u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(destroy_flow_table_in, in, opcode, + MLX5_CMD_OP_DESTROY_FLOW_TABLE); + MLX5_SET(destroy_flow_table_in, in, table_type, ft->type); + MLX5_SET(destroy_flow_table_in, in, table_id, ft->id); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, + sizeof(out)); +} + +int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + u32 out[MLX5_ST_SZ_DW(create_flow_group_out)]; + int err; + + memset(out, 0, sizeof(out)); + + MLX5_SET(create_flow_group_in, in, opcode, + MLX5_CMD_OP_CREATE_FLOW_GROUP); + MLX5_SET(create_flow_group_in, in, table_type, ft->type); + MLX5_SET(create_flow_group_in, in, table_id, ft->id); + + err = mlx5_cmd_exec_check_status(dev, in, + inlen, out, + sizeof(out)); + if (!err) + *group_id = MLX5_GET(create_flow_group_out, out, + group_id); + + return err; +} + +int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id) +{ + u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)]; + u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(destroy_flow_group_in, in, opcode, + MLX5_CMD_OP_DESTROY_FLOW_GROUP); + MLX5_SET(destroy_flow_group_in, in, table_type, ft->type); + MLX5_SET(destroy_flow_group_in, in, table_id, ft->id); + MLX5_SET(destroy_flow_group_in, in, group_id, group_id); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, + sizeof(out)); +} + +static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev, + int opmod, int modify_mask, + struct mlx5_flow_table *ft, + unsigned group_id, + struct fs_fte *fte) +{ + unsigned int inlen = MLX5_ST_SZ_BYTES(set_fte_in) + + fte->dests_size * MLX5_ST_SZ_BYTES(dest_format_struct); + u32 out[MLX5_ST_SZ_DW(set_fte_out)]; + struct mlx5_flow_rule *dst; + void *in_flow_context; + void *in_match_value; + void *in_dests; + u32 *in; + int err; + + in = mlx5_vzalloc(inlen); + if (!in) { + mlx5_core_warn(dev, "failed to allocate inbox\n"); + return -ENOMEM; + } + + MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY); + MLX5_SET(set_fte_in, in, op_mod, opmod); + MLX5_SET(set_fte_in, in, modify_enable_mask, modify_mask); + MLX5_SET(set_fte_in, in, table_type, ft->type); + MLX5_SET(set_fte_in, in, table_id, ft->id); + MLX5_SET(set_fte_in, in, flow_index, fte->index); + + in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context); + MLX5_SET(flow_context, in_flow_context, group_id, group_id); + MLX5_SET(flow_context, in_flow_context, flow_tag, fte->flow_tag); + MLX5_SET(flow_context, in_flow_context, action, fte->action); + MLX5_SET(flow_context, in_flow_context, destination_list_size, + fte->dests_size); + in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context, + match_value); + memcpy(in_match_value, &fte->val, MLX5_ST_SZ_BYTES(fte_match_param)); + + in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination); + list_for_each_entry(dst, &fte->node.children, node.list) { + unsigned int id; + + MLX5_SET(dest_format_struct, in_dests, destination_type, + dst->dest_attr.type); + if (dst->dest_attr.type == + MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE) + id = dst->dest_attr.ft->id; + else + id = dst->dest_attr.tir_num; + MLX5_SET(dest_format_struct, in_dests, destination_id, id); + in_dests += MLX5_ST_SZ_BYTES(dest_format_struct); + } + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, inlen, out, + sizeof(out)); + kvfree(in); + + return err; +} + +int mlx5_cmd_create_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned group_id, + struct fs_fte *fte) +{ + return mlx5_cmd_set_fte(dev, 0, 0, ft, group_id, fte); +} + +int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned group_id, + struct fs_fte *fte) +{ + int opmod; + int modify_mask; + int atomic_mod_cap = MLX5_CAP_FLOWTABLE(dev, + flow_table_properties_nic_receive. + flow_modify_en); + if (!atomic_mod_cap) + return -ENOTSUPP; + opmod = 1; + modify_mask = 1 << + MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST; + + return mlx5_cmd_set_fte(dev, opmod, modify_mask, ft, group_id, fte); +} + +int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int index) +{ + u32 out[MLX5_ST_SZ_DW(delete_fte_out)]; + u32 in[MLX5_ST_SZ_DW(delete_fte_in)]; + int err; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(delete_fte_in, in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY); + MLX5_SET(delete_fte_in, in, table_type, ft->type); + MLX5_SET(delete_fte_in, in, table_id, ft->id); + MLX5_SET(delete_fte_in, in, flow_index, index); + + err = mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out)); + + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h new file mode 100644 index 000000000000..f39304ede186 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -0,0 +1,65 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _MLX5_FS_CMD_ +#define _MLX5_FS_CMD_ + +int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, + enum fs_flow_table_type type, unsigned int level, + unsigned int log_size, unsigned int *table_id); + +int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft); + +int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, unsigned int *group_id); + +int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id); + +int mlx5_cmd_create_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned group_id, + struct fs_fte *fte); + +int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned group_id, + struct fs_fte *fte); + +int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int index); + +#endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h new file mode 100644 index 000000000000..e8b34a9b147b --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -0,0 +1,81 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _MLX5_FS_CORE_ +#define _MLX5_FS_CORE_ + +#include + +enum fs_node_type { + FS_TYPE_NAMESPACE, + FS_TYPE_PRIO, + FS_TYPE_FLOW_TABLE, + FS_TYPE_FLOW_GROUP, + FS_TYPE_FLOW_ENTRY, + FS_TYPE_FLOW_DEST +}; + +enum fs_flow_table_type { + FS_FT_NIC_RX = 0x0, +}; + +enum fs_fte_status { + FS_FTE_STATUS_EXISTING = 1UL << 0, +}; + +struct fs_node { + struct list_head list; + struct list_head children; + enum fs_node_type type; +}; + +struct mlx5_flow_rule { + struct fs_node node; + struct mlx5_flow_destination dest_attr; +}; + +struct mlx5_flow_table { + struct fs_node node; + u32 id; + enum fs_flow_table_type type; +}; + +struct fs_fte { + struct fs_node node; + u32 val[MLX5_ST_SZ_DW(fte_match_param)]; + u32 dests_size; + u32 flow_tag; + u32 index; + u32 action; +}; + +#endif diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h new file mode 100644 index 000000000000..34fd8dc0b3e1 --- /dev/null +++ b/include/linux/mlx5/fs.h @@ -0,0 +1,47 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _MLX5_FS_ +#define _MLX5_FS_ + +#include + +struct mlx5_flow_table; + +struct mlx5_flow_destination { + enum mlx5_flow_destination_type type; + union { + u32 tir_num; + struct mlx5_flow_table *ft; + }; +}; +#endif diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index f5d94495758a..131a2737cfa3 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -256,25 +256,27 @@ struct mlx5_ifc_flow_table_fields_supported_bits { struct mlx5_ifc_flow_table_prop_layout_bits { u8 ft_support[0x1]; - u8 reserved_0[0x1f]; + u8 reserved_0[0x2]; + u8 flow_modify_en[0x1]; + u8 reserved_1[0x1c]; - u8 reserved_1[0x2]; + u8 reserved_2[0x2]; u8 log_max_ft_size[0x6]; - u8 reserved_2[0x10]; + u8 reserved_3[0x10]; u8 max_ft_level[0x8]; - u8 reserved_3[0x20]; + u8 reserved_4[0x20]; - u8 reserved_4[0x18]; + u8 reserved_5[0x18]; u8 log_max_ft_num[0x8]; - u8 reserved_5[0x18]; + u8 reserved_6[0x18]; u8 log_max_destination[0x8]; - u8 reserved_6[0x18]; + u8 reserved_7[0x18]; u8 log_max_flow[0x8]; - u8 reserved_7[0x40]; + u8 reserved_8[0x40]; struct mlx5_ifc_flow_table_fields_supported_bits ft_field_support; @@ -2843,6 +2845,13 @@ struct mlx5_ifc_set_hca_cap_in_bits { union mlx5_ifc_hca_cap_union_bits capability; }; +enum { + MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION = 0x0, + MLX5_SET_FTE_MODIFY_ENABLE_MASK_FLOW_TAG = 0x1, + MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST = 0x2, + MLX5_SET_FTE_MODIFY_ENABLE_MASK_FLOW_COUNTERS = 0x3 +}; + struct mlx5_ifc_set_fte_out_bits { u8 status[0x8]; u8 reserved_0[0x18]; @@ -2867,11 +2876,14 @@ struct mlx5_ifc_set_fte_in_bits { u8 reserved_4[0x8]; u8 table_id[0x18]; - u8 reserved_5[0x40]; + u8 reserved_5[0x18]; + u8 modify_enable_mask[0x8]; + + u8 reserved_6[0x20]; u8 flow_index[0x20]; - u8 reserved_6[0xe0]; + u8 reserved_7[0xe0]; struct mlx5_ifc_flow_context_bits flow_context; }; -- cgit v1.2.3-59-g8ed1b From 2530236303d9e705db6a28eb9a10c8d79b288b37 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Thu, 10 Dec 2015 17:12:43 +0200 Subject: net/mlx5_core: Flow steering tree initialization Flow steering initialization is based on static tree which illustrates the flow steering tree when the driver is loaded. The initialization considers the max supported flow table level of the device, a minimum of 2 kernel flow tables(vlan and mac) are required to have kernel flow table functionality. The tree structures when the driver is loaded: root_namespace(receive nic) | priority-0 (kernel priority) | namespace(kernel namespace) | priority-0 (flow tables priority) In the following patches, When the EN driver will use the flow steering API, it create two flow tables and their flow groups under priority-0(flow tables priority). Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 374 ++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 4 + include/linux/mlx5/driver.h | 2 + include/linux/mlx5/fs.h | 8 + 4 files changed, 388 insertions(+) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 1828351102c5..4264e8b34b76 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -37,6 +37,54 @@ #include "fs_core.h" #include "fs_cmd.h" +#define INIT_TREE_NODE_ARRAY_SIZE(...) (sizeof((struct init_tree_node[]){__VA_ARGS__}) /\ + sizeof(struct init_tree_node)) + +#define INIT_PRIO(min_level_val, max_ft_val,\ + start_level_val, ...) {.type = FS_TYPE_PRIO,\ + .min_ft_level = min_level_val,\ + .start_level = start_level_val,\ + .max_ft = max_ft_val,\ + .children = (struct init_tree_node[]) {__VA_ARGS__},\ + .ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \ +} + +#define ADD_PRIO(min_level_val, max_ft_val, start_level_val, ...)\ + INIT_PRIO(min_level_val, max_ft_val, start_level_val,\ + __VA_ARGS__)\ + +#define ADD_FT_PRIO(max_ft_val, start_level_val, ...)\ + INIT_PRIO(0, max_ft_val, start_level_val,\ + __VA_ARGS__)\ + +#define ADD_NS(...) {.type = FS_TYPE_NAMESPACE,\ + .children = (struct init_tree_node[]) {__VA_ARGS__},\ + .ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \ +} + +#define KERNEL_START_LEVEL 0 +#define KERNEL_P0_START_LEVEL KERNEL_START_LEVEL +#define KERNEL_MAX_FT 2 +#define KENREL_MIN_LEVEL 2 +static struct init_tree_node { + enum fs_node_type type; + struct init_tree_node *children; + int ar_size; + int min_ft_level; + int prio; + int max_ft; + int start_level; +} root_fs = { + .type = FS_TYPE_NAMESPACE, + .ar_size = 1, + .children = (struct init_tree_node[]) { + ADD_PRIO(KENREL_MIN_LEVEL, KERNEL_MAX_FT, + KERNEL_START_LEVEL, + ADD_NS(ADD_FT_PRIO(KERNEL_MAX_FT, + KERNEL_P0_START_LEVEL))), + } +}; + static void del_rule(struct fs_node *node); static void del_flow_table(struct fs_node *node); static void del_flow_group(struct fs_node *node); @@ -671,3 +719,329 @@ static void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) mlx5_core_warn(get_dev(&fg->node), "Flow group %d wasn't destroyed, refcount > 1\n", fg->id); } + +static struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, + enum mlx5_flow_namespace_type type) +{ + struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + int prio; + static struct fs_prio *fs_prio; + struct mlx5_flow_namespace *ns; + + if (!root_ns) + return NULL; + + switch (type) { + case MLX5_FLOW_NAMESPACE_KERNEL: + prio = 0; + break; + case MLX5_FLOW_NAMESPACE_FDB: + if (dev->priv.fdb_root_ns) + return &dev->priv.fdb_root_ns->ns; + else + return NULL; + default: + return NULL; + } + + fs_prio = find_prio(&root_ns->ns, prio); + if (!fs_prio) + return NULL; + + ns = list_first_entry(&fs_prio->node.children, + typeof(*ns), + node.list); + + return ns; +} + +static struct fs_prio *fs_create_prio(struct mlx5_flow_namespace *ns, + unsigned prio, int max_ft, + int start_level) +{ + struct fs_prio *fs_prio; + + fs_prio = kzalloc(sizeof(*fs_prio), GFP_KERNEL); + if (!fs_prio) + return ERR_PTR(-ENOMEM); + + fs_prio->node.type = FS_TYPE_PRIO; + tree_init_node(&fs_prio->node, 1, NULL); + tree_add_node(&fs_prio->node, &ns->node); + fs_prio->max_ft = max_ft; + fs_prio->prio = prio; + fs_prio->start_level = start_level; + list_add_tail(&fs_prio->node.list, &ns->node.children); + + return fs_prio; +} + +static struct mlx5_flow_namespace *fs_init_namespace(struct mlx5_flow_namespace + *ns) +{ + ns->node.type = FS_TYPE_NAMESPACE; + + return ns; +} + +static struct mlx5_flow_namespace *fs_create_namespace(struct fs_prio *prio) +{ + struct mlx5_flow_namespace *ns; + + ns = kzalloc(sizeof(*ns), GFP_KERNEL); + if (!ns) + return ERR_PTR(-ENOMEM); + + fs_init_namespace(ns); + tree_init_node(&ns->node, 1, NULL); + tree_add_node(&ns->node, &prio->node); + list_add_tail(&ns->node.list, &prio->node.children); + + return ns; +} + +static int init_root_tree_recursive(int max_ft_level, struct init_tree_node *init_node, + struct fs_node *fs_parent_node, + struct init_tree_node *init_parent_node, + int index) +{ + struct mlx5_flow_namespace *fs_ns; + struct fs_prio *fs_prio; + struct fs_node *base; + int i; + int err; + + if (init_node->type == FS_TYPE_PRIO) { + if (init_node->min_ft_level > max_ft_level) + return -ENOTSUPP; + + fs_get_obj(fs_ns, fs_parent_node); + fs_prio = fs_create_prio(fs_ns, index, init_node->max_ft, + init_node->start_level); + if (IS_ERR(fs_prio)) + return PTR_ERR(fs_prio); + base = &fs_prio->node; + } else if (init_node->type == FS_TYPE_NAMESPACE) { + fs_get_obj(fs_prio, fs_parent_node); + fs_ns = fs_create_namespace(fs_prio); + if (IS_ERR(fs_ns)) + return PTR_ERR(fs_ns); + base = &fs_ns->node; + } else { + return -EINVAL; + } + for (i = 0; i < init_node->ar_size; i++) { + err = init_root_tree_recursive(max_ft_level, + &init_node->children[i], base, + init_node, i); + if (err) + return err; + } + + return 0; +} + +static int init_root_tree(int max_ft_level, struct init_tree_node *init_node, + struct fs_node *fs_parent_node) +{ + int i; + struct mlx5_flow_namespace *fs_ns; + int err; + + fs_get_obj(fs_ns, fs_parent_node); + for (i = 0; i < init_node->ar_size; i++) { + err = init_root_tree_recursive(max_ft_level, + &init_node->children[i], + &fs_ns->node, + init_node, i); + if (err) + return err; + } + return 0; +} + +static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev, + enum fs_flow_table_type + table_type) +{ + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_namespace *ns; + + /* create the root namespace */ + root_ns = mlx5_vzalloc(sizeof(*root_ns)); + if (!root_ns) + return NULL; + + root_ns->dev = dev; + root_ns->table_type = table_type; + + ns = &root_ns->ns; + fs_init_namespace(ns); + tree_init_node(&ns->node, 1, NULL); + tree_add_node(&ns->node, NULL); + + return root_ns; +} + +static int init_root_ns(struct mlx5_core_dev *dev) +{ + int max_ft_level = MLX5_CAP_FLOWTABLE(dev, + flow_table_properties_nic_receive. + max_ft_level); + + dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX); + if (IS_ERR_OR_NULL(dev->priv.root_ns)) + goto cleanup; + + if (init_root_tree(max_ft_level, &root_fs, &dev->priv.root_ns->ns.node)) + goto cleanup; + + return 0; + +cleanup: + mlx5_cleanup_fs(dev); + return -ENOMEM; +} + +static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, + struct mlx5_flow_root_namespace *root_ns) +{ + struct fs_node *prio; + + if (!root_ns) + return; + + if (!list_empty(&root_ns->ns.node.children)) { + prio = list_first_entry(&root_ns->ns.node.children, + struct fs_node, + list); + if (tree_remove_node(prio)) + mlx5_core_warn(dev, + "Flow steering priority wasn't destroyed, refcount > 1\n"); + } + if (tree_remove_node(&root_ns->ns.node)) + mlx5_core_warn(dev, + "Flow steering namespace wasn't destroyed, refcount > 1\n"); + root_ns = NULL; +} + +static void cleanup_root_ns(struct mlx5_core_dev *dev) +{ + struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct fs_prio *iter_prio; + + if (!MLX5_CAP_GEN(dev, nic_flow_table)) + return; + + if (!root_ns) + return; + + /* stage 1 */ + fs_for_each_prio(iter_prio, &root_ns->ns) { + struct fs_node *node; + struct mlx5_flow_namespace *iter_ns; + + fs_for_each_ns_or_ft(node, iter_prio) { + if (node->type == FS_TYPE_FLOW_TABLE) + continue; + fs_get_obj(iter_ns, node); + while (!list_empty(&iter_ns->node.children)) { + struct fs_prio *obj_iter_prio2; + struct fs_node *iter_prio2 = + list_first_entry(&iter_ns->node.children, + struct fs_node, + list); + + fs_get_obj(obj_iter_prio2, iter_prio2); + if (tree_remove_node(iter_prio2)) { + mlx5_core_warn(dev, + "Priority %d wasn't destroyed, refcount > 1\n", + obj_iter_prio2->prio); + return; + } + } + } + } + + /* stage 2 */ + fs_for_each_prio(iter_prio, &root_ns->ns) { + while (!list_empty(&iter_prio->node.children)) { + struct fs_node *iter_ns = + list_first_entry(&iter_prio->node.children, + struct fs_node, + list); + if (tree_remove_node(iter_ns)) { + mlx5_core_warn(dev, + "Namespace wasn't destroyed, refcount > 1\n"); + return; + } + } + } + + /* stage 3 */ + while (!list_empty(&root_ns->ns.node.children)) { + struct fs_prio *obj_prio_node; + struct fs_node *prio_node = + list_first_entry(&root_ns->ns.node.children, + struct fs_node, + list); + + fs_get_obj(obj_prio_node, prio_node); + if (tree_remove_node(prio_node)) { + mlx5_core_warn(dev, + "Priority %d wasn't destroyed, refcount > 1\n", + obj_prio_node->prio); + return; + } + } + + if (tree_remove_node(&root_ns->ns.node)) { + mlx5_core_warn(dev, + "root namespace wasn't destroyed, refcount > 1\n"); + return; + } + + dev->priv.root_ns = NULL; +} + +void mlx5_cleanup_fs(struct mlx5_core_dev *dev) +{ + cleanup_root_ns(dev); + cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); +} + +static int init_fdb_root_ns(struct mlx5_core_dev *dev) +{ + struct fs_prio *prio; + + dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB); + if (!dev->priv.fdb_root_ns) + return -ENOMEM; + + /* create 1 prio*/ + prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1, 0); + if (IS_ERR(prio)) { + cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); + return PTR_ERR(prio); + } else { + return 0; + } +} + +int mlx5_init_fs(struct mlx5_core_dev *dev) +{ + int err = 0; + + if (MLX5_CAP_GEN(dev, nic_flow_table)) { + err = init_root_ns(dev); + if (err) + return err; + } + if (MLX5_CAP_GEN(dev, eswitch_flow_table)) { + err = init_fdb_root_ns(dev); + if (err) + cleanup_root_ns(dev); + } + + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 6c27b8ef42b7..4ebb97fd5544 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -46,6 +46,7 @@ enum fs_node_type { enum fs_flow_table_type { FS_FT_NIC_RX = 0x0, + FS_FT_FDB = 0X4, }; enum fs_fte_status { @@ -125,6 +126,9 @@ struct mlx5_flow_root_namespace { struct mlx5_core_dev *dev; }; +int mlx5_init_fs(struct mlx5_core_dev *dev); +void mlx5_cleanup_fs(struct mlx5_core_dev *dev); + #define fs_get_obj(v, _node) {v = container_of((_node), typeof(*v), node); } #define fs_list_for_each_entry(pos, root) \ diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index ac098b6b97bf..2fd7019f69db 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -502,6 +502,8 @@ struct mlx5_priv { struct mlx5_eswitch *eswitch; struct mlx5_core_sriov sriov; unsigned long pci_dev_data; + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_root_namespace *fdb_root_ns; }; enum mlx5_device_state { diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 34fd8dc0b3e1..16ae5233dc7b 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -35,6 +35,13 @@ #include +#define MLX5_FS_DEFAULT_FLOW_TAG 0x0 + +enum mlx5_flow_namespace_type { + MLX5_FLOW_NAMESPACE_KERNEL, + MLX5_FLOW_NAMESPACE_FDB, +}; + struct mlx5_flow_table; struct mlx5_flow_destination { @@ -42,6 +49,7 @@ struct mlx5_flow_destination { union { u32 tir_num; struct mlx5_flow_table *ft; + u32 vport_num; }; }; #endif -- cgit v1.2.3-59-g8ed1b From 86d722ad2c3bd2f0536b196b7fd67ae2a7e2a492 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Thu, 10 Dec 2015 17:12:44 +0200 Subject: net/mlx5: Use flow steering infrastructure for mlx5_en Expose the new flow steering API and remove the old one. Few changes are required: 1. The Ethernet flow steering follows the existing implementation, but uses the new steering API. The old flow steering implementation is removed. 2. Move the E-switch FDB management to use the new API. 3. When driver is loaded call to mlx5_init_fs which initialize the flow steering tree structure, open namespaces for NIC receive and for E-switch FDB. 4. Call to mlx5_cleanup_fs when the driver is unloaded. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en.h | 23 +- .../ethernet/mellanox/mlx5/core/en_flow_table.c | 824 +++++++++++++-------- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 291 ++------ drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 15 +- .../net/ethernet/mellanox/mlx5/core/flow_table.c | 422 ----------- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 26 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 9 + include/linux/mlx5/flow_table.h | 63 -- include/linux/mlx5/fs.h | 38 + 11 files changed, 633 insertions(+), 1082 deletions(-) delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/flow_table.c delete mode 100644 include/linux/mlx5/flow_table.h (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 7fc5e2388dec..11ee062965c5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -3,6 +3,6 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o -mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o eswitch.o \ +mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \ en_txrx.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 89313d46952d..f689ce580b44 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -64,6 +64,8 @@ #define MLX5E_UPDATE_STATS_INTERVAL 200 /* msecs */ #define MLX5E_SQ_BF_BUDGET 16 +#define MLX5E_NUM_MAIN_GROUPS 9 + static const char vport_strings[][ETH_GSTRING_LEN] = { /* vport statistics */ "rx_packets", @@ -442,7 +444,7 @@ enum mlx5e_rqt_ix { struct mlx5e_eth_addr_info { u8 addr[ETH_ALEN + 2]; u32 tt_vec; - u32 ft_ix[MLX5E_NUM_TT]; /* flow table index per traffic type */ + struct mlx5_flow_rule *ft_rule[MLX5E_NUM_TT]; }; #define MLX5E_ETH_ADDR_HASH_SIZE (1 << BITS_PER_BYTE) @@ -466,15 +468,22 @@ enum { struct mlx5e_vlan_db { unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; - u32 active_vlans_ft_ix[VLAN_N_VID]; - u32 untagged_rule_ft_ix; - u32 any_vlan_rule_ft_ix; + struct mlx5_flow_rule *active_vlans_rule[VLAN_N_VID]; + struct mlx5_flow_rule *untagged_rule; + struct mlx5_flow_rule *any_vlan_rule; bool filter_disabled; }; struct mlx5e_flow_table { - void *vlan; - void *main; + int num_groups; + struct mlx5_flow_table *t; + struct mlx5_flow_group **g; +}; + +struct mlx5e_flow_tables { + struct mlx5_flow_namespace *ns; + struct mlx5e_flow_table vlan; + struct mlx5e_flow_table main; }; struct mlx5e_priv { @@ -497,7 +506,7 @@ struct mlx5e_priv { u32 rqtn[MLX5E_NUM_RQT]; u32 tirn[MLX5E_NUM_TT]; - struct mlx5e_flow_table ft; + struct mlx5e_flow_tables fts; struct mlx5e_eth_addr_db eth_addr; struct mlx5e_vlan_db vlan; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c index 5b93c9c6e341..80d81abc4820 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c @@ -34,9 +34,11 @@ #include #include #include -#include +#include #include "en.h" +#define MLX5_SET_CFG(p, f, v) MLX5_SET(create_flow_group_in, p, f, v) + enum { MLX5E_FULLMATCH = 0, MLX5E_ALLMULTI = 1, @@ -103,44 +105,38 @@ static void mlx5e_del_eth_addr_from_hash(struct mlx5e_eth_addr_hash_node *hn) static void mlx5e_del_eth_addr_from_flow_table(struct mlx5e_priv *priv, struct mlx5e_eth_addr_info *ai) { - void *ft = priv->ft.main; - if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP)) - mlx5_del_flow_table_entry(ft, - ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP)) - mlx5_del_flow_table_entry(ft, - ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH)) - mlx5_del_flow_table_entry(ft, - ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH]); if (ai->tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH)) - mlx5_del_flow_table_entry(ft, - ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH]); if (ai->tt_vec & BIT(MLX5E_TT_IPV6_TCP)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_TCP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_TCP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV4_TCP)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_TCP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_TCP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV6_UDP)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_UDP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6_UDP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV4_UDP)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_UDP]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4_UDP]); if (ai->tt_vec & BIT(MLX5E_TT_IPV6)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV6]); if (ai->tt_vec & BIT(MLX5E_TT_IPV4)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_IPV4]); if (ai->tt_vec & BIT(MLX5E_TT_ANY)) - mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_ANY]); + mlx5_del_flow_rule(ai->ft_rule[MLX5E_TT_ANY]); } static int mlx5e_get_eth_addr_type(u8 *addr) @@ -240,44 +236,34 @@ static u32 mlx5e_get_tt_vec(struct mlx5e_eth_addr_info *ai, int type) } static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv, - struct mlx5e_eth_addr_info *ai, int type, - void *flow_context, void *match_criteria) + struct mlx5e_eth_addr_info *ai, + int type, u32 *mc, u32 *mv) { + struct mlx5_flow_destination dest; u8 match_criteria_enable = 0; - void *match_value; - void *dest; - u8 *dmac; - u8 *match_criteria_dmac; - void *ft = priv->ft.main; - u32 *tirn = priv->tirn; - u32 *ft_ix; - u32 tt_vec; - int err; - - match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); - dmac = MLX5_ADDR_OF(fte_match_param, match_value, - outer_headers.dmac_47_16); - match_criteria_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, - outer_headers.dmac_47_16); - dest = MLX5_ADDR_OF(flow_context, flow_context, destination); - - MLX5_SET(flow_context, flow_context, action, - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST); - MLX5_SET(flow_context, flow_context, destination_list_size, 1); - MLX5_SET(dest_format_struct, dest, destination_type, - MLX5_FLOW_CONTEXT_DEST_TYPE_TIR); + struct mlx5_flow_rule **rule_p; + struct mlx5_flow_table *ft = priv->fts.main.t; + u8 *mc_dmac = MLX5_ADDR_OF(fte_match_param, mc, + outer_headers.dmac_47_16); + u8 *mv_dmac = MLX5_ADDR_OF(fte_match_param, mv, + outer_headers.dmac_47_16); + u32 *tirn = priv->tirn; + u32 tt_vec; + int err = 0; + + dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; switch (type) { case MLX5E_FULLMATCH: match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - memset(match_criteria_dmac, 0xff, ETH_ALEN); - ether_addr_copy(dmac, ai->addr); + eth_broadcast_addr(mc_dmac); + ether_addr_copy(mv_dmac, ai->addr); break; case MLX5E_ALLMULTI: match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - match_criteria_dmac[0] = 0x01; - dmac[0] = 0x01; + mc_dmac[0] = 0x01; + mv_dmac[0] = 0x01; break; case MLX5E_PROMISC: @@ -286,190 +272,165 @@ static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv, tt_vec = mlx5e_get_tt_vec(ai, type); - ft_ix = &ai->ft_ix[MLX5E_TT_ANY]; if (tt_vec & BIT(MLX5E_TT_ANY)) { - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_ANY]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + rule_p = &ai->ft_rule[MLX5E_TT_ANY]; + dest.tir_num = tirn[MLX5E_TT_ANY]; + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_ANY); } match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, - outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); - ft_ix = &ai->ft_ix[MLX5E_TT_IPV4]; if (tt_vec & BIT(MLX5E_TT_IPV4)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV4]; + dest.tir_num = tirn[MLX5E_TT_IPV4]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IP); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV4]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV4); } - ft_ix = &ai->ft_ix[MLX5E_TT_IPV6]; if (tt_vec & BIT(MLX5E_TT_IPV6)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV6]; + dest.tir_num = tirn[MLX5E_TT_IPV6]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IPV6); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV6]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV6); } - MLX5_SET_TO_ONES(fte_match_param, match_criteria, - outer_headers.ip_protocol); - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, - IPPROTO_UDP); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol); + MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_UDP); - ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_UDP]; if (tt_vec & BIT(MLX5E_TT_IPV4_UDP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV4_UDP]; + dest.tir_num = tirn[MLX5E_TT_IPV4_UDP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IP); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV4_UDP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV4_UDP); } - ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_UDP]; if (tt_vec & BIT(MLX5E_TT_IPV6_UDP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV6_UDP]; + dest.tir_num = tirn[MLX5E_TT_IPV6_UDP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IPV6); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV6_UDP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV6_UDP); } - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, - IPPROTO_TCP); + MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_TCP); - ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_TCP]; if (tt_vec & BIT(MLX5E_TT_IPV4_TCP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV4_TCP]; + dest.tir_num = tirn[MLX5E_TT_IPV4_TCP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IP); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV4_TCP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV4_TCP); } - ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_TCP]; if (tt_vec & BIT(MLX5E_TT_IPV6_TCP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV6_TCP]; + dest.tir_num = tirn[MLX5E_TT_IPV6_TCP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IPV6); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV6_TCP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; ai->tt_vec |= BIT(MLX5E_TT_IPV6_TCP); } - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, - IPPROTO_AH); + MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_AH); - ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_AH]; if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_AH)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_AH]; + dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_AH]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IP); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV4_IPSEC_AH]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_AH); } - ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_AH]; if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_AH)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_AH]; + dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_AH]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IPV6); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV6_IPSEC_AH]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_AH); } - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, - IPPROTO_ESP); + MLX5_SET(fte_match_param, mv, outer_headers.ip_protocol, IPPROTO_ESP); - ft_ix = &ai->ft_ix[MLX5E_TT_IPV4_IPSEC_ESP]; if (tt_vec & BIT(MLX5E_TT_IPV4_IPSEC_ESP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV4_IPSEC_ESP]; + dest.tir_num = tirn[MLX5E_TT_IPV4_IPSEC_ESP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IP); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV4_IPSEC_ESP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV4_IPSEC_ESP); } - ft_ix = &ai->ft_ix[MLX5E_TT_IPV6_IPSEC_ESP]; if (tt_vec & BIT(MLX5E_TT_IPV6_IPSEC_ESP)) { - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + rule_p = &ai->ft_rule[MLX5E_TT_IPV6_IPSEC_ESP]; + dest.tir_num = tirn[MLX5E_TT_IPV6_IPSEC_ESP]; + MLX5_SET(fte_match_param, mv, outer_headers.ethertype, ETH_P_IPV6); - MLX5_SET(dest_format_struct, dest, destination_id, - tirn[MLX5E_TT_IPV6_IPSEC_ESP]); - err = mlx5_add_flow_table_entry(ft, match_criteria_enable, - match_criteria, flow_context, - ft_ix); - if (err) + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, &dest); + if (IS_ERR_OR_NULL(*rule_p)) goto err_del_ai; - ai->tt_vec |= BIT(MLX5E_TT_IPV6_IPSEC_ESP); } return 0; err_del_ai: + err = PTR_ERR(*rule_p); + *rule_p = NULL; mlx5e_del_eth_addr_from_flow_table(priv, ai); return err; @@ -478,27 +439,25 @@ err_del_ai: static int mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv, struct mlx5e_eth_addr_info *ai, int type) { - u32 *flow_context; u32 *match_criteria; - int err; + u32 *match_value; + int err = 0; - flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + - MLX5_ST_SZ_BYTES(dest_format_struct)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!flow_context || !match_criteria) { + match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + if (!match_value || !match_criteria) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto add_eth_addr_rule_out; } - err = __mlx5e_add_eth_addr_rule(priv, ai, type, flow_context, - match_criteria); - if (err) - netdev_err(priv->netdev, "%s: failed\n", __func__); + err = __mlx5e_add_eth_addr_rule(priv, ai, type, match_criteria, + match_value); add_eth_addr_rule_out: kvfree(match_criteria); - kvfree(flow_context); + kvfree(match_value); + return err; } @@ -551,72 +510,77 @@ enum mlx5e_vlan_rule_type { MLX5E_VLAN_RULE_TYPE_MATCH_VID, }; -static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, - enum mlx5e_vlan_rule_type rule_type, u16 vid) +static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, + enum mlx5e_vlan_rule_type rule_type, + u16 vid, u32 *mc, u32 *mv) { + struct mlx5_flow_table *ft = priv->fts.vlan.t; + struct mlx5_flow_destination dest; u8 match_criteria_enable = 0; - u32 *flow_context; - void *match_value; - void *dest; - u32 *match_criteria; - u32 *ft_ix; - int err; + struct mlx5_flow_rule **rule_p; + int err = 0; - flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + - MLX5_ST_SZ_BYTES(dest_format_struct)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!flow_context || !match_criteria) { - netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_vlan_rule_out; - } - match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); - dest = MLX5_ADDR_OF(flow_context, flow_context, destination); - - MLX5_SET(flow_context, flow_context, action, - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST); - MLX5_SET(flow_context, flow_context, destination_list_size, 1); - MLX5_SET(dest_format_struct, dest, destination_type, - MLX5_FLOW_CONTEXT_DEST_TYPE_FLOW_TABLE); - MLX5_SET(dest_format_struct, dest, destination_id, - mlx5_get_flow_table_id(priv->ft.main)); + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = priv->fts.main.t; match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, - outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); switch (rule_type) { case MLX5E_VLAN_RULE_TYPE_UNTAGGED: - ft_ix = &priv->vlan.untagged_rule_ft_ix; + rule_p = &priv->vlan.untagged_rule; break; case MLX5E_VLAN_RULE_TYPE_ANY_VID: - ft_ix = &priv->vlan.any_vlan_rule_ft_ix; - MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag, - 1); + rule_p = &priv->vlan.any_vlan_rule; + MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); break; default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */ - err = mlx5e_vport_context_update_vlans(priv); - if (err) - goto add_vlan_rule_out; - - ft_ix = &priv->vlan.active_vlans_ft_ix[vid]; - MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag, - 1); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, - outer_headers.first_vid); - MLX5_SET(fte_match_param, match_value, outer_headers.first_vid, - vid); + rule_p = &priv->vlan.active_vlans_rule[vid]; + MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid); + MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid); break; } - err = mlx5_add_flow_table_entry(priv->ft.vlan, match_criteria_enable, - match_criteria, flow_context, ft_ix); - if (err) - netdev_err(priv->netdev, "%s: failed\n", __func__); + *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, + &dest); + + if (IS_ERR(*rule_p)) { + err = PTR_ERR(*rule_p); + *rule_p = NULL; + netdev_err(priv->netdev, "%s: add rule failed\n", __func__); + } + + return err; +} + +static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, + enum mlx5e_vlan_rule_type rule_type, u16 vid) +{ + u32 *match_criteria; + u32 *match_value; + int err = 0; + + match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + if (!match_value || !match_criteria) { + netdev_err(priv->netdev, "%s: alloc failed\n", __func__); + err = -ENOMEM; + goto add_vlan_rule_out; + } + + if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID) + mlx5e_vport_context_update_vlans(priv); + + err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria, + match_value); add_vlan_rule_out: kvfree(match_criteria); - kvfree(flow_context); + kvfree(match_value); + return err; } @@ -625,16 +589,23 @@ static void mlx5e_del_vlan_rule(struct mlx5e_priv *priv, { switch (rule_type) { case MLX5E_VLAN_RULE_TYPE_UNTAGGED: - mlx5_del_flow_table_entry(priv->ft.vlan, - priv->vlan.untagged_rule_ft_ix); + if (priv->vlan.untagged_rule) { + mlx5_del_flow_rule(priv->vlan.untagged_rule); + priv->vlan.untagged_rule = NULL; + } break; case MLX5E_VLAN_RULE_TYPE_ANY_VID: - mlx5_del_flow_table_entry(priv->ft.vlan, - priv->vlan.any_vlan_rule_ft_ix); + if (priv->vlan.any_vlan_rule) { + mlx5_del_flow_rule(priv->vlan.any_vlan_rule); + priv->vlan.any_vlan_rule = NULL; + } break; case MLX5E_VLAN_RULE_TYPE_MATCH_VID: - mlx5_del_flow_table_entry(priv->ft.vlan, - priv->vlan.active_vlans_ft_ix[vid]); + mlx5e_vport_context_update_vlans(priv); + if (priv->vlan.active_vlans_rule[vid]) { + mlx5_del_flow_rule(priv->vlan.active_vlans_rule[vid]); + priv->vlan.active_vlans_rule[vid] = NULL; + } mlx5e_vport_context_update_vlans(priv); break; } @@ -889,151 +860,358 @@ void mlx5e_set_rx_mode_work(struct work_struct *work) mlx5e_vport_context_update(priv); } +static void mlx5e_destroy_groups(struct mlx5e_flow_table *ft) +{ + int i; + + for (i = ft->num_groups - 1; i >= 0; i--) { + if (!IS_ERR_OR_NULL(ft->g[i])) + mlx5_destroy_flow_group(ft->g[i]); + ft->g[i] = NULL; + } + ft->num_groups = 0; +} + void mlx5e_init_eth_addr(struct mlx5e_priv *priv) { ether_addr_copy(priv->eth_addr.broadcast.addr, priv->netdev->broadcast); } -static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv) +#define MLX5E_MAIN_GROUP0_SIZE BIT(3) +#define MLX5E_MAIN_GROUP1_SIZE BIT(1) +#define MLX5E_MAIN_GROUP2_SIZE BIT(0) +#define MLX5E_MAIN_GROUP3_SIZE BIT(14) +#define MLX5E_MAIN_GROUP4_SIZE BIT(13) +#define MLX5E_MAIN_GROUP5_SIZE BIT(11) +#define MLX5E_MAIN_GROUP6_SIZE BIT(2) +#define MLX5E_MAIN_GROUP7_SIZE BIT(1) +#define MLX5E_MAIN_GROUP8_SIZE BIT(0) +#define MLX5E_MAIN_TABLE_SIZE (MLX5E_MAIN_GROUP0_SIZE +\ + MLX5E_MAIN_GROUP1_SIZE +\ + MLX5E_MAIN_GROUP2_SIZE +\ + MLX5E_MAIN_GROUP3_SIZE +\ + MLX5E_MAIN_GROUP4_SIZE +\ + MLX5E_MAIN_GROUP5_SIZE +\ + MLX5E_MAIN_GROUP6_SIZE +\ + MLX5E_MAIN_GROUP7_SIZE +\ + MLX5E_MAIN_GROUP8_SIZE) + +static int __mlx5e_create_main_groups(struct mlx5e_flow_table *ft, u32 *in, + int inlen) { - struct mlx5_flow_table_group *g; - u8 *dmac; + u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria); + u8 *dmac = MLX5_ADDR_OF(create_flow_group_in, in, + match_criteria.outer_headers.dmac_47_16); + int err; + int ix = 0; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP0_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP1_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP2_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol); + eth_broadcast_addr(dmac); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP3_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + eth_broadcast_addr(dmac); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP4_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + eth_broadcast_addr(dmac); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP5_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ip_protocol); + dmac[0] = 0x01; + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP6_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.ethertype); + dmac[0] = 0x01; + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP7_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + dmac[0] = 0x01; + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_MAIN_GROUP8_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + return 0; + +err_destroy_groups: + err = PTR_ERR(ft->g[ft->num_groups]); + ft->g[ft->num_groups] = NULL; + mlx5e_destroy_groups(ft); + + return err; +} - g = kcalloc(9, sizeof(*g), GFP_KERNEL); - if (!g) +static int mlx5e_create_main_groups(struct mlx5e_flow_table *ft) +{ + u32 *in; + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + int err; + + in = mlx5_vzalloc(inlen); + if (!in) return -ENOMEM; - g[0].log_sz = 3; - g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, - outer_headers.ethertype); - MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, - outer_headers.ip_protocol); - - g[1].log_sz = 1; - g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria, - outer_headers.ethertype); - - g[2].log_sz = 0; - - g[3].log_sz = 14; - g[3].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[3].match_criteria, - outer_headers.dmac_47_16); - memset(dmac, 0xff, ETH_ALEN); - MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria, - outer_headers.ethertype); - MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria, - outer_headers.ip_protocol); - - g[4].log_sz = 13; - g[4].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[4].match_criteria, - outer_headers.dmac_47_16); - memset(dmac, 0xff, ETH_ALEN); - MLX5_SET_TO_ONES(fte_match_param, g[4].match_criteria, - outer_headers.ethertype); - - g[5].log_sz = 11; - g[5].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[5].match_criteria, - outer_headers.dmac_47_16); - memset(dmac, 0xff, ETH_ALEN); - - g[6].log_sz = 2; - g[6].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[6].match_criteria, - outer_headers.dmac_47_16); - dmac[0] = 0x01; - MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria, - outer_headers.ethertype); - MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria, - outer_headers.ip_protocol); - - g[7].log_sz = 1; - g[7].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[7].match_criteria, - outer_headers.dmac_47_16); - dmac[0] = 0x01; - MLX5_SET_TO_ONES(fte_match_param, g[7].match_criteria, - outer_headers.ethertype); + err = __mlx5e_create_main_groups(ft, in, inlen); - g[8].log_sz = 0; - g[8].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g[8].match_criteria, - outer_headers.dmac_47_16); - dmac[0] = 0x01; - priv->ft.main = mlx5_create_flow_table(priv->mdev, 1, - MLX5_FLOW_TABLE_TYPE_NIC_RCV, - 9, g); - kfree(g); + kvfree(in); + return err; +} - return priv->ft.main ? 0 : -ENOMEM; +static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv) +{ + struct mlx5e_flow_table *ft = &priv->fts.main; + int err; + + ft->num_groups = 0; + ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_MAIN_TABLE_SIZE); + + if (IS_ERR(ft->t)) { + err = PTR_ERR(ft->t); + ft->t = NULL; + return err; + } + ft->g = kcalloc(MLX5E_NUM_MAIN_GROUPS, sizeof(*ft->g), GFP_KERNEL); + if (!ft->g) { + err = -ENOMEM; + goto err_destroy_main_flow_table; + } + + err = mlx5e_create_main_groups(ft); + if (err) + goto err_free_g; + return 0; + +err_free_g: + kfree(ft->g); + +err_destroy_main_flow_table: + mlx5_destroy_flow_table(ft->t); + ft->t = NULL; + + return err; +} + +static void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft) +{ + mlx5e_destroy_groups(ft); + kfree(ft->g); + mlx5_destroy_flow_table(ft->t); + ft->t = NULL; } static void mlx5e_destroy_main_flow_table(struct mlx5e_priv *priv) { - mlx5_destroy_flow_table(priv->ft.main); + mlx5e_destroy_flow_table(&priv->fts.main); } -static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv) +#define MLX5E_NUM_VLAN_GROUPS 2 +#define MLX5E_VLAN_GROUP0_SIZE BIT(12) +#define MLX5E_VLAN_GROUP1_SIZE BIT(1) +#define MLX5E_VLAN_TABLE_SIZE (MLX5E_VLAN_GROUP0_SIZE +\ + MLX5E_VLAN_GROUP1_SIZE) + +static int __mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft, u32 *in, + int inlen) +{ + int err; + int ix = 0; + u8 *mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria); + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_VLAN_GROUP0_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + memset(in, 0, inlen); + MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); + MLX5_SET_CFG(in, start_flow_index, ix); + ix += MLX5E_VLAN_GROUP1_SIZE; + MLX5_SET_CFG(in, end_flow_index, ix - 1); + ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in); + if (IS_ERR(ft->g[ft->num_groups])) + goto err_destroy_groups; + ft->num_groups++; + + return 0; + +err_destroy_groups: + err = PTR_ERR(ft->g[ft->num_groups]); + ft->g[ft->num_groups] = NULL; + mlx5e_destroy_groups(ft); + + return err; +} + +static int mlx5e_create_vlan_groups(struct mlx5e_flow_table *ft) { - struct mlx5_flow_table_group *g; + u32 *in; + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + int err; - g = kcalloc(2, sizeof(*g), GFP_KERNEL); - if (!g) + in = mlx5_vzalloc(inlen); + if (!in) return -ENOMEM; - g[0].log_sz = 12; - g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, - outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, - outer_headers.first_vid); - - /* untagged + any vlan id */ - g[1].log_sz = 1; - g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria, - outer_headers.vlan_tag); - - priv->ft.vlan = mlx5_create_flow_table(priv->mdev, 0, - MLX5_FLOW_TABLE_TYPE_NIC_RCV, - 2, g); - - kfree(g); - return priv->ft.vlan ? 0 : -ENOMEM; + err = __mlx5e_create_vlan_groups(ft, in, inlen); + + kvfree(in); + return err; +} + +static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv) +{ + struct mlx5e_flow_table *ft = &priv->fts.vlan; + int err; + + ft->num_groups = 0; + ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_VLAN_TABLE_SIZE); + + if (IS_ERR(ft->t)) { + err = PTR_ERR(ft->t); + ft->t = NULL; + return err; + } + ft->g = kcalloc(MLX5E_NUM_VLAN_GROUPS, sizeof(*ft->g), GFP_KERNEL); + if (!ft->g) { + err = -ENOMEM; + goto err_destroy_vlan_flow_table; + } + + err = mlx5e_create_vlan_groups(ft); + if (err) + goto err_free_g; + + return 0; + +err_free_g: + kfree(ft->g); + +err_destroy_vlan_flow_table: + mlx5_destroy_flow_table(ft->t); + ft->t = NULL; + + return err; } static void mlx5e_destroy_vlan_flow_table(struct mlx5e_priv *priv) { - mlx5_destroy_flow_table(priv->ft.vlan); + mlx5e_destroy_flow_table(&priv->fts.vlan); } int mlx5e_create_flow_tables(struct mlx5e_priv *priv) { int err; - err = mlx5e_create_main_flow_table(priv); + priv->fts.ns = mlx5_get_flow_namespace(priv->mdev, + MLX5_FLOW_NAMESPACE_KERNEL); + + if (!priv->fts.ns) + return -EINVAL; + + err = mlx5e_create_vlan_flow_table(priv); if (err) return err; - err = mlx5e_create_vlan_flow_table(priv); + err = mlx5e_create_main_flow_table(priv); if (err) - goto err_destroy_main_flow_table; + goto err_destroy_vlan_flow_table; err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0); if (err) - goto err_destroy_vlan_flow_table; + goto err_destroy_main_flow_table; return 0; -err_destroy_vlan_flow_table: - mlx5e_destroy_vlan_flow_table(priv); - err_destroy_main_flow_table: mlx5e_destroy_main_flow_table(priv); +err_destroy_vlan_flow_table: + mlx5e_destroy_vlan_flow_table(priv); return err; } @@ -1041,6 +1219,6 @@ err_destroy_main_flow_table: void mlx5e_destroy_flow_tables(struct mlx5e_priv *priv) { mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0); - mlx5e_destroy_vlan_flow_table(priv); mlx5e_destroy_main_flow_table(priv); + mlx5e_destroy_vlan_flow_table(priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index a20be56df553..d4601a564699 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -30,7 +30,7 @@ * SOFTWARE. */ -#include +#include #include "en.h" #include "eswitch.h" diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index d8939e597c54..bc3d9f8a75c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include "mlx5_core.h" #include "eswitch.h" @@ -321,220 +321,6 @@ static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index) free_l2_table_index(l2_table, index); } -/* E-Switch FDB flow steering */ -struct dest_node { - struct list_head list; - struct mlx5_flow_destination dest; -}; - -static int _mlx5_flow_rule_apply(struct mlx5_flow_rule *fr) -{ - bool was_valid = fr->valid; - struct dest_node *dest_n; - u32 dest_list_size = 0; - void *in_match_value; - u32 *flow_context; - u32 flow_index; - int err; - int i; - - if (list_empty(&fr->dest_list)) { - if (fr->valid) - mlx5_del_flow_table_entry(fr->ft, fr->fi); - fr->valid = false; - return 0; - } - - list_for_each_entry(dest_n, &fr->dest_list, list) - dest_list_size++; - - flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + - MLX5_ST_SZ_BYTES(dest_format_struct) * - dest_list_size); - if (!flow_context) - return -ENOMEM; - - MLX5_SET(flow_context, flow_context, flow_tag, fr->flow_tag); - MLX5_SET(flow_context, flow_context, action, fr->action); - MLX5_SET(flow_context, flow_context, destination_list_size, - dest_list_size); - - i = 0; - list_for_each_entry(dest_n, &fr->dest_list, list) { - void *dest_addr = MLX5_ADDR_OF(flow_context, flow_context, - destination[i++]); - - MLX5_SET(dest_format_struct, dest_addr, destination_type, - dest_n->dest.type); - MLX5_SET(dest_format_struct, dest_addr, destination_id, - dest_n->dest.vport_num); - } - - in_match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); - memcpy(in_match_value, fr->match_value, MLX5_ST_SZ_BYTES(fte_match_param)); - - err = mlx5_add_flow_table_entry(fr->ft, fr->match_criteria_enable, - fr->match_criteria, flow_context, - &flow_index); - if (!err) { - if (was_valid) - mlx5_del_flow_table_entry(fr->ft, fr->fi); - fr->fi = flow_index; - fr->valid = true; - } - kfree(flow_context); - return err; -} - -static int mlx5_flow_rule_add_dest(struct mlx5_flow_rule *fr, - struct mlx5_flow_destination *new_dest) -{ - struct dest_node *dest_n; - int err; - - dest_n = kzalloc(sizeof(*dest_n), GFP_KERNEL); - if (!dest_n) - return -ENOMEM; - - memcpy(&dest_n->dest, new_dest, sizeof(dest_n->dest)); - mutex_lock(&fr->mutex); - list_add(&dest_n->list, &fr->dest_list); - err = _mlx5_flow_rule_apply(fr); - if (err) { - list_del(&dest_n->list); - kfree(dest_n); - } - mutex_unlock(&fr->mutex); - return err; -} - -static int mlx5_flow_rule_del_dest(struct mlx5_flow_rule *fr, - struct mlx5_flow_destination *dest) -{ - struct dest_node *dest_n; - struct dest_node *n; - int err; - - mutex_lock(&fr->mutex); - list_for_each_entry_safe(dest_n, n, &fr->dest_list, list) { - if (dest->vport_num == dest_n->dest.vport_num) - goto found; - } - mutex_unlock(&fr->mutex); - return -ENOENT; - -found: - list_del(&dest_n->list); - err = _mlx5_flow_rule_apply(fr); - mutex_unlock(&fr->mutex); - kfree(dest_n); - - return err; -} - -static struct mlx5_flow_rule *find_fr(struct mlx5_eswitch *esw, - u8 match_criteria_enable, - u32 *match_value) -{ - struct hlist_head *hash = esw->mc_table; - struct esw_mc_addr *esw_mc; - u8 *dmac_v; - - dmac_v = MLX5_ADDR_OF(fte_match_param, match_value, - outer_headers.dmac_47_16); - - /* UNICAST FULL MATCH */ - if (!is_multicast_ether_addr(dmac_v)) - return NULL; - - /* MULTICAST FULL MATCH */ - esw_mc = l2addr_hash_find(hash, dmac_v, struct esw_mc_addr); - - return esw_mc ? esw_mc->uplink_rule : NULL; -} - -static struct mlx5_flow_rule *alloc_fr(void *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, - u32 action, - u32 flow_tag) -{ - struct mlx5_flow_rule *fr = kzalloc(sizeof(*fr), GFP_KERNEL); - - if (!fr) - return NULL; - - fr->match_criteria = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - fr->match_value = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!fr->match_criteria || !fr->match_value) { - kfree(fr->match_criteria); - kfree(fr->match_value); - kfree(fr); - return NULL; - } - - memcpy(fr->match_criteria, match_criteria, MLX5_ST_SZ_BYTES(fte_match_param)); - memcpy(fr->match_value, match_value, MLX5_ST_SZ_BYTES(fte_match_param)); - fr->match_criteria_enable = match_criteria_enable; - fr->flow_tag = flow_tag; - fr->action = action; - - mutex_init(&fr->mutex); - INIT_LIST_HEAD(&fr->dest_list); - atomic_set(&fr->refcount, 0); - fr->ft = ft; - return fr; -} - -static void deref_fr(struct mlx5_flow_rule *fr) -{ - if (!atomic_dec_and_test(&fr->refcount)) - return; - - kfree(fr->match_criteria); - kfree(fr->match_value); - kfree(fr); -} - -static struct mlx5_flow_rule * -mlx5_add_flow_rule(struct mlx5_eswitch *esw, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, - u32 action, - u32 flow_tag, - struct mlx5_flow_destination *dest) -{ - struct mlx5_flow_rule *fr; - int err; - - fr = find_fr(esw, match_criteria_enable, match_value); - fr = fr ? fr : alloc_fr(esw->fdb_table.fdb, match_criteria_enable, match_criteria, - match_value, action, flow_tag); - if (!fr) - return NULL; - - atomic_inc(&fr->refcount); - - err = mlx5_flow_rule_add_dest(fr, dest); - if (err) { - deref_fr(fr); - return NULL; - } - - return fr; -} - -static void mlx5_del_flow_rule(struct mlx5_flow_rule *fr, u32 vport) -{ - struct mlx5_flow_destination dest; - - dest.vport_num = vport; - mlx5_flow_rule_del_dest(fr, &dest); - deref_fr(fr); -} - /* E-Switch FDB */ static struct mlx5_flow_rule * esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport) @@ -569,7 +355,7 @@ esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport) "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n", dmac_v, dmac_c, vport); flow_rule = - mlx5_add_flow_rule(esw, + mlx5_add_flow_rule(esw->fdb_table.fdb, match_header, match_c, match_v, @@ -589,33 +375,61 @@ out: static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) { + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_core_dev *dev = esw->dev; - struct mlx5_flow_table_group g; + struct mlx5_flow_namespace *root_ns; struct mlx5_flow_table *fdb; + struct mlx5_flow_group *g; + void *match_criteria; + int table_size; + u32 *flow_group_in; u8 *dmac; + int err = 0; esw_debug(dev, "Create FDB log_max_size(%d)\n", MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); - memset(&g, 0, sizeof(g)); - /* UC MC Full match rules*/ - g.log_sz = MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size); - g.match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - dmac = MLX5_ADDR_OF(fte_match_param, g.match_criteria, - outer_headers.dmac_47_16); - /* Match criteria mask */ - memset(dmac, 0xff, 6); + root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + if (!root_ns) { + esw_warn(dev, "Failed to get FDB flow namespace\n"); + return -ENOMEM; + } - fdb = mlx5_create_flow_table(dev, 0, - MLX5_FLOW_TABLE_TYPE_ESWITCH, - 1, &g); - if (fdb) - esw_debug(dev, "ESW: FDB Table created fdb->id %d\n", mlx5_get_flow_table_id(fdb)); - else - esw_warn(dev, "ESW: Failed to create FDB Table\n"); + flow_group_in = mlx5_vzalloc(inlen); + if (!flow_group_in) + return -ENOMEM; + memset(flow_group_in, 0, inlen); + + table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); + fdb = mlx5_create_flow_table(root_ns, 0, table_size); + if (IS_ERR_OR_NULL(fdb)) { + err = PTR_ERR(fdb); + esw_warn(dev, "Failed to create FDB Table err %d\n", err); + goto out; + } + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_OUTER_HEADERS); + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1); + eth_broadcast_addr(dmac); + + g = mlx5_create_flow_group(fdb, flow_group_in); + if (IS_ERR_OR_NULL(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create flow group err(%d)\n", err); + goto out; + } + + esw->fdb_table.addr_grp = g; esw->fdb_table.fdb = fdb; - return fdb ? 0 : -ENOMEM; +out: + kfree(flow_group_in); + if (err && !IS_ERR_OR_NULL(fdb)) + mlx5_destroy_flow_table(fdb); + return err; } static void esw_destroy_fdb_table(struct mlx5_eswitch *esw) @@ -623,10 +437,11 @@ static void esw_destroy_fdb_table(struct mlx5_eswitch *esw) if (!esw->fdb_table.fdb) return; - esw_debug(esw->dev, "Destroy FDB Table fdb(%d)\n", - mlx5_get_flow_table_id(esw->fdb_table.fdb)); + esw_debug(esw->dev, "Destroy FDB Table\n"); + mlx5_destroy_flow_group(esw->fdb_table.addr_grp); mlx5_destroy_flow_table(esw->fdb_table.fdb); esw->fdb_table.fdb = NULL; + esw->fdb_table.addr_grp = NULL; } /* E-Switch vport UC/MC lists management */ @@ -689,7 +504,7 @@ static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) del_l2_table_entry(esw->dev, esw_uc->table_index); if (vaddr->flow_rule) - mlx5_del_flow_rule(vaddr->flow_rule, vport); + mlx5_del_flow_rule(vaddr->flow_rule); vaddr->flow_rule = NULL; l2addr_hash_del(esw_uc); @@ -750,14 +565,14 @@ static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) esw_mc->uplink_rule); if (vaddr->flow_rule) - mlx5_del_flow_rule(vaddr->flow_rule, vport); + mlx5_del_flow_rule(vaddr->flow_rule); vaddr->flow_rule = NULL; if (--esw_mc->refcnt) return 0; if (esw_mc->uplink_rule) - mlx5_del_flow_rule(esw_mc->uplink_rule, UPLINK_VPORT); + mlx5_del_flow_rule(esw_mc->uplink_rule); l2addr_hash_del(esw_mc); return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 02ff3eade026..3416a428f70f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -88,20 +88,6 @@ struct l2addr_node { kfree(ptr); \ }) -struct mlx5_flow_rule { - void *ft; - u32 fi; - u8 match_criteria_enable; - u32 *match_criteria; - u32 *match_value; - u32 action; - u32 flow_tag; - bool valid; - atomic_t refcount; - struct mutex mutex; /* protect flow rule updates */ - struct list_head dest_list; -}; - struct mlx5_vport { struct mlx5_core_dev *dev; int vport; @@ -126,6 +112,7 @@ struct mlx5_l2_table { struct mlx5_eswitch_fdb { void *fdb; + struct mlx5_flow_group *addr_grp; }; struct mlx5_eswitch { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c deleted file mode 100644 index ca90b9bc3b95..000000000000 --- a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c +++ /dev/null @@ -1,422 +0,0 @@ -/* - * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * OpenIB.org BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include -#include -#include -#include "mlx5_core.h" - -struct mlx5_ftg { - struct mlx5_flow_table_group g; - u32 id; - u32 start_ix; -}; - -struct mlx5_flow_table { - struct mlx5_core_dev *dev; - u8 level; - u8 type; - u32 id; - struct mutex mutex; /* sync bitmap alloc */ - u16 num_groups; - struct mlx5_ftg *group; - unsigned long *bitmap; - u32 size; -}; - -static int mlx5_set_flow_entry_cmd(struct mlx5_flow_table *ft, u32 group_ix, - u32 flow_index, void *flow_context) -{ - u32 out[MLX5_ST_SZ_DW(set_fte_out)]; - u32 *in; - void *in_flow_context; - int fcdls = - MLX5_GET(flow_context, flow_context, destination_list_size) * - MLX5_ST_SZ_BYTES(dest_format_struct); - int inlen = MLX5_ST_SZ_BYTES(set_fte_in) + fcdls; - int err; - - in = mlx5_vzalloc(inlen); - if (!in) { - mlx5_core_warn(ft->dev, "failed to allocate inbox\n"); - return -ENOMEM; - } - - MLX5_SET(set_fte_in, in, table_type, ft->type); - MLX5_SET(set_fte_in, in, table_id, ft->id); - MLX5_SET(set_fte_in, in, flow_index, flow_index); - MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY); - - in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context); - memcpy(in_flow_context, flow_context, - MLX5_ST_SZ_BYTES(flow_context) + fcdls); - - MLX5_SET(flow_context, in_flow_context, group_id, - ft->group[group_ix].id); - - memset(out, 0, sizeof(out)); - err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out, - sizeof(out)); - kvfree(in); - - return err; -} - -static void mlx5_del_flow_entry_cmd(struct mlx5_flow_table *ft, u32 flow_index) -{ - u32 in[MLX5_ST_SZ_DW(delete_fte_in)]; - u32 out[MLX5_ST_SZ_DW(delete_fte_out)]; - - memset(in, 0, sizeof(in)); - memset(out, 0, sizeof(out)); - -#define MLX5_SET_DFTEI(p, x, v) MLX5_SET(delete_fte_in, p, x, v) - MLX5_SET_DFTEI(in, table_type, ft->type); - MLX5_SET_DFTEI(in, table_id, ft->id); - MLX5_SET_DFTEI(in, flow_index, flow_index); - MLX5_SET_DFTEI(in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY); - - mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); -} - -static void mlx5_destroy_flow_group_cmd(struct mlx5_flow_table *ft, int i) -{ - u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)]; - u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)]; - - memset(in, 0, sizeof(in)); - memset(out, 0, sizeof(out)); - -#define MLX5_SET_DFGI(p, x, v) MLX5_SET(destroy_flow_group_in, p, x, v) - MLX5_SET_DFGI(in, table_type, ft->type); - MLX5_SET_DFGI(in, table_id, ft->id); - MLX5_SET_DFGI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_GROUP); - MLX5_SET_DFGI(in, group_id, ft->group[i].id); - mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); -} - -static int mlx5_create_flow_group_cmd(struct mlx5_flow_table *ft, int i) -{ - u32 out[MLX5_ST_SZ_DW(create_flow_group_out)]; - u32 *in; - void *in_match_criteria; - int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_flow_table_group *g = &ft->group[i].g; - u32 start_ix = ft->group[i].start_ix; - u32 end_ix = start_ix + (1 << g->log_sz) - 1; - int err; - - in = mlx5_vzalloc(inlen); - if (!in) { - mlx5_core_warn(ft->dev, "failed to allocate inbox\n"); - return -ENOMEM; - } - in_match_criteria = MLX5_ADDR_OF(create_flow_group_in, in, - match_criteria); - - memset(out, 0, sizeof(out)); - -#define MLX5_SET_CFGI(p, x, v) MLX5_SET(create_flow_group_in, p, x, v) - MLX5_SET_CFGI(in, table_type, ft->type); - MLX5_SET_CFGI(in, table_id, ft->id); - MLX5_SET_CFGI(in, opcode, MLX5_CMD_OP_CREATE_FLOW_GROUP); - MLX5_SET_CFGI(in, start_flow_index, start_ix); - MLX5_SET_CFGI(in, end_flow_index, end_ix); - MLX5_SET_CFGI(in, match_criteria_enable, g->match_criteria_enable); - - memcpy(in_match_criteria, g->match_criteria, - MLX5_ST_SZ_BYTES(fte_match_param)); - - err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out, - sizeof(out)); - if (!err) - ft->group[i].id = MLX5_GET(create_flow_group_out, out, - group_id); - - kvfree(in); - - return err; -} - -static void mlx5_destroy_flow_table_groups(struct mlx5_flow_table *ft) -{ - int i; - - for (i = 0; i < ft->num_groups; i++) - mlx5_destroy_flow_group_cmd(ft, i); -} - -static int mlx5_create_flow_table_groups(struct mlx5_flow_table *ft) -{ - int err; - int i; - - for (i = 0; i < ft->num_groups; i++) { - err = mlx5_create_flow_group_cmd(ft, i); - if (err) - goto err_destroy_flow_table_groups; - } - - return 0; - -err_destroy_flow_table_groups: - for (i--; i >= 0; i--) - mlx5_destroy_flow_group_cmd(ft, i); - - return err; -} - -static int mlx5_create_flow_table_cmd(struct mlx5_flow_table *ft) -{ - u32 in[MLX5_ST_SZ_DW(create_flow_table_in)]; - u32 out[MLX5_ST_SZ_DW(create_flow_table_out)]; - int err; - - memset(in, 0, sizeof(in)); - - MLX5_SET(create_flow_table_in, in, table_type, ft->type); - MLX5_SET(create_flow_table_in, in, level, ft->level); - MLX5_SET(create_flow_table_in, in, log_size, order_base_2(ft->size)); - - MLX5_SET(create_flow_table_in, in, opcode, - MLX5_CMD_OP_CREATE_FLOW_TABLE); - - memset(out, 0, sizeof(out)); - err = mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, - sizeof(out)); - if (err) - return err; - - ft->id = MLX5_GET(create_flow_table_out, out, table_id); - - return 0; -} - -static void mlx5_destroy_flow_table_cmd(struct mlx5_flow_table *ft) -{ - u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)]; - u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)]; - - memset(in, 0, sizeof(in)); - memset(out, 0, sizeof(out)); - -#define MLX5_SET_DFTI(p, x, v) MLX5_SET(destroy_flow_table_in, p, x, v) - MLX5_SET_DFTI(in, table_type, ft->type); - MLX5_SET_DFTI(in, table_id, ft->id); - MLX5_SET_DFTI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_TABLE); - - mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); -} - -static int mlx5_find_group(struct mlx5_flow_table *ft, u8 match_criteria_enable, - u32 *match_criteria, int *group_ix) -{ - void *mc_outer = MLX5_ADDR_OF(fte_match_param, match_criteria, - outer_headers); - void *mc_misc = MLX5_ADDR_OF(fte_match_param, match_criteria, - misc_parameters); - void *mc_inner = MLX5_ADDR_OF(fte_match_param, match_criteria, - inner_headers); - int mc_outer_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4); - int mc_misc_sz = MLX5_ST_SZ_BYTES(fte_match_set_misc); - int mc_inner_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4); - int i; - - for (i = 0; i < ft->num_groups; i++) { - struct mlx5_flow_table_group *g = &ft->group[i].g; - void *gmc_outer = MLX5_ADDR_OF(fte_match_param, - g->match_criteria, - outer_headers); - void *gmc_misc = MLX5_ADDR_OF(fte_match_param, - g->match_criteria, - misc_parameters); - void *gmc_inner = MLX5_ADDR_OF(fte_match_param, - g->match_criteria, - inner_headers); - - if (g->match_criteria_enable != match_criteria_enable) - continue; - - if (match_criteria_enable & MLX5_MATCH_OUTER_HEADERS) - if (memcmp(mc_outer, gmc_outer, mc_outer_sz)) - continue; - - if (match_criteria_enable & MLX5_MATCH_MISC_PARAMETERS) - if (memcmp(mc_misc, gmc_misc, mc_misc_sz)) - continue; - - if (match_criteria_enable & MLX5_MATCH_INNER_HEADERS) - if (memcmp(mc_inner, gmc_inner, mc_inner_sz)) - continue; - - *group_ix = i; - return 0; - } - - return -EINVAL; -} - -static int alloc_flow_index(struct mlx5_flow_table *ft, int group_ix, u32 *ix) -{ - struct mlx5_ftg *g = &ft->group[group_ix]; - int err = 0; - - mutex_lock(&ft->mutex); - - *ix = find_next_zero_bit(ft->bitmap, ft->size, g->start_ix); - if (*ix >= (g->start_ix + (1 << g->g.log_sz))) - err = -ENOSPC; - else - __set_bit(*ix, ft->bitmap); - - mutex_unlock(&ft->mutex); - - return err; -} - -static void mlx5_free_flow_index(struct mlx5_flow_table *ft, u32 ix) -{ - __clear_bit(ix, ft->bitmap); -} - -int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable, - void *match_criteria, void *flow_context, - u32 *flow_index) -{ - struct mlx5_flow_table *ft = flow_table; - int group_ix; - int err; - - err = mlx5_find_group(ft, match_criteria_enable, match_criteria, - &group_ix); - if (err) { - mlx5_core_warn(ft->dev, "mlx5_find_group failed\n"); - return err; - } - - err = alloc_flow_index(ft, group_ix, flow_index); - if (err) { - mlx5_core_warn(ft->dev, "alloc_flow_index failed\n"); - return err; - } - - return mlx5_set_flow_entry_cmd(ft, group_ix, *flow_index, flow_context); -} -EXPORT_SYMBOL(mlx5_add_flow_table_entry); - -void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index) -{ - struct mlx5_flow_table *ft = flow_table; - - mlx5_del_flow_entry_cmd(ft, flow_index); - mlx5_free_flow_index(ft, flow_index); -} -EXPORT_SYMBOL(mlx5_del_flow_table_entry); - -void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type, - u16 num_groups, - struct mlx5_flow_table_group *group) -{ - struct mlx5_flow_table *ft; - u32 start_ix = 0; - u32 ft_size = 0; - void *gr; - void *bm; - int err; - int i; - - for (i = 0; i < num_groups; i++) - ft_size += (1 << group[i].log_sz); - - ft = kzalloc(sizeof(*ft), GFP_KERNEL); - gr = kcalloc(num_groups, sizeof(struct mlx5_ftg), GFP_KERNEL); - bm = kcalloc(BITS_TO_LONGS(ft_size), sizeof(uintptr_t), GFP_KERNEL); - if (!ft || !gr || !bm) - goto err_free_ft; - - ft->group = gr; - ft->bitmap = bm; - ft->num_groups = num_groups; - ft->level = level; - ft->type = table_type; - ft->size = ft_size; - ft->dev = dev; - mutex_init(&ft->mutex); - - for (i = 0; i < ft->num_groups; i++) { - memcpy(&ft->group[i].g, &group[i], sizeof(*group)); - ft->group[i].start_ix = start_ix; - start_ix += 1 << group[i].log_sz; - } - - err = mlx5_create_flow_table_cmd(ft); - if (err) - goto err_free_ft; - - err = mlx5_create_flow_table_groups(ft); - if (err) - goto err_destroy_flow_table_cmd; - - return ft; - -err_destroy_flow_table_cmd: - mlx5_destroy_flow_table_cmd(ft); - -err_free_ft: - mlx5_core_warn(dev, "failed to alloc flow table\n"); - kfree(bm); - kfree(gr); - kfree(ft); - - return NULL; -} -EXPORT_SYMBOL(mlx5_create_flow_table); - -void mlx5_destroy_flow_table(void *flow_table) -{ - struct mlx5_flow_table *ft = flow_table; - - mlx5_destroy_flow_table_groups(ft); - mlx5_destroy_flow_table_cmd(ft); - kfree(ft->bitmap); - kfree(ft->group); - kfree(ft); -} -EXPORT_SYMBOL(mlx5_destroy_flow_table); - -u32 mlx5_get_flow_table_id(void *flow_table) -{ - struct mlx5_flow_table *ft = flow_table; - - return ft->id; -} -EXPORT_SYMBOL(mlx5_get_flow_table_id); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 4264e8b34b76..f7d62fe595f6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -436,9 +436,9 @@ static struct mlx5_flow_table *alloc_flow_table(int level, int max_fte, return ft; } -static struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, - int prio, - int max_fte) +struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, + int prio, + int max_fte) { struct mlx5_flow_table *ft; int err; @@ -491,8 +491,8 @@ unlock_prio: return ERR_PTR(err); } -static struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, - u32 *fg_in) +struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, + u32 *fg_in) { struct mlx5_flow_group *fg; struct mlx5_core_dev *dev = get_dev(&ft->node); @@ -669,7 +669,7 @@ unlock_fg: return rule; } -static struct mlx5_flow_rule * +struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, u8 match_criteria_enable, u32 *match_criteria, @@ -699,12 +699,12 @@ put: return rule; } -static void mlx5_del_flow_rule(struct mlx5_flow_rule *rule) +void mlx5_del_flow_rule(struct mlx5_flow_rule *rule) { tree_remove_node(&rule->node); } -static int mlx5_destroy_flow_table(struct mlx5_flow_table *ft) +int mlx5_destroy_flow_table(struct mlx5_flow_table *ft) { if (tree_remove_node(&ft->node)) mlx5_core_warn(get_dev(&ft->node), "Flow table %d wasn't destroyed, refcount > 1\n", @@ -713,15 +713,15 @@ static int mlx5_destroy_flow_table(struct mlx5_flow_table *ft) return 0; } -static void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) +void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) { if (tree_remove_node(&fg->node)) mlx5_core_warn(get_dev(&fg->node), "Flow group %d wasn't destroyed, refcount > 1\n", fg->id); } -static struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, - enum mlx5_flow_namespace_type type) +struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, + enum mlx5_flow_namespace_type type) { struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; int prio; @@ -867,7 +867,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev struct mlx5_flow_root_namespace *root_ns; struct mlx5_flow_namespace *ns; - /* create the root namespace */ + /* Create the root namespace */ root_ns = mlx5_vzalloc(sizeof(*root_ns)); if (!root_ns) return NULL; @@ -1018,7 +1018,7 @@ static int init_fdb_root_ns(struct mlx5_core_dev *dev) if (!dev->priv.fdb_root_ns) return -ENOMEM; - /* create 1 prio*/ + /* Create single prio */ prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1, 0); if (IS_ERR(prio)) { cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index c6de3240f76f..789882b7b711 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -49,6 +49,7 @@ #include #include #include "mlx5_core.h" +#include "fs_core.h" #ifdef CONFIG_MLX5_CORE_EN #include "eswitch.h" #endif @@ -1055,6 +1056,11 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_init_srq_table(dev); mlx5_init_mr_table(dev); + err = mlx5_init_fs(dev); + if (err) { + dev_err(&pdev->dev, "Failed to init flow steering\n"); + goto err_fs; + } #ifdef CONFIG_MLX5_CORE_EN err = mlx5_eswitch_init(dev); if (err) { @@ -1093,6 +1099,8 @@ err_sriov: mlx5_eswitch_cleanup(dev->priv.eswitch); #endif err_reg_dev: + mlx5_cleanup_fs(dev); +err_fs: mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); @@ -1165,6 +1173,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_eswitch_cleanup(dev->priv.eswitch); #endif + mlx5_cleanup_fs(dev); mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); diff --git a/include/linux/mlx5/flow_table.h b/include/linux/mlx5/flow_table.h deleted file mode 100644 index 0f2a15cf3317..000000000000 --- a/include/linux/mlx5/flow_table.h +++ /dev/null @@ -1,63 +0,0 @@ -/* - * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * OpenIB.org BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#ifndef MLX5_FLOW_TABLE_H -#define MLX5_FLOW_TABLE_H - -#include - -struct mlx5_flow_table_group { - u8 log_sz; - u8 match_criteria_enable; - u32 match_criteria[MLX5_ST_SZ_DW(fte_match_param)]; -}; - -struct mlx5_flow_destination { - enum mlx5_flow_destination_type type; - union { - u32 tir_num; - void *ft; - u32 vport_num; - }; -}; - -void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type, - u16 num_groups, - struct mlx5_flow_table_group *group); -void mlx5_destroy_flow_table(void *flow_table); -int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable, - void *match_criteria, void *flow_context, - u32 *flow_index); -void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index); -u32 mlx5_get_flow_table_id(void *flow_table); - -#endif /* MLX5_FLOW_TABLE_H */ diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 16ae5233dc7b..bc7ad019afde 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -33,6 +33,7 @@ #ifndef _MLX5_FS_ #define _MLX5_FS_ +#include #include #define MLX5_FS_DEFAULT_FLOW_TAG 0x0 @@ -43,6 +44,9 @@ enum mlx5_flow_namespace_type { }; struct mlx5_flow_table; +struct mlx5_flow_group; +struct mlx5_flow_rule; +struct mlx5_flow_namespace; struct mlx5_flow_destination { enum mlx5_flow_destination_type type; @@ -52,4 +56,38 @@ struct mlx5_flow_destination { u32 vport_num; }; }; + +struct mlx5_flow_namespace * +mlx5_get_flow_namespace(struct mlx5_core_dev *dev, + enum mlx5_flow_namespace_type type); + +struct mlx5_flow_table * +mlx5_create_flow_table(struct mlx5_flow_namespace *ns, + int prio, + int num_flow_table_entries); +int mlx5_destroy_flow_table(struct mlx5_flow_table *ft); + +/* inbox should be set with the following values: + * start_flow_index + * end_flow_index + * match_criteria_enable + * match_criteria + */ +struct mlx5_flow_group * +mlx5_create_flow_group(struct mlx5_flow_table *ft, u32 *in); +void mlx5_destroy_flow_group(struct mlx5_flow_group *fg); + +/* Single destination per rule. + * Group ID is implied by the match criteria. + */ +struct mlx5_flow_rule * +mlx5_add_flow_rule(struct mlx5_flow_table *ft, + u8 match_criteria_enable, + u32 *match_criteria, + u32 *match_value, + u32 action, + u32 flow_tag, + struct mlx5_flow_destination *dest); +void mlx5_del_flow_rule(struct mlx5_flow_rule *fr); + #endif -- cgit v1.2.3-59-g8ed1b From 369620a09bc5ab867342d51f1820c66b00d78a2c Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Thu, 10 Dec 2015 12:37:44 -0800 Subject: rco: Clean up casting errors Fixe a couple of cast errors found by sparse. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/net/checksum.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/checksum.h b/include/net/checksum.h index 9fcaedf994ee..10a16b5bd1c7 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -165,7 +165,8 @@ static inline __wsum remcsum_adjust(void *ptr, __wsum csum, csum = csum_sub(csum, csum_partial(ptr, start, 0)); /* Set derived checksum in packet */ - delta = csum_sub(csum_fold(csum), *psum); + delta = csum_sub((__force __wsum)csum_fold(csum), + (__force __wsum)*psum); *psum = csum_fold(csum); return delta; -- cgit v1.2.3-59-g8ed1b From abe492b4f50c3ae2ebcfaa2f5c16176aebaa1c68 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Thu, 10 Dec 2015 12:37:45 -0800 Subject: geneve: UDP checksum configuration via netlink Add support to enable and disable UDP checksums via netlink. This is similar to how VXLAN and GUE allow this. This includes support for enabling the UDP zero checksum (for both TX and RX). Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/net/geneve.c | 93 +++++++++++++++++++++++++++++++++----------- include/uapi/linux/if_link.h | 3 ++ 2 files changed, 73 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index de5c30c9f059..0750d7a93878 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -71,8 +71,14 @@ struct geneve_dev { __be16 dst_port; bool collect_md; struct gro_cells gro_cells; + u32 flags; }; +/* Geneve device flags */ +#define GENEVE_F_UDP_CSUM BIT(0) +#define GENEVE_F_UDP_ZERO_CSUM6_TX BIT(1) +#define GENEVE_F_UDP_ZERO_CSUM6_RX BIT(2) + struct geneve_sock { bool collect_md; struct list_head list; @@ -81,6 +87,7 @@ struct geneve_sock { int refcnt; struct udp_offload udp_offloads; struct hlist_head vni_list[VNI_HASH_SIZE]; + u32 flags; }; static inline __u32 geneve_net_vni_hash(u8 vni[3]) @@ -343,7 +350,7 @@ error: } static struct socket *geneve_create_sock(struct net *net, bool ipv6, - __be16 port) + __be16 port, u32 flags) { struct socket *sock; struct udp_port_cfg udp_conf; @@ -354,6 +361,8 @@ static struct socket *geneve_create_sock(struct net *net, bool ipv6, if (ipv6) { udp_conf.family = AF_INET6; udp_conf.ipv6_v6only = 1; + udp_conf.use_udp6_rx_checksums = + !(flags & GENEVE_F_UDP_ZERO_CSUM6_RX); } else { udp_conf.family = AF_INET; udp_conf.local_ip.s_addr = htonl(INADDR_ANY); @@ -480,7 +489,7 @@ static int geneve_gro_complete(struct sk_buff *skb, int nhoff, /* Create new listen socket if needed */ static struct geneve_sock *geneve_socket_create(struct net *net, __be16 port, - bool ipv6) + bool ipv6, u32 flags) { struct geneve_net *gn = net_generic(net, geneve_net_id); struct geneve_sock *gs; @@ -492,7 +501,7 @@ static struct geneve_sock *geneve_socket_create(struct net *net, __be16 port, if (!gs) return ERR_PTR(-ENOMEM); - sock = geneve_create_sock(net, ipv6, port); + sock = geneve_create_sock(net, ipv6, port, flags); if (IS_ERR(sock)) { kfree(gs); return ERR_CAST(sock); @@ -575,12 +584,13 @@ static int geneve_sock_add(struct geneve_dev *geneve, bool ipv6) goto out; } - gs = geneve_socket_create(net, geneve->dst_port, ipv6); + gs = geneve_socket_create(net, geneve->dst_port, ipv6, geneve->flags); if (IS_ERR(gs)) return PTR_ERR(gs); out: gs->collect_md = geneve->collect_md; + gs->flags = geneve->flags; #if IS_ENABLED(CONFIG_IPV6) if (ipv6) geneve->sock6 = gs; @@ -642,11 +652,12 @@ static void geneve_build_header(struct genevehdr *geneveh, static int geneve_build_skb(struct rtable *rt, struct sk_buff *skb, __be16 tun_flags, u8 vni[3], u8 opt_len, u8 *opt, - bool csum, bool xnet) + u32 flags, bool xnet) { struct genevehdr *gnvh; int min_headroom; int err; + bool udp_sum = !!(flags & GENEVE_F_UDP_CSUM); skb_scrub_packet(skb, xnet); @@ -658,7 +669,7 @@ static int geneve_build_skb(struct rtable *rt, struct sk_buff *skb, goto free_rt; } - skb = udp_tunnel_handle_offloads(skb, csum); + skb = udp_tunnel_handle_offloads(skb, udp_sum); if (IS_ERR(skb)) { err = PTR_ERR(skb); goto free_rt; @@ -678,11 +689,12 @@ free_rt: #if IS_ENABLED(CONFIG_IPV6) static int geneve6_build_skb(struct dst_entry *dst, struct sk_buff *skb, __be16 tun_flags, u8 vni[3], u8 opt_len, u8 *opt, - bool csum, bool xnet) + u32 flags, bool xnet) { struct genevehdr *gnvh; int min_headroom; int err; + bool udp_sum = !(flags & GENEVE_F_UDP_ZERO_CSUM6_TX); skb_scrub_packet(skb, xnet); @@ -694,7 +706,7 @@ static int geneve6_build_skb(struct dst_entry *dst, struct sk_buff *skb, goto free_dst; } - skb = udp_tunnel_handle_offloads(skb, csum); + skb = udp_tunnel_handle_offloads(skb, udp_sum); if (IS_ERR(skb)) { err = PTR_ERR(skb); goto free_dst; @@ -824,9 +836,9 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, struct flowi4 fl4; __u8 tos, ttl; __be16 sport; - bool udp_csum; __be16 df; bool xnet = !net_eq(geneve->net, dev_net(geneve->dev)); + u32 flags = geneve->flags; if (geneve->collect_md) { if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) { @@ -857,9 +869,13 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, if (key->tun_flags & TUNNEL_GENEVE_OPT) opts = ip_tunnel_info_opts(info); - udp_csum = !!(key->tun_flags & TUNNEL_CSUM); + if (key->tun_flags & TUNNEL_CSUM) + flags |= GENEVE_F_UDP_CSUM; + else + flags &= ~GENEVE_F_UDP_CSUM; + err = geneve_build_skb(rt, skb, key->tun_flags, vni, - info->options_len, opts, udp_csum, xnet); + info->options_len, opts, flags, xnet); if (unlikely(err)) goto err; @@ -867,9 +883,8 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, ttl = key->ttl; df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; } else { - udp_csum = false; err = geneve_build_skb(rt, skb, 0, geneve->vni, - 0, NULL, udp_csum, xnet); + 0, NULL, flags, xnet); if (unlikely(err)) goto err; @@ -883,7 +898,7 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, err = udp_tunnel_xmit_skb(rt, gs4->sock->sk, skb, fl4.saddr, fl4.daddr, tos, ttl, df, sport, geneve->dst_port, !net_eq(geneve->net, dev_net(geneve->dev)), - !udp_csum); + !(flags & GENEVE_F_UDP_CSUM)); iptunnel_xmit_stats(err, &dev->stats, dev->tstats); return NETDEV_TX_OK; @@ -912,8 +927,8 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, struct flowi6 fl6; __u8 prio, ttl; __be16 sport; - bool udp_csum; bool xnet = !net_eq(geneve->net, dev_net(geneve->dev)); + u32 flags = geneve->flags; if (geneve->collect_md) { if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) { @@ -942,19 +957,22 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, if (key->tun_flags & TUNNEL_GENEVE_OPT) opts = ip_tunnel_info_opts(info); - udp_csum = !!(key->tun_flags & TUNNEL_CSUM); + if (key->tun_flags & TUNNEL_CSUM) + flags |= GENEVE_F_UDP_CSUM; + else + flags &= ~GENEVE_F_UDP_CSUM; + err = geneve6_build_skb(dst, skb, key->tun_flags, vni, info->options_len, opts, - udp_csum, xnet); + flags, xnet); if (unlikely(err)) goto err; prio = ip_tunnel_ecn_encap(key->tos, iip, skb); ttl = key->ttl; } else { - udp_csum = false; err = geneve6_build_skb(dst, skb, 0, geneve->vni, - 0, NULL, udp_csum, xnet); + 0, NULL, flags, xnet); if (unlikely(err)) goto err; @@ -966,7 +984,8 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, } err = udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev, &fl6.saddr, &fl6.daddr, prio, ttl, - sport, geneve->dst_port, !udp_csum); + sport, geneve->dst_port, + !!(flags & GENEVE_F_UDP_ZERO_CSUM6_TX)); iptunnel_xmit_stats(err, &dev->stats, dev->tstats); return NETDEV_TX_OK; @@ -1099,6 +1118,9 @@ static const struct nla_policy geneve_policy[IFLA_GENEVE_MAX + 1] = { [IFLA_GENEVE_TOS] = { .type = NLA_U8 }, [IFLA_GENEVE_PORT] = { .type = NLA_U16 }, [IFLA_GENEVE_COLLECT_METADATA] = { .type = NLA_FLAG }, + [IFLA_GENEVE_UDP_CSUM] = { .type = NLA_U8 }, + [IFLA_GENEVE_UDP_ZERO_CSUM6_TX] = { .type = NLA_U8 }, + [IFLA_GENEVE_UDP_ZERO_CSUM6_RX] = { .type = NLA_U8 }, }; static int geneve_validate(struct nlattr *tb[], struct nlattr *data[]) @@ -1152,7 +1174,7 @@ static struct geneve_dev *geneve_find_dev(struct geneve_net *gn, static int geneve_configure(struct net *net, struct net_device *dev, union geneve_addr *remote, __u32 vni, __u8 ttl, __u8 tos, __be16 dst_port, - bool metadata) + bool metadata, u32 flags) { struct geneve_net *gn = net_generic(net, geneve_net_id); struct geneve_dev *t, *geneve = netdev_priv(dev); @@ -1183,6 +1205,7 @@ static int geneve_configure(struct net *net, struct net_device *dev, geneve->tos = tos; geneve->dst_port = dst_port; geneve->collect_md = metadata; + geneve->flags = flags; t = geneve_find_dev(gn, dst_port, remote, geneve->vni, &tun_on_same_port, &tun_collect_md); @@ -1213,6 +1236,7 @@ static int geneve_newlink(struct net *net, struct net_device *dev, bool metadata = false; union geneve_addr remote = geneve_remote_unspec; __u32 vni = 0; + u32 flags = 0; if (data[IFLA_GENEVE_REMOTE] && data[IFLA_GENEVE_REMOTE6]) return -EINVAL; @@ -1253,8 +1277,20 @@ static int geneve_newlink(struct net *net, struct net_device *dev, if (data[IFLA_GENEVE_COLLECT_METADATA]) metadata = true; + if (data[IFLA_GENEVE_UDP_CSUM] && + nla_get_u8(data[IFLA_GENEVE_UDP_CSUM])) + flags |= GENEVE_F_UDP_CSUM; + + if (data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX] && + nla_get_u8(data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX])) + flags |= GENEVE_F_UDP_ZERO_CSUM6_TX; + + if (data[IFLA_GENEVE_UDP_ZERO_CSUM6_RX] && + nla_get_u8(data[IFLA_GENEVE_UDP_ZERO_CSUM6_RX])) + flags |= GENEVE_F_UDP_ZERO_CSUM6_RX; + return geneve_configure(net, dev, &remote, vni, ttl, tos, dst_port, - metadata); + metadata, flags); } static void geneve_dellink(struct net_device *dev, struct list_head *head) @@ -1273,6 +1309,9 @@ static size_t geneve_get_size(const struct net_device *dev) nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_TOS */ nla_total_size(sizeof(__be16)) + /* IFLA_GENEVE_PORT */ nla_total_size(0) + /* IFLA_GENEVE_COLLECT_METADATA */ + nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_UDP_CSUM */ + nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_UDP_ZERO_CSUM6_TX */ + nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_UDP_ZERO_CSUM6_RX */ 0; } @@ -1309,6 +1348,14 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev) goto nla_put_failure; } + if (nla_put_u8(skb, IFLA_GENEVE_UDP_CSUM, + !!(geneve->flags & GENEVE_F_UDP_CSUM)) || + nla_put_u8(skb, IFLA_GENEVE_UDP_ZERO_CSUM6_TX, + !!(geneve->flags & GENEVE_F_UDP_ZERO_CSUM6_TX)) || + nla_put_u8(skb, IFLA_GENEVE_UDP_ZERO_CSUM6_RX, + !!(geneve->flags & GENEVE_F_UDP_ZERO_CSUM6_RX))) + goto nla_put_failure; + return 0; nla_put_failure: @@ -1342,7 +1389,7 @@ struct net_device *geneve_dev_create_fb(struct net *net, const char *name, return dev; err = geneve_configure(net, dev, &geneve_remote_unspec, - 0, 0, 0, htons(dst_port), true); + 0, 0, 0, htons(dst_port), true, 0); if (err) { free_netdev(dev); return ERR_PTR(err); diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 5ad57375a99f..2be1dd5a103f 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -462,6 +462,9 @@ enum { IFLA_GENEVE_PORT, /* destination port */ IFLA_GENEVE_COLLECT_METADATA, IFLA_GENEVE_REMOTE6, + IFLA_GENEVE_UDP_CSUM, + IFLA_GENEVE_UDP_ZERO_CSUM6_TX, + IFLA_GENEVE_UDP_ZERO_CSUM6_RX, __IFLA_GENEVE_MAX }; #define IFLA_GENEVE_MAX (__IFLA_GENEVE_MAX - 1) -- cgit v1.2.3-59-g8ed1b From 19576c9478682a398276c994ea0d2696474df32b Mon Sep 17 00:00:00 2001 From: Pablo Neira Date: Wed, 9 Dec 2015 14:07:40 +0100 Subject: netfilter: cttimeout: add netns support Add a per-netns list of timeout objects and adjust code to use it. Signed-off-by: Pablo Neira Ayuso --- include/net/net_namespace.h | 3 + include/net/netfilter/nf_conntrack_timeout.h | 2 +- net/netfilter/nf_conntrack_timeout.c | 2 +- net/netfilter/nfnetlink_cttimeout.c | 82 +++++++++++++++++----------- net/netfilter/xt_CT.c | 2 +- 5 files changed, 57 insertions(+), 34 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 2dcea635ecce..4089abc6e9c0 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -121,6 +121,9 @@ struct net { #if IS_ENABLED(CONFIG_NETFILTER_NETLINK_ACCT) struct list_head nfnl_acct_list; #endif +#if IS_ENABLED(CONFIG_NF_CT_NETLINK_TIMEOUT) + struct list_head nfct_timeout_list; +#endif #endif #ifdef CONFIG_WEXT_CORE struct sk_buff_head wext_nlevents; diff --git a/include/net/netfilter/nf_conntrack_timeout.h b/include/net/netfilter/nf_conntrack_timeout.h index f72be38860a7..5cc5e9e6171a 100644 --- a/include/net/netfilter/nf_conntrack_timeout.h +++ b/include/net/netfilter/nf_conntrack_timeout.h @@ -104,7 +104,7 @@ static inline void nf_conntrack_timeout_fini(void) #endif /* CONFIG_NF_CONNTRACK_TIMEOUT */ #ifdef CONFIG_NF_CONNTRACK_TIMEOUT -extern struct ctnl_timeout *(*nf_ct_timeout_find_get_hook)(const char *name); +extern struct ctnl_timeout *(*nf_ct_timeout_find_get_hook)(struct net *net, const char *name); extern void (*nf_ct_timeout_put_hook)(struct ctnl_timeout *timeout); #endif diff --git a/net/netfilter/nf_conntrack_timeout.c b/net/netfilter/nf_conntrack_timeout.c index 93da609d9d29..26e742006c48 100644 --- a/net/netfilter/nf_conntrack_timeout.c +++ b/net/netfilter/nf_conntrack_timeout.c @@ -25,7 +25,7 @@ #include struct ctnl_timeout * -(*nf_ct_timeout_find_get_hook)(const char *name) __read_mostly; +(*nf_ct_timeout_find_get_hook)(struct net *net, const char *name) __read_mostly; EXPORT_SYMBOL_GPL(nf_ct_timeout_find_get_hook); void (*nf_ct_timeout_put_hook)(struct ctnl_timeout *timeout) __read_mostly; diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index c7a2d0e1c462..3921d544f5ba 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -38,8 +38,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Pablo Neira Ayuso "); MODULE_DESCRIPTION("cttimeout: Extended Netfilter Connection Tracking timeout tuning"); -static LIST_HEAD(cttimeout_list); - static const struct nla_policy cttimeout_nla_policy[CTA_TIMEOUT_MAX+1] = { [CTA_TIMEOUT_NAME] = { .type = NLA_NUL_STRING, .len = CTNL_TIMEOUT_NAME_MAX - 1}, @@ -90,7 +88,7 @@ cttimeout_new_timeout(struct sock *ctnl, struct sk_buff *skb, l3num = ntohs(nla_get_be16(cda[CTA_TIMEOUT_L3PROTO])); l4num = nla_get_u8(cda[CTA_TIMEOUT_L4PROTO]); - list_for_each_entry(timeout, &cttimeout_list, head) { + list_for_each_entry(timeout, &net->nfct_timeout_list, head) { if (strncmp(timeout->name, name, CTNL_TIMEOUT_NAME_MAX) != 0) continue; @@ -145,7 +143,7 @@ cttimeout_new_timeout(struct sock *ctnl, struct sk_buff *skb, timeout->l3num = l3num; timeout->l4proto = l4proto; atomic_set(&timeout->refcnt, 1); - list_add_tail_rcu(&timeout->head, &cttimeout_list); + list_add_tail_rcu(&timeout->head, &net->nfct_timeout_list); return 0; err: @@ -209,6 +207,7 @@ nla_put_failure: static int ctnl_timeout_dump(struct sk_buff *skb, struct netlink_callback *cb) { + struct net *net = sock_net(skb->sk); struct ctnl_timeout *cur, *last; if (cb->args[2]) @@ -219,7 +218,7 @@ ctnl_timeout_dump(struct sk_buff *skb, struct netlink_callback *cb) cb->args[1] = 0; rcu_read_lock(); - list_for_each_entry_rcu(cur, &cttimeout_list, head) { + list_for_each_entry_rcu(cur, &net->nfct_timeout_list, head) { if (last) { if (cur != last) continue; @@ -245,6 +244,7 @@ cttimeout_get_timeout(struct sock *ctnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { + struct net *net = sock_net(skb->sk); int ret = -ENOENT; char *name; struct ctnl_timeout *cur; @@ -260,7 +260,7 @@ cttimeout_get_timeout(struct sock *ctnl, struct sk_buff *skb, return -EINVAL; name = nla_data(cda[CTA_TIMEOUT_NAME]); - list_for_each_entry(cur, &cttimeout_list, head) { + list_for_each_entry(cur, &net->nfct_timeout_list, head) { struct sk_buff *skb2; if (strncmp(cur->name, name, CTNL_TIMEOUT_NAME_MAX) != 0) @@ -301,17 +301,17 @@ static void untimeout(struct nf_conntrack_tuple_hash *i, RCU_INIT_POINTER(timeout_ext->timeout, NULL); } -static void ctnl_untimeout(struct ctnl_timeout *timeout) +static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout) { struct nf_conntrack_tuple_hash *h; const struct hlist_nulls_node *nn; int i; local_bh_disable(); - for (i = 0; i < init_net.ct.htable_size; i++) { + for (i = 0; i < net->ct.htable_size; i++) { spin_lock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); - if (i < init_net.ct.htable_size) { - hlist_nulls_for_each_entry(h, nn, &init_net.ct.hash[i], hnnode) + if (i < net->ct.htable_size) { + hlist_nulls_for_each_entry(h, nn, &net->ct.hash[i], hnnode) untimeout(h, timeout); } spin_unlock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); @@ -320,7 +320,7 @@ static void ctnl_untimeout(struct ctnl_timeout *timeout) } /* try to delete object, fail if it is still in use. */ -static int ctnl_timeout_try_del(struct ctnl_timeout *timeout) +static int ctnl_timeout_try_del(struct net *net, struct ctnl_timeout *timeout) { int ret = 0; @@ -329,7 +329,7 @@ static int ctnl_timeout_try_del(struct ctnl_timeout *timeout) /* We are protected by nfnl mutex. */ list_del_rcu(&timeout->head); nf_ct_l4proto_put(timeout->l4proto); - ctnl_untimeout(timeout); + ctnl_untimeout(net, timeout); kfree_rcu(timeout, rcu_head); } else { /* still in use, restore reference counter. */ @@ -344,23 +344,24 @@ cttimeout_del_timeout(struct sock *ctnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { + struct net *net = sock_net(skb->sk); char *name; struct ctnl_timeout *cur; int ret = -ENOENT; if (!cda[CTA_TIMEOUT_NAME]) { - list_for_each_entry(cur, &cttimeout_list, head) - ctnl_timeout_try_del(cur); + list_for_each_entry(cur, &net->nfct_timeout_list, head) + ctnl_timeout_try_del(net, cur); return 0; } name = nla_data(cda[CTA_TIMEOUT_NAME]); - list_for_each_entry(cur, &cttimeout_list, head) { + list_for_each_entry(cur, &net->nfct_timeout_list, head) { if (strncmp(cur->name, name, CTNL_TIMEOUT_NAME_MAX) != 0) continue; - ret = ctnl_timeout_try_del(cur); + ret = ctnl_timeout_try_del(net, cur); if (ret < 0) return ret; @@ -511,12 +512,13 @@ err: } #ifdef CONFIG_NF_CONNTRACK_TIMEOUT -static struct ctnl_timeout *ctnl_timeout_find_get(const char *name) +static struct ctnl_timeout * +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, &cttimeout_list, head) { + list_for_each_entry_rcu(timeout, &net->nfct_timeout_list, head) { if (strncmp(timeout->name, name, CTNL_TIMEOUT_NAME_MAX) != 0) continue; @@ -569,10 +571,39 @@ static const struct nfnetlink_subsystem cttimeout_subsys = { MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_CTNETLINK_TIMEOUT); +static int __net_init cttimeout_net_init(struct net *net) +{ + INIT_LIST_HEAD(&net->nfct_timeout_list); + + return 0; +} + +static void __net_exit cttimeout_net_exit(struct net *net) +{ + struct ctnl_timeout *cur, *tmp; + + ctnl_untimeout(net, NULL); + + list_for_each_entry_safe(cur, tmp, &net->nfct_timeout_list, head) { + list_del_rcu(&cur->head); + nf_ct_l4proto_put(cur->l4proto); + kfree_rcu(cur, rcu_head); + } +} + +static struct pernet_operations cttimeout_ops = { + .init = cttimeout_net_init, + .exit = cttimeout_net_exit, +}; + static int __init cttimeout_init(void) { int ret; + ret = register_pernet_subsys(&cttimeout_ops); + if (ret < 0) + return ret; + ret = nfnetlink_subsys_register(&cttimeout_subsys); if (ret < 0) { pr_err("cttimeout_init: cannot register cttimeout with " @@ -586,28 +617,17 @@ static int __init cttimeout_init(void) return 0; err_out: + unregister_pernet_subsys(&cttimeout_ops); return ret; } static void __exit cttimeout_exit(void) { - struct ctnl_timeout *cur, *tmp; - pr_info("cttimeout: unregistering from nfnetlink.\n"); nfnetlink_subsys_unregister(&cttimeout_subsys); - /* Make sure no conntrack objects refer to custom timeouts anymore. */ - ctnl_untimeout(NULL); - - list_for_each_entry_safe(cur, tmp, &cttimeout_list, head) { - list_del_rcu(&cur->head); - /* We are sure that our objects have no clients at this point, - * it's safe to release them all without checking refcnt. - */ - nf_ct_l4proto_put(cur->l4proto); - kfree_rcu(cur, rcu_head); - } + unregister_pernet_subsys(&cttimeout_ops); #ifdef CONFIG_NF_CONNTRACK_TIMEOUT RCU_INIT_POINTER(nf_ct_timeout_find_get_hook, NULL); RCU_INIT_POINTER(nf_ct_timeout_put_hook, NULL); diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index e7ac07e53b59..6669e68d589e 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -143,7 +143,7 @@ xt_ct_set_timeout(struct nf_conn *ct, const struct xt_tgchk_param *par, goto out; } - timeout = timeout_find_get(timeout_name); + timeout = timeout_find_get(par->net, timeout_name); if (timeout == NULL) { ret = -ENOENT; pr_info("No such timeout policy \"%s\"\n", timeout_name); -- cgit v1.2.3-59-g8ed1b From 4ec8ff0edccffe7a77f18e2a1e2ce86f03e08b5c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 7 Dec 2015 17:38:54 -0500 Subject: netfilter: prepare xt_cgroup for multi revisions xt_cgroup will grow cgroup2 path based match. Postfix existing symbols with _v0 and prepare for multi revision registration. Signed-off-by: Tejun Heo Cc: Daniel Borkmann Cc: Daniel Wagner CC: Neil Horman Cc: Jan Engelhardt Cc: Pablo Neira Ayuso Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/xt_cgroup.h | 2 +- net/netfilter/xt_cgroup.c | 36 +++++++++++++++++--------------- 2 files changed, 20 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/netfilter/xt_cgroup.h b/include/uapi/linux/netfilter/xt_cgroup.h index 43acb7e175f6..577c9e0b9406 100644 --- a/include/uapi/linux/netfilter/xt_cgroup.h +++ b/include/uapi/linux/netfilter/xt_cgroup.h @@ -3,7 +3,7 @@ #include -struct xt_cgroup_info { +struct xt_cgroup_info_v0 { __u32 id; __u32 invert; }; diff --git a/net/netfilter/xt_cgroup.c b/net/netfilter/xt_cgroup.c index 54eaeb45ce99..17300256772a 100644 --- a/net/netfilter/xt_cgroup.c +++ b/net/netfilter/xt_cgroup.c @@ -24,9 +24,9 @@ MODULE_DESCRIPTION("Xtables: process control group matching"); MODULE_ALIAS("ipt_cgroup"); MODULE_ALIAS("ip6t_cgroup"); -static int cgroup_mt_check(const struct xt_mtchk_param *par) +static int cgroup_mt_check_v0(const struct xt_mtchk_param *par) { - struct xt_cgroup_info *info = par->matchinfo; + struct xt_cgroup_info_v0 *info = par->matchinfo; if (info->invert & ~1) return -EINVAL; @@ -35,9 +35,9 @@ static int cgroup_mt_check(const struct xt_mtchk_param *par) } static bool -cgroup_mt(const struct sk_buff *skb, struct xt_action_param *par) +cgroup_mt_v0(const struct sk_buff *skb, struct xt_action_param *par) { - const struct xt_cgroup_info *info = par->matchinfo; + const struct xt_cgroup_info_v0 *info = par->matchinfo; if (skb->sk == NULL || !sk_fullsock(skb->sk)) return false; @@ -46,27 +46,29 @@ cgroup_mt(const struct sk_buff *skb, struct xt_action_param *par) info->invert; } -static struct xt_match cgroup_mt_reg __read_mostly = { - .name = "cgroup", - .revision = 0, - .family = NFPROTO_UNSPEC, - .checkentry = cgroup_mt_check, - .match = cgroup_mt, - .matchsize = sizeof(struct xt_cgroup_info), - .me = THIS_MODULE, - .hooks = (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_POST_ROUTING) | - (1 << NF_INET_LOCAL_IN), +static struct xt_match cgroup_mt_reg[] __read_mostly = { + { + .name = "cgroup", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = cgroup_mt_check_v0, + .match = cgroup_mt_v0, + .matchsize = sizeof(struct xt_cgroup_info_v0), + .me = THIS_MODULE, + .hooks = (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING) | + (1 << NF_INET_LOCAL_IN), + }, }; static int __init cgroup_mt_init(void) { - return xt_register_match(&cgroup_mt_reg); + return xt_register_matches(cgroup_mt_reg, ARRAY_SIZE(cgroup_mt_reg)); } static void __exit cgroup_mt_exit(void) { - xt_unregister_match(&cgroup_mt_reg); + xt_unregister_matches(cgroup_mt_reg, ARRAY_SIZE(cgroup_mt_reg)); } module_init(cgroup_mt_init); -- cgit v1.2.3-59-g8ed1b From c38c4597e4bf3e99860eac98211748e1ecb0e139 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 7 Dec 2015 17:38:55 -0500 Subject: netfilter: implement xt_cgroup cgroup2 path match This patch implements xt_cgroup path match which matches cgroup2 membership of the associated socket. The match is recursive and invertible. For rationales on introducing another cgroup based match, please refer to a preceding commit "sock, cgroup: add sock->sk_cgroup". v3: Folded into xt_cgroup as a new revision interface as suggested by Pablo. v2: Included linux/limits.h from xt_cgroup2.h for PATH_MAX. Added explicit alignment to the priv field. Both suggested by Jan. Signed-off-by: Tejun Heo Cc: Daniel Borkmann Cc: Daniel Wagner CC: Neil Horman Cc: Jan Engelhardt Cc: Pablo Neira Ayuso Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/xt_cgroup.h | 13 ++++++ net/netfilter/xt_cgroup.c | 69 ++++++++++++++++++++++++++++++++ 2 files changed, 82 insertions(+) (limited to 'include') diff --git a/include/uapi/linux/netfilter/xt_cgroup.h b/include/uapi/linux/netfilter/xt_cgroup.h index 577c9e0b9406..1e4b37b93bef 100644 --- a/include/uapi/linux/netfilter/xt_cgroup.h +++ b/include/uapi/linux/netfilter/xt_cgroup.h @@ -2,10 +2,23 @@ #define _UAPI_XT_CGROUP_H #include +#include struct xt_cgroup_info_v0 { __u32 id; __u32 invert; }; +struct xt_cgroup_info_v1 { + __u8 has_path; + __u8 has_classid; + __u8 invert_path; + __u8 invert_classid; + char path[PATH_MAX]; + __u32 classid; + + /* kernel internal data */ + void *priv __attribute__((aligned(8))); +}; + #endif /* _UAPI_XT_CGROUP_H */ diff --git a/net/netfilter/xt_cgroup.c b/net/netfilter/xt_cgroup.c index 17300256772a..a086a914865f 100644 --- a/net/netfilter/xt_cgroup.c +++ b/net/netfilter/xt_cgroup.c @@ -34,6 +34,37 @@ static int cgroup_mt_check_v0(const struct xt_mtchk_param *par) return 0; } +static int cgroup_mt_check_v1(const struct xt_mtchk_param *par) +{ + struct xt_cgroup_info_v1 *info = par->matchinfo; + struct cgroup *cgrp; + + if ((info->invert_path & ~1) || (info->invert_classid & ~1)) + return -EINVAL; + + if (!info->has_path && !info->has_classid) { + pr_info("xt_cgroup: no path or classid specified\n"); + return -EINVAL; + } + + if (info->has_path && info->has_classid) { + pr_info("xt_cgroup: both path and classid specified\n"); + return -EINVAL; + } + + if (info->has_path) { + cgrp = cgroup_get_from_path(info->path); + if (IS_ERR(cgrp)) { + pr_info("xt_cgroup: invalid path, errno=%ld\n", + PTR_ERR(cgrp)); + return -EINVAL; + } + info->priv = cgrp; + } + + return 0; +} + static bool cgroup_mt_v0(const struct sk_buff *skb, struct xt_action_param *par) { @@ -46,6 +77,31 @@ cgroup_mt_v0(const struct sk_buff *skb, struct xt_action_param *par) info->invert; } +static bool cgroup_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_cgroup_info_v1 *info = par->matchinfo; + struct sock_cgroup_data *skcd = &skb->sk->sk_cgrp_data; + struct cgroup *ancestor = info->priv; + + if (!skb->sk || !sk_fullsock(skb->sk)) + return false; + + if (ancestor) + return cgroup_is_descendant(sock_cgroup_ptr(skcd), ancestor) ^ + info->invert_path; + else + return (info->classid == sock_cgroup_classid(skcd)) ^ + info->invert_classid; +} + +static void cgroup_mt_destroy_v1(const struct xt_mtdtor_param *par) +{ + struct xt_cgroup_info_v1 *info = par->matchinfo; + + if (info->priv) + cgroup_put(info->priv); +} + static struct xt_match cgroup_mt_reg[] __read_mostly = { { .name = "cgroup", @@ -59,6 +115,19 @@ static struct xt_match cgroup_mt_reg[] __read_mostly = { (1 << NF_INET_POST_ROUTING) | (1 << NF_INET_LOCAL_IN), }, + { + .name = "cgroup", + .revision = 1, + .family = NFPROTO_UNSPEC, + .checkentry = cgroup_mt_check_v1, + .match = cgroup_mt_v1, + .matchsize = sizeof(struct xt_cgroup_info_v1), + .destroy = cgroup_mt_destroy_v1, + .me = THIS_MODULE, + .hooks = (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING) | + (1 << NF_INET_LOCAL_IN), + }, }; static int __init cgroup_mt_init(void) -- cgit v1.2.3-59-g8ed1b From bda13fed677bdb423b97dcf054f68b9eb4c6dbfb Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Sun, 13 Dec 2015 16:53:02 +0900 Subject: net: Fix typo in skb_fclone_busy This patch fix a typo found within comment of skb_fclone_busy. Signed-off-by: Masanari Iida Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9b9b9ead7bb3..af4f6ac025b6 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -833,7 +833,7 @@ struct sk_buff_fclones { * skb_fclone_busy - check if fclone is busy * @skb: buffer * - * Returns true is skb is a fast clone, and its clone is not freed. + * Returns true if skb is a fast clone, and its clone is not freed. * Some drivers call skb_orphan() in their ndo_start_xmit(), * so we also check that this didnt happen. */ -- cgit v1.2.3-59-g8ed1b From 6ff64f6f9242d7e50f3e99cb280f69d1927a5fa6 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 15 Dec 2015 16:03:35 +0100 Subject: switchdev: Pass original device to port netdev driver switchdev drivers need to know the netdev on which the switchdev op was invoked. For example, the STP state of a VLAN interface configured on top of a port can change while being member in a bridge. In this case, the underlying driver should only change the STP state of that particular VLAN and not of all the VLANs configured on the port. However, current switchdev infrastructure only passes the port netdev down to the driver. Solve that by passing the original device down to the driver as part of the required switchdev object / attribute. This doesn't entail any change in current switchdev drivers. It simply enables those supporting stacked devices to know the originating device and act accordingly. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/switchdev.h | 2 ++ net/bridge/br_fdb.c | 1 + net/bridge/br_stp.c | 2 ++ net/bridge/br_stp_if.c | 1 + net/bridge/br_vlan.c | 2 ++ net/core/net-sysfs.c | 1 + net/core/rtnetlink.c | 1 + net/switchdev/switchdev.c | 12 ++++++++++++ 8 files changed, 22 insertions(+) (limited to 'include') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 1d22ce9f352e..6612946167fe 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -50,6 +50,7 @@ enum switchdev_attr_id { }; struct switchdev_attr { + struct net_device *orig_dev; enum switchdev_attr_id id; u32 flags; union { @@ -68,6 +69,7 @@ enum switchdev_obj_id { }; struct switchdev_obj { + struct net_device *orig_dev; enum switchdev_obj_id id; u32 flags; }; diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index a642bb829d09..82e3e9705017 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -135,6 +135,7 @@ 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, }, diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c index 5f3f64553179..b3cca126b103 100644 --- a/net/bridge/br_stp.c +++ b/net/bridge/br_stp.c @@ -40,6 +40,7 @@ void br_log_state(const struct net_bridge_port *p) void br_set_state(struct net_bridge_port *p, unsigned int state) { struct switchdev_attr attr = { + .orig_dev = p->dev, .id = SWITCHDEV_ATTR_ID_PORT_STP_STATE, .flags = SWITCHDEV_F_DEFER, .u.stp_state = state, @@ -570,6 +571,7 @@ int br_set_max_age(struct net_bridge *br, unsigned long val) int br_set_ageing_time(struct net_bridge *br, u32 ageing_time) { struct switchdev_attr attr = { + .orig_dev = br->dev, .id = SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME, .flags = SWITCHDEV_F_SKIP_EOPNOTSUPP, .u.ageing_time = ageing_time, diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c index 5396ff08af32..775e00fbeb1e 100644 --- a/net/bridge/br_stp_if.c +++ b/net/bridge/br_stp_if.c @@ -37,6 +37,7 @@ static inline port_id br_make_port_id(__u8 priority, __u16 port_no) void br_init_port(struct net_bridge_port *p) { struct switchdev_attr attr = { + .orig_dev = p->dev, .id = SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME, .flags = SWITCHDEV_F_SKIP_EOPNOTSUPP | SWITCHDEV_F_DEFER, .u.ageing_time = p->br->ageing_time, diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 1394da63614a..66c4549efbbb 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -73,6 +73,7 @@ static int __vlan_vid_add(struct net_device *dev, struct net_bridge *br, u16 vid, u16 flags) { struct switchdev_obj_port_vlan v = { + .obj.orig_dev = dev, .obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, .flags = flags, .vid_begin = vid, @@ -120,6 +121,7 @@ static int __vlan_vid_del(struct net_device *dev, struct net_bridge *br, u16 vid) { struct switchdev_obj_port_vlan v = { + .obj.orig_dev = dev, .obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, .vid_begin = vid, .vid_end = vid, diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index f88a62ab019d..bca8c350e7f3 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -471,6 +471,7 @@ static ssize_t phys_switch_id_show(struct device *dev, if (dev_isalive(netdev)) { struct switchdev_attr attr = { + .orig_dev = netdev, .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, .flags = SWITCHDEV_F_NO_RECURSE, }; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 34ba7a08876d..d8b0113d3eec 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1027,6 +1027,7 @@ static int rtnl_phys_switch_id_fill(struct sk_buff *skb, struct net_device *dev) { int err; struct switchdev_attr attr = { + .orig_dev = dev, .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, .flags = SWITCHDEV_F_NO_RECURSE, }; diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index f34e535e93bd..df790d3385a2 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -723,6 +723,7 @@ 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, @@ -757,6 +758,7 @@ int switchdev_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, int nlflags) { struct switchdev_attr attr = { + .orig_dev = dev, .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, }; u16 mode = BRIDGE_MODE_UNDEF; @@ -778,6 +780,7 @@ static int switchdev_port_br_setflag(struct net_device *dev, unsigned long brport_flag) { struct switchdev_attr attr = { + .orig_dev = dev, .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, }; u8 flag = nla_get_u8(nlattr); @@ -853,6 +856,7 @@ static int switchdev_port_br_afspec(struct net_device *dev, 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; @@ -975,6 +979,7 @@ int switchdev_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], u16 vid, u16 nlm_flags) { struct switchdev_obj_port_fdb fdb = { + .obj.orig_dev = dev, .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, .vid = vid, }; @@ -1000,6 +1005,7 @@ int switchdev_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], u16 vid) { struct switchdev_obj_port_fdb fdb = { + .obj.orig_dev = dev, .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, .vid = vid, }; @@ -1077,6 +1083,7 @@ int switchdev_port_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb, 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, @@ -1135,6 +1142,7 @@ static struct net_device *switchdev_get_dev_by_nhs(struct fib_info *fi) if (!dev) return NULL; + attr.orig_dev = dev; if (switchdev_port_attr_get(dev, &attr)) return NULL; @@ -1194,6 +1202,7 @@ int switchdev_fib_ipv4_add(u32 dst, int dst_len, struct fib_info *fi, if (!dev) return 0; + ipv4_fib.obj.orig_dev = dev; err = switchdev_port_obj_add(dev, &ipv4_fib.obj); if (!err) fi->fib_flags |= RTNH_F_OFFLOAD; @@ -1238,6 +1247,7 @@ int switchdev_fib_ipv4_del(u32 dst, int dst_len, struct fib_info *fi, if (!dev) return 0; + ipv4_fib.obj.orig_dev = dev; err = switchdev_port_obj_del(dev, &ipv4_fib.obj); if (!err) fi->fib_flags &= ~RTNH_F_OFFLOAD; @@ -1270,10 +1280,12 @@ static bool switchdev_port_same_parent_id(struct net_device *a, struct net_device *b) { struct switchdev_attr a_attr = { + .orig_dev = a, .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, .flags = SWITCHDEV_F_NO_RECURSE, }; struct switchdev_attr b_attr = { + .orig_dev = b, .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, .flags = SWITCHDEV_F_NO_RECURSE, }; -- cgit v1.2.3-59-g8ed1b From 55dc5a9f2f2afd32d7b1bda44a5fc95e67a3371f Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:40 -0800 Subject: net: Add skb_inner_transport_offset function Same thing as skb_transport_offset but returns the offset of the inner transport header (when skb->encpasulation is set). Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index af4f6ac025b6..2393373c9d08 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1939,6 +1939,11 @@ static inline unsigned char *skb_inner_transport_header(const struct sk_buff return skb->head + skb->inner_transport_header; } +static inline int skb_inner_transport_offset(const struct sk_buff *skb) +{ + return skb_inner_transport_header(skb) - skb->data; +} + static inline void skb_reset_inner_transport_header(struct sk_buff *skb) { skb->inner_transport_header = skb->data - skb->head; -- cgit v1.2.3-59-g8ed1b From 53692b1de419c1b59106909c7f6b4dd3dbc768ac Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:41 -0800 Subject: sctp: Rename NETIF_F_SCTP_CSUM to NETIF_F_SCTP_CRC The SCTP checksum is really a CRC and is very different from the standards 1's complement checksum that serves as the checksum for IP protocols. This offload interface is also very different. Rename NETIF_F_SCTP_CSUM to NETIF_F_SCTP_CRC to highlight these differences. The term CSUM should be reserved in the stack to refer to the standard 1's complement IP checksum. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +- drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +- drivers/net/ethernet/intel/igb/igb_main.c | 4 ++-- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 ++-- drivers/net/loopback.c | 2 +- include/linux/netdev_features.h | 4 ++-- net/8021q/vlan_dev.c | 2 +- net/core/ethtool.c | 4 ++-- net/netfilter/ipvs/ip_vs_proto_sctp.c | 2 +- net/sctp/output.c | 2 +- 10 files changed, 14 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index b118deb08ce6..a63d980f478e 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -8842,7 +8842,7 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) netdev->features = NETIF_F_SG | NETIF_F_IP_CSUM | - NETIF_F_SCTP_CSUM | + NETIF_F_SCTP_CRC | NETIF_F_HIGHDMA | NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE | diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index 455394cf7f80..4d05ff6f0423 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -2321,7 +2321,7 @@ int i40evf_process_config(struct i40evf_adapter *adapter) netdev->features |= NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_IP_CSUM | - NETIF_F_SCTP_CSUM | + NETIF_F_SCTP_CRC | NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO6 | diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index 7afde455326d..31e5f3942839 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -2379,8 +2379,8 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } if (hw->mac.type >= e1000_82576) { - netdev->hw_features |= NETIF_F_SCTP_CSUM; - netdev->features |= NETIF_F_SCTP_CSUM; + netdev->hw_features |= NETIF_F_SCTP_CRC; + netdev->features |= NETIF_F_SCTP_CRC; } netdev->priv_flags |= IFF_UNICAST_FLT; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 66c64a376719..9f27001cac1f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8995,8 +8995,8 @@ skip_sriov: case ixgbe_mac_X540: case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: - netdev->features |= NETIF_F_SCTP_CSUM; - netdev->hw_features |= NETIF_F_SCTP_CSUM | + netdev->features |= NETIF_F_SCTP_CRC; + netdev->hw_features |= NETIF_F_SCTP_CRC | NETIF_F_NTUPLE; break; default: diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index dc7d970bd1c0..a400288cb37b 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -175,7 +175,7 @@ static void loopback_setup(struct net_device *dev) | NETIF_F_UFO | NETIF_F_HW_CSUM | NETIF_F_RXCSUM - | NETIF_F_SCTP_CSUM + | NETIF_F_SCTP_CRC | NETIF_F_HIGHDMA | NETIF_F_LLTX | NETIF_F_NETNS_LOCAL diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index f0d87347df19..6395f8309393 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -52,7 +52,7 @@ enum { NETIF_F_GSO_TUNNEL_REMCSUM_BIT, NETIF_F_FCOE_CRC_BIT, /* FCoE CRC32 */ - NETIF_F_SCTP_CSUM_BIT, /* SCTP checksum offload */ + NETIF_F_SCTP_CRC_BIT, /* SCTP checksum offload */ NETIF_F_FCOE_MTU_BIT, /* Supports max FCoE MTU, 2158 bytes*/ NETIF_F_NTUPLE_BIT, /* N-tuple filters supported */ NETIF_F_RXHASH_BIT, /* Receive hashing offload */ @@ -103,7 +103,7 @@ enum { #define NETIF_F_NTUPLE __NETIF_F(NTUPLE) #define NETIF_F_RXCSUM __NETIF_F(RXCSUM) #define NETIF_F_RXHASH __NETIF_F(RXHASH) -#define NETIF_F_SCTP_CSUM __NETIF_F(SCTP_CSUM) +#define NETIF_F_SCTP_CRC __NETIF_F(SCTP_CRC) #define NETIF_F_SG __NETIF_F(SG) #define NETIF_F_TSO6 __NETIF_F(TSO6) #define NETIF_F_TSO_ECN __NETIF_F(TSO_ECN) diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 9f4bd137e045..45b74e875381 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -545,7 +545,7 @@ static int vlan_dev_init(struct net_device *dev) dev->hw_features = NETIF_F_ALL_CSUM | NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE | - NETIF_F_HIGHDMA | NETIF_F_SCTP_CSUM | + NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC | NETIF_F_ALL_FCOE; dev->features |= real_dev->vlan_features | NETIF_F_LLTX | diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 29edf74846fc..4a0cab85d67d 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -87,7 +87,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] [NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation", [NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc", - [NETIF_F_SCTP_CSUM_BIT] = "tx-checksum-sctp", + [NETIF_F_SCTP_CRC_BIT] = "tx-checksum-sctp", [NETIF_F_FCOE_MTU_BIT] = "fcoe-mtu", [NETIF_F_NTUPLE_BIT] = "rx-ntuple-filter", [NETIF_F_RXHASH_BIT] = "rx-hashing", @@ -235,7 +235,7 @@ static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd) switch (eth_cmd) { case ETHTOOL_GTXCSUM: case ETHTOOL_STXCSUM: - return NETIF_F_ALL_CSUM | NETIF_F_SCTP_CSUM; + return NETIF_F_ALL_CSUM | NETIF_F_SCTP_CRC; case ETHTOOL_GRXCSUM: case ETHTOOL_SRXCSUM: return NETIF_F_RXCSUM; diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 010ddeec135f..d952d67f904d 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -169,7 +169,7 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp, /* Only update csum if we really have to */ if (sctph->dest != cp->dport || payload_csum || (skb->ip_summed == CHECKSUM_PARTIAL && - !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CSUM))) { + !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CRC))) { sctph->dest = cp->dport; sctp_nat_csum(skb, sctph, sctphoff); } else if (skb->ip_summed != CHECKSUM_PARTIAL) { diff --git a/net/sctp/output.c b/net/sctp/output.c index abe7c2db2412..9d610eddd19e 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -534,7 +534,7 @@ int sctp_packet_transmit(struct sctp_packet *packet) * by CRC32-C as described in . */ if (!sctp_checksum_disable) { - if (!(dst->dev->features & NETIF_F_SCTP_CSUM) || + if (!(dst->dev->features & NETIF_F_SCTP_CRC) || (dst_xfrm(dst) != NULL) || packet->ipfragok) { sh->checksum = sctp_compute_cksum(nskb, 0); } else { -- cgit v1.2.3-59-g8ed1b From a188222b6ed29404ac2d4232d35d1fe0e77af370 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:43 -0800 Subject: net: Rename NETIF_F_ALL_CSUM to NETIF_F_CSUM_MASK The name NETIF_F_ALL_CSUM is a misnomer. This does not correspond to the set of features for offloading all checksums. This is a mask of the checksum offload related features bits. It is incorrect to set both NETIF_F_HW_CSUM and NETIF_F_IP_CSUM or NETIF_F_IPV6 at the same time for features of a device. This patch: - Changes instances of NETIF_F_ALL_CSUM to NETIF_F_CSUM_MASK (where NETIF_F_ALL_CSUM is being used as a mask). - Changes bonding, sfc/efx, ipvlan, macvlan, vlan, and team drivers to use NEITF_F_HW_CSUM in features list instead of NETIF_F_ALL_CSUM. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 7 +++---- drivers/net/ethernet/emulex/benet/be_main.c | 2 +- drivers/net/ethernet/ibm/ibmveth.c | 5 +++-- drivers/net/ethernet/intel/fm10k/fm10k_netdev.c | 2 +- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +- drivers/net/ethernet/jme.c | 2 +- drivers/net/ethernet/marvell/sky2.c | 2 +- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 4 ++-- drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c | 2 +- drivers/net/ethernet/sfc/efx.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 4 ++-- drivers/net/ipvlan/ipvlan_main.c | 2 +- drivers/net/macvlan.c | 2 +- drivers/net/macvtap.c | 2 +- drivers/net/team/team.c | 3 +-- drivers/net/usb/r8152.c | 2 +- drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib.c | 2 +- include/linux/netdev_features.h | 7 ++++++- include/linux/netdevice.h | 6 +++--- include/net/vxlan.h | 2 +- net/8021q/vlan_dev.c | 2 +- net/core/dev.c | 10 +++++----- net/core/ethtool.c | 2 +- net/ipv4/tcp.c | 4 ++-- 25 files changed, 43 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index fe0e7a6f4d72..cab99fd44c8e 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1067,12 +1067,12 @@ static netdev_features_t bond_fix_features(struct net_device *dev, return features; } -#define BOND_VLAN_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | \ +#define BOND_VLAN_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ NETIF_F_HIGHDMA | NETIF_F_LRO) -#define BOND_ENC_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | NETIF_F_RXCSUM |\ - NETIF_F_ALL_TSO) +#define BOND_ENC_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ + NETIF_F_RXCSUM | NETIF_F_ALL_TSO) static void bond_compute_features(struct bonding *bond) { @@ -4182,7 +4182,6 @@ void bond_setup(struct net_device *bond_dev) NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER; - bond_dev->hw_features &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_HW_CSUM); bond_dev->hw_features |= NETIF_F_GSO_ENCAP_ALL; bond_dev->features |= bond_dev->hw_features; } diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 4cab8879f5ae..34e324f20d80 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -5289,7 +5289,7 @@ static netdev_features_t be_features_check(struct sk_buff *skb, skb->inner_protocol != htons(ETH_P_TEB) || skb_inner_mac_header(skb) - skb_transport_header(skb) != sizeof(struct udphdr) + sizeof(struct vxlanhdr)) - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); return features; } diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index 7af870a3c549..6691b5a45b9d 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -763,7 +763,7 @@ static netdev_features_t ibmveth_fix_features(struct net_device *dev, */ if (!(features & NETIF_F_RXCSUM)) - features &= ~NETIF_F_ALL_CSUM; + features &= ~NETIF_F_CSUM_MASK; return features; } @@ -928,7 +928,8 @@ static int ibmveth_set_features(struct net_device *dev, rc1 = ibmveth_set_csum_offload(dev, rx_csum); if (rc1 && !adapter->rx_csum) dev->features = - features & ~(NETIF_F_ALL_CSUM | NETIF_F_RXCSUM); + features & ~(NETIF_F_CSUM_MASK | + NETIF_F_RXCSUM); } if (large_send != adapter->large_send) { diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c index d9854d39576d..83ddf362ea77 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c @@ -1357,7 +1357,7 @@ static netdev_features_t fm10k_features_check(struct sk_buff *skb, if (!skb->encapsulation || fm10k_tx_encap_offload(skb)) return features; - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); } static const struct net_device_ops fm10k_netdev_ops = { diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index a63d980f478e..c284e4341c7c 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -8766,7 +8766,7 @@ static netdev_features_t i40e_features_check(struct sk_buff *skb, if (skb->encapsulation && (skb_inner_mac_header(skb) - skb_transport_header(skb) > I40E_MAX_TUNNEL_HDR_LEN)) - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); return features; } diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 9f27001cac1f..fca35aa90d0f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8598,7 +8598,7 @@ ixgbe_features_check(struct sk_buff *skb, struct net_device *dev, if (unlikely(skb_inner_mac_header(skb) - skb_transport_header(skb) > IXGBE_MAX_TUNNEL_HDR_LEN)) - return features & ~NETIF_F_ALL_CSUM; + return features & ~NETIF_F_CSUM_MASK; return features; } diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c index 060dd3922974..b1de7afd4116 100644 --- a/drivers/net/ethernet/jme.c +++ b/drivers/net/ethernet/jme.c @@ -2753,7 +2753,7 @@ static netdev_features_t jme_fix_features(struct net_device *netdev, netdev_features_t features) { if (netdev->mtu > 1900) - features &= ~(NETIF_F_ALL_TSO | NETIF_F_ALL_CSUM); + features &= ~(NETIF_F_ALL_TSO | NETIF_F_CSUM_MASK); return features; } diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index 5606a043063e..ec0a22119e09 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -4380,7 +4380,7 @@ static netdev_features_t sky2_fix_features(struct net_device *dev, */ if (dev->mtu > ETH_DATA_LEN && hw->chip_id == CHIP_ID_YUKON_EC_U) { netdev_info(dev, "checksum offload not possible with jumbo frames\n"); - features &= ~(NETIF_F_TSO|NETIF_F_SG|NETIF_F_ALL_CSUM); + features &= ~(NETIF_F_TSO | NETIF_F_SG | NETIF_F_CSUM_MASK); } /* Some hardware requires receive checksum for RSS to work. */ diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 038ac6b14a60..7060539d276a 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -2071,7 +2071,7 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev, l4_hdr = ipv6_hdr(skb)->nexthdr; break; default: - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); } if (skb->inner_protocol_type != ENCAP_TYPE_ETHER || @@ -2080,7 +2080,7 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev, (l4_hdr == IPPROTO_UDP && (skb_inner_mac_header(skb) - skb_transport_header(skb) != sizeof(struct udphdr) + sizeof(struct vxlanhdr)))) - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); return features; } diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c index 08d4be616064..e097e6baaac4 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c @@ -500,7 +500,7 @@ void pch_gbe_check_options(struct pch_gbe_adapter *adapter) val = XsumTX; pch_gbe_validate_option(&val, &opt, adapter); if (!val) - dev->features &= ~NETIF_F_ALL_CSUM; + dev->features &= ~NETIF_F_CSUM_MASK; } { /* Flow Control */ static const struct pch_gbe_option opt = { diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index b405349a570c..1fe13c733c1e 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -3131,7 +3131,7 @@ static int efx_pci_probe(struct pci_dev *pci_dev, if (efx->type->offload_features & NETIF_F_V6_CSUM) net_dev->features |= NETIF_F_TSO6; /* Mask for features that also apply to VLAN devices */ - net_dev->vlan_features |= (NETIF_F_ALL_CSUM | NETIF_F_SG | + net_dev->vlan_features |= (NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_ALL_TSO | NETIF_F_RXCSUM); /* All offloads can be toggled */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 3c6549aee11d..0b0fea73a7a7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2402,7 +2402,7 @@ static netdev_features_t stmmac_fix_features(struct net_device *dev, features &= ~NETIF_F_RXCSUM; if (!priv->plat->tx_coe) - features &= ~NETIF_F_ALL_CSUM; + features &= ~NETIF_F_CSUM_MASK; /* Some GMAC devices have a bugged Jumbo frame support that * needs to have the Tx COE disabled for oversized frames @@ -2410,7 +2410,7 @@ static netdev_features_t stmmac_fix_features(struct net_device *dev, * the TX csum insertionin the TDES and not use SF. */ if (priv->plat->bugged_jumbo && (dev->mtu > ETH_DATA_LEN)) - features &= ~NETIF_F_ALL_CSUM; + features &= ~NETIF_F_CSUM_MASK; return features; } diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index a9268db4e349..f94392d07126 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -88,7 +88,7 @@ static struct lock_class_key ipvlan_netdev_xmit_lock_key; static struct lock_class_key ipvlan_netdev_addr_lock_key; #define IPVLAN_FEATURES \ - (NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ + (NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ NETIF_F_GSO | NETIF_F_TSO | NETIF_F_UFO | NETIF_F_GSO_ROBUST | \ NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_GRO | NETIF_F_RXCSUM | \ NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_FILTER) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 06c8bfeaccd6..ae3b486fb663 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -762,7 +762,7 @@ static struct lock_class_key macvlan_netdev_addr_lock_key; NETIF_F_GSO_ROBUST) #define MACVLAN_FEATURES \ - (NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ + (NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ NETIF_F_GSO | NETIF_F_TSO | NETIF_F_UFO | NETIF_F_LRO | \ NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_GRO | NETIF_F_RXCSUM | \ NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_FILTER) diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 0fc521941c71..d636d051fac8 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -388,7 +388,7 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb) * check, we either support them all or none. */ if (skb->ip_summed == CHECKSUM_PARTIAL && - !(features & NETIF_F_ALL_CSUM) && + !(features & NETIF_F_CSUM_MASK) && skb_checksum_help(skb)) goto drop; skb_queue_tail(&q->sk.sk_receive_queue, skb); diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 059c0f60a2b2..915f60fce186 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -981,7 +981,7 @@ static void team_port_disable(struct team *team, team_lower_state_changed(port); } -#define TEAM_VLAN_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | \ +#define TEAM_VLAN_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ NETIF_F_HIGHDMA | NETIF_F_LRO) @@ -2091,7 +2091,6 @@ static void team_setup(struct net_device *dev) NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER; - dev->hw_features &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_HW_CSUM); dev->features |= dev->hw_features; } diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index d9427ca3dba7..34642a9583e0 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -1986,7 +1986,7 @@ rtl8152_features_check(struct sk_buff *skb, struct net_device *dev, int offset = skb_transport_offset(skb); if ((mss || skb->ip_summed == CHECKSUM_PARTIAL) && offset > max_offset) - features &= ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); else if ((skb->len + sizeof(struct tx_desc)) > agg_buf_sz) features &= ~NETIF_F_GSO_MASK; diff --git a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib.c b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib.c index 679785b0209c..9de4f23910d8 100644 --- a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib.c +++ b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib.c @@ -69,7 +69,7 @@ ksocknal_lib_zc_capable(ksock_conn_t *conn) /* ZC if the socket supports scatter/gather and doesn't need software * checksums */ - return ((caps & NETIF_F_SG) != 0 && (caps & NETIF_F_ALL_CSUM) != 0); + return ((caps & NETIF_F_SG) != 0 && (caps & NETIF_F_CSUM_MASK) != 0); } int diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index 6395f8309393..2c4e94ab88da 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -149,7 +149,12 @@ enum { #define NETIF_F_GEN_CSUM NETIF_F_HW_CSUM #define NETIF_F_V4_CSUM (NETIF_F_GEN_CSUM | NETIF_F_IP_CSUM) #define NETIF_F_V6_CSUM (NETIF_F_GEN_CSUM | NETIF_F_IPV6_CSUM) -#define NETIF_F_ALL_CSUM (NETIF_F_V4_CSUM | NETIF_F_V6_CSUM) + +/* List of IP checksum features. Note that NETIF_HW_CSUM should not be + * set in features when NETIF_F_IP_CSUM or NETIF_F_IPV6_CSUM are set-- + * this would be contradictory + */ +#define NETIF_F_CSUM_MASK (NETIF_F_V4_CSUM | NETIF_F_V6_CSUM) #define NETIF_F_ALL_TSO (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1bb21ff0fa64..a54223a113b1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3763,12 +3763,12 @@ static inline netdev_features_t netdev_intersect_features(netdev_features_t f1, netdev_features_t f2) { if (f1 & NETIF_F_GEN_CSUM) - f1 |= (NETIF_F_ALL_CSUM & ~NETIF_F_GEN_CSUM); + f1 |= (NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); if (f2 & NETIF_F_GEN_CSUM) - f2 |= (NETIF_F_ALL_CSUM & ~NETIF_F_GEN_CSUM); + f2 |= (NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); f1 &= f2; if (f1 & NETIF_F_GEN_CSUM) - f1 &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_GEN_CSUM); + f1 &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); return f1; } diff --git a/include/net/vxlan.h b/include/net/vxlan.h index c1c899c3a51b..b5a1aec1a167 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -232,7 +232,7 @@ static inline netdev_features_t vxlan_features_check(struct sk_buff *skb, skb->inner_protocol != htons(ETH_P_TEB) || (skb_inner_mac_header(skb) - skb_transport_header(skb) != sizeof(struct udphdr) + sizeof(struct vxlanhdr)))) - return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK); + return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); return features; } diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 45b74e875381..ad5e2fd1012c 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -543,7 +543,7 @@ static int vlan_dev_init(struct net_device *dev) (1<<__LINK_STATE_DORMANT))) | (1<<__LINK_STATE_PRESENT); - dev->hw_features = NETIF_F_ALL_CSUM | NETIF_F_SG | + dev->hw_features = NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE | NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC | NETIF_F_ALL_FCOE; diff --git a/net/core/dev.c b/net/core/dev.c index 8f705fcedb94..5a3b5a404642 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2645,7 +2645,7 @@ static netdev_features_t harmonize_features(struct sk_buff *skb, if (skb->ip_summed != CHECKSUM_NONE && !can_checksum_protocol(features, type)) { - features &= ~NETIF_F_ALL_CSUM; + features &= ~NETIF_F_CSUM_MASK; } else if (illegal_highdma(skb->dev, skb)) { features &= ~NETIF_F_SG; } @@ -2792,7 +2792,7 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device else skb_set_transport_header(skb, skb_checksum_start_offset(skb)); - if (!(features & NETIF_F_ALL_CSUM) && + if (!(features & NETIF_F_CSUM_MASK) && skb_checksum_help(skb)) goto out_kfree_skb; } @@ -7572,15 +7572,15 @@ netdev_features_t netdev_increment_features(netdev_features_t all, netdev_features_t one, netdev_features_t mask) { if (mask & NETIF_F_GEN_CSUM) - mask |= NETIF_F_ALL_CSUM; + mask |= NETIF_F_CSUM_MASK; mask |= NETIF_F_VLAN_CHALLENGED; - all |= one & (NETIF_F_ONE_FOR_ALL|NETIF_F_ALL_CSUM) & mask; + all |= one & (NETIF_F_ONE_FOR_ALL | NETIF_F_CSUM_MASK) & mask; all &= one | ~NETIF_F_ALL_FOR_ALL; /* If one device supports hw checksumming, set for all. */ if (all & NETIF_F_GEN_CSUM) - all &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_GEN_CSUM); + all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); return all; } diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 4a0cab85d67d..09948a726347 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -235,7 +235,7 @@ static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd) switch (eth_cmd) { case ETHTOOL_GTXCSUM: case ETHTOOL_STXCSUM: - return NETIF_F_ALL_CSUM | NETIF_F_SCTP_CRC; + return NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC; case ETHTOOL_GRXCSUM: case ETHTOOL_SRXCSUM: return NETIF_F_RXCSUM; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c82cca18c90f..cf7ef7be79f0 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1018,7 +1018,7 @@ int tcp_sendpage(struct sock *sk, struct page *page, int offset, ssize_t res; if (!(sk->sk_route_caps & NETIF_F_SG) || - !(sk->sk_route_caps & NETIF_F_ALL_CSUM)) + !(sk->sk_route_caps & NETIF_F_CSUM_MASK)) return sock_no_sendpage(sk->sk_socket, page, offset, size, flags); @@ -1175,7 +1175,7 @@ new_segment: /* * Check whether we can use HW checksum. */ - if (sk->sk_route_caps & NETIF_F_ALL_CSUM) + if (sk->sk_route_caps & NETIF_F_CSUM_MASK) skb->ip_summed = CHECKSUM_PARTIAL; skb_entail(sk, skb); -- cgit v1.2.3-59-g8ed1b From c8cd0989bd151fda87bbf10887b3df18021284bc Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:44 -0800 Subject: net: Eliminate NETIF_F_GEN_CSUM and NETIF_F_V[46]_CSUM These netif flags are unnecessary convolutions. It is more straightforward to just use NETIF_F_HW_CSUM, NETIF_F_IP_CSUM, and NETIF_F_IPV6_CSUM directly. This patch also: - Cleans up can_checksum_protocol - Simplifies netdev_intersect_features Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/net/ethernet/sfc/efx.c | 2 +- drivers/net/macvlan.c | 2 +- include/linux/if_vlan.h | 2 +- include/linux/netdev_features.h | 9 +++---- include/linux/netdevice.h | 40 ++++++++++++++++++++------------ net/core/dev.c | 12 +++++----- net/core/pktgen.c | 4 ++-- net/ipv4/ip_output.c | 2 +- net/ipv4/netfilter/nf_nat_l3proto_ipv4.c | 3 ++- net/ipv4/udp.c | 3 ++- net/ipv4/udp_offload.c | 5 ++-- net/ipv6/ip6_output.c | 2 +- net/ipv6/netfilter/nf_nat_l3proto_ipv6.c | 3 ++- 13 files changed, 50 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 1fe13c733c1e..6f697438545d 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -3128,7 +3128,7 @@ static int efx_pci_probe(struct pci_dev *pci_dev, net_dev->features |= (efx->type->offload_features | NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_RXCSUM); - if (efx->type->offload_features & NETIF_F_V6_CSUM) + if (efx->type->offload_features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM)) net_dev->features |= NETIF_F_TSO6; /* Mask for features that also apply to VLAN devices */ net_dev->vlan_features |= (NETIF_F_HW_CSUM | NETIF_F_SG | diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index ae3b486fb663..6a57a005e0ca 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -758,7 +758,7 @@ static struct lock_class_key macvlan_netdev_xmit_lock_key; static struct lock_class_key macvlan_netdev_addr_lock_key; #define ALWAYS_ON_FEATURES \ - (NETIF_F_SG | NETIF_F_GEN_CSUM | NETIF_F_GSO_SOFTWARE | NETIF_F_LLTX | \ + (NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE | NETIF_F_LLTX | \ NETIF_F_GSO_ROBUST) #define MACVLAN_FEATURES \ diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 05f5879821b8..a5f6ce6b578c 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -621,7 +621,7 @@ static inline netdev_features_t vlan_features_check(const struct sk_buff *skb, NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | - NETIF_F_GEN_CSUM | + NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX); diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index 2c4e94ab88da..d9654f0eecb3 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -146,15 +146,12 @@ enum { #define NETIF_F_GSO_SOFTWARE (NETIF_F_TSO | NETIF_F_TSO_ECN | \ NETIF_F_TSO6 | NETIF_F_UFO) -#define NETIF_F_GEN_CSUM NETIF_F_HW_CSUM -#define NETIF_F_V4_CSUM (NETIF_F_GEN_CSUM | NETIF_F_IP_CSUM) -#define NETIF_F_V6_CSUM (NETIF_F_GEN_CSUM | NETIF_F_IPV6_CSUM) - -/* List of IP checksum features. Note that NETIF_HW_CSUM should not be +/* List of IP checksum features. Note that NETIF_F_ HW_CSUM should not be * set in features when NETIF_F_IP_CSUM or NETIF_F_IPV6_CSUM are set-- * this would be contradictory */ -#define NETIF_F_CSUM_MASK (NETIF_F_V4_CSUM | NETIF_F_V6_CSUM) +#define NETIF_F_CSUM_MASK (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | \ + NETIF_F_HW_CSUM) #define NETIF_F_ALL_TSO (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a54223a113b1..283984b67cd9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3691,13 +3691,24 @@ __be16 skb_network_protocol(struct sk_buff *skb, int *depth); static inline bool can_checksum_protocol(netdev_features_t features, __be16 protocol) { - return ((features & NETIF_F_GEN_CSUM) || - ((features & NETIF_F_V4_CSUM) && - protocol == htons(ETH_P_IP)) || - ((features & NETIF_F_V6_CSUM) && - protocol == htons(ETH_P_IPV6)) || - ((features & NETIF_F_FCOE_CRC) && - protocol == htons(ETH_P_FCOE))); + if (protocol == htons(ETH_P_FCOE)) + return !!(features & NETIF_F_FCOE_CRC); + + /* Assume this is an IP checksum (not SCTP CRC) */ + + if (features & NETIF_F_HW_CSUM) { + /* Can checksum everything */ + return true; + } + + switch (protocol) { + case htons(ETH_P_IP): + return !!(features & NETIF_F_IP_CSUM); + case htons(ETH_P_IPV6): + return !!(features & NETIF_F_IPV6_CSUM); + default: + return false; + } } #ifdef CONFIG_BUG @@ -3762,15 +3773,14 @@ void linkwatch_run_queue(void); static inline netdev_features_t netdev_intersect_features(netdev_features_t f1, netdev_features_t f2) { - if (f1 & NETIF_F_GEN_CSUM) - f1 |= (NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); - if (f2 & NETIF_F_GEN_CSUM) - f2 |= (NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); - f1 &= f2; - if (f1 & NETIF_F_GEN_CSUM) - f1 &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); + if ((f1 ^ f2) & NETIF_F_HW_CSUM) { + if (f1 & NETIF_F_HW_CSUM) + f1 |= (NETIF_F_IP_CSUM|NETIF_F_IP_CSUM); + else + f2 |= (NETIF_F_IP_CSUM|NETIF_F_IP_CSUM); + } - return f1; + return f1 & f2; } static inline netdev_features_t netdev_get_wanted_features( diff --git a/net/core/dev.c b/net/core/dev.c index 5a3b5a404642..45b013f27625 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6467,9 +6467,9 @@ static netdev_features_t netdev_fix_features(struct net_device *dev, /* UFO needs SG and checksumming */ if (features & NETIF_F_UFO) { /* maybe split UFO into V4 and V6? */ - if (!((features & NETIF_F_GEN_CSUM) || - (features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM)) - == (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))) { + 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; @@ -7571,7 +7571,7 @@ static int dev_cpu_callback(struct notifier_block *nfb, netdev_features_t netdev_increment_features(netdev_features_t all, netdev_features_t one, netdev_features_t mask) { - if (mask & NETIF_F_GEN_CSUM) + if (mask & NETIF_F_HW_CSUM) mask |= NETIF_F_CSUM_MASK; mask |= NETIF_F_VLAN_CHALLENGED; @@ -7579,8 +7579,8 @@ netdev_features_t netdev_increment_features(netdev_features_t all, all &= one | ~NETIF_F_ALL_FOR_ALL; /* If one device supports hw checksumming, set for all. */ - if (all & NETIF_F_GEN_CSUM) - all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_GEN_CSUM); + if (all & NETIF_F_HW_CSUM) + all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_HW_CSUM); return all; } diff --git a/net/core/pktgen.c b/net/core/pktgen.c index de8d5cc5eb24..2be144498bcf 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2898,7 +2898,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev, if (!(pkt_dev->flags & F_UDPCSUM)) { skb->ip_summed = CHECKSUM_NONE; - } else if (odev->features & NETIF_F_V4_CSUM) { + } else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM)) { skb->ip_summed = CHECKSUM_PARTIAL; skb->csum = 0; udp4_hwcsum(skb, iph->saddr, iph->daddr); @@ -3032,7 +3032,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev, if (!(pkt_dev->flags & F_UDPCSUM)) { skb->ip_summed = CHECKSUM_NONE; - } else if (odev->features & NETIF_F_V6_CSUM) { + } else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM)) { skb->ip_summed = CHECKSUM_PARTIAL; skb->csum_start = skb_transport_header(skb) - skb->head; skb->csum_offset = offsetof(struct udphdr, check); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index e0b94cd843d7..568e2bc0d93d 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -911,7 +911,7 @@ static int __ip_append_data(struct sock *sk, */ if (transhdrlen && length + fragheaderlen <= mtu && - rt->dst.dev->features & NETIF_F_V4_CSUM && + rt->dst.dev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM) && !(flags & MSG_MORE) && !exthdrlen) csummode = CHECKSUM_PARTIAL; diff --git a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c index 5075b7ecd26d..61c7cc22ea68 100644 --- a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c @@ -132,7 +132,8 @@ static void nf_nat_ipv4_csum_recalc(struct sk_buff *skb, if (skb->ip_summed != CHECKSUM_PARTIAL) { if (!(rt->rt_flags & RTCF_LOCAL) && - (!skb->dev || skb->dev->features & NETIF_F_V4_CSUM)) { + (!skb->dev || skb->dev->features & + (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) { skb->ip_summed = CHECKSUM_PARTIAL; skb->csum_start = skb_headroom(skb) + skb_network_offset(skb) + diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 0c7b0e61b917..8841e984f8bf 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -772,7 +772,8 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb, else if (skb_is_gso(skb)) uh->check = ~udp_v4_check(len, saddr, daddr, 0); else if (skb_dst(skb) && skb_dst(skb)->dev && - (skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) { + (skb_dst(skb)->dev->features & + (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) { BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL); diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index f9386160cbee..130042660181 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -60,8 +60,9 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, /* Try to offload checksum if possible */ offload_csum = !!(need_csum && - (skb->dev->features & - (is_ipv6 ? NETIF_F_V6_CSUM : NETIF_F_V4_CSUM))); + ((skb->dev->features & NETIF_F_HW_CSUM) || + (skb->dev->features & (is_ipv6 ? + NETIF_F_IPV6_CSUM : NETIF_F_IP_CSUM)))); /* segment inner packet. */ enc_features = skb->dev->hw_enc_features & features; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index e6a7bd15b9b7..2f748452b4aa 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1322,7 +1322,7 @@ emsgsize: headersize == sizeof(struct ipv6hdr) && length < mtu - headersize && !(flags & MSG_MORE) && - rt->dst.dev->features & NETIF_F_V6_CSUM) + rt->dst.dev->features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM)) csummode = CHECKSUM_PARTIAL; if (sk->sk_type == SOCK_DGRAM || sk->sk_type == SOCK_RAW) { diff --git a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c index 238e70c3f7b7..6ce309928841 100644 --- a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c @@ -136,7 +136,8 @@ static void nf_nat_ipv6_csum_recalc(struct sk_buff *skb, if (skb->ip_summed != CHECKSUM_PARTIAL) { if (!(rt->rt6i_flags & RTF_LOCAL) && - (!skb->dev || skb->dev->features & NETIF_F_V6_CSUM)) { + (!skb->dev || skb->dev->features & + (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))) { skb->ip_summed = CHECKSUM_PARTIAL; skb->csum_start = skb_headroom(skb) + skb_network_offset(skb) + -- cgit v1.2.3-59-g8ed1b From 9a49850d0af7b9fd14d091dfe61ef6cb369f86b9 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:45 -0800 Subject: tcp: Fix conditions to determine checksum offload In tcp_send_sendpage and tcp_sendmsg we check the route capabilities to determine if checksum offload can be performed. This check currently does not take the IP protocol into account for devices that advertise only one of NETIF_F_IPV6_CSUM or NETIF_F_IP_CSUM. This patch adds a function to check capabilities for checksum offload with a socket called sk_check_csum_caps. This function checks for specific IPv4 or IPv6 offload support based on the family of the socket. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/net/sock.h | 9 +++++++++ net/ipv4/tcp.c | 4 ++-- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 0ca22b014de1..ab0269f4b2cc 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1791,6 +1791,15 @@ static inline void sk_nocaps_add(struct sock *sk, netdev_features_t flags) sk->sk_route_caps &= ~flags; } +static inline bool sk_check_csum_caps(struct sock *sk) +{ + return (sk->sk_route_caps & NETIF_F_HW_CSUM) || + (sk->sk_family == PF_INET && + (sk->sk_route_caps & NETIF_F_IP_CSUM)) || + (sk->sk_family == PF_INET6 && + (sk->sk_route_caps & NETIF_F_IPV6_CSUM)); +} + static inline int skb_do_copy_data_nocache(struct sock *sk, struct sk_buff *skb, struct iov_iter *from, char *to, int copy, int offset) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index cf7ef7be79f0..92b3e61b847d 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1018,7 +1018,7 @@ int tcp_sendpage(struct sock *sk, struct page *page, int offset, ssize_t res; if (!(sk->sk_route_caps & NETIF_F_SG) || - !(sk->sk_route_caps & NETIF_F_CSUM_MASK)) + !sk_check_csum_caps(sk)) return sock_no_sendpage(sk->sk_socket, page, offset, size, flags); @@ -1175,7 +1175,7 @@ new_segment: /* * Check whether we can use HW checksum. */ - if (sk->sk_route_caps & NETIF_F_CSUM_MASK) + if (sk_check_csum_caps(sk)) skb->ip_summed = CHECKSUM_PARTIAL; skb_entail(sk, skb); -- cgit v1.2.3-59-g8ed1b From 6ae23ad36253a8033c5714c52b691b84456487c5 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:46 -0800 Subject: net: Add driver helper functions to determine checksum offloadability Add skb_csum_offload_chk driver helper function to determine if a device with limited checksum offload capabilities is able to offload the checksum for a given packet. This patch includes: - The skb_csum_offload_chk function. Returns true if checksum is offloadable, else false. Optionally, in the case that the checksum is not offloable, the function can call skb_checksum_help to resolve the checksum. skb_csum_offload_chk also returns whether the checksum refers to an encapsulated checksum. - Definition of skb_csum_offl_spec structure that caller uses to indicate rules about what it can offload (e.g. IPv4/v6, TCP/UDP only, whether encapsulated checksums can be offloaded, whether checksum with IPv6 extension headers can be offloaded). - Ancilary functions called skb_csum_offload_chk_help, skb_csum_off_chk_help_cmn, skb_csum_off_chk_help_cmn_v4_only. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/netdevice.h | 78 ++++++++++++++++++++++++++ net/core/dev.c | 136 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 214 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 283984b67cd9..9fb6395967de 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2522,6 +2522,71 @@ static inline void skb_gro_remcsum_cleanup(struct sk_buff *skb, remcsum_unadjust((__sum16 *)ptr, grc->delta); } +struct skb_csum_offl_spec { + __u16 ipv4_okay:1, + ipv6_okay:1, + encap_okay:1, + ip_options_okay:1, + ext_hdrs_okay:1, + tcp_okay:1, + udp_okay:1, + sctp_okay:1, + vlan_okay:1, + no_encapped_ipv6:1, + no_not_encapped:1; +}; + +bool __skb_csum_offload_chk(struct sk_buff *skb, + const struct skb_csum_offl_spec *spec, + bool *csum_encapped, + bool csum_help); + +static inline bool skb_csum_offload_chk(struct sk_buff *skb, + const struct skb_csum_offl_spec *spec, + bool *csum_encapped, + bool csum_help) +{ + if (skb->ip_summed != CHECKSUM_PARTIAL) + return false; + + return __skb_csum_offload_chk(skb, spec, csum_encapped, csum_help); +} + +static inline bool skb_csum_offload_chk_help(struct sk_buff *skb, + const struct skb_csum_offl_spec *spec) +{ + bool csum_encapped; + + return skb_csum_offload_chk(skb, spec, &csum_encapped, true); +} + +static inline bool skb_csum_off_chk_help_cmn(struct sk_buff *skb) +{ + static const struct skb_csum_offl_spec csum_offl_spec = { + .ipv4_okay = 1, + .ip_options_okay = 1, + .ipv6_okay = 1, + .vlan_okay = 1, + .tcp_okay = 1, + .udp_okay = 1, + }; + + return skb_csum_offload_chk_help(skb, &csum_offl_spec); +} + +static inline bool skb_csum_off_chk_help_cmn_v4_only(struct sk_buff *skb) +{ + static const struct skb_csum_offl_spec csum_offl_spec = { + .ipv4_okay = 1, + .ip_options_okay = 1, + .tcp_okay = 1, + .udp_okay = 1, + .vlan_okay = 1, + }; + + return skb_csum_offload_chk_help(skb, &csum_offl_spec); +} + static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, const void *daddr, const void *saddr, @@ -3711,6 +3776,19 @@ static inline bool can_checksum_protocol(netdev_features_t features, } } +/* Map an ethertype into IP protocol if possible */ +static inline int eproto_to_ipproto(int eproto) +{ + switch (eproto) { + case htons(ETH_P_IP): + return IPPROTO_IP; + case htons(ETH_P_IPV6): + return IPPROTO_IPV6; + default: + return -1; + } +} + #ifdef CONFIG_BUG void netdev_rx_csum_fault(struct net_device *dev); #else diff --git a/net/core/dev.c b/net/core/dev.c index 45b013f27625..914b4a24c654 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -138,6 +138,7 @@ #include #include #include +#include #include "net-sysfs.h" @@ -2471,6 +2472,141 @@ out: } EXPORT_SYMBOL(skb_checksum_help); +/* skb_csum_offload_check - Driver helper function to determine if a device + * with limited checksum offload capabilities is able to offload the checksum + * for a given packet. + * + * Arguments: + * skb - sk_buff for the packet in question + * spec - contains the description of what device can offload + * csum_encapped - returns true if the checksum being offloaded is + * encpasulated. That is it is checksum for the transport header + * in the inner headers. + * checksum_help - when set indicates that helper function should + * call skb_checksum_help if offload checks fail + * + * Returns: + * true: Packet has passed the checksum checks and should be offloadable to + * the device (a driver may still need to check for additional + * restrictions of its device) + * false: Checksum is not offloadable. If checksum_help was set then + * skb_checksum_help was called to resolve checksum for non-GSO + * packets and when IP protocol is not SCTP + */ +bool __skb_csum_offload_chk(struct sk_buff *skb, + const struct skb_csum_offl_spec *spec, + bool *csum_encapped, + bool csum_help) +{ + struct iphdr *iph; + struct ipv6hdr *ipv6; + void *nhdr; + int protocol; + u8 ip_proto; + + if (skb->protocol == htons(ETH_P_8021Q) || + skb->protocol == htons(ETH_P_8021AD)) { + if (!spec->vlan_okay) + goto need_help; + } + + /* We check whether the checksum refers to a transport layer checksum in + * the outermost header or an encapsulated transport layer checksum that + * corresponds to the inner headers of the skb. If the checksum is for + * something else in the packet we need help. + */ + if (skb_checksum_start_offset(skb) == skb_transport_offset(skb)) { + /* Non-encapsulated checksum */ + protocol = eproto_to_ipproto(vlan_get_protocol(skb)); + nhdr = skb_network_header(skb); + *csum_encapped = false; + if (spec->no_not_encapped) + goto need_help; + } else if (skb->encapsulation && spec->encap_okay && + skb_checksum_start_offset(skb) == + skb_inner_transport_offset(skb)) { + /* Encapsulated checksum */ + *csum_encapped = true; + switch (skb->inner_protocol_type) { + case ENCAP_TYPE_ETHER: + protocol = eproto_to_ipproto(skb->inner_protocol); + break; + case ENCAP_TYPE_IPPROTO: + protocol = skb->inner_protocol; + break; + } + nhdr = skb_inner_network_header(skb); + } else { + goto need_help; + } + + switch (protocol) { + case IPPROTO_IP: + if (!spec->ipv4_okay) + goto need_help; + iph = nhdr; + ip_proto = iph->protocol; + if (iph->ihl != 5 && !spec->ip_options_okay) + goto need_help; + break; + case IPPROTO_IPV6: + if (!spec->ipv6_okay) + goto need_help; + if (spec->no_encapped_ipv6 && *csum_encapped) + goto need_help; + ipv6 = nhdr; + nhdr += sizeof(*ipv6); + ip_proto = ipv6->nexthdr; + break; + default: + goto need_help; + } + +ip_proto_again: + switch (ip_proto) { + case IPPROTO_TCP: + if (!spec->tcp_okay || + skb->csum_offset != offsetof(struct tcphdr, check)) + goto need_help; + break; + case IPPROTO_UDP: + if (!spec->udp_okay || + skb->csum_offset != offsetof(struct udphdr, check)) + goto need_help; + break; + case IPPROTO_SCTP: + if (!spec->sctp_okay || + skb->csum_offset != offsetof(struct sctphdr, checksum)) + goto cant_help; + break; + case NEXTHDR_HOP: + case NEXTHDR_ROUTING: + case NEXTHDR_DEST: { + u8 *opthdr = nhdr; + + if (protocol != IPPROTO_IPV6 || !spec->ext_hdrs_okay) + goto need_help; + + ip_proto = opthdr[0]; + nhdr += (opthdr[1] + 1) << 3; + + goto ip_proto_again; + } + default: + goto need_help; + } + + /* Passed the tests for offloading checksum */ + return true; + +need_help: + if (csum_help && !skb_shinfo(skb)->gso_size) + skb_checksum_help(skb); +cant_help: + return false; +} +EXPORT_SYMBOL(__skb_csum_offload_chk); + __be16 skb_network_protocol(struct sk_buff *skb, int *depth) { __be16 type = skb->protocol; -- cgit v1.2.3-59-g8ed1b From 7a6ae71b2490586ed55105893a18dfc648e5fcbb Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 14 Dec 2015 11:19:47 -0800 Subject: net: Elaborate on checksum offload interface description Add specifics and details the description of the interface between the stack and drivers for doing checksum offload. This description is meant to be as specific and complete as possible. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/skbuff.h | 138 ++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 109 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2393373c9d08..6b6bd42d6134 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -39,11 +39,55 @@ #include #include -/* A. Checksumming of received packets by device. +/* The interface for checksum offload between the stack and networking drivers + * is as follows... + * + * A. IP checksum related features + * + * Drivers advertise checksum offload capabilities in the features of a device. + * From the stack's point of view these are capabilities offered by the driver, + * a driver typically only advertises features that it is capable of offloading + * to its device. + * + * The checksum related features are: + * + * NETIF_F_HW_CSUM - The driver (or its device) is able to compute one + * IP (one's complement) checksum for any combination + * of protocols or protocol layering. The checksum is + * computed and set in a packet per the CHECKSUM_PARTIAL + * interface (see below). + * + * NETIF_F_IP_CSUM - Driver (device) is only able to checksum plain + * TCP or UDP packets over IPv4. These are specifically + * unencapsulated packets of the form IPv4|TCP or + * IPv4|UDP where the Protocol field in the IPv4 header + * is TCP or UDP. The IPv4 header may contain IP options + * This feature cannot be set in features for a device + * with NETIF_F_HW_CSUM also set. This feature is being + * DEPRECATED (see below). + * + * NETIF_F_IPV6_CSUM - Driver (device) is only able to checksum plain + * TCP or UDP packets over IPv6. These are specifically + * unencapsulated packets of the form IPv6|TCP or + * IPv4|UDP where the Next Header field in the IPv6 + * header is either TCP or UDP. IPv6 extension headers + * are not supported with this feature. This feature + * cannot be set in features for a device with + * NETIF_F_HW_CSUM also set. This feature is being + * DEPRECATED (see below). + * + * NETIF_F_RXCSUM - Driver (device) performs receive checksum offload. + * This flag is used only used to disable the RX checksum + * feature for a device. The stack will accept receive + * checksum indication in packets received on a device + * regardless of whether NETIF_F_RXCSUM is set. + * + * B. Checksumming of received packets by device. Indication of checksum + * verification is in set skb->ip_summed. Possible values are: * * CHECKSUM_NONE: * - * Device failed to checksum this packet e.g. due to lack of capabilities. + * Device did not checksum this packet e.g. due to lack of capabilities. * The packet contains full (though not verified) checksum in packet but * not in skb->csum. Thus, skb->csum is undefined in this case. * @@ -53,9 +97,8 @@ * (as in CHECKSUM_COMPLETE), but it does parse headers and verify checksums * for specific protocols. For such packets it will set CHECKSUM_UNNECESSARY * if their checksums are okay. skb->csum is still undefined in this case - * though. It is a bad option, but, unfortunately, nowadays most vendors do - * this. Apparently with the secret goal to sell you new devices, when you - * will add new protocol to your host, f.e. IPv6 8) + * though. A driver or device must never modify the checksum field in the + * packet even if checksum is verified. * * CHECKSUM_UNNECESSARY is applicable to following protocols: * TCP: IPv6 and IPv4. @@ -96,40 +139,77 @@ * packet that are after the checksum being offloaded are not considered to * be verified. * - * B. Checksumming on output. - * - * CHECKSUM_NONE: - * - * The skb was already checksummed by the protocol, or a checksum is not - * required. + * C. Checksumming on transmit for non-GSO. The stack requests checksum offload + * in the skb->ip_summed for a packet. Values are: * * CHECKSUM_PARTIAL: * - * The device is required to checksum the packet as seen by hard_start_xmit() + * The driver is required to checksum the packet as seen by hard_start_xmit() * from skb->csum_start up to the end, and to record/write the checksum at - * offset skb->csum_start + skb->csum_offset. + * offset skb->csum_start + skb->csum_offset. A driver may verify that the + * csum_start and csum_offset values are valid values given the length and + * offset of the packet, however they should not attempt to validate that the + * checksum refers to a legitimate transport layer checksum-- it is the + * purview of the stack to validate that csum_start and csum_offset are set + * correctly. + * + * When the stack requests checksum offload for a packet, the driver MUST + * ensure that the checksum is set correctly. A driver can either offload the + * checksum calculation to the device, or call skb_checksum_help (in the case + * that the device does not support offload for a particular checksum). + * + * NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM are being deprecated in favor of + * NETIF_F_HW_CSUM. New devices should use NETIF_F_HW_CSUM to indicate + * checksum offload capability. If a device has limited checksum capabilities + * (for instance can only perform NETIF_F_IP_CSUM or NETIF_F_IPV6_CSUM as + * described above) a helper function can be called to resolve + * CHECKSUM_PARTIAL. The helper functions are skb_csum_off_chk*. The helper + * function takes a spec argument that describes the protocol layer that is + * supported for checksum offload and can be called for each packet. If a + * packet does not match the specification for offload, skb_checksum_help + * is called to resolve the checksum. * - * The device must show its capabilities in dev->features, set up at device - * setup time, e.g. netdev_features.h: + * CHECKSUM_NONE: * - * NETIF_F_HW_CSUM - It's a clever device, it's able to checksum everything. - * NETIF_F_IP_CSUM - Device is dumb, it's able to checksum only TCP/UDP over - * IPv4. Sigh. Vendors like this way for an unknown reason. - * Though, see comment above about CHECKSUM_UNNECESSARY. 8) - * NETIF_F_IPV6_CSUM - About as dumb as the last one but does IPv6 instead. - * NETIF_F_... - Well, you get the picture. + * The skb was already checksummed by the protocol, or a checksum is not + * required. * * CHECKSUM_UNNECESSARY: * - * Normally, the device will do per protocol specific checksumming. Protocol - * implementations that do not want the NIC to perform the checksum - * calculation should use this flag in their outgoing skbs. + * This has the same meaning on as CHECKSUM_NONE for checksum offload on + * output. * - * NETIF_F_FCOE_CRC - This indicates that the device can do FCoE FC CRC - * offload. Correspondingly, the FCoE protocol driver - * stack should use CHECKSUM_UNNECESSARY. - * - * Any questions? No questions, good. --ANK + * CHECKSUM_COMPLETE: + * Not used in checksum output. If a driver observes a packet with this value + * set in skbuff, if should treat as CHECKSUM_NONE being set. + * + * D. Non-IP checksum (CRC) offloads + * + * NETIF_F_SCTP_CRC - This feature indicates that a device is capable of + * offloading the SCTP CRC in a packet. To perform this offload the stack + * will set ip_summed to CHECKSUM_PARTIAL and set csum_start and csum_offset + * accordingly. Note the there is no indication in the skbuff that the + * CHECKSUM_PARTIAL refers to an SCTP checksum, a driver that supports + * both IP checksum offload and SCTP CRC offload must verify which offload + * is configured for a packet presumably by inspecting packet headers. + * + * NETIF_F_FCOE_CRC - This feature indicates that a device is capable of + * offloading the FCOE CRC in a packet. To perform this offload the stack + * will set ip_summed to CHECKSUM_PARTIAL and set csum_start and csum_offset + * accordingly. Note the there is no indication in the skbuff that the + * CHECKSUM_PARTIAL refers to an FCOE checksum, a driver that supports + * both IP checksum offload and FCOE CRC offload must verify which offload + * is configured for a packet presumably by inspecting packet headers. + * + * E. Checksumming on output with GSO. + * + * In the case of a GSO packet (skb_is_gso(skb) is true), checksum offload + * is implied by the SKB_GSO_* flags in gso_type. Most obviously, if the + * gso_type is SKB_GSO_TCPV4 or SKB_GSO_TCPV6, TCP checksum offload as + * part of the GSO operation is implied. If a checksum is being offloaded + * with GSO then ip_summed is CHECKSUM_PARTIAL, csum_start and csum_offset + * are set to refer to the outermost checksum being offload (two offloaded + * checksums are possible with UDP encapsulation). */ /* Don't change this without changing skb_csum_unnecessary! */ -- cgit v1.2.3-59-g8ed1b From 3502cad73c4bbf8f6365d539e814159275252c59 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 15 Dec 2015 15:41:36 -0800 Subject: rhashtable: add function to replace an element Add the rhashtable_replace_fast function. This replaces one object in the table with another atomically. The hashes of the new and old objects must be equal. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/rhashtable.h | 82 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 82 insertions(+) (limited to 'include') diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h index 843ceca9a21e..77deece15fb3 100644 --- a/include/linux/rhashtable.h +++ b/include/linux/rhashtable.h @@ -819,4 +819,86 @@ out: return err; } +/* Internal function, please use rhashtable_replace_fast() instead */ +static inline int __rhashtable_replace_fast( + struct rhashtable *ht, struct bucket_table *tbl, + struct rhash_head *obj_old, struct rhash_head *obj_new, + const struct rhashtable_params params) +{ + struct rhash_head __rcu **pprev; + struct rhash_head *he; + spinlock_t *lock; + unsigned int hash; + int err = -ENOENT; + + /* Minimally, the old and new objects must have same hash + * (which should mean identifiers are the same). + */ + hash = rht_head_hashfn(ht, tbl, obj_old, params); + if (hash != rht_head_hashfn(ht, tbl, obj_new, params)) + return -EINVAL; + + lock = rht_bucket_lock(tbl, hash); + + spin_lock_bh(lock); + + pprev = &tbl->buckets[hash]; + rht_for_each(he, tbl, hash) { + if (he != obj_old) { + pprev = &he->next; + continue; + } + + rcu_assign_pointer(obj_new->next, obj_old->next); + rcu_assign_pointer(*pprev, obj_new); + err = 0; + break; + } + + spin_unlock_bh(lock); + + return err; +} + +/** + * rhashtable_replace_fast - replace an object in hash table + * @ht: hash table + * @obj_old: pointer to hash head inside object being replaced + * @obj_new: pointer to hash head inside object which is new + * @params: hash table parameters + * + * Replacing an object doesn't affect the number of elements in the hash table + * or bucket, so we don't need to worry about shrinking or expanding the + * table here. + * + * Returns zero on success, -ENOENT if the entry could not be found, + * -EINVAL if hash is not the same for the old and new objects. + */ +static inline int rhashtable_replace_fast( + struct rhashtable *ht, struct rhash_head *obj_old, + struct rhash_head *obj_new, + const struct rhashtable_params params) +{ + struct bucket_table *tbl; + int err; + + rcu_read_lock(); + + tbl = rht_dereference_rcu(ht->tbl, ht); + + /* Because we have already taken (and released) the bucket + * lock in old_tbl, if we find that future_tbl is not yet + * visible then that guarantees the entry to still be in + * the old tbl if it exists. + */ + while ((err = __rhashtable_replace_fast(ht, tbl, obj_old, + obj_new, params)) && + (tbl = rht_dereference_rcu(tbl->future_tbl, ht))) + ; + + rcu_read_unlock(); + + return err; +} + #endif /* _LINUX_RHASHTABLE_H */ -- cgit v1.2.3-59-g8ed1b From fc9e50f5a5a4e1fa9ba2756f745a13e693cf6a06 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 15 Dec 2015 15:41:37 -0800 Subject: netlink: add a start callback for starting a netlink dump The start callback allows the caller to set up a context for the dump callbacks. Presumably, the context can then be destroyed in the done callback. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/netlink.h | 2 ++ include/net/genetlink.h | 2 ++ net/netlink/af_netlink.c | 4 ++++ net/netlink/genetlink.c | 16 ++++++++++++++++ 4 files changed, 24 insertions(+) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 639e9b8b0e4d..0b41959aab9f 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -131,6 +131,7 @@ netlink_skb_clone(struct sk_buff *skb, gfp_t gfp_mask) struct netlink_callback { struct sk_buff *skb; const struct nlmsghdr *nlh; + int (*start)(struct netlink_callback *); int (*dump)(struct sk_buff * skb, struct netlink_callback *cb); int (*done)(struct netlink_callback *cb); @@ -153,6 +154,7 @@ struct nlmsghdr * __nlmsg_put(struct sk_buff *skb, u32 portid, u32 seq, int type, int len, int flags); struct netlink_dump_control { + int (*start)(struct netlink_callback *); int (*dump)(struct sk_buff *skb, struct netlink_callback *); int (*done)(struct netlink_callback *); void *data; diff --git a/include/net/genetlink.h b/include/net/genetlink.h index 1b6b6dcb018d..43c0e771f417 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -114,6 +114,7 @@ static inline void genl_info_net_set(struct genl_info *info, struct net *net) * @flags: flags * @policy: attribute validation policy * @doit: standard command callback + * @start: start callback for dumps * @dumpit: callback for dumpers * @done: completion callback for dumps * @ops_list: operations list @@ -122,6 +123,7 @@ struct genl_ops { const struct nla_policy *policy; int (*doit)(struct sk_buff *skb, struct genl_info *info); + int (*start)(struct netlink_callback *cb); int (*dumpit)(struct sk_buff *skb, struct netlink_callback *cb); int (*done)(struct netlink_callback *cb); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 59651af8cc27..81dc1bb6e016 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2915,6 +2915,7 @@ int __netlink_dump_start(struct sock *ssk, struct sk_buff *skb, cb = &nlk->cb; memset(cb, 0, sizeof(*cb)); + cb->start = control->start; cb->dump = control->dump; cb->done = control->done; cb->nlh = nlh; @@ -2927,6 +2928,9 @@ int __netlink_dump_start(struct sock *ssk, struct sk_buff *skb, mutex_unlock(nlk->cb_mutex); + if (cb->start) + cb->start(cb); + ret = netlink_dump(sk); sock_put(sk); diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index bc0e504f33a6..8e63662c6fb0 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -513,6 +513,20 @@ void *genlmsg_put(struct sk_buff *skb, u32 portid, u32 seq, } EXPORT_SYMBOL(genlmsg_put); +static int genl_lock_start(struct netlink_callback *cb) +{ + /* our ops are always const - netlink API doesn't propagate that */ + const struct genl_ops *ops = cb->data; + int rc = 0; + + if (ops->start) { + genl_lock(); + rc = ops->start(cb); + genl_unlock(); + } + return rc; +} + static int genl_lock_dumpit(struct sk_buff *skb, struct netlink_callback *cb) { /* our ops are always const - netlink API doesn't propagate that */ @@ -577,6 +591,7 @@ static int genl_family_rcv_msg(struct genl_family *family, .module = family->module, /* we have const, but the netlink API doesn't */ .data = (void *)ops, + .start = genl_lock_start, .dump = genl_lock_dumpit, .done = genl_lock_done, }; @@ -588,6 +603,7 @@ static int genl_family_rcv_msg(struct genl_family *family, } else { struct netlink_dump_control c = { .module = family->module, + .start = ops->start, .dump = ops->dumpit, .done = ops->done, }; -- cgit v1.2.3-59-g8ed1b From 7f00feaf107645d95a6d87e99b4d141ac0a08efd Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 15 Dec 2015 15:41:38 -0800 Subject: ila: Add generic ILA translation facility This patch implements an ILA tanslation table. This table can be configured with identifier to locator mappings, and can be be queried to resolve a mapping. Queries can be parameterized based on interface, direction (incoming or outoing), and matching locator. The table is implemented using rhashtable and is configured via netlink (through "ip ila .." in iproute). The table may be used as alternative means to do do ILA tanslations other than the lw tunnels Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/net/ila.h | 18 ++ include/uapi/linux/ila.h | 22 ++ net/ipv6/ila/Makefile | 2 +- net/ipv6/ila/ila.h | 2 + net/ipv6/ila/ila_common.c | 8 + net/ipv6/ila/ila_xlat.c | 680 ++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 731 insertions(+), 1 deletion(-) create mode 100644 include/net/ila.h create mode 100644 net/ipv6/ila/ila_xlat.c (limited to 'include') diff --git a/include/net/ila.h b/include/net/ila.h new file mode 100644 index 000000000000..9f4f43e94ae4 --- /dev/null +++ b/include/net/ila.h @@ -0,0 +1,18 @@ +/* + * ILA kernel interface + * + * Copyright (c) 2015 Tom Herbert + * + * 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 _NET_ILA_H +#define _NET_ILA_H + +int ila_xlat_outgoing(struct sk_buff *skb); +int ila_xlat_incoming(struct sk_buff *skb); + +#endif /* _NET_ILA_H */ diff --git a/include/uapi/linux/ila.h b/include/uapi/linux/ila.h index 7ed9e670814e..abde7bbd6f3b 100644 --- a/include/uapi/linux/ila.h +++ b/include/uapi/linux/ila.h @@ -3,13 +3,35 @@ #ifndef _UAPI_LINUX_ILA_H #define _UAPI_LINUX_ILA_H +/* NETLINK_GENERIC related info */ +#define ILA_GENL_NAME "ila" +#define ILA_GENL_VERSION 0x1 + enum { ILA_ATTR_UNSPEC, ILA_ATTR_LOCATOR, /* u64 */ + ILA_ATTR_IDENTIFIER, /* u64 */ + ILA_ATTR_LOCATOR_MATCH, /* u64 */ + ILA_ATTR_IFINDEX, /* s32 */ + ILA_ATTR_DIR, /* u32 */ __ILA_ATTR_MAX, }; #define ILA_ATTR_MAX (__ILA_ATTR_MAX - 1) +enum { + ILA_CMD_UNSPEC, + ILA_CMD_ADD, + ILA_CMD_DEL, + ILA_CMD_GET, + + __ILA_CMD_MAX, +}; + +#define ILA_CMD_MAX (__ILA_CMD_MAX - 1) + +#define ILA_DIR_IN (1 << 0) +#define ILA_DIR_OUT (1 << 1) + #endif /* _UAPI_LINUX_ILA_H */ diff --git a/net/ipv6/ila/Makefile b/net/ipv6/ila/Makefile index 31d136be2f21..4b32e5921e5c 100644 --- a/net/ipv6/ila/Makefile +++ b/net/ipv6/ila/Makefile @@ -4,4 +4,4 @@ obj-$(CONFIG_IPV6_ILA) += ila.o -ila-objs := ila_common.o ila_lwt.o +ila-objs := ila_common.o ila_lwt.o ila_xlat.o diff --git a/net/ipv6/ila/ila.h b/net/ipv6/ila/ila.h index b94081ff2f8a..28542cb2b387 100644 --- a/net/ipv6/ila/ila.h +++ b/net/ipv6/ila/ila.h @@ -42,5 +42,7 @@ void update_ipv6_locator(struct sk_buff *skb, struct ila_params *p); int ila_lwt_init(void); void ila_lwt_fini(void); +int ila_xlat_init(void); +void ila_xlat_fini(void); #endif /* __ILA_H */ diff --git a/net/ipv6/ila/ila_common.c b/net/ipv6/ila/ila_common.c index 64e1904991ac..32dc9aab7297 100644 --- a/net/ipv6/ila/ila_common.c +++ b/net/ipv6/ila/ila_common.c @@ -80,12 +80,20 @@ static int __init ila_init(void) if (ret) goto fail_lwt; + ret = ila_xlat_init(); + if (ret) + goto fail_xlat; + + return 0; +fail_xlat: + ila_lwt_fini(); fail_lwt: return ret; } static void __exit ila_fini(void) { + ila_xlat_fini(); ila_lwt_fini(); } diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c new file mode 100644 index 000000000000..295ca29a23c3 --- /dev/null +++ b/net/ipv6/ila/ila_xlat.c @@ -0,0 +1,680 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ila.h" + +struct ila_xlat_params { + struct ila_params ip; + __be64 identifier; + int ifindex; + unsigned int dir; +}; + +struct ila_map { + struct ila_xlat_params p; + struct rhash_head node; + struct ila_map __rcu *next; + struct rcu_head rcu; +}; + +static unsigned int ila_net_id; + +struct ila_net { + struct rhashtable rhash_table; + spinlock_t *locks; /* Bucket locks for entry manipulation */ + unsigned int locks_mask; + bool hooks_registered; +}; + +#define LOCKS_PER_CPU 10 + +static int alloc_ila_locks(struct ila_net *ilan) +{ + unsigned int i, size; + unsigned int nr_pcpus = num_possible_cpus(); + + nr_pcpus = min_t(unsigned int, nr_pcpus, 32UL); + size = roundup_pow_of_two(nr_pcpus * LOCKS_PER_CPU); + + if (sizeof(spinlock_t) != 0) { +#ifdef CONFIG_NUMA + if (size * sizeof(spinlock_t) > PAGE_SIZE) + ilan->locks = vmalloc(size * sizeof(spinlock_t)); + else +#endif + ilan->locks = kmalloc_array(size, sizeof(spinlock_t), + GFP_KERNEL); + if (!ilan->locks) + return -ENOMEM; + for (i = 0; i < size; i++) + spin_lock_init(&ilan->locks[i]); + } + ilan->locks_mask = size - 1; + + return 0; +} + +static u32 hashrnd __read_mostly; +static __always_inline void __ila_hash_secret_init(void) +{ + net_get_random_once(&hashrnd, sizeof(hashrnd)); +} + +static inline u32 ila_identifier_hash(__be64 identifier) +{ + u32 *v = (u32 *)&identifier; + + return jhash_2words(v[0], v[1], hashrnd); +} + +static inline spinlock_t *ila_get_lock(struct ila_net *ilan, __be64 identifier) +{ + return &ilan->locks[ila_identifier_hash(identifier) & ilan->locks_mask]; +} + +static inline int ila_cmp_wildcards(struct ila_map *ila, __be64 loc, + int ifindex, unsigned int dir) +{ + return (ila->p.ip.locator_match && ila->p.ip.locator_match != loc) || + (ila->p.ifindex && ila->p.ifindex != ifindex) || + !(ila->p.dir & dir); +} + +static inline int ila_cmp_params(struct ila_map *ila, struct ila_xlat_params *p) +{ + return (ila->p.ip.locator_match != p->ip.locator_match) || + (ila->p.ifindex != p->ifindex) || + (ila->p.dir != p->dir); +} + +static int ila_cmpfn(struct rhashtable_compare_arg *arg, + const void *obj) +{ + const struct ila_map *ila = obj; + + return (ila->p.identifier != *(__be64 *)arg->key); +} + +static inline int ila_order(struct ila_map *ila) +{ + int score = 0; + + if (ila->p.ip.locator_match) + score += 1 << 0; + + if (ila->p.ifindex) + score += 1 << 1; + + return score; +} + +static const struct rhashtable_params rht_params = { + .nelem_hint = 1024, + .head_offset = offsetof(struct ila_map, node), + .key_offset = offsetof(struct ila_map, p.identifier), + .key_len = sizeof(u64), /* identifier */ + .max_size = 1048576, + .min_size = 256, + .automatic_shrinking = true, + .obj_cmpfn = ila_cmpfn, +}; + +static struct genl_family ila_nl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = ILA_GENL_NAME, + .version = ILA_GENL_VERSION, + .maxattr = ILA_ATTR_MAX, + .netnsok = true, + .parallel_ops = true, +}; + +static struct nla_policy ila_nl_policy[ILA_ATTR_MAX + 1] = { + [ILA_ATTR_IDENTIFIER] = { .type = NLA_U64, }, + [ILA_ATTR_LOCATOR] = { .type = NLA_U64, }, + [ILA_ATTR_LOCATOR_MATCH] = { .type = NLA_U64, }, + [ILA_ATTR_IFINDEX] = { .type = NLA_U32, }, + [ILA_ATTR_DIR] = { .type = NLA_U32, }, +}; + +static int parse_nl_config(struct genl_info *info, + struct ila_xlat_params *p) +{ + memset(p, 0, sizeof(*p)); + + if (info->attrs[ILA_ATTR_IDENTIFIER]) + p->identifier = (__force __be64)nla_get_u64( + info->attrs[ILA_ATTR_IDENTIFIER]); + + if (info->attrs[ILA_ATTR_LOCATOR]) + p->ip.locator = (__force __be64)nla_get_u64( + info->attrs[ILA_ATTR_LOCATOR]); + + if (info->attrs[ILA_ATTR_LOCATOR_MATCH]) + p->ip.locator_match = (__force __be64)nla_get_u64( + info->attrs[ILA_ATTR_LOCATOR_MATCH]); + + if (info->attrs[ILA_ATTR_IFINDEX]) + p->ifindex = nla_get_s32(info->attrs[ILA_ATTR_IFINDEX]); + + if (info->attrs[ILA_ATTR_DIR]) + p->dir = nla_get_u32(info->attrs[ILA_ATTR_DIR]); + + return 0; +} + +/* Must be called with rcu readlock */ +static inline struct ila_map *ila_lookup_wildcards(__be64 id, __be64 loc, + int ifindex, + unsigned int dir, + struct ila_net *ilan) +{ + struct ila_map *ila; + + ila = rhashtable_lookup_fast(&ilan->rhash_table, &id, rht_params); + while (ila) { + if (!ila_cmp_wildcards(ila, loc, ifindex, dir)) + return ila; + ila = rcu_access_pointer(ila->next); + } + + return NULL; +} + +/* Must be called with rcu readlock */ +static inline struct ila_map *ila_lookup_by_params(struct ila_xlat_params *p, + struct ila_net *ilan) +{ + struct ila_map *ila; + + ila = rhashtable_lookup_fast(&ilan->rhash_table, &p->identifier, + rht_params); + while (ila) { + if (!ila_cmp_params(ila, p)) + return ila; + ila = rcu_access_pointer(ila->next); + } + + return NULL; +} + +static inline void ila_release(struct ila_map *ila) +{ + kfree_rcu(ila, rcu); +} + +static void ila_free_cb(void *ptr, void *arg) +{ + struct ila_map *ila = (struct ila_map *)ptr, *next; + + /* Assume rcu_readlock held */ + while (ila) { + next = rcu_access_pointer(ila->next); + ila_release(ila); + ila = next; + } +} + +static int ila_xlat_addr(struct sk_buff *skb, int dir); + +static unsigned int +ila_nf_input(void *priv, + struct sk_buff *skb, + const struct nf_hook_state *state) +{ + ila_xlat_addr(skb, ILA_DIR_IN); + return NF_ACCEPT; +} + +static struct nf_hook_ops ila_nf_hook_ops[] __read_mostly = { + { + .hook = ila_nf_input, + .pf = NFPROTO_IPV6, + .hooknum = NF_INET_PRE_ROUTING, + .priority = -1, + }, +}; + +static int ila_add_mapping(struct net *net, struct ila_xlat_params *p) +{ + struct ila_net *ilan = net_generic(net, ila_net_id); + struct ila_map *ila, *head; + spinlock_t *lock = ila_get_lock(ilan, p->identifier); + int err = 0, order; + + if (!ilan->hooks_registered) { + /* We defer registering net hooks in the namespace until the + * first mapping is added. + */ + err = nf_register_net_hooks(net, ila_nf_hook_ops, + ARRAY_SIZE(ila_nf_hook_ops)); + if (err) + return err; + + ilan->hooks_registered = true; + } + + ila = kzalloc(sizeof(*ila), GFP_KERNEL); + if (!ila) + return -ENOMEM; + + ila->p = *p; + + if (p->ip.locator_match) { + /* Precompute checksum difference for translation since we + * know both the old identifier and the new one. + */ + ila->p.ip.csum_diff = compute_csum_diff8( + (__be32 *)&p->ip.locator_match, + (__be32 *)&p->ip.locator); + } + + order = ila_order(ila); + + spin_lock(lock); + + head = rhashtable_lookup_fast(&ilan->rhash_table, &p->identifier, + rht_params); + if (!head) { + /* New entry for the rhash_table */ + err = rhashtable_lookup_insert_fast(&ilan->rhash_table, + &ila->node, rht_params); + } else { + struct ila_map *tila = head, *prev = NULL; + + do { + if (!ila_cmp_params(tila, p)) { + err = -EEXIST; + goto out; + } + + if (order > ila_order(tila)) + break; + + prev = tila; + tila = rcu_dereference_protected(tila->next, + lockdep_is_held(lock)); + } while (tila); + + if (prev) { + /* Insert in sub list of head */ + RCU_INIT_POINTER(ila->next, tila); + rcu_assign_pointer(prev->next, ila); + } else { + /* Make this ila new head */ + RCU_INIT_POINTER(ila->next, head); + err = rhashtable_replace_fast(&ilan->rhash_table, + &head->node, + &ila->node, rht_params); + if (err) + goto out; + } + } + +out: + spin_unlock(lock); + + if (err) + kfree(ila); + + return err; +} + +static int ila_del_mapping(struct net *net, struct ila_xlat_params *p) +{ + struct ila_net *ilan = net_generic(net, ila_net_id); + struct ila_map *ila, *head, *prev; + spinlock_t *lock = ila_get_lock(ilan, p->identifier); + int err = -ENOENT; + + spin_lock(lock); + + head = rhashtable_lookup_fast(&ilan->rhash_table, + &p->identifier, rht_params); + ila = head; + + prev = NULL; + + while (ila) { + if (ila_cmp_params(ila, p)) { + prev = ila; + ila = rcu_dereference_protected(ila->next, + lockdep_is_held(lock)); + continue; + } + + err = 0; + + if (prev) { + /* Not head, just delete from list */ + rcu_assign_pointer(prev->next, ila->next); + } else { + /* It is the head. If there is something in the + * sublist we need to make a new head. + */ + head = rcu_dereference_protected(ila->next, + lockdep_is_held(lock)); + if (head) { + /* Put first entry in the sublist into the + * table + */ + err = rhashtable_replace_fast( + &ilan->rhash_table, &ila->node, + &head->node, rht_params); + if (err) + goto out; + } else { + /* Entry no longer used */ + err = rhashtable_remove_fast(&ilan->rhash_table, + &ila->node, + rht_params); + } + } + + ila_release(ila); + + break; + } + +out: + spin_unlock(lock); + + return err; +} + +static int ila_nl_cmd_add_mapping(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct ila_xlat_params p; + int err; + + err = parse_nl_config(info, &p); + if (err) + return err; + + return ila_add_mapping(net, &p); +} + +static int ila_nl_cmd_del_mapping(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct ila_xlat_params p; + int err; + + err = parse_nl_config(info, &p); + if (err) + return err; + + ila_del_mapping(net, &p); + + return 0; +} + +static int ila_fill_info(struct ila_map *ila, struct sk_buff *msg) +{ + if (nla_put_u64(msg, ILA_ATTR_IDENTIFIER, + (__force u64)ila->p.identifier) || + nla_put_u64(msg, ILA_ATTR_LOCATOR, + (__force u64)ila->p.ip.locator) || + nla_put_u64(msg, ILA_ATTR_LOCATOR_MATCH, + (__force u64)ila->p.ip.locator_match) || + nla_put_s32(msg, ILA_ATTR_IFINDEX, ila->p.ifindex) || + nla_put_u32(msg, ILA_ATTR_DIR, ila->p.dir)) + return -1; + + return 0; +} + +static int ila_dump_info(struct ila_map *ila, + u32 portid, u32 seq, u32 flags, + struct sk_buff *skb, u8 cmd) +{ + void *hdr; + + hdr = genlmsg_put(skb, portid, seq, &ila_nl_family, flags, cmd); + if (!hdr) + return -ENOMEM; + + if (ila_fill_info(ila, skb) < 0) + goto nla_put_failure; + + genlmsg_end(skb, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int ila_nl_cmd_get_mapping(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct ila_net *ilan = net_generic(net, ila_net_id); + struct sk_buff *msg; + struct ila_xlat_params p; + struct ila_map *ila; + int ret; + + ret = parse_nl_config(info, &p); + if (ret) + return ret; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + rcu_read_lock(); + + ila = ila_lookup_by_params(&p, ilan); + if (ila) { + ret = ila_dump_info(ila, + info->snd_portid, + info->snd_seq, 0, msg, + info->genlhdr->cmd); + } + + rcu_read_unlock(); + + if (ret < 0) + goto out_free; + + return genlmsg_reply(msg, info); + +out_free: + nlmsg_free(msg); + return ret; +} + +struct ila_dump_iter { + struct rhashtable_iter rhiter; +}; + +static int ila_nl_dump_start(struct netlink_callback *cb) +{ + struct net *net = sock_net(cb->skb->sk); + struct ila_net *ilan = net_generic(net, ila_net_id); + struct ila_dump_iter *iter = (struct ila_dump_iter *)cb->args; + + return rhashtable_walk_init(&ilan->rhash_table, &iter->rhiter); +} + +static int ila_nl_dump_done(struct netlink_callback *cb) +{ + struct ila_dump_iter *iter = (struct ila_dump_iter *)cb->args; + + rhashtable_walk_exit(&iter->rhiter); + + return 0; +} + +static int ila_nl_dump(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct ila_dump_iter *iter = (struct ila_dump_iter *)cb->args; + struct rhashtable_iter *rhiter = &iter->rhiter; + struct ila_map *ila; + int ret; + + ret = rhashtable_walk_start(rhiter); + if (ret && ret != -EAGAIN) + goto done; + + for (;;) { + ila = rhashtable_walk_next(rhiter); + + if (IS_ERR(ila)) { + if (PTR_ERR(ila) == -EAGAIN) + continue; + ret = PTR_ERR(ila); + goto done; + } else if (!ila) { + break; + } + + while (ila) { + ret = ila_dump_info(ila, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + skb, ILA_CMD_GET); + if (ret) + goto done; + + ila = rcu_access_pointer(ila->next); + } + } + + ret = skb->len; + +done: + rhashtable_walk_stop(rhiter); + return ret; +} + +static const struct genl_ops ila_nl_ops[] = { + { + .cmd = ILA_CMD_ADD, + .doit = ila_nl_cmd_add_mapping, + .policy = ila_nl_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = ILA_CMD_DEL, + .doit = ila_nl_cmd_del_mapping, + .policy = ila_nl_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = ILA_CMD_GET, + .doit = ila_nl_cmd_get_mapping, + .start = ila_nl_dump_start, + .dumpit = ila_nl_dump, + .done = ila_nl_dump_done, + .policy = ila_nl_policy, + }, +}; + +#define ILA_HASH_TABLE_SIZE 1024 + +static __net_init int ila_init_net(struct net *net) +{ + int err; + struct ila_net *ilan = net_generic(net, ila_net_id); + + err = alloc_ila_locks(ilan); + if (err) + return err; + + rhashtable_init(&ilan->rhash_table, &rht_params); + + return 0; +} + +static __net_exit void ila_exit_net(struct net *net) +{ + struct ila_net *ilan = net_generic(net, ila_net_id); + + rhashtable_free_and_destroy(&ilan->rhash_table, ila_free_cb, NULL); + + kvfree(ilan->locks); + + if (ilan->hooks_registered) + nf_unregister_net_hooks(net, ila_nf_hook_ops, + ARRAY_SIZE(ila_nf_hook_ops)); +} + +static struct pernet_operations ila_net_ops = { + .init = ila_init_net, + .exit = ila_exit_net, + .id = &ila_net_id, + .size = sizeof(struct ila_net), +}; + +static int ila_xlat_addr(struct sk_buff *skb, int dir) +{ + struct ila_map *ila; + struct ipv6hdr *ip6h = ipv6_hdr(skb); + struct net *net = dev_net(skb->dev); + struct ila_net *ilan = net_generic(net, ila_net_id); + __be64 identifier, locator_match; + size_t nhoff; + + /* Assumes skb contains a valid IPv6 header that is pulled */ + + identifier = *(__be64 *)&ip6h->daddr.in6_u.u6_addr8[8]; + locator_match = *(__be64 *)&ip6h->daddr.in6_u.u6_addr8[0]; + nhoff = sizeof(struct ipv6hdr); + + rcu_read_lock(); + + ila = ila_lookup_wildcards(identifier, locator_match, + skb->dev->ifindex, dir, ilan); + if (ila) + update_ipv6_locator(skb, &ila->p.ip); + + rcu_read_unlock(); + + return 0; +} + +int ila_xlat_incoming(struct sk_buff *skb) +{ + return ila_xlat_addr(skb, ILA_DIR_IN); +} +EXPORT_SYMBOL(ila_xlat_incoming); + +int ila_xlat_outgoing(struct sk_buff *skb) +{ + return ila_xlat_addr(skb, ILA_DIR_OUT); +} +EXPORT_SYMBOL(ila_xlat_outgoing); + +int ila_xlat_init(void) +{ + int ret; + + ret = register_pernet_device(&ila_net_ops); + if (ret) + goto exit; + + ret = genl_register_family_with_ops(&ila_nl_family, + ila_nl_ops); + if (ret < 0) + goto unregister; + + return 0; + +unregister: + unregister_pernet_device(&ila_net_ops); +exit: + return ret; +} + +void ila_xlat_fini(void) +{ + genl_unregister_family(&ila_nl_family); + unregister_pernet_device(&ila_net_ops); +} -- cgit v1.2.3-59-g8ed1b From b613f56ec9baf30edf5d9d607b822532a273dad7 Mon Sep 17 00:00:00 2001 From: Lorenzo Colitti Date: Wed, 16 Dec 2015 12:30:02 +0900 Subject: net: diag: split inet_diag_dump_one_icsk into two Currently, inet_diag_dump_one_icsk finds a socket and then dumps its information to userspace. Split it into a part that finds the socket and a part that dumps the information. Signed-off-by: Lorenzo Colitti Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/inet_diag.h | 5 +++++ net/ipv4/inet_diag.c | 42 +++++++++++++++++++++++++++--------------- 2 files changed, 32 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h index 0e707f0c1a3e..e7032f041982 100644 --- a/include/linux/inet_diag.h +++ b/include/linux/inet_diag.h @@ -3,6 +3,7 @@ #include +struct net; struct sock; struct inet_hashinfo; struct nlattr; @@ -41,6 +42,10 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *in_skb, const struct nlmsghdr *nlh, const struct inet_diag_req_v2 *req); +struct sock *inet_diag_find_one_icsk(struct net *net, + struct inet_hashinfo *hashinfo, + const struct inet_diag_req_v2 *req); + int inet_diag_bc_sk(const struct nlattr *_bc, struct sock *sk); extern int inet_diag_register(const struct inet_diag_handler *handler); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index ab9f8a66615d..cfabb8f8f0a0 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -350,17 +350,12 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, nlmsg_flags, unlh); } -int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, - struct sk_buff *in_skb, - const struct nlmsghdr *nlh, - const struct inet_diag_req_v2 *req) +struct sock *inet_diag_find_one_icsk(struct net *net, + struct inet_hashinfo *hashinfo, + const struct inet_diag_req_v2 *req) { - struct net *net = sock_net(in_skb->sk); - struct sk_buff *rep; struct sock *sk; - int err; - err = -EINVAL; if (req->sdiag_family == AF_INET) sk = inet_lookup(net, hashinfo, req->id.idiag_dst[0], req->id.idiag_dport, req->id.idiag_src[0], @@ -375,15 +370,33 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, req->id.idiag_if); #endif else - goto out_nosk; + return ERR_PTR(-EINVAL); - err = -ENOENT; if (!sk) - goto out_nosk; + return ERR_PTR(-ENOENT); - err = sock_diag_check_cookie(sk, req->id.idiag_cookie); - if (err) - goto out; + if (sock_diag_check_cookie(sk, req->id.idiag_cookie)) { + sock_gen_put(sk); + return ERR_PTR(-ENOENT); + } + + return sk; +} +EXPORT_SYMBOL_GPL(inet_diag_find_one_icsk); + +int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, + struct sk_buff *in_skb, + const struct nlmsghdr *nlh, + const struct inet_diag_req_v2 *req) +{ + struct net *net = sock_net(in_skb->sk); + struct sk_buff *rep; + struct sock *sk; + int err; + + sk = inet_diag_find_one_icsk(net, hashinfo, req); + if (IS_ERR(sk)) + return PTR_ERR(sk); rep = nlmsg_new(inet_sk_attr_size(), GFP_KERNEL); if (!rep) { @@ -409,7 +422,6 @@ out: if (sk) sock_gen_put(sk); -out_nosk: return err; } EXPORT_SYMBOL_GPL(inet_diag_dump_one_icsk); -- cgit v1.2.3-59-g8ed1b From 64be0aed59ad519d6f2160868734f7e278290ac1 Mon Sep 17 00:00:00 2001 From: Lorenzo Colitti Date: Wed, 16 Dec 2015 12:30:03 +0900 Subject: net: diag: Add the ability to destroy a socket. This patch adds a SOCK_DESTROY operation, a destroy function pointer to sock_diag_handler, and a diag_destroy function pointer. It does not include any implementation code. Signed-off-by: Lorenzo Colitti Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/sock_diag.h | 2 ++ include/net/sock.h | 1 + include/uapi/linux/sock_diag.h | 1 + net/core/sock_diag.c | 23 ++++++++++++++++++++--- 4 files changed, 24 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/sock_diag.h b/include/linux/sock_diag.h index fddebc617469..4018b48f2b3b 100644 --- a/include/linux/sock_diag.h +++ b/include/linux/sock_diag.h @@ -15,6 +15,7 @@ struct sock_diag_handler { __u8 family; int (*dump)(struct sk_buff *skb, struct nlmsghdr *nlh); int (*get_info)(struct sk_buff *skb, struct sock *sk); + int (*destroy)(struct sk_buff *skb, struct nlmsghdr *nlh); }; int sock_diag_register(const struct sock_diag_handler *h); @@ -68,4 +69,5 @@ bool sock_diag_has_destroy_listeners(const struct sock *sk) } void sock_diag_broadcast_destroy(struct sock *sk); +int sock_diag_destroy(struct sock *sk, int err); #endif diff --git a/include/net/sock.h b/include/net/sock.h index ab0269f4b2cc..6e6e8a25d997 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1060,6 +1060,7 @@ struct proto { void (*destroy_cgroup)(struct mem_cgroup *memcg); struct cg_proto *(*proto_cgroup)(struct mem_cgroup *memcg); #endif + int (*diag_destroy)(struct sock *sk, int err); }; int proto_register(struct proto *prot, int alloc_slab); diff --git a/include/uapi/linux/sock_diag.h b/include/uapi/linux/sock_diag.h index 49230d36f9ce..bae2d80034d4 100644 --- a/include/uapi/linux/sock_diag.h +++ b/include/uapi/linux/sock_diag.h @@ -4,6 +4,7 @@ #include #define SOCK_DIAG_BY_FAMILY 20 +#define SOCK_DESTROY 21 struct sock_diag_req { __u8 sdiag_family; diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index 0c1d58d43f67..a996ce8c8fb2 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -214,7 +214,7 @@ void sock_diag_unregister(const struct sock_diag_handler *hnld) } EXPORT_SYMBOL_GPL(sock_diag_unregister); -static int __sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) +static int __sock_diag_cmd(struct sk_buff *skb, struct nlmsghdr *nlh) { int err; struct sock_diag_req *req = nlmsg_data(nlh); @@ -234,8 +234,12 @@ static int __sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) hndl = sock_diag_handlers[req->sdiag_family]; if (hndl == NULL) err = -ENOENT; - else + else if (nlh->nlmsg_type == SOCK_DIAG_BY_FAMILY) err = hndl->dump(skb, nlh); + else if (nlh->nlmsg_type == SOCK_DESTROY && hndl->destroy) + err = hndl->destroy(skb, nlh); + else + err = -EOPNOTSUPP; mutex_unlock(&sock_diag_table_mutex); return err; @@ -261,7 +265,8 @@ static int sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return ret; case SOCK_DIAG_BY_FAMILY: - return __sock_diag_rcv_msg(skb, nlh); + case SOCK_DESTROY: + return __sock_diag_cmd(skb, nlh); default: return -EINVAL; } @@ -295,6 +300,18 @@ static int sock_diag_bind(struct net *net, int group) return 0; } +int sock_diag_destroy(struct sock *sk, int err) +{ + if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + if (!sk->sk_prot->diag_destroy) + return -EOPNOTSUPP; + + return sk->sk_prot->diag_destroy(sk, err); +} +EXPORT_SYMBOL_GPL(sock_diag_destroy); + static int __net_init diag_net_init(struct net *net) { struct netlink_kernel_cfg cfg = { -- cgit v1.2.3-59-g8ed1b From 6eb5d2e08f071c05ecbe135369c9ad418826cab2 Mon Sep 17 00:00:00 2001 From: Lorenzo Colitti Date: Wed, 16 Dec 2015 12:30:04 +0900 Subject: net: diag: Support SOCK_DESTROY for inet sockets. This passes the SOCK_DESTROY operation to the underlying protocol diag handler, or returns -EOPNOTSUPP if that handler does not define a destroy operation. Most of this patch is just renaming functions. This is not strictly necessary, but it would be fairly counterintuitive to have the code to destroy inet sockets be in a function whose name starts with inet_diag_get. Signed-off-by: Lorenzo Colitti Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/inet_diag.h | 4 ++++ net/ipv4/inet_diag.c | 23 +++++++++++++++-------- 2 files changed, 19 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h index e7032f041982..7c27fa1030e8 100644 --- a/include/linux/inet_diag.h +++ b/include/linux/inet_diag.h @@ -24,6 +24,10 @@ struct inet_diag_handler { void (*idiag_get_info)(struct sock *sk, struct inet_diag_msg *r, void *info); + + int (*destroy)(struct sk_buff *in_skb, + const struct inet_diag_req_v2 *req); + __u16 idiag_type; __u16 idiag_info_size; }; diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index cfabb8f8f0a0..8bb8e7ad8548 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -426,7 +426,7 @@ out: } EXPORT_SYMBOL_GPL(inet_diag_dump_one_icsk); -static int inet_diag_get_exact(struct sk_buff *in_skb, +static int inet_diag_cmd_exact(int cmd, struct sk_buff *in_skb, const struct nlmsghdr *nlh, const struct inet_diag_req_v2 *req) { @@ -436,8 +436,12 @@ static int inet_diag_get_exact(struct sk_buff *in_skb, handler = inet_diag_lock_handler(req->sdiag_protocol); if (IS_ERR(handler)) err = PTR_ERR(handler); - else + else if (cmd == SOCK_DIAG_BY_FAMILY) err = handler->dump_one(in_skb, nlh, req); + else if (cmd == SOCK_DESTROY && handler->destroy) + err = handler->destroy(in_skb, req); + else + err = -EOPNOTSUPP; inet_diag_unlock_handler(handler); return err; @@ -950,7 +954,7 @@ static int inet_diag_get_exact_compat(struct sk_buff *in_skb, req.idiag_states = rc->idiag_states; req.id = rc->id; - return inet_diag_get_exact(in_skb, nlh, &req); + return inet_diag_cmd_exact(SOCK_DIAG_BY_FAMILY, in_skb, nlh, &req); } static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh) @@ -984,7 +988,7 @@ static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh) return inet_diag_get_exact_compat(skb, nlh); } -static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h) +static int inet_diag_handler_cmd(struct sk_buff *skb, struct nlmsghdr *h) { int hdrlen = sizeof(struct inet_diag_req_v2); struct net *net = sock_net(skb->sk); @@ -992,7 +996,8 @@ static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h) if (nlmsg_len(h) < hdrlen) return -EINVAL; - if (h->nlmsg_flags & NLM_F_DUMP) { + if (h->nlmsg_type == SOCK_DIAG_BY_FAMILY && + h->nlmsg_flags & NLM_F_DUMP) { if (nlmsg_attrlen(h, hdrlen)) { struct nlattr *attr; @@ -1011,7 +1016,7 @@ static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h) } } - return inet_diag_get_exact(skb, h, nlmsg_data(h)); + return inet_diag_cmd_exact(h->nlmsg_type, skb, h, nlmsg_data(h)); } static @@ -1062,14 +1067,16 @@ int inet_diag_handler_get_info(struct sk_buff *skb, struct sock *sk) static const struct sock_diag_handler inet_diag_handler = { .family = AF_INET, - .dump = inet_diag_handler_dump, + .dump = inet_diag_handler_cmd, .get_info = inet_diag_handler_get_info, + .destroy = inet_diag_handler_cmd, }; static const struct sock_diag_handler inet6_diag_handler = { .family = AF_INET6, - .dump = inet_diag_handler_dump, + .dump = inet_diag_handler_cmd, .get_info = inet_diag_handler_get_info, + .destroy = inet_diag_handler_cmd, }; int inet_diag_register(const struct inet_diag_handler *h) -- cgit v1.2.3-59-g8ed1b From c1e64e298b8cad309091b95d8436a0255c84f54a Mon Sep 17 00:00:00 2001 From: Lorenzo Colitti Date: Wed, 16 Dec 2015 12:30:05 +0900 Subject: net: diag: Support destroying TCP sockets. This implements SOCK_DESTROY for TCP sockets. It causes all blocking calls on the socket to fail fast with ECONNABORTED and causes a protocol close of the socket. It informs the other end of the connection by sending a RST, i.e., initiating a TCP ABORT as per RFC 793. ECONNABORTED was chosen for consistency with FreeBSD. Signed-off-by: Lorenzo Colitti Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 2 ++ net/ipv4/Kconfig | 13 +++++++++++++ net/ipv4/tcp.c | 32 ++++++++++++++++++++++++++++++++ net/ipv4/tcp_diag.c | 19 +++++++++++++++++++ net/ipv4/tcp_ipv4.c | 1 + net/ipv6/tcp_ipv6.c | 1 + 6 files changed, 68 insertions(+) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index f80e74c5ad18..3077735b348d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1170,6 +1170,8 @@ void tcp_set_state(struct sock *sk, int state); void tcp_done(struct sock *sk); +int tcp_abort(struct sock *sk, int err); + static inline void tcp_sack_reset(struct tcp_options_received *rx_opt) { rx_opt->dsack = 0; diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 416dfa004cfb..c22920525e5d 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -436,6 +436,19 @@ config INET_UDP_DIAG Support for UDP socket monitoring interface used by the ss tool. If unsure, say Y. +config INET_DIAG_DESTROY + bool "INET: allow privileged process to administratively close sockets" + depends on INET_DIAG + default n + ---help--- + Provides a SOCK_DESTROY operation that allows privileged processes + (e.g., a connection manager or a network administration tool such as + ss) to close sockets opened by other processes. Closing a socket in + this way interrupts any blocking read/write/connect operations on + the socket and causes future socket calls to behave as if the socket + had been disconnected. + If unsure, say N. + menuconfig TCP_CONG_ADVANCED bool "TCP: advanced congestion control" ---help--- diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 92b3e61b847d..2c0e340518d2 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3080,6 +3080,38 @@ void tcp_done(struct sock *sk) } EXPORT_SYMBOL_GPL(tcp_done); +int tcp_abort(struct sock *sk, int err) +{ + if (!sk_fullsock(sk)) { + sock_gen_put(sk); + return -EOPNOTSUPP; + } + + /* Don't race with userspace socket closes such as tcp_close. */ + lock_sock(sk); + + /* Don't race with BH socket closes such as inet_csk_listen_stop. */ + local_bh_disable(); + bh_lock_sock(sk); + + if (!sock_flag(sk, SOCK_DEAD)) { + sk->sk_err = err; + /* This barrier is coupled with smp_rmb() in tcp_poll() */ + smp_wmb(); + sk->sk_error_report(sk); + if (tcp_need_reset(sk->sk_state)) + tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_done(sk); + } + + bh_unlock_sock(sk); + local_bh_enable(); + release_sock(sk); + sock_put(sk); + return 0; +} +EXPORT_SYMBOL_GPL(tcp_abort); + extern struct tcp_congestion_ops tcp_reno; static __initdata unsigned long thash_entries; diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index b31604086edd..4d610934fb39 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -10,6 +10,8 @@ */ #include +#include +#include #include #include @@ -46,12 +48,29 @@ static int tcp_diag_dump_one(struct sk_buff *in_skb, const struct nlmsghdr *nlh, return inet_diag_dump_one_icsk(&tcp_hashinfo, in_skb, nlh, req); } +#ifdef CONFIG_INET_DIAG_DESTROY +static int tcp_diag_destroy(struct sk_buff *in_skb, + const struct inet_diag_req_v2 *req) +{ + struct net *net = sock_net(in_skb->sk); + struct sock *sk = inet_diag_find_one_icsk(net, &tcp_hashinfo, req); + + if (IS_ERR(sk)) + return PTR_ERR(sk); + + return sock_diag_destroy(sk, ECONNABORTED); +} +#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), +#ifdef CONFIG_INET_DIAG_DESTROY + .destroy = tcp_diag_destroy, +#endif }; static int __init tcp_diag_init(void) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index db003438aaf5..7aa13bd3de29 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2342,6 +2342,7 @@ struct proto tcp_prot = { .destroy_cgroup = tcp_destroy_cgroup, .proto_cgroup = tcp_proto_cgroup, #endif + .diag_destroy = tcp_abort, }; EXPORT_SYMBOL(tcp_prot); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index c16e3fbf6854..5382c2662fa2 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1890,6 +1890,7 @@ struct proto tcpv6_prot = { .proto_cgroup = tcp_proto_cgroup, #endif .clear_sk = tcp_v6_clear_sk, + .diag_destroy = tcp_abort, }; static const struct inet6_protocol tcpv6_protocol = { -- cgit v1.2.3-59-g8ed1b From 541c9a84cd85203244307d9ebb821102eed82789 Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Wed, 9 Dec 2015 23:36:51 +0100 Subject: ssb: pick SoC invariants code from MIPS BCM47xx arch MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is code in ssb fetching "invariants" that is basically a set of board specific data. Every host requires its own implementation of reading function. In ssb we have support for PCI, PCMCIA & SDIO. For some (historical?) reason code reading "invariants" for SoC was placed in arch code and provided by a callback. This is not needed nowadays, so lets move that into ssb. This way we keep all "invariants" functions in a single module making code cleaner. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo --- arch/mips/bcm47xx/setup.c | 39 +-------------------------------------- drivers/ssb/Kconfig | 2 +- drivers/ssb/host_soc.c | 37 +++++++++++++++++++++++++++++++++++++ drivers/ssb/main.c | 5 ++--- drivers/ssb/ssb_private.h | 3 +++ include/linux/ssb/ssb.h | 10 +++------- 6 files changed, 47 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/arch/mips/bcm47xx/setup.c b/arch/mips/bcm47xx/setup.c index 6d38948f0f1e..c807e32d6d81 100644 --- a/arch/mips/bcm47xx/setup.c +++ b/arch/mips/bcm47xx/setup.c @@ -101,50 +101,13 @@ static void bcm47xx_machine_halt(void) } #ifdef CONFIG_BCM47XX_SSB -static int bcm47xx_get_invariants(struct ssb_bus *bus, - struct ssb_init_invariants *iv) -{ - char buf[20]; - int len, err; - - /* Fill boardinfo structure */ - memset(&iv->boardinfo, 0 , sizeof(struct ssb_boardinfo)); - - len = bcm47xx_nvram_getenv("boardvendor", buf, sizeof(buf)); - if (len > 0) { - err = kstrtou16(strim(buf), 0, &iv->boardinfo.vendor); - if (err) - pr_warn("Couldn't parse nvram board vendor entry with value \"%s\"\n", - buf); - } - if (!iv->boardinfo.vendor) - iv->boardinfo.vendor = SSB_BOARDVENDOR_BCM; - - len = bcm47xx_nvram_getenv("boardtype", buf, sizeof(buf)); - if (len > 0) { - err = kstrtou16(strim(buf), 0, &iv->boardinfo.type); - if (err) - pr_warn("Couldn't parse nvram board type entry with value \"%s\"\n", - buf); - } - - memset(&iv->sprom, 0, sizeof(struct ssb_sprom)); - bcm47xx_fill_sprom(&iv->sprom, NULL, false); - - if (bcm47xx_nvram_getenv("cardbus", buf, sizeof(buf)) >= 0) - iv->has_cardbus_slot = !!simple_strtoul(buf, NULL, 10); - - return 0; -} - static void __init bcm47xx_register_ssb(void) { int err; char buf[100]; struct ssb_mipscore *mcore; - err = ssb_bus_ssbbus_register(&bcm47xx_bus.ssb, SSB_ENUM_BASE, - bcm47xx_get_invariants); + err = ssb_bus_host_soc_register(&bcm47xx_bus.ssb, SSB_ENUM_BASE); if (err) panic("Failed to initialize SSB bus (err %d)", err); diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index 149214beeda9..0c675861623f 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -82,7 +82,7 @@ config SSB_SDIOHOST config SSB_HOST_SOC bool "Support for SSB bus on SoC" - depends on SSB + depends on SSB && BCM47XX_NVRAM help Host interface for a SSB directly mapped into memory. This is for some Broadcom SoCs from the BCM47xx and BCM53xx lines. diff --git a/drivers/ssb/host_soc.c b/drivers/ssb/host_soc.c index c809f255af34..d62992dc08b2 100644 --- a/drivers/ssb/host_soc.c +++ b/drivers/ssb/host_soc.c @@ -8,6 +8,7 @@ * Licensed under the GNU/GPL. See COPYING for details. */ +#include #include #include "ssb_private.h" @@ -171,3 +172,39 @@ const struct ssb_bus_ops ssb_host_soc_ops = { .block_write = ssb_host_soc_block_write, #endif }; + +int ssb_host_soc_get_invariants(struct ssb_bus *bus, + struct ssb_init_invariants *iv) +{ + char buf[20]; + int len, err; + + /* Fill boardinfo structure */ + memset(&iv->boardinfo, 0, sizeof(struct ssb_boardinfo)); + + len = bcm47xx_nvram_getenv("boardvendor", buf, sizeof(buf)); + if (len > 0) { + err = kstrtou16(strim(buf), 0, &iv->boardinfo.vendor); + if (err) + pr_warn("Couldn't parse nvram board vendor entry with value \"%s\"\n", + buf); + } + if (!iv->boardinfo.vendor) + iv->boardinfo.vendor = SSB_BOARDVENDOR_BCM; + + len = bcm47xx_nvram_getenv("boardtype", buf, sizeof(buf)); + if (len > 0) { + err = kstrtou16(strim(buf), 0, &iv->boardinfo.type); + if (err) + pr_warn("Couldn't parse nvram board type entry with value \"%s\"\n", + buf); + } + + memset(&iv->sprom, 0, sizeof(struct ssb_sprom)); + ssb_fill_sprom_with_fallback(bus, &iv->sprom); + + if (bcm47xx_nvram_getenv("cardbus", buf, sizeof(buf)) >= 0) + iv->has_cardbus_slot = !!simple_strtoul(buf, NULL, 10); + + return 0; +} diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index 5d1e9a0fc389..cde5ff7529eb 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -762,15 +762,14 @@ EXPORT_SYMBOL(ssb_bus_sdiobus_register); #endif /* CONFIG_SSB_PCMCIAHOST */ #ifdef CONFIG_SSB_HOST_SOC -int ssb_bus_ssbbus_register(struct ssb_bus *bus, unsigned long baseaddr, - ssb_invariants_func_t get_invariants) +int ssb_bus_host_soc_register(struct ssb_bus *bus, unsigned long baseaddr) { int err; bus->bustype = SSB_BUSTYPE_SSB; bus->ops = &ssb_host_soc_ops; - err = ssb_bus_register(bus, get_invariants, baseaddr); + err = ssb_bus_register(bus, ssb_host_soc_get_invariants, baseaddr); if (!err) { ssb_info("Sonics Silicon Backplane found at address 0x%08lX\n", baseaddr); diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h index 15bfd5c7d2d7..c2f5d3969c8b 100644 --- a/drivers/ssb/ssb_private.h +++ b/drivers/ssb/ssb_private.h @@ -163,6 +163,9 @@ static inline int ssb_sdio_init(struct ssb_bus *bus) #ifdef CONFIG_SSB_HOST_SOC extern const struct ssb_bus_ops ssb_host_soc_ops; + +extern int ssb_host_soc_get_invariants(struct ssb_bus *bus, + struct ssb_init_invariants *iv); #endif /* scan.c */ diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index c3d1a525bacc..26a0b3c3ce5f 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -524,13 +524,9 @@ struct ssb_init_invariants { typedef int (*ssb_invariants_func_t)(struct ssb_bus *bus, struct ssb_init_invariants *iv); -/* Register a SSB system bus. get_invariants() is called after the - * basic system devices are initialized. - * The invariants are usually fetched from some NVRAM. - * Put the invariants into the struct pointed to by iv. */ -extern int ssb_bus_ssbbus_register(struct ssb_bus *bus, - unsigned long baseaddr, - ssb_invariants_func_t get_invariants); +/* Register SoC bus. */ +extern int ssb_bus_host_soc_register(struct ssb_bus *bus, + unsigned long baseaddr); #ifdef CONFIG_SSB_PCIHOST extern int ssb_bus_pcibus_register(struct ssb_bus *bus, struct pci_dev *host_pci); -- cgit v1.2.3-59-g8ed1b From 566178f853c1aa57be9c16007c7cca07df5d51b6 Mon Sep 17 00:00:00 2001 From: Zhu Yanjun Date: Wed, 16 Dec 2015 13:55:04 +0800 Subject: net: sctp: dynamically enable or disable pf state As we all know, the value of pf_retrans >= max_retrans_path can disable pf state. The variables of pf_retrans and max_retrans_path can be changed by the userspace application. Sometimes the user expects to disable pf state while the 2 variables are changed to enable pf state. So it is necessary to introduce a new variable to disable pf state. According to the suggestions from Vlad Yasevich, extra1 and extra2 are removed. The initialization of pf_enable is added. Acked-by: Vlad Yasevich Signed-off-by: Zhu Yanjun Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 23 ++++++++++++++++++++++- include/net/netns/sctp.h | 7 +++++++ net/sctp/protocol.c | 3 +++ net/sctp/sm_sideeffect.c | 5 ++++- net/sctp/sysctl.c | 7 +++++++ 5 files changed, 43 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 2ea4c45cf1c8..5de632ed0ec0 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -1723,6 +1723,25 @@ addip_enable - BOOLEAN Default: 0 +pf_enable - INTEGER + Enable or disable pf (pf is short for potentially failed) state. A value + of pf_retrans > path_max_retrans also disables pf state. That is, one of + both pf_enable and pf_retrans > path_max_retrans can disable pf state. + Since pf_retrans and path_max_retrans can be changed by userspace + application, sometimes user expects to disable pf state by the value of + pf_retrans > path_max_retrans, but occasionally the value of pf_retrans + or path_max_retrans is changed by the user application, this pf state is + enabled. As such, it is necessary to add this to dynamically enable + and disable pf state. See: + https://datatracker.ietf.org/doc/draft-ietf-tsvwg-sctp-failover for + details. + + 1: Enable pf. + + 0: Disable pf. + + Default: 1 + addip_noauth_enable - BOOLEAN Dynamic Address Reconfiguration (ADD-IP) requires the use of authentication to protect the operations of adding or removing new @@ -1799,7 +1818,9 @@ pf_retrans - INTEGER having to reduce path_max_retrans to a very low value. See: http://www.ietf.org/id/draft-nishida-tsvwg-sctp-failover-05.txt for details. Note also that a value of pf_retrans > path_max_retrans - disables this feature + disables this feature. Since both pf_retrans and path_max_retrans can + be changed by userspace application, a variable pf_enable is used to + disable pf state. Default: 0 diff --git a/include/net/netns/sctp.h b/include/net/netns/sctp.h index 8ba379f9e467..c501d67172b1 100644 --- a/include/net/netns/sctp.h +++ b/include/net/netns/sctp.h @@ -88,6 +88,13 @@ struct netns_sctp { */ int pf_retrans; + /* + * Disable Potentially-Failed feature, the feature is enabled by default + * pf_enable - 0 : disable pf + * - >0 : enable pf + */ + int pf_enable; + /* * Policy for preforming sctp/socket accounting * 0 - do socket level accounting, all assocs share sk_sndbuf diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 6c2c0accc6a0..010aced44b6b 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1223,6 +1223,9 @@ static int __net_init sctp_defaults_init(struct net *net) /* Max.Burst - 4 */ net->sctp.max_burst = SCTP_DEFAULT_MAX_BURST; + /* Enable pf state by default */ + net->sctp.pf_enable = 1; + /* Association.Max.Retrans - 10 attempts * Path.Max.Retrans - 5 attempts (per destination address) * Max.Init.Retransmits - 8 attempts diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 6098d4c42fa9..05cd16400e0b 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -477,6 +477,8 @@ static void sctp_do_8_2_transport_strike(sctp_cmd_seq_t *commands, struct sctp_transport *transport, int is_hb) { + struct net *net = sock_net(asoc->base.sk); + /* The check for association's overall error counter exceeding the * threshold is done in the state function. */ @@ -503,7 +505,8 @@ static void sctp_do_8_2_transport_strike(sctp_cmd_seq_t *commands, * is SCTP_ACTIVE, then mark this transport as Partially Failed, * see SCTP Quick Failover Draft, section 5.1 */ - if ((transport->state == SCTP_ACTIVE) && + if (net->sctp.pf_enable && + (transport->state == SCTP_ACTIVE) && (asoc->pf_retrans < transport->pathmaxrxt) && (transport->error_count > asoc->pf_retrans)) { diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 26d50c565f54..ccbfc93fb8fe 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -308,6 +308,13 @@ static struct ctl_table sctp_net_table[] = { .extra1 = &max_autoclose_min, .extra2 = &max_autoclose_max, }, + { + .procname = "pf_enable", + .data = &init_net.sctp.pf_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { /* sentinel */ } }; -- cgit v1.2.3-59-g8ed1b From a8170d2b9e8d38a1f3fa3b40b6f8cd34a87d5382 Mon Sep 17 00:00:00 2001 From: "Singhai, Anjali" Date: Mon, 14 Dec 2015 12:21:17 -0800 Subject: geneve: Add geneve udp port offload for ethernet devices Add ndo_ops to add/del UDP ports to a device that supports geneve offload. v2: Comment fix. Signed-off-by: Anjali Singhai Jain Signed-off-by: Kiran Patil Signed-off-by: David S. Miller --- drivers/net/geneve.c | 23 +++++++++++++++++++++++ include/linux/netdevice.h | 20 +++++++++++++++++++- 2 files changed, 42 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 0750d7a93878..89325e483ecf 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -380,8 +380,11 @@ static struct socket *geneve_create_sock(struct net *net, bool ipv6, static void geneve_notify_add_rx_port(struct geneve_sock *gs) { + struct net_device *dev; struct sock *sk = gs->sock->sk; + struct net *net = sock_net(sk); sa_family_t sa_family = sk->sk_family; + __be16 port = inet_sk(sk)->inet_sport; int err; if (sa_family == AF_INET) { @@ -390,6 +393,14 @@ static void geneve_notify_add_rx_port(struct geneve_sock *gs) pr_warn("geneve: udp_add_offload failed with status %d\n", err); } + + rcu_read_lock(); + for_each_netdev_rcu(net, dev) { + if (dev->netdev_ops->ndo_add_geneve_port) + dev->netdev_ops->ndo_add_geneve_port(dev, sa_family, + port); + } + rcu_read_unlock(); } static int geneve_hlen(struct genevehdr *gh) @@ -530,8 +541,20 @@ static struct geneve_sock *geneve_socket_create(struct net *net, __be16 port, static void geneve_notify_del_rx_port(struct geneve_sock *gs) { + struct net_device *dev; struct sock *sk = gs->sock->sk; + struct net *net = sock_net(sk); sa_family_t sa_family = sk->sk_family; + __be16 port = inet_sk(sk)->inet_sport; + + rcu_read_lock(); + for_each_netdev_rcu(net, dev) { + if (dev->netdev_ops->ndo_del_geneve_port) + dev->netdev_ops->ndo_del_geneve_port(dev, sa_family, + port); + } + + rcu_read_unlock(); if (sa_family == AF_INET) udp_del_offload(&gs->udp_offloads); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9fb6395967de..81b26a543a3c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1013,6 +1013,19 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * a new port starts listening. The operation is protected by the * vxlan_net->sock_lock. * + * void (*ndo_add_geneve_port)(struct net_device *dev, + * sa_family_t sa_family, __be16 port); + * Called by geneve to notify a driver about the UDP port and socket + * address family that geneve is listnening to. It is called only when + * a new port starts listening. The operation is protected by the + * geneve_net->sock_lock. + * + * void (*ndo_del_geneve_port)(struct net_device *dev, + * sa_family_t sa_family, __be16 port); + * Called by geneve to notify the driver about a UDP port and socket + * address family that geneve is not listening to anymore. The operation + * is protected by the geneve_net->sock_lock. + * * void (*ndo_del_vxlan_port)(struct net_device *dev, * sa_family_t sa_family, __be16 port); * Called by vxlan to notify the driver about a UDP port and socket @@ -1217,7 +1230,12 @@ struct net_device_ops { void (*ndo_del_vxlan_port)(struct net_device *dev, sa_family_t sa_family, __be16 port); - + void (*ndo_add_geneve_port)(struct net_device *dev, + sa_family_t sa_family, + __be16 port); + void (*ndo_del_geneve_port)(struct net_device *dev, + sa_family_t sa_family, + __be16 port); void* (*ndo_dfwd_add_station)(struct net_device *pdev, struct net_device *dev); void (*ndo_dfwd_del_station)(struct net_device *pdev, -- cgit v1.2.3-59-g8ed1b From 05ca4029b25c65c860be5baa9288ec8f7e0a97e6 Mon Sep 17 00:00:00 2001 From: "Singhai, Anjali" Date: Mon, 14 Dec 2015 12:21:20 -0800 Subject: geneve: Add geneve_get_rx_port support This patch adds an op that the drivers can call into to get existing geneve ports. Signed-off-by: Anjali Singhai Jain Signed-off-by: David S. Miller --- drivers/net/geneve.c | 24 ++++++++++++++++++++++++ include/net/geneve.h | 8 ++++++++ 2 files changed, 32 insertions(+) (limited to 'include') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 89325e483ecf..31b19fdf659d 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1109,6 +1109,30 @@ static struct device_type geneve_type = { .name = "geneve", }; +/* Calls the ndo_add_geneve_port of the caller in order to + * supply the listening GENEVE udp ports. Callers are expected + * to implement the ndo_add_geneve_port. + */ +void geneve_get_rx_port(struct net_device *dev) +{ + struct net *net = dev_net(dev); + struct geneve_net *gn = net_generic(net, geneve_net_id); + struct geneve_sock *gs; + sa_family_t sa_family; + struct sock *sk; + __be16 port; + + rcu_read_lock(); + list_for_each_entry_rcu(gs, &gn->sock_list, list) { + sk = gs->sock->sk; + sa_family = sk->sk_family; + port = inet_sk(sk)->inet_sport; + dev->netdev_ops->ndo_add_geneve_port(dev, sa_family, port); + } + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(geneve_get_rx_port); + /* Initialize the device structure. */ static void geneve_setup(struct net_device *dev) { diff --git a/include/net/geneve.h b/include/net/geneve.h index 3106ed6eae0d..e6c23dc765f7 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -62,6 +62,14 @@ struct genevehdr { struct geneve_opt options[]; }; +#if IS_ENABLED(CONFIG_GENEVE) +void geneve_get_rx_port(struct net_device *netdev); +#else +static inline void geneve_get_rx_port(struct net_device *netdev) +{ +} +#endif + #ifdef CONFIG_INET struct net_device *geneve_dev_create_fb(struct net *net, const char *name, u8 name_assign_type, u16 dst_port); -- cgit v1.2.3-59-g8ed1b From 32bc201e1974976b7d3fea9a9b17bb7392ca6394 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 16 Dec 2015 17:50:11 +0800 Subject: ipv6: allow routes to be configured with expire values Add the support for adding expire value to routes, requested by Tom Gundersen for systemd-networkd, and NetworkManager wants it too. implement it by adding the new RTNETLINK attribute RTA_EXPIRES. Signed-off-by: Xin Long Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/uapi/linux/rtnetlink.h | 1 + net/ipv6/route.c | 10 ++++++++++ 2 files changed, 11 insertions(+) (limited to 'include') diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h index 123a5af4e8bb..ca764b5da86d 100644 --- a/include/uapi/linux/rtnetlink.h +++ b/include/uapi/linux/rtnetlink.h @@ -311,6 +311,7 @@ enum rtattr_type_t { RTA_PREF, RTA_ENCAP_TYPE, RTA_ENCAP, + RTA_EXPIRES, __RTA_MAX }; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index c83b6a5b3604..3c8834bc822d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2709,6 +2709,7 @@ static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = { [RTA_PREF] = { .type = NLA_U8 }, [RTA_ENCAP_TYPE] = { .type = NLA_U16 }, [RTA_ENCAP] = { .type = NLA_NESTED }, + [RTA_EXPIRES] = { .type = NLA_U32 }, }; static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -2809,6 +2810,15 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, if (tb[RTA_ENCAP_TYPE]) cfg->fc_encap_type = nla_get_u16(tb[RTA_ENCAP_TYPE]); + if (tb[RTA_EXPIRES]) { + unsigned long timeout = addrconf_timeout_fixup(nla_get_u32(tb[RTA_EXPIRES]), HZ); + + if (addrconf_finite_timeout(timeout)) { + cfg->fc_expires = jiffies_to_clock_t(timeout * HZ); + cfg->fc_flags |= RTF_EXPIRES; + } + } + err = 0; errout: return err; -- cgit v1.2.3-59-g8ed1b From 715f504b118998c41a2079a17e16bf5a8a114885 Mon Sep 17 00:00:00 2001 From: Hannes Frederic Sowa Date: Wed, 16 Dec 2015 17:22:47 +0100 Subject: ipv6: add IPV6_HDRINCL option for raw sockets Same as in Windows, we miss IPV6_HDRINCL for SOL_IPV6 and SOL_RAW. The SOL_IP/IP_HDRINCL is not available for IPv6 sockets. Signed-off-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/uapi/linux/in6.h | 1 + net/ipv6/raw.c | 20 ++++++++++++++++---- 2 files changed, 17 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/in6.h b/include/uapi/linux/in6.h index 79b12b004ade..318a4828bf98 100644 --- a/include/uapi/linux/in6.h +++ b/include/uapi/linux/in6.h @@ -196,6 +196,7 @@ struct in6_flowlabel_req { #define IPV6_IPSEC_POLICY 34 #define IPV6_XFRM_POLICY 35 +#define IPV6_HDRINCL 36 #endif /* diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 99140986e887..fa59dd7a427e 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -972,6 +972,11 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, return -EFAULT; switch (optname) { + case IPV6_HDRINCL: + if (sk->sk_type != SOCK_RAW) + return -EINVAL; + inet_sk(sk)->hdrincl = !!val; + return 0; case IPV6_CHECKSUM: if (inet_sk(sk)->inet_num == IPPROTO_ICMPV6 && level == IPPROTO_IPV6) { @@ -1016,7 +1021,8 @@ static int rawv6_setsockopt(struct sock *sk, int level, int optname, return -EOPNOTSUPP; return rawv6_seticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: - if (optname == IPV6_CHECKSUM) + if (optname == IPV6_CHECKSUM || + optname == IPV6_HDRINCL) break; default: return ipv6_setsockopt(sk, level, optname, optval, optlen); @@ -1037,7 +1043,8 @@ static int compat_rawv6_setsockopt(struct sock *sk, int level, int optname, return -EOPNOTSUPP; return rawv6_seticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: - if (optname == IPV6_CHECKSUM) + if (optname == IPV6_CHECKSUM || + optname == IPV6_HDRINCL) break; default: return compat_ipv6_setsockopt(sk, level, optname, @@ -1057,6 +1064,9 @@ static int do_rawv6_getsockopt(struct sock *sk, int level, int optname, return -EFAULT; switch (optname) { + case IPV6_HDRINCL: + val = inet_sk(sk)->hdrincl; + break; case IPV6_CHECKSUM: /* * We allow getsockopt() for IPPROTO_IPV6-level @@ -1094,7 +1104,8 @@ static int rawv6_getsockopt(struct sock *sk, int level, int optname, return -EOPNOTSUPP; return rawv6_geticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: - if (optname == IPV6_CHECKSUM) + if (optname == IPV6_CHECKSUM || + optname == IPV6_HDRINCL) break; default: return ipv6_getsockopt(sk, level, optname, optval, optlen); @@ -1115,7 +1126,8 @@ static int compat_rawv6_getsockopt(struct sock *sk, int level, int optname, return -EOPNOTSUPP; return rawv6_geticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: - if (optname == IPV6_CHECKSUM) + if (optname == IPV6_CHECKSUM || + optname == IPV6_HDRINCL) break; default: return compat_ipv6_getsockopt(sk, level, optname, -- cgit v1.2.3-59-g8ed1b From cc9da6cc4f56e05cc9e591459fe0192727ff58b3 Mon Sep 17 00:00:00 2001 From: Bjørn Mork Date: Wed, 16 Dec 2015 16:44:38 +0100 Subject: ipv6: addrconf: use stable address generator for ARPHRD_NONE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a new address generator mode, using the stable address generator with an automatically generated secret. This is intended as a default address generator mode for device types with no EUI64 implementation. The new generator is used for ARPHRD_NONE interfaces initially, adding default IPv6 autoconf support to e.g. tun interfaces. If the addrgenmode is set to 'random', either by default or manually, and no stable secret is available, then a random secret is used as input for the stable-privacy address generator. The secret can be read and modified like manually configured secrets, using the proc interface. Modifying the secret will change the addrgen mode to 'stable-privacy' to indicate that it operates on a known secret. Existing behaviour of the 'stable-privacy' mode is kept unchanged. If a known secret is available when the device is created, then the mode will default to 'stable-privacy' as before. The mode can be manually set to 'random' but it will behave exactly like 'stable-privacy' in this case. The secret will not change. Cc: Hannes Frederic Sowa Cc: 吉藤英明 Signed-off-by: Bjørn Mork Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/uapi/linux/if_link.h | 1 + net/ipv6/addrconf.c | 45 ++++++++++++++++++++++++++++++++++++++------ 2 files changed, 40 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 2be1dd5a103f..a30b78090594 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -218,6 +218,7 @@ enum in6_addr_gen_mode { IN6_ADDR_GEN_MODE_EUI64, IN6_ADDR_GEN_MODE_NONE, IN6_ADDR_GEN_MODE_STABLE_PRIVACY, + IN6_ADDR_GEN_MODE_RANDOM, }; /* Bridge section */ diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 233efa67dc3d..819b7777f3cb 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2319,6 +2319,12 @@ static void manage_tempaddrs(struct inet6_dev *idev, } } +static bool is_addr_mode_generate_stable(struct inet6_dev *idev) +{ + return idev->addr_gen_mode == IN6_ADDR_GEN_MODE_STABLE_PRIVACY || + idev->addr_gen_mode == IN6_ADDR_GEN_MODE_RANDOM; +} + void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) { struct prefix_info *pinfo; @@ -2432,8 +2438,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) in6_dev->token.s6_addr + 8, 8); read_unlock_bh(&in6_dev->lock); tokenized = true; - } else if (in6_dev->addr_gen_mode == - IN6_ADDR_GEN_MODE_STABLE_PRIVACY && + } else if (is_addr_mode_generate_stable(in6_dev) && !ipv6_generate_stable_address(&addr, 0, in6_dev)) { addr_flags |= IFA_F_STABLE_PRIVACY; @@ -3033,6 +3038,17 @@ retry: return 0; } +static void ipv6_gen_mode_random_init(struct inet6_dev *idev) +{ + struct ipv6_stable_secret *s = &idev->cnf.stable_secret; + + if (s->initialized) + return; + s = &idev->cnf.stable_secret; + get_random_bytes(&s->secret, sizeof(s->secret)); + s->initialized = true; +} + static void addrconf_addr_gen(struct inet6_dev *idev, bool prefix_route) { struct in6_addr addr; @@ -3043,13 +3059,18 @@ static void addrconf_addr_gen(struct inet6_dev *idev, bool prefix_route) ipv6_addr_set(&addr, htonl(0xFE800000), 0, 0, 0); - if (idev->addr_gen_mode == IN6_ADDR_GEN_MODE_STABLE_PRIVACY) { + switch (idev->addr_gen_mode) { + case IN6_ADDR_GEN_MODE_RANDOM: + ipv6_gen_mode_random_init(idev); + /* fallthrough */ + case IN6_ADDR_GEN_MODE_STABLE_PRIVACY: if (!ipv6_generate_stable_address(&addr, 0, idev)) addrconf_add_linklocal(idev, &addr, IFA_F_STABLE_PRIVACY); else if (prefix_route) addrconf_prefix_route(&addr, 64, idev->dev, 0, 0); - } else if (idev->addr_gen_mode == IN6_ADDR_GEN_MODE_EUI64) { + break; + case IN6_ADDR_GEN_MODE_EUI64: /* addrconf_add_linklocal also adds a prefix_route and we * only need to care about prefix routes if ipv6_generate_eui64 * couldn't generate one. @@ -3058,6 +3079,11 @@ static void addrconf_addr_gen(struct inet6_dev *idev, bool prefix_route) addrconf_add_linklocal(idev, &addr, 0); else if (prefix_route) addrconf_prefix_route(&addr, 64, idev->dev, 0, 0); + break; + case IN6_ADDR_GEN_MODE_NONE: + default: + /* will not add any link local address */ + break; } } @@ -3073,7 +3099,8 @@ static void addrconf_dev_config(struct net_device *dev) (dev->type != ARPHRD_INFINIBAND) && (dev->type != ARPHRD_IEEE1394) && (dev->type != ARPHRD_TUNNEL6) && - (dev->type != ARPHRD_6LOWPAN)) { + (dev->type != ARPHRD_6LOWPAN) && + (dev->type != ARPHRD_NONE)) { /* Alas, we support only Ethernet autoconfiguration. */ return; } @@ -3082,6 +3109,11 @@ static void addrconf_dev_config(struct net_device *dev) if (IS_ERR(idev)) return; + /* this device type has no EUI support */ + if (dev->type == ARPHRD_NONE && + idev->addr_gen_mode == IN6_ADDR_GEN_MODE_EUI64) + idev->addr_gen_mode = IN6_ADDR_GEN_MODE_RANDOM; + addrconf_addr_gen(idev, false); } @@ -4926,7 +4958,8 @@ static int inet6_set_link_af(struct net_device *dev, const struct nlattr *nla) if (mode != IN6_ADDR_GEN_MODE_EUI64 && mode != IN6_ADDR_GEN_MODE_NONE && - mode != IN6_ADDR_GEN_MODE_STABLE_PRIVACY) + mode != IN6_ADDR_GEN_MODE_STABLE_PRIVACY && + mode != IN6_ADDR_GEN_MODE_RANDOM) return -EINVAL; if (mode == IN6_ADDR_GEN_MODE_STABLE_PRIVACY && -- cgit v1.2.3-59-g8ed1b From 1a8524794fc7c70f44ac28e3a6e8fd637bc41f14 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 16 Dec 2015 13:20:43 -0800 Subject: net: l3mdev: Add master device lookup by index Add helper to lookup l3mdev master index given a device index. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/net/l3mdev.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'include') diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index 774d85b2d5d9..786226f8e77b 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -51,6 +51,24 @@ static inline int l3mdev_master_ifindex(struct net_device *dev) return ifindex; } +static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex) +{ + struct net_device *dev; + int rc = 0; + + if (likely(ifindex)) { + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, ifindex); + if (dev) + rc = l3mdev_master_ifindex_rcu(dev); + + rcu_read_unlock(); + } + + return rc; +} + /* get index of an interface to use for FIB lookups. For devices * enslaved to an L3 master device FIB lookups are based on the * master index @@ -167,6 +185,11 @@ static inline int l3mdev_master_ifindex(struct net_device *dev) return 0; } +static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex) +{ + return 0; +} + static inline int l3mdev_fib_oif_rcu(struct net_device *dev) { return dev ? dev->ifindex : 0; -- cgit v1.2.3-59-g8ed1b From 6dd9a14e92e54895e143f10fef4d0b9abe109aa9 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 16 Dec 2015 13:20:44 -0800 Subject: net: Allow accepted sockets to be bound to l3mdev domain Allow accepted sockets to derive their sk_bound_dev_if setting from the l3mdev domain in which the packets originated. A sysctl setting is added to control the behavior which is similar to sk_mark and sysctl_tcp_fwmark_accept. This effectively allow a process to have a "VRF-global" listen socket, with child sockets bound to the VRF device in which the packet originated. A similar behavior can be achieved using sk_mark, but a solution using marks is incomplete as it does not handle duplicate addresses in different L3 domains/VRFs. Allowing sockets to inherit the sk_bound_dev_if from l3mdev domain provides a complete solution. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 8 ++++++++ include/net/inet_sock.h | 14 ++++++++++++++ include/net/netns/ipv4.h | 3 +++ net/ipv4/syncookies.c | 4 ++-- net/ipv4/sysctl_net_ipv4.c | 11 +++++++++++ net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_ipv4.c | 1 + net/ipv6/syncookies.c | 4 ++-- 8 files changed, 42 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 5de632ed0ec0..ceb44a095a27 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -335,6 +335,14 @@ tcp_keepalive_intvl - INTEGER after probes started. Default value: 75sec i.e. connection will be aborted after ~11 minutes of retries. +tcp_l3mdev_accept - BOOLEAN + Enables child sockets to inherit the L3 master device index. + Enabling this option allows a "global" listen socket to work + across L3 master domains (e.g., VRFs) with connected sockets + derived from the listen socket to be bound to the L3 domain in + which the packets originated. Only valid when the kernel was + compiled with CONFIG_NET_L3_MASTER_DEV. + tcp_low_latency - BOOLEAN If set, the TCP stack makes decisions that prefer lower latency as opposed to higher throughput. By default, this diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 625bdf95d673..012b1f91f3ec 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -28,6 +28,7 @@ #include #include #include +#include /** struct ip_options - IP Options * @@ -113,6 +114,19 @@ static inline u32 inet_request_mark(const struct sock *sk, struct sk_buff *skb) return sk->sk_mark; } +static inline int inet_request_bound_dev_if(const struct sock *sk, + struct sk_buff *skb) +{ +#ifdef CONFIG_NET_L3_MASTER_DEV + struct net *net = sock_net(sk); + + if (!sk->sk_bound_dev_if && net->ipv4.sysctl_tcp_l3mdev_accept) + return l3mdev_master_ifindex_by_index(net, skb->skb_iif); +#endif + + return sk->sk_bound_dev_if; +} + struct inet_cork { unsigned int flags; __be32 addr; diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index c68926b4899c..d75be32650ba 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -86,6 +86,9 @@ struct netns_ipv4 { int sysctl_fwmark_reflect; int sysctl_tcp_fwmark_accept; +#ifdef CONFIG_NET_L3_MASTER_DEV + int sysctl_tcp_l3mdev_accept; +#endif int sysctl_tcp_mtu_probing; int sysctl_tcp_base_mss; int sysctl_tcp_probe_threshold; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 4cbe9f0a4281..643a86c49020 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -351,7 +351,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) treq->snt_synack.v64 = 0; treq->tfo_listener = false; - ireq->ir_iif = sk->sk_bound_dev_if; + ireq->ir_iif = inet_request_bound_dev_if(sk, 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) @@ -371,7 +371,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) * hasn't changed since we received the original syn, but I see * no easy way to do this. */ - flowi4_init_output(&fl4, sk->sk_bound_dev_if, ireq->ir_mark, + flowi4_init_output(&fl4, ireq->ir_iif, ireq->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP, inet_sk_flowi_flags(sk), opt->srr ? opt->faddr : ireq->ir_rmt_addr, diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a0bd7a55193e..41ff1f87dfd7 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -915,6 +915,17 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, +#ifdef CONFIG_NET_L3_MASTER_DEV + { + .procname = "tcp_l3mdev_accept", + .data = &init_net.ipv4.sysctl_tcp_l3mdev_accept, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, +#endif { .procname = "tcp_mtu_probing", .data = &init_net.ipv4.sysctl_tcp_mtu_probing, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2d656eef7f8e..7b1fddc47019 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6204,7 +6204,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, tcp_openreq_init(req, &tmp_opt, skb, sk); /* Note: tcp_v6_init_req() might override ir_iif for link locals */ - inet_rsk(req)->ir_iif = sk->sk_bound_dev_if; + inet_rsk(req)->ir_iif = inet_request_bound_dev_if(sk, skb); af_ops->init_req(req, sk, skb); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 205e6745393f..46e92fbd26a8 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1276,6 +1276,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, ireq = inet_rsk(req); sk_daddr_set(newsk, ireq->ir_rmt_addr); sk_rcv_saddr_set(newsk, ireq->ir_loc_addr); + newsk->sk_bound_dev_if = ireq->ir_iif; newinet->inet_saddr = ireq->ir_loc_addr; inet_opt = ireq->opt; rcu_assign_pointer(newinet->inet_opt, inet_opt); diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index eaf7ac496d50..2906ef20795e 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -193,7 +193,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) ireq->pktopts = skb; } - ireq->ir_iif = sk->sk_bound_dev_if; + ireq->ir_iif = inet_request_bound_dev_if(sk, skb); /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL) @@ -224,7 +224,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) fl6.daddr = ireq->ir_v6_rmt_addr; final_p = fl6_update_dst(&fl6, rcu_dereference(np->opt), &final); fl6.saddr = ireq->ir_v6_loc_addr; - fl6.flowi6_oif = sk->sk_bound_dev_if; + fl6.flowi6_oif = ireq->ir_iif; fl6.flowi6_mark = ireq->ir_mark; fl6.fl6_dport = ireq->ir_rmt_port; fl6.fl6_sport = inet_sk(sk)->inet_sport; -- cgit v1.2.3-59-g8ed1b From 05c74e5e53f6cb07502c3e6a820f33e2777b6605 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 17 Dec 2015 23:51:53 +0100 Subject: bpf: add bpf_skb_load_bytes helper When hacking tc programs with eBPF, one of the issues that come up from time to time is to load addresses from headers. In eBPF as in classic BPF, we have BPF_LD | BPF_ABS | BPF_{B,H,W} instructions that extract a byte, half-word or word out of the skb data though helpers such as bpf_load_pointer() (interpreter case). F.e. extracting a whole IPv6 address could possibly look like ... union v6addr { struct { __u32 p1; __u32 p2; __u32 p3; __u32 p4; }; __u8 addr[16]; }; [...] a.p1 = htonl(load_word(skb, off)); a.p2 = htonl(load_word(skb, off + 4)); a.p3 = htonl(load_word(skb, off + 8)); a.p4 = htonl(load_word(skb, off + 12)); [...] /* access to a.addr[...] */ This work adds a complementary helper bpf_skb_load_bytes() (we also have bpf_skb_store_bytes()) as an alternative where the same call would look like from an eBPF program: ret = bpf_skb_load_bytes(skb, off, addr, sizeof(addr)); Same verifier restrictions apply as in ffeedafbf023 ("bpf: introduce current->pid, tgid, uid, gid, comm accessors") case, where stack memory access needs to be statically verified and thus guaranteed to be initialized in first use (otherwise verifier cannot tell whether a subsequent access to it is valid or not as it's runtime dependent). Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 1 + net/core/filter.c | 35 ++++++++++++++++++++++++++++++++++- 2 files changed, 35 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 9ea2d22fa2cb..8bed7f1176b8 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -269,6 +269,7 @@ enum bpf_func_id { * Return: 0 on success */ BPF_FUNC_perf_event_output, + BPF_FUNC_skb_load_bytes, __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index 672eefbfbe99..34bf6fc77c1d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1245,6 +1245,7 @@ int sk_attach_bpf(u32 ufd, struct sock *sk) } #define BPF_RECOMPUTE_CSUM(flags) ((flags) & 1) +#define BPF_LDST_LEN 16U static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) { @@ -1252,7 +1253,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) int offset = (int) r2; void *from = (void *) (long) r3; unsigned int len = (unsigned int) r4; - char buf[16]; + char buf[BPF_LDST_LEN]; void *ptr; /* bpf verifier guarantees that: @@ -1299,6 +1300,36 @@ const struct bpf_func_proto bpf_skb_store_bytes_proto = { .arg5_type = ARG_ANYTHING, }; +static u64 bpf_skb_load_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + const struct sk_buff *skb = (const struct sk_buff *)(unsigned long) r1; + int offset = (int) r2; + void *to = (void *)(unsigned long) r3; + unsigned int len = (unsigned int) r4; + void *ptr; + + if (unlikely((u32) offset > 0xffff || len > BPF_LDST_LEN)) + return -EFAULT; + + ptr = skb_header_pointer(skb, offset, len, to); + if (unlikely(!ptr)) + return -EFAULT; + if (ptr != to) + memcpy(to, ptr, len); + + return 0; +} + +const struct bpf_func_proto bpf_skb_load_bytes_proto = { + .func = bpf_skb_load_bytes, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_PTR_TO_STACK, + .arg4_type = ARG_CONST_STACK_SIZE, +}; + #define BPF_HEADER_FIELD_SIZE(flags) ((flags) & 0x0f) #define BPF_IS_PSEUDO_HEADER(flags) ((flags) & 0x10) @@ -1654,6 +1685,8 @@ tc_cls_act_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_l3_csum_replace: return &bpf_l3_csum_replace_proto; case BPF_FUNC_l4_csum_replace: -- cgit v1.2.3-59-g8ed1b From b1f0a0e99c58fbd7ea053ca36ba623718272b618 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 21 Dec 2015 21:29:24 +0100 Subject: net: add inet_sk_transparent() helper Avoids cluttering tcp_v4_send_reset when followup patch extends it to deal with timewait sockets. Suggested-by: Eric Dumazet Signed-off-by: Florian Westphal Acked-by: Eric Dumazet Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/net/request_sock.h | 2 +- include/net/tcp.h | 12 ++++++++++++ 2 files changed, 13 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index a0dde04eb178..f49759decb28 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -68,7 +68,7 @@ struct request_sock { u32 peer_secid; }; -static inline struct request_sock *inet_reqsk(struct sock *sk) +static inline struct request_sock *inet_reqsk(const struct sock *sk) { return (struct request_sock *)sk; } diff --git a/include/net/tcp.h b/include/net/tcp.h index 3077735b348d..f33fecf4e282 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1620,6 +1620,18 @@ static inline void tcp_highest_sack_combine(struct sock *sk, tcp_sk(sk)->highest_sack = new; } +/* This helper checks if socket has IP_TRANSPARENT set */ +static inline bool inet_sk_transparent(const struct sock *sk) +{ + switch (sk->sk_state) { + case TCP_TIME_WAIT: + return inet_twsk(sk)->tw_transparent; + case TCP_NEW_SYN_RECV: + return inet_rsk(inet_reqsk(sk))->no_srccheck; + } + return inet_sk(sk)->transparent; +} + /* Determines whether this is a thin stream (which may suffer from * increased latency). Used to trigger latency-reducing mechanisms. */ -- cgit v1.2.3-59-g8ed1b From 039f50629b7f860f36644ed1f34b27da9aa62f43 Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Thu, 24 Dec 2015 14:34:54 -0800 Subject: ip_tunnel: Move stats update to iptunnel_xmit() By moving stats update into iptunnel_xmit(), we can simplify iptunnel_xmit() usage. With this change there is no need to call another function (iptunnel_xmit_stats()) to update stats in tunnel xmit code path. Signed-off-by: Pravin B Shelar Signed-off-by: David S. Miller --- drivers/net/geneve.c | 17 ++++++++--------- drivers/net/vxlan.c | 9 ++++----- include/net/ip6_tunnel.h | 17 ++++------------- include/net/ip_tunnels.h | 28 +++++++++++++++------------- include/net/udp_tunnel.h | 8 ++++---- net/ipv4/ip_gre.c | 7 +++---- net/ipv4/ip_tunnel.c | 7 ++----- net/ipv4/ip_tunnel_core.c | 9 +++++---- net/ipv4/ip_vti.c | 2 +- net/ipv4/udp_tunnel.c | 11 +++++------ net/ipv6/sit.c | 7 ++----- net/tipc/udp_media.c | 12 +++--------- 12 files changed, 56 insertions(+), 78 deletions(-) (limited to 'include') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index e6e00924f8ef..20dd66423ec8 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -918,12 +918,11 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, ttl = ttl ? : ip4_dst_hoplimit(&rt->dst); df = 0; } - err = udp_tunnel_xmit_skb(rt, gs4->sock->sk, skb, fl4.saddr, fl4.daddr, - tos, ttl, df, sport, geneve->dst_port, - !net_eq(geneve->net, dev_net(geneve->dev)), - !(flags & GENEVE_F_UDP_CSUM)); + udp_tunnel_xmit_skb(rt, gs4->sock->sk, skb, fl4.saddr, fl4.daddr, + tos, ttl, df, sport, geneve->dst_port, + !net_eq(geneve->net, dev_net(geneve->dev)), + !(flags & GENEVE_F_UDP_CSUM)); - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); return NETDEV_TX_OK; tx_error: @@ -1005,10 +1004,10 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, ttl = 1; ttl = ttl ? : ip6_dst_hoplimit(dst); } - err = udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev, - &fl6.saddr, &fl6.daddr, prio, ttl, - sport, geneve->dst_port, - !!(flags & GENEVE_F_UDP_ZERO_CSUM6_TX)); + udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev, + &fl6.saddr, &fl6.daddr, prio, ttl, + sport, geneve->dst_port, + !!(flags & GENEVE_F_UDP_ZERO_CSUM6_TX)); return NETDEV_TX_OK; tx_error: diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index ba363cedef80..fecf7b6c732e 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -1841,9 +1841,10 @@ static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *sk skb_set_inner_protocol(skb, htons(ETH_P_TEB)); - return udp_tunnel_xmit_skb(rt, sk, skb, src, dst, tos, - ttl, df, src_port, dst_port, xnet, - !(vxflags & VXLAN_F_UDP_CSUM)); + udp_tunnel_xmit_skb(rt, sk, skb, src, dst, tos, ttl, df, + src_port, dst_port, xnet, + !(vxflags & VXLAN_F_UDP_CSUM)); + return 0; } #if IS_ENABLED(CONFIG_IPV6) @@ -2056,8 +2057,6 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, skb = NULL; goto rt_tx_error; } - - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); #if IS_ENABLED(CONFIG_IPV6) } else { struct dst_entry *ndst; diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index ff788b665277..ae07e94778d8 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -5,6 +5,7 @@ #include #include #include +#include #define IP6TUNNEL_ERR_TIMEO (30*HZ) @@ -83,22 +84,12 @@ int ip6_tnl_get_iflink(const struct net_device *dev); static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb, struct net_device *dev) { - struct net_device_stats *stats = &dev->stats; int pkt_len, err; pkt_len = skb->len - skb_inner_network_offset(skb); err = ip6_local_out(dev_net(skb_dst(skb)->dev), sk, skb); - - if (net_xmit_eval(err) == 0) { - struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats); - u64_stats_update_begin(&tstats->syncp); - tstats->tx_bytes += pkt_len; - tstats->tx_packets++; - u64_stats_update_end(&tstats->syncp); - put_cpu_ptr(tstats); - } else { - stats->tx_errors++; - stats->tx_aborted_errors++; - } + if (unlikely(net_xmit_eval(err))) + pkt_len = -1; + iptunnel_xmit_stats(dev, pkt_len); } #endif diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 62a750a6a8f8..6db96ea0144f 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -273,32 +273,34 @@ static inline u8 ip_tunnel_ecn_encap(u8 tos, const struct iphdr *iph, } int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto); -int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, - __be32 src, __be32 dst, u8 proto, - u8 tos, u8 ttl, __be16 df, bool xnet); +void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, + __be32 src, __be32 dst, u8 proto, + u8 tos, u8 ttl, __be16 df, bool xnet); struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, gfp_t flags); struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, bool gre_csum, int gso_type_mask); -static inline void iptunnel_xmit_stats(int err, - struct net_device_stats *err_stats, - struct pcpu_sw_netstats __percpu *stats) +static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len) { - if (err > 0) { - struct pcpu_sw_netstats *tstats = get_cpu_ptr(stats); + if (pkt_len > 0) { + struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats); u64_stats_update_begin(&tstats->syncp); - tstats->tx_bytes += err; + tstats->tx_bytes += pkt_len; tstats->tx_packets++; u64_stats_update_end(&tstats->syncp); put_cpu_ptr(tstats); - } else if (err < 0) { - err_stats->tx_errors++; - err_stats->tx_aborted_errors++; } else { - err_stats->tx_dropped++; + struct net_device_stats *err_stats = &dev->stats; + + if (pkt_len < 0) { + err_stats->tx_errors++; + err_stats->tx_aborted_errors++; + } else { + err_stats->tx_dropped++; + } } } diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index cb2f89f20f5c..cca2ad3082c3 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -78,10 +78,10 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, struct udp_tunnel_sock_cfg *sock_cfg); /* Transmit the skb using UDP encapsulation. */ -int udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, - __be32 src, __be32 dst, __u8 tos, __u8 ttl, - __be16 df, __be16 src_port, __be16 dst_port, - bool xnet, bool nocheck); +void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, + __be32 src, __be32 dst, __u8 tos, __u8 ttl, + __be16 df, __be16 src_port, __be16 dst_port, + bool xnet, bool nocheck); #if IS_ENABLED(CONFIG_IPV6) int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk, diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 04a48c0159cc..7c51c4e1661f 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -561,10 +561,9 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev) tunnel_id_to_key(tun_info->key.tun_id), 0); df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; - err = iptunnel_xmit(skb->sk, rt, skb, fl.saddr, - key->u.ipv4.dst, IPPROTO_GRE, - key->tos, key->ttl, df, false); - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); + + iptunnel_xmit(skb->sk, rt, skb, fl.saddr, key->u.ipv4.dst, IPPROTO_GRE, + key->tos, key->ttl, df, false); return; err_free_rt: diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 0f6e9ee031c4..c7bd72e9b544 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -656,7 +656,6 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, struct rtable *rt; /* Route to the other host */ unsigned int max_headroom; /* The extra header space needed */ __be32 dst; - int err; bool connected; inner_iph = (const struct iphdr *)skb_inner_network_header(skb); @@ -794,10 +793,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, return; } - err = iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, protocol, - tos, ttl, df, !net_eq(tunnel->net, dev_net(dev))); - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); - + iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, protocol, tos, ttl, + df, !net_eq(tunnel->net, dev_net(dev))); return; #if IS_ENABLED(CONFIG_IPV6) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 1db8418aa62e..eb52ce950c27 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -47,12 +47,13 @@ #include #include -int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, - __be32 src, __be32 dst, __u8 proto, - __u8 tos, __u8 ttl, __be16 df, bool xnet) +void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, + __be32 src, __be32 dst, __u8 proto, + __u8 tos, __u8 ttl, __be16 df, bool xnet) { int pkt_len = skb->len - skb_inner_network_offset(skb); struct net *net = dev_net(rt->dst.dev); + struct net_device *dev = skb->dev; struct iphdr *iph; int err; @@ -81,7 +82,7 @@ int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, err = ip_local_out(net, sk, skb); if (unlikely(net_xmit_eval(err))) pkt_len = 0; - return pkt_len; + iptunnel_xmit_stats(dev, pkt_len); } EXPORT_SYMBOL_GPL(iptunnel_xmit); diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 02d9c21e2953..5cf10b777b7e 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -199,7 +199,7 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev, err = dst_output(tunnel->net, skb->sk, skb); if (net_xmit_eval(err) == 0) err = skb->len; - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); + iptunnel_xmit_stats(dev, err); return NETDEV_TX_OK; tx_error_icmp: diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index aba428626b52..0ec08814f37d 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -74,10 +74,10 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, } EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); -int udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, - __be32 src, __be32 dst, __u8 tos, __u8 ttl, - __be16 df, __be16 src_port, __be16 dst_port, - bool xnet, bool nocheck) +void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, + __be32 src, __be32 dst, __u8 tos, __u8 ttl, + __be16 df, __be16 src_port, __be16 dst_port, + bool xnet, bool nocheck) { struct udphdr *uh; @@ -91,8 +91,7 @@ int udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, udp_set_csum(nocheck, skb, src, dst, skb->len); - return iptunnel_xmit(sk, rt, skb, src, dst, IPPROTO_UDP, - tos, ttl, df, xnet); + iptunnel_xmit(sk, rt, skb, src, dst, IPPROTO_UDP, tos, ttl, df, xnet); } EXPORT_SYMBOL_GPL(udp_tunnel_xmit_skb); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index dcccae86190f..e794ef66a401 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -820,7 +820,6 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, const struct in6_addr *addr6; int addr_type; u8 ttl; - int err; u8 protocol = IPPROTO_IPV6; int t_hlen = tunnel->hlen + sizeof(struct iphdr); @@ -983,10 +982,8 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, skb_set_inner_ipproto(skb, IPPROTO_IPV6); - err = iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, - protocol, tos, ttl, df, - !net_eq(tunnel->net, dev_net(dev))); - iptunnel_xmit_stats(err, &dev->stats, dev->tstats); + iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, protocol, tos, ttl, + df, !net_eq(tunnel->net, dev_net(dev))); return NETDEV_TX_OK; tx_error_icmp: diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c index 6af78c6276b4..d63a911e7fe2 100644 --- a/net/tipc/udp_media.c +++ b/net/tipc/udp_media.c @@ -182,15 +182,9 @@ static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb, goto tx_error; } ttl = ip4_dst_hoplimit(&rt->dst); - err = udp_tunnel_xmit_skb(rt, ub->ubsock->sk, skb, - src->ipv4.s_addr, - dst->ipv4.s_addr, 0, ttl, 0, - src->udp_port, dst->udp_port, - false, true); - if (err < 0) { - ip_rt_put(rt); - goto tx_error; - } + udp_tunnel_xmit_skb(rt, ub->ubsock->sk, skb, src->ipv4.s_addr, + dst->ipv4.s_addr, 0, ttl, 0, src->udp_port, + dst->udp_port, false, true); #if IS_ENABLED(CONFIG_IPV6) } else { struct dst_entry *ndst; -- cgit v1.2.3-59-g8ed1b From df05ef874b284d833c2d9795a6350c6a373ab6c9 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 15 Dec 2015 19:39:32 +0100 Subject: netfilter: nf_tables: release objects on netns destruction We have to release the existing objects on netns removal otherwise we leak them. Chains are unregistered in first place to make sure no packets are walking on our rules and sets anymore. The object release happens by when we unregister the family via nft_release_afinfo() which is called from nft_unregister_afinfo() from the corresponding __net_exit path in every family. Reported-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 2 +- net/bridge/netfilter/nf_tables_bridge.c | 2 +- net/ipv4/netfilter/nf_tables_arp.c | 2 +- net/ipv4/netfilter/nf_tables_ipv4.c | 2 +- net/ipv6/netfilter/nf_tables_ipv6.c | 2 +- net/netfilter/nf_tables_api.c | 47 +++++++++++++++++++++++++++++++-- net/netfilter/nf_tables_inet.c | 2 +- net/netfilter/nf_tables_netdev.c | 2 +- 8 files changed, 52 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index b313cda49194..a50f139ce087 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -880,7 +880,7 @@ struct nft_af_info { }; int nft_register_afinfo(struct net *, struct nft_af_info *); -void nft_unregister_afinfo(struct nft_af_info *); +void nft_unregister_afinfo(struct net *, struct nft_af_info *); int nft_register_chain_type(const struct nf_chain_type *); void nft_unregister_chain_type(const struct nf_chain_type *); diff --git a/net/bridge/netfilter/nf_tables_bridge.c b/net/bridge/netfilter/nf_tables_bridge.c index 62f6b1b19589..7fcdd7261d88 100644 --- a/net/bridge/netfilter/nf_tables_bridge.c +++ b/net/bridge/netfilter/nf_tables_bridge.c @@ -141,7 +141,7 @@ err: static void nf_tables_bridge_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.bridge); + nft_unregister_afinfo(net, net->nft.bridge); kfree(net->nft.bridge); } diff --git a/net/ipv4/netfilter/nf_tables_arp.c b/net/ipv4/netfilter/nf_tables_arp.c index 9d09d4f59545..cd84d4295a20 100644 --- a/net/ipv4/netfilter/nf_tables_arp.c +++ b/net/ipv4/netfilter/nf_tables_arp.c @@ -57,7 +57,7 @@ err: static void nf_tables_arp_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.arp); + nft_unregister_afinfo(net, net->nft.arp); kfree(net->nft.arp); } diff --git a/net/ipv4/netfilter/nf_tables_ipv4.c b/net/ipv4/netfilter/nf_tables_ipv4.c index ca9dc3c46c4f..e44ba3b12fbb 100644 --- a/net/ipv4/netfilter/nf_tables_ipv4.c +++ b/net/ipv4/netfilter/nf_tables_ipv4.c @@ -78,7 +78,7 @@ err: static void nf_tables_ipv4_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.ipv4); + nft_unregister_afinfo(net, net->nft.ipv4); kfree(net->nft.ipv4); } diff --git a/net/ipv6/netfilter/nf_tables_ipv6.c b/net/ipv6/netfilter/nf_tables_ipv6.c index 120ea9131be0..30b22f4dff55 100644 --- a/net/ipv6/netfilter/nf_tables_ipv6.c +++ b/net/ipv6/netfilter/nf_tables_ipv6.c @@ -77,7 +77,7 @@ err: static void nf_tables_ipv6_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.ipv6); + nft_unregister_afinfo(net, net->nft.ipv6); kfree(net->nft.ipv6); } diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 4a23f77c363a..852273110275 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -41,6 +41,8 @@ int nft_register_afinfo(struct net *net, struct nft_af_info *afi) } EXPORT_SYMBOL_GPL(nft_register_afinfo); +static void __nft_release_afinfo(struct net *net, struct nft_af_info *afi); + /** * nft_unregister_afinfo - unregister nf_tables address family info * @@ -48,9 +50,10 @@ EXPORT_SYMBOL_GPL(nft_register_afinfo); * * Unregister the address family for use with nf_tables. */ -void nft_unregister_afinfo(struct nft_af_info *afi) +void nft_unregister_afinfo(struct net *net, struct nft_af_info *afi) { nfnl_lock(NFNL_SUBSYS_NFTABLES); + __nft_release_afinfo(net, afi); list_del_rcu(&afi->list); nfnl_unlock(NFNL_SUBSYS_NFTABLES); } @@ -4579,7 +4582,7 @@ int nft_data_dump(struct sk_buff *skb, int attr, const struct nft_data *data, } EXPORT_SYMBOL_GPL(nft_data_dump); -static int nf_tables_init_net(struct net *net) +static int __net_init nf_tables_init_net(struct net *net) { INIT_LIST_HEAD(&net->nft.af_info); INIT_LIST_HEAD(&net->nft.commit_list); @@ -4587,6 +4590,46 @@ static int nf_tables_init_net(struct net *net) return 0; } +/* Called by nft_unregister_afinfo() from __net_exit path, nfnl_lock is held. */ +static void __nft_release_afinfo(struct net *net, struct nft_af_info *afi) +{ + struct nft_table *table, *nt; + struct nft_chain *chain, *nc; + struct nft_rule *rule, *nr; + struct nft_set *set, *ns; + struct nft_ctx ctx = { + .net = net, + .afi = afi, + }; + + list_for_each_entry_safe(table, nt, &afi->tables, list) { + list_for_each_entry(chain, &table->chains, list) + nf_tables_unregister_hooks(table, chain, afi->nops); + /* No packets are walking on these chains anymore. */ + ctx.table = table; + list_for_each_entry(chain, &table->chains, list) { + ctx.chain = chain; + list_for_each_entry_safe(rule, nr, &chain->rules, list) { + list_del(&rule->list); + chain->use--; + nf_tables_rule_destroy(&ctx, rule); + } + } + list_for_each_entry_safe(set, ns, &table->sets, list) { + list_del(&set->list); + table->use--; + nft_set_destroy(set); + } + list_for_each_entry_safe(chain, nc, &table->chains, list) { + list_del(&chain->list); + table->use--; + nf_tables_chain_destroy(chain); + } + list_del(&table->list); + nf_tables_table_destroy(&ctx); + } +} + static struct pernet_operations nf_tables_net_ops = { .init = nf_tables_init_net, }; diff --git a/net/netfilter/nf_tables_inet.c b/net/netfilter/nf_tables_inet.c index 9dd2d216cfc1..6b5f76295d3d 100644 --- a/net/netfilter/nf_tables_inet.c +++ b/net/netfilter/nf_tables_inet.c @@ -57,7 +57,7 @@ err: static void __net_exit nf_tables_inet_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.inet); + nft_unregister_afinfo(net, net->nft.inet); kfree(net->nft.inet); } diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c index 7b9c053ba750..2bfd1fbccec8 100644 --- a/net/netfilter/nf_tables_netdev.c +++ b/net/netfilter/nf_tables_netdev.c @@ -139,7 +139,7 @@ err: static void nf_tables_netdev_exit_net(struct net *net) { - nft_unregister_afinfo(net->nft.netdev); + nft_unregister_afinfo(net, net->nft.netdev); kfree(net->nft.netdev); } -- cgit v1.2.3-59-g8ed1b From 5ebe0b0eec9d6f703b137f9b938c52f7b91dd9d6 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 15 Dec 2015 19:40:49 +0100 Subject: netfilter: nf_tables: destroy basechain and rules on netdevice removal If the netdevice is destroyed, the resources that are attached should be released too as they belong to the device that is now gone. Suggested-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 +---- net/netfilter/nf_tables_api.c | 31 +++++++++++++++++++++------ net/netfilter/nf_tables_netdev.c | 45 ++++++++++++++++----------------------- 3 files changed, 44 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index a50f139ce087..0191fbb33a2f 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -821,10 +821,7 @@ static inline struct nft_base_chain *nft_base_chain(const struct nft_chain *chai return container_of(chain, struct nft_base_chain, chain); } -int nft_register_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops); -void nft_unregister_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops); +int __nft_release_basechain(struct nft_ctx *ctx); unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 852273110275..5729844e1d46 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -131,8 +131,8 @@ static void nft_trans_destroy(struct nft_trans *trans) kfree(trans); } -int nft_register_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops) +static int nft_register_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops) { struct net *net = read_pnet(&basechain->pnet); @@ -141,10 +141,9 @@ int nft_register_basechain(struct nft_base_chain *basechain, return nf_register_net_hooks(net, basechain->ops, hook_nops); } -EXPORT_SYMBOL_GPL(nft_register_basechain); -void nft_unregister_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops) +static void nft_unregister_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops) { struct net *net = read_pnet(&basechain->pnet); @@ -153,7 +152,6 @@ void nft_unregister_basechain(struct nft_base_chain *basechain, nf_unregister_net_hooks(net, basechain->ops, hook_nops); } -EXPORT_SYMBOL_GPL(nft_unregister_basechain); static int nf_tables_register_hooks(const struct nft_table *table, struct nft_chain *chain, @@ -4590,6 +4588,27 @@ static int __net_init nf_tables_init_net(struct net *net) return 0; } +int __nft_release_basechain(struct nft_ctx *ctx) +{ + struct nft_rule *rule, *nr; + + BUG_ON(!(ctx->chain->flags & NFT_BASE_CHAIN)); + + nf_tables_unregister_hooks(ctx->chain->table, ctx->chain, + ctx->afi->nops); + list_for_each_entry_safe(rule, nr, &ctx->chain->rules, list) { + list_del(&rule->list); + ctx->chain->use--; + nf_tables_rule_destroy(ctx, rule); + } + list_del(&ctx->chain->list); + ctx->table->use--; + nf_tables_chain_destroy(ctx->chain); + + return 0; +} +EXPORT_SYMBOL_GPL(__nft_release_basechain); + /* Called by nft_unregister_afinfo() from __net_exit path, nfnl_lock is held. */ static void __nft_release_afinfo(struct net *net, struct nft_af_info *afi) { diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c index 2bfd1fbccec8..3e9c87b961ba 100644 --- a/net/netfilter/nf_tables_netdev.c +++ b/net/netfilter/nf_tables_netdev.c @@ -156,35 +156,17 @@ static const struct nf_chain_type nft_filter_chain_netdev = { .hook_mask = (1 << NF_NETDEV_INGRESS), }; -static void nft_netdev_event(unsigned long event, struct nft_af_info *afi, - struct net_device *dev, struct nft_table *table, - struct nft_base_chain *basechain) +static void nft_netdev_event(unsigned long event, struct net_device *dev, + struct nft_ctx *ctx) { - switch (event) { - case NETDEV_REGISTER: - if (strcmp(basechain->dev_name, dev->name) != 0) - return; + struct nft_base_chain *basechain = nft_base_chain(ctx->chain); - BUG_ON(!(basechain->flags & NFT_BASECHAIN_DISABLED)); - - dev_hold(dev); - basechain->ops[0].dev = dev; - basechain->flags &= ~NFT_BASECHAIN_DISABLED; - if (!(table->flags & NFT_TABLE_F_DORMANT)) - nft_register_basechain(basechain, afi->nops); - break; + switch (event) { case NETDEV_UNREGISTER: if (strcmp(basechain->dev_name, dev->name) != 0) return; - BUG_ON(basechain->flags & NFT_BASECHAIN_DISABLED); - - if (!(table->flags & NFT_TABLE_F_DORMANT)) - nft_unregister_basechain(basechain, afi->nops); - - dev_put(basechain->ops[0].dev); - basechain->ops[0].dev = NULL; - basechain->flags |= NFT_BASECHAIN_DISABLED; + __nft_release_basechain(ctx); break; case NETDEV_CHANGENAME: if (dev->ifindex != basechain->ops[0].dev->ifindex) @@ -201,20 +183,29 @@ static int nf_tables_netdev_event(struct notifier_block *this, struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct nft_af_info *afi; struct nft_table *table; - struct nft_chain *chain; + struct nft_chain *chain, *nr; + struct nft_ctx ctx = { + .net = dev_net(dev), + }; + + if (event != NETDEV_UNREGISTER && + event != NETDEV_CHANGENAME) + return NOTIFY_DONE; nfnl_lock(NFNL_SUBSYS_NFTABLES); list_for_each_entry(afi, &dev_net(dev)->nft.af_info, list) { + ctx.afi = afi; if (afi->family != NFPROTO_NETDEV) continue; list_for_each_entry(table, &afi->tables, list) { - list_for_each_entry(chain, &table->chains, list) { + ctx.table = table; + list_for_each_entry_safe(chain, nr, &table->chains, list) { if (!(chain->flags & NFT_BASE_CHAIN)) continue; - nft_netdev_event(event, afi, dev, table, - nft_base_chain(chain)); + ctx.chain = chain; + nft_netdev_event(event, dev, &ctx); } } } -- cgit v1.2.3-59-g8ed1b From 7b8002a1511fcbcb0596cac90d67ad5c8182d0aa Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 15 Dec 2015 18:41:56 +0100 Subject: netfilter: nfnetlink: pass down netns pointer to call() and call_rcu() Adapt callsites to avoid recurrent lookup of the netns pointer. Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nfnetlink.h | 8 +-- net/netfilter/ipset/ip_set_core.c | 108 +++++++++++++++-------------------- net/netfilter/nf_conntrack_netlink.c | 96 ++++++++++++++----------------- net/netfilter/nf_tables_api.c | 30 ++++------ net/netfilter/nfnetlink.c | 6 +- net/netfilter/nfnetlink_acct.c | 21 +++---- net/netfilter/nfnetlink_cthelper.c | 18 +++--- net/netfilter/nfnetlink_cttimeout.c | 42 ++++++-------- net/netfilter/nfnetlink_log.c | 15 ++--- net/netfilter/nfnetlink_queue.c | 36 +++++------- net/netfilter/nft_compat.c | 6 +- net/netfilter/xt_osf.c | 7 ++- 12 files changed, 173 insertions(+), 220 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 5646b24bfc64..ceacbf5dcb73 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -8,12 +8,12 @@ #include struct nfnl_callback { - int (*call)(struct sock *nl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]); - int (*call_rcu)(struct sock *nl, struct sk_buff *skb, + int (*call)(struct net *net, struct sock *nl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]); + int (*call_rcu)(struct net *net, struct sock *nl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]); int (*call_batch)(struct net *net, struct sock *nl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]); diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 54f3d7cb23e6..95db43fc0303 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -825,20 +825,17 @@ find_free_id(struct ip_set_net *inst, const char *name, ip_set_id_t *index, return 0; } -static int -ip_set_none(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_none(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { return -EOPNOTSUPP; } -static int -ip_set_create(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_create(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct net *net = sock_net(ctnl); struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set, *clash = NULL; ip_set_id_t index = IPSET_INVALID_ID; @@ -976,12 +973,11 @@ ip_set_destroy_set(struct ip_set *set) kfree(set); } -static int -ip_set_destroy(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_destroy(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *s; ip_set_id_t i; int ret = 0; @@ -1052,12 +1048,11 @@ ip_set_flush_set(struct ip_set *set) spin_unlock_bh(&set->lock); } -static int -ip_set_flush(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_flush(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *s; ip_set_id_t i; @@ -1092,12 +1087,11 @@ ip_set_setname2_policy[IPSET_ATTR_CMD_MAX + 1] = { .len = IPSET_MAXNAMELEN - 1 }, }; -static int -ip_set_rename(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_rename(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set, *s; const char *name2; ip_set_id_t i; @@ -1142,12 +1136,11 @@ out: * so the ip_set_list always contains valid pointers to the sets. */ -static int -ip_set_swap(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_swap(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *from, *to; ip_set_id_t from_id, to_id; char from_name[IPSET_MAXNAMELEN]; @@ -1413,10 +1406,9 @@ out: return ret < 0 ? ret : skb->len; } -static int -ip_set_dump(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_dump(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { if (unlikely(protocol_failed(attr))) return -IPSET_ERR_PROTOCOL; @@ -1500,12 +1492,11 @@ call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, return ret; } -static int -ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_uadd(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set; struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; const struct nlattr *nla; @@ -1555,12 +1546,11 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, return ret; } -static int -ip_set_udel(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_udel(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set; struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; const struct nlattr *nla; @@ -1610,12 +1600,11 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, return ret; } -static int -ip_set_utest(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_utest(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set; struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; int ret = 0; @@ -1646,12 +1635,11 @@ ip_set_utest(struct sock *ctnl, struct sk_buff *skb, /* Get headed data of a set */ -static int -ip_set_header(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_header(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { - struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); + struct ip_set_net *inst = ip_set_pernet(net); const struct ip_set *set; struct sk_buff *skb2; struct nlmsghdr *nlh2; @@ -1703,10 +1691,9 @@ static const struct nla_policy ip_set_type_policy[IPSET_ATTR_CMD_MAX + 1] = { [IPSET_ATTR_FAMILY] = { .type = NLA_U8 }, }; -static int -ip_set_type(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_type(struct net *net, struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { struct sk_buff *skb2; struct nlmsghdr *nlh2; @@ -1762,10 +1749,9 @@ ip_set_protocol_policy[IPSET_ATTR_CMD_MAX + 1] = { [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, }; -static int -ip_set_protocol(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const attr[]) +static int ip_set_protocol(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) { struct sk_buff *skb2; struct nlmsghdr *nlh2; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 9f5272968abb..dbb1bb3edb45 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1113,12 +1113,11 @@ static int ctnetlink_flush_conntrack(struct net *net, return 0; } -static int -ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_del_conntrack(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_tuple_hash *h; struct nf_conntrack_tuple tuple; struct nf_conn *ct; @@ -1168,12 +1167,11 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, return 0; } -static int -ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_get_conntrack(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_tuple_hash *h; struct nf_conntrack_tuple tuple; struct nf_conn *ct; @@ -1330,10 +1328,10 @@ ctnetlink_dump_dying(struct sk_buff *skb, struct netlink_callback *cb) return ctnetlink_dump_list(skb, cb, true); } -static int -ctnetlink_get_ct_dying(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_get_ct_dying(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -1352,10 +1350,10 @@ ctnetlink_dump_unconfirmed(struct sk_buff *skb, struct netlink_callback *cb) return ctnetlink_dump_list(skb, cb, false); } -static int -ctnetlink_get_ct_unconfirmed(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_get_ct_unconfirmed(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -1865,12 +1863,11 @@ err1: return ERR_PTR(err); } -static int -ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_new_conntrack(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_tuple otuple, rtuple; struct nf_conntrack_tuple_hash *h = NULL; struct nfgenmsg *nfmsg = nlmsg_data(nlh); @@ -2034,10 +2031,10 @@ ctnetlink_ct_stat_cpu_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static int -ctnetlink_stat_ct_cpu(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_stat_ct_cpu(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -2080,10 +2077,9 @@ nlmsg_failure: return -1; } -static int -ctnetlink_stat_ct(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_stat_ct(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { struct sk_buff *skb2; int err; @@ -2729,12 +2725,12 @@ out: return skb->len; } -static int ctnetlink_dump_exp_ct(struct sock *ctnl, struct sk_buff *skb, +static int ctnetlink_dump_exp_ct(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { int err; - struct net *net = sock_net(ctnl); struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; struct nf_conntrack_tuple tuple; @@ -2768,12 +2764,10 @@ static int ctnetlink_dump_exp_ct(struct sock *ctnl, struct sk_buff *skb, return err; } -static int -ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_get_expect(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_tuple tuple; struct nf_conntrack_expect *exp; struct sk_buff *skb2; @@ -2784,7 +2778,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (nlh->nlmsg_flags & NLM_F_DUMP) { if (cda[CTA_EXPECT_MASTER]) - return ctnetlink_dump_exp_ct(ctnl, skb, nlh, cda); + return ctnetlink_dump_exp_ct(net, ctnl, skb, nlh, cda); else { struct netlink_dump_control c = { .dump = ctnetlink_exp_dump_table, @@ -2850,12 +2844,10 @@ out: return err == -EAGAIN ? -ENOBUFS : err; } -static int -ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_del_expect(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_expect *exp; struct nf_conntrack_tuple tuple; struct nfgenmsg *nfmsg = nlmsg_data(nlh); @@ -3136,12 +3128,10 @@ err_ct: return err; } -static int -ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_new_expect(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(ctnl); struct nf_conntrack_tuple tuple; struct nf_conntrack_expect *exp; struct nfgenmsg *nfmsg = nlmsg_data(nlh); @@ -3242,10 +3232,10 @@ ctnetlink_exp_stat_cpu_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static int -ctnetlink_stat_exp_cpu(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int ctnetlink_stat_exp_cpu(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 28cbc457f1f3..69cb5be9a174 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -543,15 +543,14 @@ done: return skb->len; } -static int nf_tables_gettable(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_gettable(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); const struct nft_af_info *afi; const struct nft_table *table; struct sk_buff *skb2; - struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; int err; @@ -1097,8 +1096,8 @@ done: return skb->len; } -static int nf_tables_getchain(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_getchain(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); @@ -1106,7 +1105,6 @@ static int nf_tables_getchain(struct sock *nlsk, struct sk_buff *skb, const struct nft_table *table; const struct nft_chain *chain; struct sk_buff *skb2; - struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; int err; @@ -1923,8 +1921,8 @@ done: return skb->len; } -static int nf_tables_getrule(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_getrule(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); @@ -1933,7 +1931,6 @@ static int nf_tables_getrule(struct sock *nlsk, struct sk_buff *skb, const struct nft_chain *chain; const struct nft_rule *rule; struct sk_buff *skb2; - struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; int err; @@ -2604,11 +2601,10 @@ static int nf_tables_dump_sets_done(struct netlink_callback *cb) return 0; } -static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_getset(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { - struct net *net = sock_net(skb->sk); const struct nft_set *set; struct nft_ctx ctx; struct sk_buff *skb2; @@ -3190,11 +3186,10 @@ nla_put_failure: return -ENOSPC; } -static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_getsetelem(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { - struct net *net = sock_net(skb->sk); const struct nft_set *set; struct nft_ctx ctx; int err; @@ -3723,11 +3718,10 @@ err: return err; } -static int nf_tables_getgen(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int nf_tables_getgen(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { - struct net *net = sock_net(skb->sk); struct sk_buff *skb2; int err; diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 9ed453465167..7012154b28ca 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -206,7 +206,7 @@ replay: } if (nc->call_rcu) { - err = nc->call_rcu(net->nfnl, skb, nlh, + err = nc->call_rcu(net, net->nfnl, skb, nlh, (const struct nlattr **)cda); rcu_read_unlock(); } else { @@ -216,8 +216,8 @@ replay: nfnetlink_find_client(type, ss) != nc) err = -EAGAIN; else if (nc->call) - err = nc->call(net->nfnl, skb, nlh, - (const struct nlattr **)cda); + err = nc->call(net, net->nfnl, skb, nlh, + (const struct nlattr **)cda); else err = -EINVAL; nfnl_unlock(subsys_id); diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index fefbf5f0b28d..5274b04c42a6 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -46,12 +46,11 @@ struct nfacct_filter { #define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS | NFACCT_F_QUOTA_BYTES) #define NFACCT_OVERQUOTA_BIT 2 /* NFACCT_F_OVERQUOTA */ -static int -nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_acct_new(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { struct nf_acct *nfacct, *matching = NULL; - struct net *net = sock_net(nfnl); char *acct_name; unsigned int size = 0; u32 flags = 0; @@ -253,11 +252,10 @@ nfacct_filter_alloc(const struct nlattr * const attr) return filter; } -static int -nfnl_acct_get(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_acct_get(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { - struct net *net = sock_net(nfnl); int ret = -ENOENT; struct nf_acct *cur; char *acct_name; @@ -333,11 +331,10 @@ static int nfnl_acct_try_del(struct nf_acct *cur) return ret; } -static int -nfnl_acct_del(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_acct_del(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { - struct net *net = sock_net(nfnl); char *acct_name; struct nf_acct *cur; int ret = -ENOENT; diff --git a/net/netfilter/nfnetlink_cthelper.c b/net/netfilter/nfnetlink_cthelper.c index 54330fb5efaf..e924e95fcc7f 100644 --- a/net/netfilter/nfnetlink_cthelper.c +++ b/net/netfilter/nfnetlink_cthelper.c @@ -286,9 +286,9 @@ nfnl_cthelper_update(const struct nlattr * const tb[], return 0; } -static int -nfnl_cthelper_new(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_cthelper_new(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { const char *helper_name; struct nf_conntrack_helper *cur, *helper = NULL; @@ -498,9 +498,9 @@ out: return skb->len; } -static int -nfnl_cthelper_get(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_cthelper_get(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { int ret = -ENOENT, i; struct nf_conntrack_helper *cur; @@ -570,9 +570,9 @@ nfnl_cthelper_get(struct sock *nfnl, struct sk_buff *skb, return ret; } -static int -nfnl_cthelper_del(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_cthelper_del(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { char *helper_name = NULL; struct nf_conntrack_helper *cur; diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 3921d544f5ba..5d010f27ac01 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -65,16 +65,15 @@ ctnl_timeout_parse_policy(void *timeouts, struct nf_conntrack_l4proto *l4proto, return ret; } -static int -cttimeout_new_timeout(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int cttimeout_new_timeout(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { __u16 l3num; __u8 l4num; struct nf_conntrack_l4proto *l4proto; struct ctnl_timeout *timeout, *matching = NULL; - struct net *net = sock_net(skb->sk); char *name; int ret; @@ -239,12 +238,11 @@ ctnl_timeout_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static int -cttimeout_get_timeout(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int cttimeout_get_timeout(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(skb->sk); int ret = -ENOENT; char *name; struct ctnl_timeout *cur; @@ -339,15 +337,14 @@ static int ctnl_timeout_try_del(struct net *net, struct ctnl_timeout *timeout) return ret; } -static int -cttimeout_del_timeout(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int cttimeout_del_timeout(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { - struct net *net = sock_net(skb->sk); - char *name; struct ctnl_timeout *cur; int ret = -ENOENT; + char *name; if (!cda[CTA_TIMEOUT_NAME]) { list_for_each_entry(cur, &net->nfct_timeout_list, head) @@ -370,15 +367,14 @@ cttimeout_del_timeout(struct sock *ctnl, struct sk_buff *skb, return ret; } -static int -cttimeout_default_set(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const cda[]) +static int cttimeout_default_set(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const cda[]) { __u16 l3num; __u8 l4num; struct nf_conntrack_l4proto *l4proto; - struct net *net = sock_net(skb->sk); unsigned int *timeouts; int ret; @@ -460,14 +456,14 @@ nla_put_failure: return -1; } -static int cttimeout_default_get(struct sock *ctnl, struct sk_buff *skb, +static int cttimeout_default_get(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { __u16 l3num; __u8 l4num; struct nf_conntrack_l4proto *l4proto; - struct net *net = sock_net(skb->sk); struct sk_buff *skb2; int ret, err; diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 70b6bd3b781e..6a57f10a4e0b 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -785,10 +785,9 @@ static struct notifier_block nfulnl_rtnl_notifier = { .notifier_call = nfulnl_rcv_nl_event, }; -static int -nfulnl_recv_unsupp(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfqa[]) +static int nfulnl_recv_unsupp(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nfqa[]) { return -ENOTSUPP; } @@ -809,16 +808,14 @@ static const struct nla_policy nfula_cfg_policy[NFULA_CFG_MAX+1] = { [NFULA_CFG_FLAGS] = { .type = NLA_U16 }, }; -static int -nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfula[]) +static int nfulnl_recv_config(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nfula[]) { struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int16_t group_num = ntohs(nfmsg->res_id); struct nfulnl_instance *inst; struct nfulnl_msg_config_cmd *cmd = NULL; - struct net *net = sock_net(ctnl); struct nfnl_log_net *log = nfnl_log_pernet(net); int ret = 0; u16 flags = 0; diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 861c6615253b..3d1f16cf5cd0 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -957,10 +957,10 @@ static int nfq_id_after(unsigned int id, unsigned int max) return (int)(id - max) > 0; } -static int -nfqnl_recv_verdict_batch(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfqa[]) +static int nfqnl_recv_verdict_batch(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const nfqa[]) { struct nfgenmsg *nfmsg = nlmsg_data(nlh); struct nf_queue_entry *entry, *tmp; @@ -969,8 +969,6 @@ nfqnl_recv_verdict_batch(struct sock *ctnl, struct sk_buff *skb, struct nfqnl_instance *queue; LIST_HEAD(batch_list); u16 queue_num = ntohs(nfmsg->res_id); - - struct net *net = sock_net(ctnl); struct nfnl_queue_net *q = nfnl_queue_pernet(net); queue = verdict_instance_lookup(q, queue_num, @@ -1029,14 +1027,13 @@ static struct nf_conn *nfqnl_ct_parse(struct nfnl_ct_hook *nfnl_ct, return ct; } -static int -nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfqa[]) +static int nfqnl_recv_verdict(struct net *net, struct sock *ctnl, + struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const nfqa[]) { struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int16_t queue_num = ntohs(nfmsg->res_id); - struct nfqnl_msg_verdict_hdr *vhdr; struct nfqnl_instance *queue; unsigned int verdict; @@ -1044,8 +1041,6 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, enum ip_conntrack_info uninitialized_var(ctinfo); struct nfnl_ct_hook *nfnl_ct; struct nf_conn *ct = NULL; - - struct net *net = sock_net(ctnl); struct nfnl_queue_net *q = nfnl_queue_pernet(net); queue = instance_lookup(q, queue_num); @@ -1092,10 +1087,9 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, return 0; } -static int -nfqnl_recv_unsupp(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfqa[]) +static int nfqnl_recv_unsupp(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nfqa[]) { return -ENOTSUPP; } @@ -1110,16 +1104,14 @@ static const struct nf_queue_handler nfqh = { .nf_hook_drop = &nfqnl_nf_hook_drop, }; -static int -nfqnl_recv_config(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nfqa[]) +static int nfqnl_recv_config(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nfqa[]) { struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int16_t queue_num = ntohs(nfmsg->res_id); struct nfqnl_instance *queue; struct nfqnl_msg_config_cmd *cmd = NULL; - struct net *net = sock_net(ctnl); struct nfnl_queue_net *q = nfnl_queue_pernet(net); int ret = 0; diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 9c8fab00164b..454841baa4d0 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -519,9 +519,9 @@ nla_put_failure: return -1; } -static int -nfnl_compat_get(struct sock *nfnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, const struct nlattr * const tb[]) +static int nfnl_compat_get(struct net *net, struct sock *nfnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const tb[]) { int ret = 0, target; struct nfgenmsg *nfmsg; diff --git a/net/netfilter/xt_osf.c b/net/netfilter/xt_osf.c index df8801e02a32..4e3c3affd285 100644 --- a/net/netfilter/xt_osf.c +++ b/net/netfilter/xt_osf.c @@ -61,8 +61,8 @@ static const struct nla_policy xt_osf_policy[OSF_ATTR_MAX + 1] = { [OSF_ATTR_FINGER] = { .len = sizeof(struct xt_osf_user_finger) }, }; -static int xt_osf_add_callback(struct sock *ctnl, struct sk_buff *skb, - const struct nlmsghdr *nlh, +static int xt_osf_add_callback(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const osf_attrs[]) { struct xt_osf_user_finger *f; @@ -104,7 +104,8 @@ static int xt_osf_add_callback(struct sock *ctnl, struct sk_buff *skb, return err; } -static int xt_osf_remove_callback(struct sock *ctnl, struct sk_buff *skb, +static int xt_osf_remove_callback(struct net *net, struct sock *ctnl, + struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const osf_attrs[]) { -- cgit v1.2.3-59-g8ed1b From 5913beaf0d70f97135ed7191c028fd88b3848864 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 15 Dec 2015 19:41:57 +0100 Subject: netfilter: nfnetlink: pass down netns pointer to commit() and abort() callbacks Adapt callsites to avoid recurrent lookup of the netns pointer. Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nfnetlink.h | 4 ++-- net/netfilter/nf_tables_api.c | 6 ++---- net/netfilter/nfnetlink.c | 6 +++--- 3 files changed, 7 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index ceacbf5dcb73..ba0d9789eb6e 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -26,8 +26,8 @@ struct nfnetlink_subsystem { __u8 subsys_id; /* nfnetlink subsystem ID */ __u8 cb_count; /* number of callbacks */ const struct nfnl_callback *cb; /* callback for individual types */ - int (*commit)(struct sk_buff *skb); - int (*abort)(struct sk_buff *skb); + int (*commit)(struct net *net, struct sk_buff *skb); + int (*abort)(struct net *net, struct sk_buff *skb); }; int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 69cb5be9a174..f5c397158e29 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3865,9 +3865,8 @@ static void nf_tables_commit_release(struct nft_trans *trans) kfree(trans); } -static int nf_tables_commit(struct sk_buff *skb) +static int nf_tables_commit(struct net *net, struct sk_buff *skb) { - struct net *net = sock_net(skb->sk); struct nft_trans *trans, *next; struct nft_trans_elem *te; @@ -4002,9 +4001,8 @@ static void nf_tables_abort_release(struct nft_trans *trans) kfree(trans); } -static int nf_tables_abort(struct sk_buff *skb) +static int nf_tables_abort(struct net *net, struct sk_buff *skb) { - struct net *net = sock_net(skb->sk); struct nft_trans *trans, *next; struct nft_trans_elem *te; diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 7012154b28ca..a7ba23353dab 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -425,15 +425,15 @@ next: } done: if (status & NFNL_BATCH_REPLAY) { - ss->abort(oskb); + ss->abort(net, oskb); nfnl_err_reset(&err_list); nfnl_unlock(subsys_id); kfree_skb(skb); goto replay; } else if (status == NFNL_BATCH_DONE) { - ss->commit(oskb); + ss->commit(net, oskb); } else { - ss->abort(oskb); + ss->abort(net, oskb); } nfnl_err_deliver(&err_list, oskb); -- cgit v1.2.3-59-g8ed1b From 9afec6d3866b8451abcf1a7a1a381a3be6c83386 Mon Sep 17 00:00:00 2001 From: Christophe Ricard Date: Wed, 23 Dec 2015 23:45:18 +0100 Subject: nfc: netlink: HCI event connectivity implementation Add support for missing HCI event EVT_CONNECTIVITY and forward it to userspace. Signed-off-by: Christophe Ricard Signed-off-by: Samuel Ortiz --- include/net/nfc/nfc.h | 1 + net/nfc/core.c | 13 +++++++++++++ net/nfc/netlink.c | 37 +++++++++++++++++++++++++++++++++++++ net/nfc/nfc.h | 1 + 4 files changed, 52 insertions(+) (limited to 'include') diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index dcfcfc9c00bf..1a3de8b34ad2 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -299,6 +299,7 @@ void nfc_driver_failure(struct nfc_dev *dev, int err); int nfc_se_transaction(struct nfc_dev *dev, u8 se_idx, struct nfc_evt_transaction *evt_transaction); +int nfc_se_connectivity(struct nfc_dev *dev, u8 se_idx); int nfc_add_se(struct nfc_dev *dev, u32 se_idx, u16 type); int nfc_remove_se(struct nfc_dev *dev, u32 se_idx); struct nfc_se *nfc_find_se(struct nfc_dev *dev, u32 se_idx); diff --git a/net/nfc/core.c b/net/nfc/core.c index 1fe3d3b362c0..122bb81da918 100644 --- a/net/nfc/core.c +++ b/net/nfc/core.c @@ -953,6 +953,19 @@ out: } EXPORT_SYMBOL(nfc_se_transaction); +int nfc_se_connectivity(struct nfc_dev *dev, u8 se_idx) +{ + int rc; + + pr_debug("connectivity: %x\n", se_idx); + + device_lock(&dev->dev); + rc = nfc_genl_se_connectivity(dev, se_idx); + device_unlock(&dev->dev); + return rc; +} +EXPORT_SYMBOL(nfc_se_connectivity); + static void nfc_release(struct device *d) { struct nfc_dev *dev = to_nfc_dev(d); diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index f58c1fba1026..ea023b35f1c2 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -552,6 +552,43 @@ free_msg: return -EMSGSIZE; } +int nfc_genl_se_connectivity(struct nfc_dev *dev, u8 se_idx) +{ + struct nfc_se *se; + struct sk_buff *msg; + void *hdr; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + hdr = genlmsg_put(msg, 0, 0, &nfc_genl_family, 0, + NFC_EVENT_SE_CONNECTIVITY); + if (!hdr) + goto free_msg; + + se = nfc_find_se(dev, se_idx); + if (!se) + goto free_msg; + + if (nla_put_u32(msg, NFC_ATTR_DEVICE_INDEX, dev->idx) || + nla_put_u32(msg, NFC_ATTR_SE_INDEX, se_idx) || + nla_put_u8(msg, NFC_ATTR_SE_TYPE, se->type)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); +free_msg: + nlmsg_free(msg); + return -EMSGSIZE; +} + static int nfc_genl_send_device(struct sk_buff *msg, struct nfc_dev *dev, u32 portid, u32 seq, struct netlink_callback *cb, diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h index c20b784ad720..6c6f76b370b1 100644 --- a/net/nfc/nfc.h +++ b/net/nfc/nfc.h @@ -105,6 +105,7 @@ int nfc_genl_se_added(struct nfc_dev *dev, u32 se_idx, u16 type); int nfc_genl_se_removed(struct nfc_dev *dev, u32 se_idx); int nfc_genl_se_transaction(struct nfc_dev *dev, u8 se_idx, struct nfc_evt_transaction *evt_transaction); +int nfc_genl_se_connectivity(struct nfc_dev *dev, u8 se_idx); struct nfc_dev *nfc_get_device(unsigned int idx); -- cgit v1.2.3-59-g8ed1b From 7be4fb643ef2d1058b897ba9dbe17bf5ced04391 Mon Sep 17 00:00:00 2001 From: Christophe Ricard Date: Wed, 23 Dec 2015 23:45:24 +0100 Subject: nfc: microread: Fix header comment microread platform_data header had an NXP header. Signed-off-by: Christophe Ricard Signed-off-by: Samuel Ortiz --- include/linux/platform_data/microread.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/platform_data/microread.h b/include/linux/platform_data/microread.h index cfda59b226ee..ca13992089b8 100644 --- a/include/linux/platform_data/microread.h +++ b/include/linux/platform_data/microread.h @@ -1,5 +1,5 @@ /* - * Driver include for the PN544 NFC chip. + * Driver include for the Inside Secure microread NFC Chip. * * Copyright (C) 2011 Tieto Poland * Copyright (C) 2012 Intel Corporation. All rights reserved. -- cgit v1.2.3-59-g8ed1b From f3a4094558ddf8afa8bb58250d548e15e059c65a Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 30 Dec 2015 16:28:25 +0100 Subject: ethtool: Add phy statistics Ethernet PHYs can maintain statistics, for example errors while idle and receive errors. Add an ethtool mechanism to retrieve these statistics, using the same model as MAC statistics. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/phy.h | 6 ++++ include/uapi/linux/ethtool.h | 3 ++ net/core/ethtool.c | 81 +++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 89 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/phy.h b/include/linux/phy.h index 05fde31b6dc6..a89cb0eef911 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -589,6 +589,12 @@ struct phy_driver { int (*module_eeprom)(struct phy_device *dev, struct ethtool_eeprom *ee, u8 *data); + /* Get statistics from the phy using ethtool */ + int (*get_sset_count)(struct phy_device *dev); + void (*get_strings)(struct phy_device *dev, u8 *data); + void (*get_stats)(struct phy_device *dev, + struct ethtool_stats *stats, u64 *data); + struct device_driver driver; }; #define to_phy_driver(d) container_of(d, struct phy_driver, driver) diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index cd1629170103..57fa39005e79 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -542,6 +542,7 @@ struct ethtool_pauseparam { * now deprecated * @ETH_SS_FEATURES: Device feature names * @ETH_SS_RSS_HASH_FUNCS: RSS hush function names + * @ETH_SS_PHY_STATS: Statistic names, for use with %ETHTOOL_GPHYSTATS */ enum ethtool_stringset { ETH_SS_TEST = 0, @@ -551,6 +552,7 @@ enum ethtool_stringset { ETH_SS_FEATURES, ETH_SS_RSS_HASH_FUNCS, ETH_SS_TUNABLES, + ETH_SS_PHY_STATS, }; /** @@ -1225,6 +1227,7 @@ enum ethtool_sfeatures_retval_bits { #define ETHTOOL_SRSSH 0x00000047 /* Set RX flow hash configuration */ #define ETHTOOL_GTUNABLE 0x00000048 /* Get tunable configuration */ #define ETHTOOL_STUNABLE 0x00000049 /* Set tunable configuration */ +#define ETHTOOL_GPHYSTATS 0x0000004a /* get PHY-specific statistics */ /* compatibility with older code */ #define SPARC_ETH_GSET ETHTOOL_GSET diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 09948a726347..daf04709dd3c 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -191,6 +191,23 @@ static int ethtool_set_features(struct net_device *dev, void __user *useraddr) return ret; } +static int phy_get_sset_count(struct phy_device *phydev) +{ + int ret; + + if (phydev->drv->get_sset_count && + phydev->drv->get_strings && + phydev->drv->get_stats) { + mutex_lock(&phydev->lock); + ret = phydev->drv->get_sset_count(phydev); + mutex_unlock(&phydev->lock); + + return ret; + } + + return -EOPNOTSUPP; +} + static int __ethtool_get_sset_count(struct net_device *dev, int sset) { const struct ethtool_ops *ops = dev->ethtool_ops; @@ -204,6 +221,13 @@ static int __ethtool_get_sset_count(struct net_device *dev, int sset) if (sset == ETH_SS_TUNABLES) return ARRAY_SIZE(tunable_strings); + if (sset == ETH_SS_PHY_STATS) { + if (dev->phydev) + return phy_get_sset_count(dev->phydev); + else + return -EOPNOTSUPP; + } + if (ops->get_sset_count && ops->get_strings) return ops->get_sset_count(dev, sset); else @@ -223,7 +247,17 @@ static void __ethtool_get_strings(struct net_device *dev, sizeof(rss_hash_func_strings)); else if (stringset == ETH_SS_TUNABLES) memcpy(data, tunable_strings, sizeof(tunable_strings)); - else + else if (stringset == ETH_SS_PHY_STATS) { + struct phy_device *phydev = dev->phydev; + + if (phydev) { + mutex_lock(&phydev->lock); + phydev->drv->get_strings(phydev, data); + mutex_unlock(&phydev->lock); + } else { + return; + } + } else /* ops->get_strings is valid because checked earlier */ ops->get_strings(dev, stringset, data); } @@ -1401,6 +1435,47 @@ static int ethtool_get_stats(struct net_device *dev, void __user *useraddr) return ret; } +static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) +{ + struct ethtool_stats stats; + struct phy_device *phydev = dev->phydev; + u64 *data; + int ret, n_stats; + + if (!phydev) + return -EOPNOTSUPP; + + n_stats = phy_get_sset_count(phydev); + + if (n_stats < 0) + return n_stats; + WARN_ON(n_stats == 0); + + if (copy_from_user(&stats, useraddr, sizeof(stats))) + return -EFAULT; + + stats.n_stats = n_stats; + data = kmalloc_array(n_stats, sizeof(u64), GFP_USER); + if (!data) + return -ENOMEM; + + mutex_lock(&phydev->lock); + phydev->drv->get_stats(phydev, &stats, data); + mutex_unlock(&phydev->lock); + + ret = -EFAULT; + if (copy_to_user(useraddr, &stats, sizeof(stats))) + goto out; + useraddr += sizeof(stats); + if (copy_to_user(useraddr, data, stats.n_stats * sizeof(u64))) + goto out; + ret = 0; + + out: + kfree(data); + return ret; +} + static int ethtool_get_perm_addr(struct net_device *dev, void __user *useraddr) { struct ethtool_perm_addr epaddr; @@ -1779,6 +1854,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GSSET_INFO: case ETHTOOL_GSTRINGS: case ETHTOOL_GSTATS: + case ETHTOOL_GPHYSTATS: case ETHTOOL_GTSO: case ETHTOOL_GPERMADDR: case ETHTOOL_GUFO: @@ -1991,6 +2067,9 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_STUNABLE: rc = ethtool_set_tunable(dev, useraddr); break; + case ETHTOOL_GPHYSTATS: + rc = ethtool_get_phy_stats(dev, useraddr); + break; default: rc = -EOPNOTSUPP; } -- cgit v1.2.3-59-g8ed1b From c7862a5f0de5f521c545f3436f0aa190964342dd Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 28 Dec 2015 18:21:44 +0100 Subject: netfilter: nft_limit: allow to invert matching criteria This patch allows you to invert the ratelimit matching criteria, so you can match packets over the ratelimit. This is required to support what hashlimit does. Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 6 ++++++ net/netfilter/nft_limit.c | 16 +++++++++++++--- 2 files changed, 19 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index b48a3ab761f8..22043ce95ae6 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -780,6 +780,10 @@ enum nft_limit_type { NFT_LIMIT_PKT_BYTES }; +enum nft_limit_flags { + NFT_LIMIT_F_INV = (1 << 0), +}; + /** * enum nft_limit_attributes - nf_tables limit expression netlink attributes * @@ -787,6 +791,7 @@ enum nft_limit_type { * @NFTA_LIMIT_UNIT: refill unit (NLA_U64) * @NFTA_LIMIT_BURST: burst (NLA_U32) * @NFTA_LIMIT_TYPE: type of limit (NLA_U32: enum nft_limit_type) + * @NFTA_LIMIT_FLAGS: flags (NLA_U32: enum nft_limit_flags) */ enum nft_limit_attributes { NFTA_LIMIT_UNSPEC, @@ -794,6 +799,7 @@ enum nft_limit_attributes { NFTA_LIMIT_UNIT, NFTA_LIMIT_BURST, NFTA_LIMIT_TYPE, + NFTA_LIMIT_FLAGS, __NFTA_LIMIT_MAX }; #define NFTA_LIMIT_MAX (__NFTA_LIMIT_MAX - 1) diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c index 5d67938f8b2f..99d18578afc6 100644 --- a/net/netfilter/nft_limit.c +++ b/net/netfilter/nft_limit.c @@ -26,6 +26,7 @@ struct nft_limit { u64 rate; u64 nsecs; u32 burst; + bool invert; }; static inline bool nft_limit_eval(struct nft_limit *limit, u64 cost) @@ -44,11 +45,11 @@ static inline bool nft_limit_eval(struct nft_limit *limit, u64 cost) if (delta >= 0) { limit->tokens = delta; spin_unlock_bh(&limit_lock); - return false; + return limit->invert; } limit->tokens = tokens; spin_unlock_bh(&limit_lock); - return true; + return !limit->invert; } static int nft_limit_init(struct nft_limit *limit, @@ -78,6 +79,12 @@ static int nft_limit_init(struct nft_limit *limit, limit->rate = rate; } + if (tb[NFTA_LIMIT_FLAGS]) { + u32 flags = ntohl(nla_get_be32(tb[NFTA_LIMIT_FLAGS])); + + if (flags & NFT_LIMIT_F_INV) + limit->invert = true; + } limit->last = ktime_get_ns(); return 0; @@ -86,13 +93,15 @@ static int nft_limit_init(struct nft_limit *limit, static int nft_limit_dump(struct sk_buff *skb, const struct nft_limit *limit, enum nft_limit_type type) { + 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)) || nla_put_be64(skb, NFTA_LIMIT_UNIT, cpu_to_be64(secs)) || nla_put_be32(skb, NFTA_LIMIT_BURST, htonl(limit->burst)) || - nla_put_be32(skb, NFTA_LIMIT_TYPE, htonl(type))) + nla_put_be32(skb, NFTA_LIMIT_TYPE, htonl(type)) || + nla_put_be32(skb, NFTA_LIMIT_FLAGS, htonl(flags))) goto nla_put_failure; return 0; @@ -120,6 +129,7 @@ static const struct nla_policy nft_limit_policy[NFTA_LIMIT_MAX + 1] = { [NFTA_LIMIT_UNIT] = { .type = NLA_U64 }, [NFTA_LIMIT_BURST] = { .type = NLA_U32 }, [NFTA_LIMIT_TYPE] = { .type = NLA_U32 }, + [NFTA_LIMIT_FLAGS] = { .type = NLA_U32 }, }; static int nft_limit_pkts_init(const struct nft_ctx *ctx, -- cgit v1.2.3-59-g8ed1b From 502061f81d3eb4518d2e72178e494a8547788ad0 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 3 Jan 2016 21:02:18 +0100 Subject: netfilter: nf_tables: add packet duplication to the netdev family You can use this to duplicate packets and inject them at the egress path of the specified interface. This duplication allows you to inspect traffic from the dummy or any other interface dedicated to this purpose. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_dup_netdev.h | 6 +++ net/netfilter/Kconfig | 16 ++++++ net/netfilter/Makefile | 6 +++ net/netfilter/nf_dup_netdev.c | 40 +++++++++++++++ net/netfilter/nft_dup_netdev.c | 97 +++++++++++++++++++++++++++++++++++ 5 files changed, 165 insertions(+) create mode 100644 include/net/netfilter/nf_dup_netdev.h create mode 100644 net/netfilter/nf_dup_netdev.c create mode 100644 net/netfilter/nft_dup_netdev.c (limited to 'include') diff --git a/include/net/netfilter/nf_dup_netdev.h b/include/net/netfilter/nf_dup_netdev.h new file mode 100644 index 000000000000..397dcae349f9 --- /dev/null +++ b/include/net/netfilter/nf_dup_netdev.h @@ -0,0 +1,6 @@ +#ifndef _NF_DUP_NETDEV_H_ +#define _NF_DUP_NETDEV_H_ + +void nf_dup_netdev_egress(const struct nft_pktinfo *pkt, int oif); + +#endif diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 4692782b5280..8514cc4b22a8 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -563,6 +563,22 @@ config NFT_COMPAT x_tables match/target extensions over the nf_tables framework. +if NF_TABLES_NETDEV + +config NF_DUP_NETDEV + tristate "Netfilter packet duplication support" + help + This option enables the generic packet duplication infrastructure + for Netfilter. + +config NFT_DUP_NETDEV + tristate "Netfilter nf_tables netdev packet duplication support" + select NF_DUP_NETDEV + help + This option enables packet duplication for the "netdev" family. + +endif # NF_TABLES_NETDEV + endif # NF_TABLES config NETFILTER_XTABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 22934846b5d1..5c9913359537 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -66,6 +66,9 @@ obj-$(CONFIG_NF_NAT_TFTP) += nf_nat_tftp.o # SYNPROXY obj-$(CONFIG_NETFILTER_SYNPROXY) += nf_synproxy_core.o +# generic packet duplication from netdev family +obj-$(CONFIG_NF_DUP_NETDEV) += nf_dup_netdev.o + # nf_tables nf_tables-objs += nf_tables_core.o nf_tables_api.o nf_tables_trace.o nf_tables-objs += nft_immediate.o nft_cmp.o nft_lookup.o nft_dynset.o @@ -90,6 +93,9 @@ obj-$(CONFIG_NFT_LOG) += nft_log.o obj-$(CONFIG_NFT_MASQ) += nft_masq.o obj-$(CONFIG_NFT_REDIR) += nft_redir.o +# nf_tables netdev +obj-$(CONFIG_NFT_DUP_NETDEV) += nft_dup_netdev.o + # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_dup_netdev.c b/net/netfilter/nf_dup_netdev.c new file mode 100644 index 000000000000..8414ee1a0319 --- /dev/null +++ b/net/netfilter/nf_dup_netdev.c @@ -0,0 +1,40 @@ +/* + * Copyright (c) 2015 Pablo Neira Ayuso + * + * 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 +#include +#include +#include +#include +#include +#include + +void nf_dup_netdev_egress(const struct nft_pktinfo *pkt, int oif) +{ + struct net_device *dev; + struct sk_buff *skb; + + dev = dev_get_by_index_rcu(pkt->net, oif); + if (dev == NULL) + return; + + skb = skb_clone(pkt->skb, GFP_ATOMIC); + if (skb == NULL) + return; + + if (skb_mac_header_was_set(skb)) + skb_push(skb, skb->mac_len); + + skb->dev = dev; + skb_sender_cpu_clear(skb); + dev_queue_xmit(skb); +} +EXPORT_SYMBOL_GPL(nf_dup_netdev_egress); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Pablo Neira Ayuso "); diff --git a/net/netfilter/nft_dup_netdev.c b/net/netfilter/nft_dup_netdev.c new file mode 100644 index 000000000000..2cc1e0ef56e8 --- /dev/null +++ b/net/netfilter/nft_dup_netdev.c @@ -0,0 +1,97 @@ +/* + * Copyright (c) 2015 Pablo Neira Ayuso + * + * 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 +#include +#include +#include +#include +#include +#include +#include + +struct nft_dup_netdev { + enum nft_registers sreg_dev:8; +}; + +static void nft_dup_netdev_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + struct nft_dup_netdev *priv = nft_expr_priv(expr); + int oif = regs->data[priv->sreg_dev]; + + nf_dup_netdev_egress(pkt, oif); +} + +static const struct nla_policy nft_dup_netdev_policy[NFTA_DUP_MAX + 1] = { + [NFTA_DUP_SREG_DEV] = { .type = NLA_U32 }, +}; + +static int nft_dup_netdev_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_dup_netdev *priv = nft_expr_priv(expr); + + if (tb[NFTA_DUP_SREG_DEV] == NULL) + return -EINVAL; + + priv->sreg_dev = nft_parse_register(tb[NFTA_DUP_SREG_DEV]); + return nft_validate_register_load(priv->sreg_dev, sizeof(int)); +} + +static const struct nft_expr_ops nft_dup_netdev_ingress_ops; + +static int nft_dup_netdev_dump(struct sk_buff *skb, const struct nft_expr *expr) +{ + struct nft_dup_netdev *priv = nft_expr_priv(expr); + + if (nft_dump_register(skb, NFTA_DUP_SREG_DEV, priv->sreg_dev)) + goto nla_put_failure; + + return 0; + +nla_put_failure: + return -1; +} + +static struct nft_expr_type nft_dup_netdev_type; +static const struct nft_expr_ops nft_dup_netdev_ops = { + .type = &nft_dup_netdev_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_dup_netdev)), + .eval = nft_dup_netdev_eval, + .init = nft_dup_netdev_init, + .dump = nft_dup_netdev_dump, +}; + +static struct nft_expr_type nft_dup_netdev_type __read_mostly = { + .family = NFPROTO_NETDEV, + .name = "dup", + .ops = &nft_dup_netdev_ops, + .policy = nft_dup_netdev_policy, + .maxattr = NFTA_DUP_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_dup_netdev_module_init(void) +{ + return nft_register_expr(&nft_dup_netdev_type); +} + +static void __exit nft_dup_netdev_module_exit(void) +{ + nft_unregister_expr(&nft_dup_netdev_type); +} + +module_init(nft_dup_netdev_module_init); +module_exit(nft_dup_netdev_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Pablo Neira Ayuso "); +MODULE_ALIAS_NFT_AF_EXPR(5, "dup"); -- cgit v1.2.3-59-g8ed1b From 39e6dea28adc874f7021e5580c13cab0b58407ea Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Nov 2015 13:39:38 +0100 Subject: netfilter: nf_tables: add forward expression to the netdev family You can use this to forward packets from ingress to the egress path of the specified interface. This provides a fast path to bounce packets from one interface to another specific destination interface. Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 12 ++++ net/netfilter/Kconfig | 6 ++ net/netfilter/Makefile | 1 + net/netfilter/nft_fwd_netdev.c | 98 ++++++++++++++++++++++++++++++++ 4 files changed, 117 insertions(+) create mode 100644 net/netfilter/nft_fwd_netdev.c (limited to 'include') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 22043ce95ae6..731288a039f6 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -983,6 +983,18 @@ enum nft_dup_attributes { }; #define NFTA_DUP_MAX (__NFTA_DUP_MAX - 1) +/** + * enum nft_fwd_attributes - nf_tables fwd expression netlink attributes + * + * @NFTA_FWD_SREG_DEV: source register of output interface (NLA_U32: nft_register) + */ +enum nft_fwd_attributes { + NFTA_FWD_UNSPEC, + NFTA_FWD_SREG_DEV, + __NFTA_FWD_MAX +}; +#define NFTA_FWD_MAX (__NFTA_FWD_MAX - 1) + /** * enum nft_gen_attributes - nf_tables ruleset generation attributes * diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 8514cc4b22a8..8c067e6663a1 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -577,6 +577,12 @@ config NFT_DUP_NETDEV help This option enables packet duplication for the "netdev" family. +config NFT_FWD_NETDEV + tristate "Netfilter nf_tables netdev packet forwarding support" + select NF_DUP_NETDEV + help + This option enables packet forwarding for the "netdev" family. + endif # NF_TABLES_NETDEV endif # NF_TABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 5c9913359537..69134541d65b 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -95,6 +95,7 @@ obj-$(CONFIG_NFT_REDIR) += nft_redir.o # nf_tables netdev obj-$(CONFIG_NFT_DUP_NETDEV) += nft_dup_netdev.o +obj-$(CONFIG_NFT_FWD_NETDEV) += nft_fwd_netdev.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nft_fwd_netdev.c b/net/netfilter/nft_fwd_netdev.c new file mode 100644 index 000000000000..763ebc3e0b2b --- /dev/null +++ b/net/netfilter/nft_fwd_netdev.c @@ -0,0 +1,98 @@ +/* + * Copyright (c) 2015 Pablo Neira Ayuso + * + * 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 +#include +#include +#include +#include +#include +#include +#include + +struct nft_fwd_netdev { + enum nft_registers sreg_dev:8; +}; + +static void nft_fwd_netdev_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) +{ + struct nft_fwd_netdev *priv = nft_expr_priv(expr); + int oif = regs->data[priv->sreg_dev]; + + nf_dup_netdev_egress(pkt, oif); + regs->verdict.code = NF_DROP; +} + +static const struct nla_policy nft_fwd_netdev_policy[NFTA_FWD_MAX + 1] = { + [NFTA_FWD_SREG_DEV] = { .type = NLA_U32 }, +}; + +static int nft_fwd_netdev_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_fwd_netdev *priv = nft_expr_priv(expr); + + if (tb[NFTA_FWD_SREG_DEV] == NULL) + return -EINVAL; + + priv->sreg_dev = nft_parse_register(tb[NFTA_FWD_SREG_DEV]); + return nft_validate_register_load(priv->sreg_dev, sizeof(int)); +} + +static const struct nft_expr_ops nft_fwd_netdev_ingress_ops; + +static int nft_fwd_netdev_dump(struct sk_buff *skb, const struct nft_expr *expr) +{ + struct nft_fwd_netdev *priv = nft_expr_priv(expr); + + if (nft_dump_register(skb, NFTA_FWD_SREG_DEV, priv->sreg_dev)) + goto nla_put_failure; + + return 0; + +nla_put_failure: + return -1; +} + +static struct nft_expr_type nft_fwd_netdev_type; +static const struct nft_expr_ops nft_fwd_netdev_ops = { + .type = &nft_fwd_netdev_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_fwd_netdev)), + .eval = nft_fwd_netdev_eval, + .init = nft_fwd_netdev_init, + .dump = nft_fwd_netdev_dump, +}; + +static struct nft_expr_type nft_fwd_netdev_type __read_mostly = { + .family = NFPROTO_NETDEV, + .name = "fwd", + .ops = &nft_fwd_netdev_ops, + .policy = nft_fwd_netdev_policy, + .maxattr = NFTA_FWD_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_fwd_netdev_module_init(void) +{ + return nft_register_expr(&nft_fwd_netdev_type); +} + +static void __exit nft_fwd_netdev_module_exit(void) +{ + nft_unregister_expr(&nft_fwd_netdev_type); +} + +module_init(nft_fwd_netdev_module_init); +module_exit(nft_fwd_netdev_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Pablo Neira Ayuso "); +MODULE_ALIAS_NFT_AF_EXPR(5, "fwd"); -- cgit v1.2.3-59-g8ed1b From 29663b0cc1d5b9b6e2f6caf41e86c599a0310def Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Wed, 23 Dec 2015 22:36:32 +0100 Subject: mac802154: constify ieee802154_llsec_ops structure The ieee802154_llsec_ops structure is never modified, so declare it as const. Done with the help of Coccinelle. Signed-off-by: Julia Lawall Acked-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/net/ieee802154_netdev.h | 2 +- net/mac802154/mac_cmd.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ieee802154_netdev.h b/include/net/ieee802154_netdev.h index a62a051a3a2f..c4b31601cd53 100644 --- a/include/net/ieee802154_netdev.h +++ b/include/net/ieee802154_netdev.h @@ -337,7 +337,7 @@ struct ieee802154_mlme_ops { void (*get_mac_params)(struct net_device *dev, struct ieee802154_mac_params *params); - struct ieee802154_llsec_ops *llsec; + const struct ieee802154_llsec_ops *llsec; }; static inline struct ieee802154_mlme_ops * diff --git a/net/mac802154/mac_cmd.c b/net/mac802154/mac_cmd.c index 8606da459ff3..3db16346cab3 100644 --- a/net/mac802154/mac_cmd.c +++ b/net/mac802154/mac_cmd.c @@ -126,7 +126,7 @@ static void mac802154_get_mac_params(struct net_device *dev, params->lbt = wpan_dev->lbt; } -static struct ieee802154_llsec_ops mac802154_llsec_ops = { +static const struct ieee802154_llsec_ops mac802154_llsec_ops = { .get_params = mac802154_get_params, .set_params = mac802154_set_params, .add_key = mac802154_add_key, -- cgit v1.2.3-59-g8ed1b From 888cc8c20cf265fcd1302f6c5d6be07628ba66c7 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Mon, 28 Dec 2015 02:07:08 +0300 Subject: sh_eth: remove EDMAC_BIG_ENDIAN Commit 71557a37adb5 ("[netdrvr] sh_eth: Add SH7619 support") added support for the big-endian EDMAC descriptors. However, it was never used and never worked right until the recent driver fixes. I think we now can just remove this support, it was only burdening the driver from the start. It should be easy to do without disturbing the SH platform code, at least for now... Signed-off-by: Sergei Shtylyov Acked-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/sh_eth.c | 18 ++---------------- drivers/net/ethernet/renesas/sh_eth.h | 1 - include/linux/sh_eth.h | 2 +- 3 files changed, 3 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 8ca040113a09..2c7dd8a68d84 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -971,24 +971,12 @@ static void sh_eth_set_receive_align(struct sk_buff *skb) /* CPU <-> EDMAC endian convert */ static inline __u32 cpu_to_edmac(struct sh_eth_private *mdp, u32 x) { - switch (mdp->edmac_endian) { - case EDMAC_LITTLE_ENDIAN: - return cpu_to_le32(x); - case EDMAC_BIG_ENDIAN: - return cpu_to_be32(x); - } - return x; + return cpu_to_le32(x); } static inline __u32 edmac_to_cpu(struct sh_eth_private *mdp, u32 x) { - switch (mdp->edmac_endian) { - case EDMAC_LITTLE_ENDIAN: - return le32_to_cpu(x); - case EDMAC_BIG_ENDIAN: - return be32_to_cpu(x); - } - return x; + return le32_to_cpu(x); } /* Program the hardware MAC address from dev->dev_addr. */ @@ -3097,8 +3085,6 @@ static int sh_eth_drv_probe(struct platform_device *pdev) /* get PHY ID */ mdp->phy_id = pd->phy; mdp->phy_interface = pd->phy_interface; - /* EDMAC endian */ - mdp->edmac_endian = pd->edmac_endian; mdp->no_ether_link = pd->no_ether_link; mdp->ether_link_active_low = pd->ether_link_active_low; diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h index 72fcfc924589..8fa4ef3a7fdd 100644 --- a/drivers/net/ethernet/renesas/sh_eth.h +++ b/drivers/net/ethernet/renesas/sh_eth.h @@ -513,7 +513,6 @@ struct sh_eth_private { u32 cur_rx, dirty_rx; /* Producer/consumer ring indices */ u32 cur_tx, dirty_tx; u32 rx_buf_sz; /* Based on MTU+slack. */ - int edmac_endian; struct napi_struct napi; bool irq_enabled; /* MII transceiver section. */ diff --git a/include/linux/sh_eth.h b/include/linux/sh_eth.h index 8c9131db2b25..f2e27e078362 100644 --- a/include/linux/sh_eth.h +++ b/include/linux/sh_eth.h @@ -4,7 +4,7 @@ #include #include -enum {EDMAC_LITTLE_ENDIAN, EDMAC_BIG_ENDIAN}; +enum {EDMAC_LITTLE_ENDIAN}; struct sh_eth_plat_data { int phy; -- cgit v1.2.3-59-g8ed1b From 0efeff2905d0ea14f2ee83e6cefbda35ee8cf6fc Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 1 Jan 2016 13:18:48 +0100 Subject: net: make ip6tunnel_xmit definition conditional Moving the caller of iptunnel_xmit_stats causes a build error in randconfig builds that disable CONFIG_INET: In file included from ../net/xfrm/xfrm_input.c:17:0: ../include/net/ip6_tunnel.h: In function 'ip6tunnel_xmit': ../include/net/ip6_tunnel.h:93:2: error: implicit declaration of function 'iptunnel_xmit_stats' [-Werror=implicit-function-declaration] iptunnel_xmit_stats(dev, pkt_len); The reason is that the iptunnel_xmit_stats definition is hidden inside #ifdef CONFIG_INET but the caller is not. We can change one or the other to fix it, and this patch adds a second #ifdef around ip6tunnel_xmit() to avoid seeing the invalid call. Signed-off-by: Arnd Bergmann Fixes: 039f50629b7f ("ip_tunnel: Move stats update to iptunnel_xmit()") Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- include/net/ip6_tunnel.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index ae07e94778d8..0d0ce0b2d870 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -81,6 +81,7 @@ __u32 ip6_tnl_get_cap(struct ip6_tnl *t, const struct in6_addr *laddr, struct net *ip6_tnl_get_link_net(const struct net_device *dev); int ip6_tnl_get_iflink(const struct net_device *dev); +#ifdef CONFIG_INET static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb, struct net_device *dev) { @@ -93,3 +94,4 @@ static inline void ip6tunnel_xmit(struct sock *sk, struct sk_buff *skb, iptunnel_xmit_stats(dev, pkt_len); } #endif +#endif -- cgit v1.2.3-59-g8ed1b From ef456144da8ef507c8cf504284b6042e9201a05c Mon Sep 17 00:00:00 2001 From: Craig Gallek Date: Mon, 4 Jan 2016 17:41:45 -0500 Subject: soreuseport: define reuseport groups struct sock_reuseport is an optional shared structure referenced by each socket belonging to a reuseport group. When a socket is bound to an address/port not yet in use and the reuseport flag has been set, the structure will be allocated and attached to the newly bound socket. When subsequent calls to bind are made for the same address/port, the shared structure will be updated to include the new socket and the newly bound socket will reference the group structure. Usually, when an incoming packet was destined for a reuseport group, all sockets in the same group needed to be considered before a dispatching decision was made. With this structure, an appropriate socket can be found after looking up just one socket in the group. This shared structure will also allow for more complicated decisions to be made when selecting a socket (eg a BPF filter). This work is based off a similar implementation written by Ying Cai for implementing policy-based reuseport selection. Signed-off-by: Craig Gallek Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 2 + include/net/sock_reuseport.h | 20 +++++ net/core/Makefile | 2 +- net/core/sock_reuseport.c | 173 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 196 insertions(+), 1 deletion(-) create mode 100644 include/net/sock_reuseport.h create mode 100644 net/core/sock_reuseport.c (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 3794cdde837a..e830c1006935 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -318,6 +318,7 @@ struct cg_proto; * @sk_error_report: callback to indicate errors (e.g. %MSG_ERRQUEUE) * @sk_backlog_rcv: callback to process the backlog * @sk_destruct: called at sock freeing time, i.e. when all refcnt == 0 + * @sk_reuseport_cb: reuseport group container */ struct sock { /* @@ -453,6 +454,7 @@ struct sock { int (*sk_backlog_rcv)(struct sock *sk, struct sk_buff *skb); void (*sk_destruct)(struct sock *sk); + struct sock_reuseport __rcu *sk_reuseport_cb; }; #define __sk_user_data(sk) ((*((void __rcu **)&(sk)->sk_user_data))) diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h new file mode 100644 index 000000000000..67d1eb8fd7af --- /dev/null +++ b/include/net/sock_reuseport.h @@ -0,0 +1,20 @@ +#ifndef _SOCK_REUSEPORT_H +#define _SOCK_REUSEPORT_H + +#include +#include + +struct sock_reuseport { + struct rcu_head rcu; + + u16 max_socks; /* length of socks */ + u16 num_socks; /* elements in socks */ + struct sock *socks[0]; /* array of sock pointers */ +}; + +extern int reuseport_alloc(struct sock *sk); +extern int reuseport_add_sock(struct sock *sk, const struct sock *sk2); +extern void reuseport_detach_sock(struct sock *sk); +extern struct sock *reuseport_select_sock(struct sock *sk, u32 hash); + +#endif /* _SOCK_REUSEPORT_H */ diff --git a/net/core/Makefile b/net/core/Makefile index 086b01fbe1bd..0b835de04de3 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -9,7 +9,7 @@ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o obj-y += dev.o ethtool.o dev_addr_lists.o dst.o netevent.o \ neighbour.o rtnetlink.o utils.o link_watch.o filter.o \ - sock_diag.o dev_ioctl.o tso.o + sock_diag.o dev_ioctl.o tso.o sock_reuseport.o obj-$(CONFIG_XFRM) += flow.o obj-y += net-sysfs.o diff --git a/net/core/sock_reuseport.c b/net/core/sock_reuseport.c new file mode 100644 index 000000000000..963c8d5f3027 --- /dev/null +++ b/net/core/sock_reuseport.c @@ -0,0 +1,173 @@ +/* + * To speed up listener socket lookup, create an array to store all sockets + * listening on the same port. This allows a decision to be made after finding + * the first socket. + */ + +#include +#include + +#define INIT_SOCKS 128 + +static DEFINE_SPINLOCK(reuseport_lock); + +static struct sock_reuseport *__reuseport_alloc(u16 max_socks) +{ + size_t size = sizeof(struct sock_reuseport) + + sizeof(struct sock *) * max_socks; + struct sock_reuseport *reuse = kzalloc(size, GFP_ATOMIC); + + if (!reuse) + return NULL; + + reuse->max_socks = max_socks; + + return reuse; +} + +int reuseport_alloc(struct sock *sk) +{ + struct sock_reuseport *reuse; + + /* bh lock used since this function call may precede hlist lock in + * soft irq of receive path or setsockopt from process context + */ + spin_lock_bh(&reuseport_lock); + WARN_ONCE(rcu_dereference_protected(sk->sk_reuseport_cb, + lockdep_is_held(&reuseport_lock)), + "multiple allocations for the same socket"); + reuse = __reuseport_alloc(INIT_SOCKS); + if (!reuse) { + spin_unlock_bh(&reuseport_lock); + return -ENOMEM; + } + + reuse->socks[0] = sk; + reuse->num_socks = 1; + rcu_assign_pointer(sk->sk_reuseport_cb, reuse); + + spin_unlock_bh(&reuseport_lock); + + return 0; +} +EXPORT_SYMBOL(reuseport_alloc); + +static struct sock_reuseport *reuseport_grow(struct sock_reuseport *reuse) +{ + struct sock_reuseport *more_reuse; + u32 more_socks_size, i; + + more_socks_size = reuse->max_socks * 2U; + if (more_socks_size > U16_MAX) + return NULL; + + more_reuse = __reuseport_alloc(more_socks_size); + if (!more_reuse) + return NULL; + + more_reuse->max_socks = more_socks_size; + more_reuse->num_socks = reuse->num_socks; + + memcpy(more_reuse->socks, reuse->socks, + reuse->num_socks * sizeof(struct sock *)); + + for (i = 0; i < reuse->num_socks; ++i) + rcu_assign_pointer(reuse->socks[i]->sk_reuseport_cb, + more_reuse); + + kfree_rcu(reuse, rcu); + return more_reuse; +} + +/** + * reuseport_add_sock - Add a socket to the reuseport group of another. + * @sk: New socket to add to the group. + * @sk2: Socket belonging to the existing reuseport group. + * May return ENOMEM and not add socket to group under memory pressure. + */ +int reuseport_add_sock(struct sock *sk, const struct sock *sk2) +{ + struct sock_reuseport *reuse; + + spin_lock_bh(&reuseport_lock); + reuse = rcu_dereference_protected(sk2->sk_reuseport_cb, + lockdep_is_held(&reuseport_lock)), + WARN_ONCE(rcu_dereference_protected(sk->sk_reuseport_cb, + lockdep_is_held(&reuseport_lock)), + "socket already in reuseport group"); + + if (reuse->num_socks == reuse->max_socks) { + reuse = reuseport_grow(reuse); + if (!reuse) { + spin_unlock_bh(&reuseport_lock); + return -ENOMEM; + } + } + + reuse->socks[reuse->num_socks] = sk; + /* paired with smp_rmb() in reuseport_select_sock() */ + smp_wmb(); + reuse->num_socks++; + rcu_assign_pointer(sk->sk_reuseport_cb, reuse); + + spin_unlock_bh(&reuseport_lock); + + return 0; +} +EXPORT_SYMBOL(reuseport_add_sock); + +void reuseport_detach_sock(struct sock *sk) +{ + struct sock_reuseport *reuse; + int i; + + spin_lock_bh(&reuseport_lock); + reuse = rcu_dereference_protected(sk->sk_reuseport_cb, + lockdep_is_held(&reuseport_lock)); + rcu_assign_pointer(sk->sk_reuseport_cb, NULL); + + for (i = 0; i < reuse->num_socks; i++) { + if (reuse->socks[i] == sk) { + reuse->socks[i] = reuse->socks[reuse->num_socks - 1]; + reuse->num_socks--; + if (reuse->num_socks == 0) + kfree_rcu(reuse, rcu); + break; + } + } + spin_unlock_bh(&reuseport_lock); +} +EXPORT_SYMBOL(reuseport_detach_sock); + +/** + * reuseport_select_sock - Select a socket from an SO_REUSEPORT group. + * @sk: First socket in the group. + * @hash: Use this hash to select. + * Returns a socket that should receive the packet (or NULL on error). + */ +struct sock *reuseport_select_sock(struct sock *sk, u32 hash) +{ + struct sock_reuseport *reuse; + struct sock *sk2 = NULL; + u16 socks; + + rcu_read_lock(); + reuse = rcu_dereference(sk->sk_reuseport_cb); + + /* if memory allocation failed or add call is not yet complete */ + if (!reuse) + goto out; + + socks = READ_ONCE(reuse->num_socks); + if (likely(socks)) { + /* paired with smp_wmb() in reuseport_add_sock() */ + smp_rmb(); + + sk2 = reuse->socks[reciprocal_scale(hash, socks)]; + } + +out: + rcu_read_unlock(); + return sk2; +} +EXPORT_SYMBOL(reuseport_select_sock); -- cgit v1.2.3-59-g8ed1b From e32ea7e747271a0abcd37e265005e97cc81d9df5 Mon Sep 17 00:00:00 2001 From: Craig Gallek Date: Mon, 4 Jan 2016 17:41:46 -0500 Subject: soreuseport: fast reuseport UDP socket selection Include a struct sock_reuseport instance when a UDP socket binds to a specific address for the first time with the reuseport flag set. When selecting a socket for an incoming UDP packet, use the information available in sock_reuseport if present. This required adding an additional field to the UDP source address equality function to differentiate between exact and wildcard matches. The original use case allowed wildcard matches when checking for existing port uses during bind. The new use case of adding a socket to a reuseport group requires exact address matching. Performance test (using a machine with 2 CPU sockets and a total of 48 cores): Create reuseport groups of varying size. Use one socket from this group per user thread (pinning each thread to a different core) calling recvmmsg in a tight loop. Record number of messages received per second while saturating a 10G link. 10 sockets: 18% increase (~2.8M -> 3.3M pkts/s) 20 sockets: 14% increase (~2.9M -> 3.3M pkts/s) 40 sockets: 13% increase (~3.0M -> 3.4M pkts/s) This work is based off a similar implementation written by Ying Cai for implementing policy-based reuseport selection. Signed-off-by: Craig Gallek Signed-off-by: David S. Miller --- include/net/addrconf.h | 3 +- include/net/udp.h | 2 +- net/ipv4/udp.c | 119 +++++++++++++++++++++++++++++++-------- net/ipv6/inet6_connection_sock.c | 4 +- net/ipv6/udp.c | 48 +++++++++++++--- 5 files changed, 141 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 78003dfb8539..47f52d3cd8df 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -87,7 +87,8 @@ int __ipv6_get_lladdr(struct inet6_dev *idev, struct in6_addr *addr, u32 banned_flags); int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr, u32 banned_flags); -int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2); +int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, + bool match_wildcard); void addrconf_join_solict(struct net_device *dev, const struct in6_addr *addr); void addrconf_leave_solict(struct inet6_dev *idev, const struct in6_addr *addr); diff --git a/include/net/udp.h b/include/net/udp.h index 6d4ed18e1427..3b5d7f93bc23 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -191,7 +191,7 @@ static inline void udp_lib_close(struct sock *sk, long timeout) } int udp_lib_get_port(struct sock *sk, unsigned short snum, - int (*)(const struct sock *, const struct sock *), + int (*)(const struct sock *, const struct sock *, bool), unsigned int hash2_nulladdr); u32 udp_flow_hashrnd(void); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index ac14ae44390d..762b01f55707 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -113,6 +113,7 @@ #include #include #include "udp_impl.h" +#include struct udp_table udp_table __read_mostly; EXPORT_SYMBOL(udp_table); @@ -137,7 +138,8 @@ static int udp_lib_lport_inuse(struct net *net, __u16 num, unsigned long *bitmap, struct sock *sk, int (*saddr_comp)(const struct sock *sk1, - const struct sock *sk2), + const struct sock *sk2, + bool match_wildcard), unsigned int log) { struct sock *sk2; @@ -152,8 +154,9 @@ static int udp_lib_lport_inuse(struct net *net, __u16 num, (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && (!sk2->sk_reuseport || !sk->sk_reuseport || + rcu_access_pointer(sk->sk_reuseport_cb) || !uid_eq(uid, sock_i_uid(sk2))) && - saddr_comp(sk, sk2)) { + saddr_comp(sk, sk2, true)) { if (!bitmap) return 1; __set_bit(udp_sk(sk2)->udp_port_hash >> log, bitmap); @@ -170,7 +173,8 @@ static int udp_lib_lport_inuse2(struct net *net, __u16 num, struct udp_hslot *hslot2, struct sock *sk, int (*saddr_comp)(const struct sock *sk1, - const struct sock *sk2)) + const struct sock *sk2, + bool match_wildcard)) { struct sock *sk2; struct hlist_nulls_node *node; @@ -186,8 +190,9 @@ static int udp_lib_lport_inuse2(struct net *net, __u16 num, (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && (!sk2->sk_reuseport || !sk->sk_reuseport || + rcu_access_pointer(sk->sk_reuseport_cb) || !uid_eq(uid, sock_i_uid(sk2))) && - saddr_comp(sk, sk2)) { + saddr_comp(sk, sk2, true)) { res = 1; break; } @@ -196,6 +201,35 @@ static int udp_lib_lport_inuse2(struct net *net, __u16 num, return res; } +static int udp_reuseport_add_sock(struct sock *sk, struct udp_hslot *hslot, + int (*saddr_same)(const struct sock *sk1, + const struct sock *sk2, + bool match_wildcard)) +{ + struct net *net = sock_net(sk); + struct hlist_nulls_node *node; + kuid_t uid = sock_i_uid(sk); + struct sock *sk2; + + sk_nulls_for_each(sk2, node, &hslot->head) { + if (net_eq(sock_net(sk2), net) && + sk2 != sk && + sk2->sk_family == sk->sk_family && + ipv6_only_sock(sk2) == ipv6_only_sock(sk) && + (udp_sk(sk2)->udp_port_hash == udp_sk(sk)->udp_port_hash) && + (sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && + sk2->sk_reuseport && uid_eq(uid, sock_i_uid(sk2)) && + (*saddr_same)(sk, sk2, false)) { + return reuseport_add_sock(sk, sk2); + } + } + + /* Initial allocation may have already happened via setsockopt */ + if (!rcu_access_pointer(sk->sk_reuseport_cb)) + return reuseport_alloc(sk); + return 0; +} + /** * udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6 * @@ -207,7 +241,8 @@ static int udp_lib_lport_inuse2(struct net *net, __u16 num, */ int udp_lib_get_port(struct sock *sk, unsigned short snum, int (*saddr_comp)(const struct sock *sk1, - const struct sock *sk2), + const struct sock *sk2, + bool match_wildcard), unsigned int hash2_nulladdr) { struct udp_hslot *hslot, *hslot2; @@ -290,6 +325,14 @@ found: udp_sk(sk)->udp_port_hash = snum; udp_sk(sk)->udp_portaddr_hash ^= snum; if (sk_unhashed(sk)) { + if (sk->sk_reuseport && + udp_reuseport_add_sock(sk, hslot, saddr_comp)) { + inet_sk(sk)->inet_num = 0; + udp_sk(sk)->udp_port_hash = 0; + udp_sk(sk)->udp_portaddr_hash ^= snum; + goto fail_unlock; + } + sk_nulls_add_node_rcu(sk, &hslot->head); hslot->count++; sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); @@ -309,13 +352,22 @@ fail: } EXPORT_SYMBOL(udp_lib_get_port); -static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) +/* match_wildcard == true: 0.0.0.0 equals to any IPv4 addresses + * match_wildcard == false: addresses must be exactly the same, i.e. + * 0.0.0.0 only equals to 0.0.0.0 + */ +static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2, + bool match_wildcard) { struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); - return (!ipv6_only_sock(sk2) && - (!inet1->inet_rcv_saddr || !inet2->inet_rcv_saddr || - inet1->inet_rcv_saddr == inet2->inet_rcv_saddr)); + if (!ipv6_only_sock(sk2)) { + if (inet1->inet_rcv_saddr == inet2->inet_rcv_saddr) + return 1; + if (!inet1->inet_rcv_saddr || !inet2->inet_rcv_saddr) + return match_wildcard; + } + return 0; } static u32 udp4_portaddr_hash(const struct net *net, __be32 saddr, @@ -459,8 +511,14 @@ begin: badness = score; reuseport = sk->sk_reuseport; if (reuseport) { + struct sock *sk2; hash = udp_ehashfn(net, daddr, hnum, saddr, sport); + sk2 = reuseport_select_sock(sk, hash); + if (sk2) { + result = sk2; + goto found; + } matches = 1; } } else if (score == badness && reuseport) { @@ -478,6 +536,7 @@ begin: if (get_nulls_value(node) != slot2) goto begin; if (result) { +found: if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2))) result = NULL; else if (unlikely(compute_score2(result, net, saddr, sport, @@ -540,8 +599,14 @@ begin: badness = score; reuseport = sk->sk_reuseport; if (reuseport) { + struct sock *sk2; hash = udp_ehashfn(net, daddr, hnum, saddr, sport); + sk2 = reuseport_select_sock(sk, hash); + if (sk2) { + result = sk2; + goto found; + } matches = 1; } } else if (score == badness && reuseport) { @@ -560,6 +625,7 @@ begin: goto begin; if (result) { +found: if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2))) result = NULL; else if (unlikely(compute_score(result, net, saddr, hnum, sport, @@ -587,7 +653,8 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, int dif) { - return __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table); + return __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, + &udp_table); } EXPORT_SYMBOL_GPL(udp4_lib_lookup); @@ -1398,6 +1465,8 @@ void udp_lib_unhash(struct sock *sk) hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); spin_lock_bh(&hslot->lock); + if (rcu_access_pointer(sk->sk_reuseport_cb)) + reuseport_detach_sock(sk); if (sk_nulls_del_node_init_rcu(sk)) { hslot->count--; inet_sk(sk)->inet_num = 0; @@ -1425,22 +1494,28 @@ void udp_lib_rehash(struct sock *sk, u16 newhash) hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); nhslot2 = udp_hashslot2(udptable, newhash); udp_sk(sk)->udp_portaddr_hash = newhash; - if (hslot2 != nhslot2) { + + if (hslot2 != nhslot2 || + rcu_access_pointer(sk->sk_reuseport_cb)) { hslot = udp_hashslot(udptable, sock_net(sk), udp_sk(sk)->udp_port_hash); /* we must lock primary chain too */ spin_lock_bh(&hslot->lock); - - spin_lock(&hslot2->lock); - hlist_nulls_del_init_rcu(&udp_sk(sk)->udp_portaddr_node); - hslot2->count--; - spin_unlock(&hslot2->lock); - - spin_lock(&nhslot2->lock); - hlist_nulls_add_head_rcu(&udp_sk(sk)->udp_portaddr_node, - &nhslot2->head); - nhslot2->count++; - spin_unlock(&nhslot2->lock); + if (rcu_access_pointer(sk->sk_reuseport_cb)) + reuseport_detach_sock(sk); + + if (hslot2 != nhslot2) { + spin_lock(&hslot2->lock); + hlist_nulls_del_init_rcu(&udp_sk(sk)->udp_portaddr_node); + hslot2->count--; + spin_unlock(&hslot2->lock); + + spin_lock(&nhslot2->lock); + hlist_nulls_add_head_rcu(&udp_sk(sk)->udp_portaddr_node, + &nhslot2->head); + nhslot2->count++; + spin_unlock(&nhslot2->lock); + } spin_unlock_bh(&hslot->lock); } diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index a7ca2cde2ecb..36c3f0155010 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -51,12 +51,12 @@ int inet6_csk_bind_conflict(const struct sock *sk, (sk2->sk_state != TCP_TIME_WAIT && !uid_eq(uid, sock_i_uid((struct sock *)sk2))))) { - if (ipv6_rcv_saddr_equal(sk, sk2)) + if (ipv6_rcv_saddr_equal(sk, sk2, true)) break; } if (!relax && reuse && sk2->sk_reuse && sk2->sk_state != TCP_LISTEN && - ipv6_rcv_saddr_equal(sk, sk2)) + ipv6_rcv_saddr_equal(sk, sk2, true)) break; } } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 00775ee27d86..6204b8992de4 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include @@ -76,7 +77,14 @@ static u32 udp6_ehashfn(const struct net *net, udp_ipv6_hash_secret + net_hash_mix(net)); } -int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2) +/* match_wildcard == true: IPV6_ADDR_ANY equals to any IPv6 addresses if IPv6 + * only, and any IPv4 addresses if not IPv6 only + * match_wildcard == false: addresses must be exactly the same, i.e. + * IPV6_ADDR_ANY only equals to IPV6_ADDR_ANY, + * and 0.0.0.0 equals to 0.0.0.0 only + */ +int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, + bool match_wildcard) { const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2); int sk2_ipv6only = inet_v6_ipv6only(sk2); @@ -84,16 +92,24 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2) int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED; /* if both are mapped, treat as IPv4 */ - if (addr_type == IPV6_ADDR_MAPPED && addr_type2 == IPV6_ADDR_MAPPED) - return (!sk2_ipv6only && - (!sk->sk_rcv_saddr || !sk2->sk_rcv_saddr || - sk->sk_rcv_saddr == sk2->sk_rcv_saddr)); + if (addr_type == IPV6_ADDR_MAPPED && addr_type2 == IPV6_ADDR_MAPPED) { + if (!sk2_ipv6only) { + if (sk->sk_rcv_saddr == sk2->sk_rcv_saddr) + return 1; + if (!sk->sk_rcv_saddr || !sk2->sk_rcv_saddr) + return match_wildcard; + } + return 0; + } + + if (addr_type == IPV6_ADDR_ANY && addr_type2 == IPV6_ADDR_ANY) + return 1; - if (addr_type2 == IPV6_ADDR_ANY && + if (addr_type2 == IPV6_ADDR_ANY && match_wildcard && !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED)) return 1; - if (addr_type == IPV6_ADDR_ANY && + if (addr_type == IPV6_ADDR_ANY && match_wildcard && !(ipv6_only_sock(sk) && addr_type2 == IPV6_ADDR_MAPPED)) return 1; @@ -253,8 +269,14 @@ begin: badness = score; reuseport = sk->sk_reuseport; if (reuseport) { + struct sock *sk2; hash = udp6_ehashfn(net, daddr, hnum, saddr, sport); + sk2 = reuseport_select_sock(sk, hash); + if (sk2) { + result = sk2; + goto found; + } matches = 1; } } else if (score == badness && reuseport) { @@ -273,6 +295,7 @@ begin: goto begin; if (result) { +found: if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2))) result = NULL; else if (unlikely(compute_score2(result, net, saddr, sport, @@ -332,8 +355,14 @@ begin: badness = score; reuseport = sk->sk_reuseport; if (reuseport) { + struct sock *sk2; hash = udp6_ehashfn(net, daddr, hnum, saddr, sport); + sk2 = reuseport_select_sock(sk, hash); + if (sk2) { + result = sk2; + goto found; + } matches = 1; } } else if (score == badness && reuseport) { @@ -352,6 +381,7 @@ begin: goto begin; if (result) { +found: if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2))) result = NULL; else if (unlikely(compute_score(result, net, hnum, saddr, sport, @@ -549,8 +579,8 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, int err; struct net *net = dev_net(skb->dev); - sk = __udp6_lib_lookup(net, daddr, uh->dest, - saddr, uh->source, inet6_iif(skb), udptable); + sk = __udp6_lib_lookup(net, daddr, uh->dest, saddr, uh->source, + inet6_iif(skb), udptable); if (!sk) { ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev), ICMP6_MIB_INERRORS); -- cgit v1.2.3-59-g8ed1b From 538950a1b7527a0a52ccd9337e3fcd304f027f13 Mon Sep 17 00:00:00 2001 From: Craig Gallek Date: Mon, 4 Jan 2016 17:41:47 -0500 Subject: soreuseport: setsockopt SO_ATTACH_REUSEPORT_[CE]BPF Expose socket options for setting a classic or extended BPF program for use when selecting sockets in an SO_REUSEPORT group. These options can be used on the first socket to belong to a group before bind or on any socket in the group after bind. This change includes refactoring of the existing sk_filter code to allow reuse of the existing BPF filter validation checks. Signed-off-by: Craig Gallek Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- arch/alpha/include/uapi/asm/socket.h | 3 + arch/avr32/include/uapi/asm/socket.h | 3 + arch/frv/include/uapi/asm/socket.h | 3 + arch/ia64/include/uapi/asm/socket.h | 3 + arch/m32r/include/uapi/asm/socket.h | 3 + arch/mips/include/uapi/asm/socket.h | 3 + arch/mn10300/include/uapi/asm/socket.h | 3 + arch/parisc/include/uapi/asm/socket.h | 3 + arch/powerpc/include/uapi/asm/socket.h | 3 + arch/s390/include/uapi/asm/socket.h | 3 + arch/sparc/include/uapi/asm/socket.h | 3 + arch/xtensa/include/uapi/asm/socket.h | 3 + include/linux/filter.h | 2 + include/net/sock_reuseport.h | 10 ++- include/net/udp.h | 5 +- include/uapi/asm-generic/socket.h | 3 + net/core/filter.c | 121 +++++++++++++++++++++++++++------ net/core/sock.c | 29 ++++++++ net/core/sock_reuseport.c | 88 ++++++++++++++++++++++-- net/ipv4/udp.c | 14 ++-- net/ipv4/udp_diag.c | 4 +- net/ipv6/udp.c | 14 ++-- 22 files changed, 282 insertions(+), 44 deletions(-) (limited to 'include') diff --git a/arch/alpha/include/uapi/asm/socket.h b/arch/alpha/include/uapi/asm/socket.h index 9a20821b111c..c5fb9e6bc3a5 100644 --- a/arch/alpha/include/uapi/asm/socket.h +++ b/arch/alpha/include/uapi/asm/socket.h @@ -92,4 +92,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _UAPI_ASM_SOCKET_H */ diff --git a/arch/avr32/include/uapi/asm/socket.h b/arch/avr32/include/uapi/asm/socket.h index 2b65ed6b277c..9de0796240a0 100644 --- a/arch/avr32/include/uapi/asm/socket.h +++ b/arch/avr32/include/uapi/asm/socket.h @@ -85,4 +85,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _UAPI__ASM_AVR32_SOCKET_H */ diff --git a/arch/frv/include/uapi/asm/socket.h b/arch/frv/include/uapi/asm/socket.h index 4823ad125578..f02e4849ae83 100644 --- a/arch/frv/include/uapi/asm/socket.h +++ b/arch/frv/include/uapi/asm/socket.h @@ -85,5 +85,8 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_SOCKET_H */ diff --git a/arch/ia64/include/uapi/asm/socket.h b/arch/ia64/include/uapi/asm/socket.h index 59be3d87f86d..bce29166de1b 100644 --- a/arch/ia64/include/uapi/asm/socket.h +++ b/arch/ia64/include/uapi/asm/socket.h @@ -94,4 +94,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_IA64_SOCKET_H */ diff --git a/arch/m32r/include/uapi/asm/socket.h b/arch/m32r/include/uapi/asm/socket.h index 7bc4cb273856..14aa4a6bccf1 100644 --- a/arch/m32r/include/uapi/asm/socket.h +++ b/arch/m32r/include/uapi/asm/socket.h @@ -85,4 +85,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_M32R_SOCKET_H */ diff --git a/arch/mips/include/uapi/asm/socket.h b/arch/mips/include/uapi/asm/socket.h index dec3c850f36b..5910fe294e93 100644 --- a/arch/mips/include/uapi/asm/socket.h +++ b/arch/mips/include/uapi/asm/socket.h @@ -103,4 +103,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _UAPI_ASM_SOCKET_H */ diff --git a/arch/mn10300/include/uapi/asm/socket.h b/arch/mn10300/include/uapi/asm/socket.h index cab7d6d50051..58b1aa01ab9f 100644 --- a/arch/mn10300/include/uapi/asm/socket.h +++ b/arch/mn10300/include/uapi/asm/socket.h @@ -85,4 +85,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_SOCKET_H */ diff --git a/arch/parisc/include/uapi/asm/socket.h b/arch/parisc/include/uapi/asm/socket.h index a5cd40cd8ee1..f9cf1223422c 100644 --- a/arch/parisc/include/uapi/asm/socket.h +++ b/arch/parisc/include/uapi/asm/socket.h @@ -84,4 +84,7 @@ #define SO_ATTACH_BPF 0x402B #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 0x402C +#define SO_ATTACH_REUSEPORT_EBPF 0x402D + #endif /* _UAPI_ASM_SOCKET_H */ diff --git a/arch/powerpc/include/uapi/asm/socket.h b/arch/powerpc/include/uapi/asm/socket.h index c046666038f8..dd54f28ecdec 100644 --- a/arch/powerpc/include/uapi/asm/socket.h +++ b/arch/powerpc/include/uapi/asm/socket.h @@ -92,4 +92,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_POWERPC_SOCKET_H */ diff --git a/arch/s390/include/uapi/asm/socket.h b/arch/s390/include/uapi/asm/socket.h index 296942d56e6a..d02e89d14fef 100644 --- a/arch/s390/include/uapi/asm/socket.h +++ b/arch/s390/include/uapi/asm/socket.h @@ -91,4 +91,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _ASM_SOCKET_H */ diff --git a/arch/sparc/include/uapi/asm/socket.h b/arch/sparc/include/uapi/asm/socket.h index e6a16c40be5f..d270ee91968e 100644 --- a/arch/sparc/include/uapi/asm/socket.h +++ b/arch/sparc/include/uapi/asm/socket.h @@ -81,6 +81,9 @@ #define SO_ATTACH_BPF 0x0034 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 0x0035 +#define SO_ATTACH_REUSEPORT_EBPF 0x0036 + /* Security levels - as per NRL IPv6 - don't actually do anything */ #define SO_SECURITY_AUTHENTICATION 0x5001 #define SO_SECURITY_ENCRYPTION_TRANSPORT 0x5002 diff --git a/arch/xtensa/include/uapi/asm/socket.h b/arch/xtensa/include/uapi/asm/socket.h index 4120af086160..fd3b96d1153f 100644 --- a/arch/xtensa/include/uapi/asm/socket.h +++ b/arch/xtensa/include/uapi/asm/socket.h @@ -96,4 +96,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* _XTENSA_SOCKET_H */ diff --git a/include/linux/filter.h b/include/linux/filter.h index 4165e9ac9e36..294c3cdf07b3 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -447,6 +447,8 @@ void bpf_prog_destroy(struct bpf_prog *fp); int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk); int sk_attach_bpf(u32 ufd, struct sock *sk); +int sk_reuseport_attach_filter(struct sock_fprog *fprog, struct sock *sk); +int sk_reuseport_attach_bpf(u32 ufd, struct sock *sk); int sk_detach_filter(struct sock *sk); int sk_get_filter(struct sock *sk, struct sock_filter __user *filter, unsigned int len); diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h index 67d1eb8fd7af..7dda3d7adba8 100644 --- a/include/net/sock_reuseport.h +++ b/include/net/sock_reuseport.h @@ -1,6 +1,8 @@ #ifndef _SOCK_REUSEPORT_H #define _SOCK_REUSEPORT_H +#include +#include #include #include @@ -9,12 +11,18 @@ struct sock_reuseport { u16 max_socks; /* length of socks */ u16 num_socks; /* elements in socks */ + struct bpf_prog __rcu *prog; /* optional BPF sock selector */ struct sock *socks[0]; /* array of sock pointers */ }; extern int reuseport_alloc(struct sock *sk); extern int reuseport_add_sock(struct sock *sk, const struct sock *sk2); extern void reuseport_detach_sock(struct sock *sk); -extern struct sock *reuseport_select_sock(struct sock *sk, u32 hash); +extern struct sock *reuseport_select_sock(struct sock *sk, + u32 hash, + struct sk_buff *skb, + int hdr_len); +extern struct bpf_prog *reuseport_attach_prog(struct sock *sk, + struct bpf_prog *prog); #endif /* _SOCK_REUSEPORT_H */ diff --git a/include/net/udp.h b/include/net/udp.h index 3b5d7f93bc23..2842541e28e7 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -258,7 +258,7 @@ struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, int dif); struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, int dif, - struct udp_table *tbl); + struct udp_table *tbl, struct sk_buff *skb); struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be16 sport, const struct in6_addr *daddr, __be16 dport, @@ -266,7 +266,8 @@ struct sock *udp6_lib_lookup(struct net *net, 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 *tbl); + int dif, struct udp_table *tbl, + struct sk_buff *skb); /* * SNMP statistics for UDP and UDP-Lite diff --git a/include/uapi/asm-generic/socket.h b/include/uapi/asm-generic/socket.h index 5c15c2a5c123..fb8a41668382 100644 --- a/include/uapi/asm-generic/socket.h +++ b/include/uapi/asm-generic/socket.h @@ -87,4 +87,7 @@ #define SO_ATTACH_BPF 50 #define SO_DETACH_BPF SO_DETACH_FILTER +#define SO_ATTACH_REUSEPORT_CBPF 51 +#define SO_ATTACH_REUSEPORT_EBPF 52 + #endif /* __ASM_GENERIC_SOCKET_H */ diff --git a/net/core/filter.c b/net/core/filter.c index c770196ae8d5..35e6fed28709 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -50,6 +50,7 @@ #include #include #include +#include /** * sk_filter - run a packet through a socket filter @@ -1167,17 +1168,32 @@ static int __sk_attach_prog(struct bpf_prog *prog, struct sock *sk) return 0; } -/** - * sk_attach_filter - attach a socket filter - * @fprog: the filter program - * @sk: the socket to use - * - * Attach the user's filter code. We first run some sanity checks on - * it to make sure it does not explode on us later. If an error - * occurs or there is insufficient memory for the filter a negative - * errno code is returned. On success the return is zero. - */ -int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) +static int __reuseport_attach_prog(struct bpf_prog *prog, struct sock *sk) +{ + struct bpf_prog *old_prog; + int err; + + if (bpf_prog_size(prog->len) > sysctl_optmem_max) + return -ENOMEM; + + if (sk_unhashed(sk)) { + err = reuseport_alloc(sk); + if (err) + return err; + } else if (!rcu_access_pointer(sk->sk_reuseport_cb)) { + /* The socket wasn't bound with SO_REUSEPORT */ + return -EINVAL; + } + + old_prog = reuseport_attach_prog(sk, prog); + if (old_prog) + bpf_prog_destroy(old_prog); + + return 0; +} + +static +struct bpf_prog *__get_filter(struct sock_fprog *fprog, struct sock *sk) { unsigned int fsize = bpf_classic_proglen(fprog); unsigned int bpf_fsize = bpf_prog_size(fprog->len); @@ -1185,19 +1201,19 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) int err; if (sock_flag(sk, SOCK_FILTER_LOCKED)) - return -EPERM; + return ERR_PTR(-EPERM); /* Make sure new filter is there and in the right amounts. */ if (fprog->filter == NULL) - return -EINVAL; + return ERR_PTR(-EINVAL); prog = bpf_prog_alloc(bpf_fsize, 0); if (!prog) - return -ENOMEM; + return ERR_PTR(-ENOMEM); if (copy_from_user(prog->insns, fprog->filter, fsize)) { __bpf_prog_free(prog); - return -EFAULT; + return ERR_PTR(-EFAULT); } prog->len = fprog->len; @@ -1205,13 +1221,30 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) err = bpf_prog_store_orig_filter(prog, fprog); if (err) { __bpf_prog_free(prog); - return -ENOMEM; + return ERR_PTR(-ENOMEM); } /* bpf_prepare_filter() already takes care of freeing * memory in case something goes wrong. */ - prog = bpf_prepare_filter(prog, NULL); + return bpf_prepare_filter(prog, NULL); +} + +/** + * sk_attach_filter - attach a socket filter + * @fprog: the filter program + * @sk: the socket to use + * + * Attach the user's filter code. We first run some sanity checks on + * it to make sure it does not explode on us later. If an error + * occurs or there is insufficient memory for the filter a negative + * errno code is returned. On success the return is zero. + */ +int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) +{ + struct bpf_prog *prog = __get_filter(fprog, sk); + int err; + if (IS_ERR(prog)) return PTR_ERR(prog); @@ -1225,23 +1258,50 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk) } EXPORT_SYMBOL_GPL(sk_attach_filter); -int sk_attach_bpf(u32 ufd, struct sock *sk) +int sk_reuseport_attach_filter(struct sock_fprog *fprog, struct sock *sk) { - struct bpf_prog *prog; + struct bpf_prog *prog = __get_filter(fprog, sk); int err; + if (IS_ERR(prog)) + return PTR_ERR(prog); + + err = __reuseport_attach_prog(prog, sk); + if (err < 0) { + __bpf_prog_release(prog); + return err; + } + + return 0; +} + +static struct bpf_prog *__get_bpf(u32 ufd, struct sock *sk) +{ + struct bpf_prog *prog; + if (sock_flag(sk, SOCK_FILTER_LOCKED)) - return -EPERM; + return ERR_PTR(-EPERM); prog = bpf_prog_get(ufd); if (IS_ERR(prog)) - return PTR_ERR(prog); + return prog; if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) { bpf_prog_put(prog); - return -EINVAL; + return ERR_PTR(-EINVAL); } + return prog; +} + +int sk_attach_bpf(u32 ufd, struct sock *sk) +{ + struct bpf_prog *prog = __get_bpf(ufd, sk); + int err; + + if (IS_ERR(prog)) + return PTR_ERR(prog); + err = __sk_attach_prog(prog, sk); if (err < 0) { bpf_prog_put(prog); @@ -1251,6 +1311,23 @@ int sk_attach_bpf(u32 ufd, struct sock *sk) return 0; } +int sk_reuseport_attach_bpf(u32 ufd, struct sock *sk) +{ + struct bpf_prog *prog = __get_bpf(ufd, sk); + int err; + + if (IS_ERR(prog)) + return PTR_ERR(prog); + + err = __reuseport_attach_prog(prog, sk); + if (err < 0) { + bpf_prog_put(prog); + return err; + } + + return 0; +} + #define BPF_RECOMPUTE_CSUM(flags) ((flags) & 1) #define BPF_LDST_LEN 16U diff --git a/net/core/sock.c b/net/core/sock.c index 565bab7baca9..51270238e269 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -134,6 +134,7 @@ #include #include +#include #include @@ -932,6 +933,32 @@ set_rcvbuf: } break; + case SO_ATTACH_REUSEPORT_CBPF: + ret = -EINVAL; + if (optlen == sizeof(struct sock_fprog)) { + struct sock_fprog fprog; + + ret = -EFAULT; + if (copy_from_user(&fprog, optval, sizeof(fprog))) + break; + + ret = sk_reuseport_attach_filter(&fprog, sk); + } + break; + + case SO_ATTACH_REUSEPORT_EBPF: + ret = -EINVAL; + if (optlen == sizeof(u32)) { + u32 ufd; + + ret = -EFAULT; + if (copy_from_user(&ufd, optval, sizeof(ufd))) + break; + + ret = sk_reuseport_attach_bpf(ufd, sk); + } + break; + case SO_DETACH_FILTER: ret = sk_detach_filter(sk); break; @@ -1443,6 +1470,8 @@ void sk_destruct(struct sock *sk) sk_filter_uncharge(sk, filter); RCU_INIT_POINTER(sk->sk_filter, NULL); } + if (rcu_access_pointer(sk->sk_reuseport_cb)) + reuseport_detach_sock(sk); sock_disable_timestamp(sk, SK_FLAGS_TIMESTAMP); diff --git a/net/core/sock_reuseport.c b/net/core/sock_reuseport.c index 963c8d5f3027..ae0969c0fc2e 100644 --- a/net/core/sock_reuseport.c +++ b/net/core/sock_reuseport.c @@ -1,10 +1,12 @@ /* * To speed up listener socket lookup, create an array to store all sockets * listening on the same port. This allows a decision to be made after finding - * the first socket. + * the first socket. An optional BPF program can also be configured for + * selecting the socket index from the array of available sockets. */ #include +#include #include #define INIT_SOCKS 128 @@ -22,6 +24,7 @@ static struct sock_reuseport *__reuseport_alloc(u16 max_socks) reuse->max_socks = max_socks; + RCU_INIT_POINTER(reuse->prog, NULL); return reuse; } @@ -67,6 +70,7 @@ static struct sock_reuseport *reuseport_grow(struct sock_reuseport *reuse) more_reuse->max_socks = more_socks_size; more_reuse->num_socks = reuse->num_socks; + more_reuse->prog = reuse->prog; memcpy(more_reuse->socks, reuse->socks, reuse->num_socks * sizeof(struct sock *)); @@ -75,6 +79,10 @@ static struct sock_reuseport *reuseport_grow(struct sock_reuseport *reuse) rcu_assign_pointer(reuse->socks[i]->sk_reuseport_cb, more_reuse); + /* Note: we use kfree_rcu here instead of reuseport_free_rcu so + * that reuse and more_reuse can temporarily share a reference + * to prog. + */ kfree_rcu(reuse, rcu); return more_reuse; } @@ -116,6 +124,16 @@ int reuseport_add_sock(struct sock *sk, const struct sock *sk2) } EXPORT_SYMBOL(reuseport_add_sock); +static void reuseport_free_rcu(struct rcu_head *head) +{ + struct sock_reuseport *reuse; + + reuse = container_of(head, struct sock_reuseport, rcu); + if (reuse->prog) + bpf_prog_destroy(reuse->prog); + kfree(reuse); +} + void reuseport_detach_sock(struct sock *sk) { struct sock_reuseport *reuse; @@ -131,7 +149,7 @@ void reuseport_detach_sock(struct sock *sk) reuse->socks[i] = reuse->socks[reuse->num_socks - 1]; reuse->num_socks--; if (reuse->num_socks == 0) - kfree_rcu(reuse, rcu); + call_rcu(&reuse->rcu, reuseport_free_rcu); break; } } @@ -139,15 +157,53 @@ void reuseport_detach_sock(struct sock *sk) } EXPORT_SYMBOL(reuseport_detach_sock); +static struct sock *run_bpf(struct sock_reuseport *reuse, u16 socks, + struct bpf_prog *prog, struct sk_buff *skb, + int hdr_len) +{ + struct sk_buff *nskb = NULL; + u32 index; + + if (skb_shared(skb)) { + nskb = skb_clone(skb, GFP_ATOMIC); + if (!nskb) + return NULL; + skb = nskb; + } + + /* temporarily advance data past protocol header */ + if (!pskb_pull(skb, hdr_len)) { + consume_skb(nskb); + return NULL; + } + index = bpf_prog_run_save_cb(prog, skb); + __skb_push(skb, hdr_len); + + consume_skb(nskb); + + if (index >= socks) + return NULL; + + return reuse->socks[index]; +} + /** * reuseport_select_sock - Select a socket from an SO_REUSEPORT group. * @sk: First socket in the group. - * @hash: Use this hash to select. + * @hash: When no BPF filter is available, use this hash to select. + * @skb: skb to run through BPF filter. + * @hdr_len: BPF filter expects skb data pointer at payload data. If + * the skb does not yet point at the payload, this parameter represents + * how far the pointer needs to advance to reach the payload. * Returns a socket that should receive the packet (or NULL on error). */ -struct sock *reuseport_select_sock(struct sock *sk, u32 hash) +struct sock *reuseport_select_sock(struct sock *sk, + u32 hash, + struct sk_buff *skb, + int hdr_len) { struct sock_reuseport *reuse; + struct bpf_prog *prog; struct sock *sk2 = NULL; u16 socks; @@ -158,12 +214,16 @@ struct sock *reuseport_select_sock(struct sock *sk, u32 hash) if (!reuse) goto out; + prog = rcu_dereference(reuse->prog); socks = READ_ONCE(reuse->num_socks); if (likely(socks)) { /* paired with smp_wmb() in reuseport_add_sock() */ smp_rmb(); - sk2 = reuse->socks[reciprocal_scale(hash, socks)]; + if (prog && skb) + sk2 = run_bpf(reuse, socks, prog, skb, hdr_len); + else + sk2 = reuse->socks[reciprocal_scale(hash, socks)]; } out: @@ -171,3 +231,21 @@ out: return sk2; } EXPORT_SYMBOL(reuseport_select_sock); + +struct bpf_prog * +reuseport_attach_prog(struct sock *sk, struct bpf_prog *prog) +{ + struct sock_reuseport *reuse; + struct bpf_prog *old_prog; + + spin_lock_bh(&reuseport_lock); + reuse = rcu_dereference_protected(sk->sk_reuseport_cb, + lockdep_is_held(&reuseport_lock)); + old_prog = rcu_dereference_protected(reuse->prog, + lockdep_is_held(&reuseport_lock)); + rcu_assign_pointer(reuse->prog, prog); + spin_unlock_bh(&reuseport_lock); + + return old_prog; +} +EXPORT_SYMBOL(reuseport_attach_prog); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 762b01f55707..835378365f25 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -514,7 +514,7 @@ begin: struct sock *sk2; hash = udp_ehashfn(net, daddr, hnum, saddr, sport); - sk2 = reuseport_select_sock(sk, hash); + sk2 = reuseport_select_sock(sk, hash, NULL, 0); if (sk2) { result = sk2; goto found; @@ -553,7 +553,7 @@ found: */ struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, - int dif, struct udp_table *udptable) + int dif, struct udp_table *udptable, struct sk_buff *skb) { struct sock *sk, *result; struct hlist_nulls_node *node; @@ -602,7 +602,8 @@ begin: struct sock *sk2; hash = udp_ehashfn(net, daddr, hnum, saddr, sport); - sk2 = reuseport_select_sock(sk, hash); + sk2 = reuseport_select_sock(sk, hash, skb, + sizeof(struct udphdr)); if (sk2) { result = sk2; goto found; @@ -647,14 +648,14 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, return __udp4_lib_lookup(dev_net(skb_dst(skb)->dev), iph->saddr, sport, iph->daddr, dport, inet_iif(skb), - udptable); + udptable, skb); } struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, int dif) { return __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, - &udp_table); + &udp_table, NULL); } EXPORT_SYMBOL_GPL(udp4_lib_lookup); @@ -702,7 +703,8 @@ 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); + iph->saddr, uh->source, skb->dev->ifindex, udptable, + NULL); if (!sk) { ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS); return; /* No socket for error */ diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c index 6116604bf6e8..df1966f3b6ec 100644 --- a/net/ipv4/udp_diag.c +++ b/net/ipv4/udp_diag.c @@ -44,7 +44,7 @@ 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); + req->id.idiag_if, tbl, NULL); #if IS_ENABLED(CONFIG_IPV6) else if (req->sdiag_family == AF_INET6) sk = __udp6_lib_lookup(net, @@ -52,7 +52,7 @@ 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); + req->id.idiag_if, tbl, NULL); #endif else goto out_nosk; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 6204b8992de4..56fcb55fda31 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -272,7 +272,7 @@ begin: struct sock *sk2; hash = udp6_ehashfn(net, daddr, hnum, saddr, sport); - sk2 = reuseport_select_sock(sk, hash); + sk2 = reuseport_select_sock(sk, hash, NULL, 0); if (sk2) { result = sk2; goto found; @@ -310,7 +310,8 @@ found: 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) + int dif, struct udp_table *udptable, + struct sk_buff *skb) { struct sock *sk, *result; struct hlist_nulls_node *node; @@ -358,7 +359,8 @@ begin: struct sock *sk2; hash = udp6_ehashfn(net, daddr, hnum, saddr, sport); - sk2 = reuseport_select_sock(sk, hash); + sk2 = reuseport_select_sock(sk, hash, skb, + sizeof(struct udphdr)); if (sk2) { result = sk2; goto found; @@ -407,13 +409,13 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, return sk; return __udp6_lib_lookup(dev_net(skb_dst(skb)->dev), &iph->saddr, sport, &iph->daddr, dport, inet6_iif(skb), - udptable); + udptable, skb); } struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be16 sport, const struct in6_addr *daddr, __be16 dport, int dif) { - return __udp6_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table); + return __udp6_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table, NULL); } EXPORT_SYMBOL_GPL(udp6_lib_lookup); @@ -580,7 +582,7 @@ 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); + inet6_iif(skb), udptable, skb); if (!sk) { ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev), ICMP6_MIB_INERRORS); -- cgit v1.2.3-59-g8ed1b From 0d3b7f64c84d53658daf28e2f9772e38acb9340d Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Tue, 5 Jan 2016 13:19:31 +0200 Subject: Bluetooth: Change eir_has_data_type() to more generic eir_get_data() To make the EIR parsing helper more general purpose, make it return the found data and its length rather than just saying whether the data was present or not. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 30 ++++++++++++++++++++---------- net/bluetooth/hci_event.c | 6 +++--- net/bluetooth/mgmt.c | 3 ++- 3 files changed, 25 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index c95e0326c41a..372e2a7c4ada 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1283,31 +1283,41 @@ static inline void hci_role_switch_cfm(struct hci_conn *conn, __u8 status, mutex_unlock(&hci_cb_list_lock); } -static inline bool eir_has_data_type(u8 *data, size_t data_len, u8 type) +static inline void *eir_get_data(u8 *eir, size_t eir_len, u8 type, + size_t *data_len) { size_t parsed = 0; - if (data_len < 2) - return false; + if (eir_len < 2) + return NULL; - while (parsed < data_len - 1) { - u8 field_len = data[0]; + while (parsed < eir_len - 1) { + u8 field_len = eir[0]; if (field_len == 0) break; parsed += field_len + 1; - if (parsed > data_len) + if (parsed > eir_len) break; - if (data[1] == type) - return true; + if (eir[1] != type) { + eir += field_len + 1; + continue; + } + + /* Zero length data */ + if (field_len == 1) + return NULL; - data += field_len + 1; + if (data_len) + *data_len = field_len - 1; + + return &eir[2]; } - return false; + return NULL; } static inline bool hci_bdaddr_is_rpa(bdaddr_t *bdaddr, u8 addr_type) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 7554da5b7a8f..c162af5d16bf 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -3833,9 +3833,9 @@ static void hci_extended_inquiry_result_evt(struct hci_dev *hdev, data.ssp_mode = 0x01; if (hci_dev_test_flag(hdev, HCI_MGMT)) - name_known = eir_has_data_type(info->data, - sizeof(info->data), - EIR_NAME_COMPLETE); + name_known = eir_get_data(info->data, + sizeof(info->data), + EIR_NAME_COMPLETE, NULL); else name_known = true; diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 621f6fdd0dd1..3297a4ecc05e 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -7266,7 +7266,8 @@ void mgmt_device_found(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type, /* Copy EIR or advertising data into event */ memcpy(ev->eir, eir, eir_len); - if (dev_class && !eir_has_data_type(ev->eir, eir_len, EIR_CLASS_OF_DEV)) + if (dev_class && !eir_get_data(ev->eir, eir_len, EIR_CLASS_OF_DEV, + NULL)) eir_len = eir_append_data(ev->eir, eir_len, EIR_CLASS_OF_DEV, dev_class, 3); -- cgit v1.2.3-59-g8ed1b From 78b781ca0d35191ebf8d8cad8beec810270f0f2e Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Tue, 5 Jan 2016 13:19:32 +0200 Subject: Bluetooth: Add support for Start Limited Discovery command This patch implements the mgmt Start Limited Discovery command. Most of existing Start Discovery code is reused since the only difference is the presence of a 'limited' flag as part of the discovery state. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 + include/net/bluetooth/mgmt.h | 2 ++ net/bluetooth/hci_request.c | 11 ++++++--- net/bluetooth/mgmt.c | 53 +++++++++++++++++++++++++++++++++------- 4 files changed, 55 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 372e2a7c4ada..d4f82edb5cff 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -77,6 +77,7 @@ struct discovery_state { u8 last_adv_data_len; bool report_invalid_rssi; bool result_filtering; + bool limited; s8 rssi; u16 uuid_count; u8 (*uuids)[16]; diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index af17774c9416..ea73e0826aa7 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -584,6 +584,8 @@ struct mgmt_rp_get_adv_size_info { __u8 max_scan_rsp_len; } __packed; +#define MGMT_OP_START_LIMITED_DISCOVERY 0x0041 + #define MGMT_EV_CMD_COMPLETE 0x0001 struct mgmt_ev_cmd_complete { __le16 opcode; diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 9997c31ef987..41b5f3813f02 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -1737,8 +1737,8 @@ static int le_scan_disable(struct hci_request *req, unsigned long opt) static int bredr_inquiry(struct hci_request *req, unsigned long opt) { u8 length = opt; - /* General inquiry access code (GIAC) */ - u8 lap[3] = { 0x33, 0x8b, 0x9e }; + const u8 giac[3] = { 0x33, 0x8b, 0x9e }; + const u8 liac[3] = { 0x00, 0x8b, 0x9e }; struct hci_cp_inquiry cp; BT_DBG("%s", req->hdev->name); @@ -1748,7 +1748,12 @@ static int bredr_inquiry(struct hci_request *req, unsigned long opt) hci_dev_unlock(req->hdev); memset(&cp, 0, sizeof(cp)); - memcpy(&cp.lap, lap, sizeof(cp.lap)); + + if (req->hdev->discovery.limited) + memcpy(&cp.lap, liac, sizeof(cp.lap)); + else + memcpy(&cp.lap, giac, sizeof(cp.lap)); + cp.length = length; hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 3297a4ecc05e..5a5089cb6570 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -103,6 +103,7 @@ static const u16 mgmt_commands[] = { MGMT_OP_ADD_ADVERTISING, MGMT_OP_REMOVE_ADVERTISING, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_OP_START_LIMITED_DISCOVERY, }; static const u16 mgmt_events[] = { @@ -3283,6 +3284,9 @@ void mgmt_start_discovery_complete(struct hci_dev *hdev, u8 status) if (!cmd) cmd = pending_find(MGMT_OP_START_SERVICE_DISCOVERY, hdev); + if (!cmd) + cmd = pending_find(MGMT_OP_START_LIMITED_DISCOVERY, hdev); + if (cmd) { cmd->cmd_complete(cmd, mgmt_status(status)); mgmt_pending_remove(cmd); @@ -3318,8 +3322,8 @@ static bool discovery_type_is_valid(struct hci_dev *hdev, uint8_t type, return true; } -static int start_discovery(struct sock *sk, struct hci_dev *hdev, - void *data, u16 len) +static int start_discovery_internal(struct sock *sk, struct hci_dev *hdev, + u16 op, void *data, u16 len) { struct mgmt_cp_start_discovery *cp = data; struct mgmt_pending_cmd *cmd; @@ -3331,7 +3335,7 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, hci_dev_lock(hdev); if (!hdev_is_powered(hdev)) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, + err = mgmt_cmd_complete(sk, hdev->id, op, MGMT_STATUS_NOT_POWERED, &cp->type, sizeof(cp->type)); goto failed; @@ -3339,15 +3343,14 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, if (hdev->discovery.state != DISCOVERY_STOPPED || hci_dev_test_flag(hdev, HCI_PERIODIC_INQ)) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, - MGMT_STATUS_BUSY, &cp->type, - sizeof(cp->type)); + err = mgmt_cmd_complete(sk, hdev->id, op, MGMT_STATUS_BUSY, + &cp->type, sizeof(cp->type)); goto failed; } if (!discovery_type_is_valid(hdev, cp->type, &status)) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, - status, &cp->type, sizeof(cp->type)); + err = mgmt_cmd_complete(sk, hdev->id, op, status, + &cp->type, sizeof(cp->type)); goto failed; } @@ -3358,8 +3361,12 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, hdev->discovery.type = cp->type; hdev->discovery.report_invalid_rssi = false; + if (op == MGMT_OP_START_LIMITED_DISCOVERY) + hdev->discovery.limited = true; + else + hdev->discovery.limited = false; - cmd = mgmt_pending_add(sk, MGMT_OP_START_DISCOVERY, hdev, data, len); + cmd = mgmt_pending_add(sk, op, hdev, data, len); if (!cmd) { err = -ENOMEM; goto failed; @@ -3376,6 +3383,21 @@ failed: return err; } +static int start_discovery(struct sock *sk, struct hci_dev *hdev, + void *data, u16 len) +{ + return start_discovery_internal(sk, hdev, MGMT_OP_START_DISCOVERY, + data, len); +} + +static int start_limited_discovery(struct sock *sk, struct hci_dev *hdev, + void *data, u16 len) +{ + return start_discovery_internal(sk, hdev, + MGMT_OP_START_LIMITED_DISCOVERY, + data, len); +} + static int service_discovery_cmd_complete(struct mgmt_pending_cmd *cmd, u8 status) { @@ -6313,6 +6335,7 @@ static const struct hci_mgmt_handler mgmt_handlers[] = { HCI_MGMT_VAR_LEN }, { remove_advertising, MGMT_REMOVE_ADVERTISING_SIZE }, { get_adv_size_info, MGMT_GET_ADV_SIZE_INFO_SIZE }, + { start_limited_discovery, MGMT_START_DISCOVERY_SIZE }, }; void mgmt_index_added(struct hci_dev *hdev) @@ -7237,6 +7260,18 @@ void mgmt_device_found(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type, return; } + if (hdev->discovery.limited) { + /* Check for limited discoverable bit */ + if (dev_class) { + if (!(dev_class[1] & 0x20)) + return; + } else { + u8 *flags = eir_get_data(eir, eir_len, EIR_FLAGS, NULL); + if (!flags || !(flags[0] & LE_AD_LIMITED)) + return; + } + } + /* Make sure that the buffer is big enough. The 5 extra bytes * are for the potential CoD field. */ -- cgit v1.2.3-59-g8ed1b From d6c0256a60e685214cc8cc2b886809f11efc0084 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 30 Dec 2015 23:50:46 +0800 Subject: sctp: add the rhashtable apis for sctp global transport hashtable tranport hashtbale will replace the association hashtable to do the lookup for transport, and then get association by t->assoc, rhashtable apis will be used because of it's resizable, scalable and using rcu. lport + rport + paddr will be the base hashkey to locate the chain, with net to protect one netns from another, then plus the laddr to compare to get the target. this patch will provider the lookup functions: - sctp_epaddr_lookup_transport - sctp_addrs_lookup_transport hash/unhash functions: - sctp_hash_transport - sctp_unhash_transport init/destroy functions: - sctp_transport_hashtable_init - sctp_transport_hashtable_destroy Signed-off-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 11 ++++ include/net/sctp/structs.h | 5 ++ net/sctp/input.c | 131 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 147 insertions(+) (limited to 'include') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index ce13cf20f625..7bbdfbab2efa 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -143,6 +143,17 @@ void sctp_icmp_proto_unreachable(struct sock *sk, struct sctp_transport *t); void sctp_backlog_migrate(struct sctp_association *assoc, struct sock *oldsk, struct sock *newsk); +int sctp_transport_hashtable_init(void); +void sctp_transport_hashtable_destroy(void); +void sctp_hash_transport(struct sctp_transport *t); +void sctp_unhash_transport(struct sctp_transport *t); +struct sctp_transport *sctp_addrs_lookup_transport( + struct net *net, + const union sctp_addr *laddr, + const union sctp_addr *paddr); +struct sctp_transport *sctp_epaddr_lookup_transport( + const struct sctp_endpoint *ep, + const union sctp_addr *paddr); /* * sctp/proc.c diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index eea9bdeecba2..4ab87d08e766 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -48,6 +48,7 @@ #define __sctp_structs_h__ #include +#include #include /* linux/in.h needs this!! */ #include /* We get struct sockaddr_in. */ #include /* We get struct in6_addr */ @@ -123,6 +124,8 @@ extern struct sctp_globals { struct sctp_hashbucket *assoc_hashtable; /* This is the sctp port control hash. */ struct sctp_bind_hashbucket *port_hashtable; + /* This is the hash of all transports. */ + struct rhashtable transport_hashtable; /* Sizes of above hashtables. */ int ep_hashsize; @@ -147,6 +150,7 @@ extern struct sctp_globals { #define sctp_assoc_hashtable (sctp_globals.assoc_hashtable) #define sctp_port_hashsize (sctp_globals.port_hashsize) #define sctp_port_hashtable (sctp_globals.port_hashtable) +#define sctp_transport_hashtable (sctp_globals.transport_hashtable) #define sctp_checksum_disable (sctp_globals.checksum_disable) /* SCTP Socket type: UDP or TCP style. */ @@ -753,6 +757,7 @@ static inline int sctp_packet_empty(struct sctp_packet *packet) struct sctp_transport { /* A list of transports. */ struct list_head transports; + struct rhash_head node; /* Reference counting. */ atomic_t refcnt; diff --git a/net/sctp/input.c b/net/sctp/input.c index b6493b3f11a9..bac8278b176b 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -782,6 +782,137 @@ hit: return ep; } +/* rhashtable for transport */ +struct sctp_hash_cmp_arg { + const union sctp_addr *laddr; + const union sctp_addr *paddr; + const struct net *net; +}; + +static inline int sctp_hash_cmp(struct rhashtable_compare_arg *arg, + const void *ptr) +{ + const struct sctp_hash_cmp_arg *x = arg->key; + const struct sctp_transport *t = ptr; + struct sctp_association *asoc = t->asoc; + const struct net *net = x->net; + + if (x->laddr->v4.sin_port != htons(asoc->base.bind_addr.port)) + return 1; + if (!sctp_cmp_addr_exact(&t->ipaddr, x->paddr)) + return 1; + if (!net_eq(sock_net(asoc->base.sk), net)) + return 1; + if (!sctp_bind_addr_match(&asoc->base.bind_addr, + x->laddr, sctp_sk(asoc->base.sk))) + return 1; + + return 0; +} + +static inline u32 sctp_hash_obj(const void *data, u32 len, u32 seed) +{ + const struct sctp_transport *t = data; + const union sctp_addr *paddr = &t->ipaddr; + const struct net *net = sock_net(t->asoc->base.sk); + u16 lport = htons(t->asoc->base.bind_addr.port); + u32 addr; + + if (paddr->sa.sa_family == AF_INET6) + addr = jhash(&paddr->v6.sin6_addr, 16, seed); + else + addr = paddr->v4.sin_addr.s_addr; + + return jhash_3words(addr, ((__u32)paddr->v4.sin_port) << 16 | + (__force __u32)lport, net_hash_mix(net), seed); +} + +static inline u32 sctp_hash_key(const void *data, u32 len, u32 seed) +{ + const struct sctp_hash_cmp_arg *x = data; + const union sctp_addr *paddr = x->paddr; + const struct net *net = x->net; + u16 lport = x->laddr->v4.sin_port; + u32 addr; + + if (paddr->sa.sa_family == AF_INET6) + addr = jhash(&paddr->v6.sin6_addr, 16, seed); + else + addr = paddr->v4.sin_addr.s_addr; + + return jhash_3words(addr, ((__u32)paddr->v4.sin_port) << 16 | + (__force __u32)lport, net_hash_mix(net), seed); +} + +static const struct rhashtable_params sctp_hash_params = { + .head_offset = offsetof(struct sctp_transport, node), + .hashfn = sctp_hash_key, + .obj_hashfn = sctp_hash_obj, + .obj_cmpfn = sctp_hash_cmp, + .automatic_shrinking = true, +}; + +int sctp_transport_hashtable_init(void) +{ + return rhashtable_init(&sctp_transport_hashtable, &sctp_hash_params); +} + +void sctp_transport_hashtable_destroy(void) +{ + rhashtable_destroy(&sctp_transport_hashtable); +} + +void sctp_hash_transport(struct sctp_transport *t) +{ + struct sctp_sockaddr_entry *addr; + struct sctp_hash_cmp_arg arg; + + addr = list_entry(t->asoc->base.bind_addr.address_list.next, + struct sctp_sockaddr_entry, list); + arg.laddr = &addr->a; + arg.paddr = &t->ipaddr; + arg.net = sock_net(t->asoc->base.sk); + +reinsert: + if (rhashtable_lookup_insert_key(&sctp_transport_hashtable, &arg, + &t->node, sctp_hash_params) == -EBUSY) + goto reinsert; +} + +void sctp_unhash_transport(struct sctp_transport *t) +{ + rhashtable_remove_fast(&sctp_transport_hashtable, &t->node, + sctp_hash_params); +} + +struct sctp_transport *sctp_addrs_lookup_transport( + struct net *net, + const union sctp_addr *laddr, + const union sctp_addr *paddr) +{ + struct sctp_hash_cmp_arg arg = { + .laddr = laddr, + .paddr = paddr, + .net = net, + }; + + return rhashtable_lookup_fast(&sctp_transport_hashtable, &arg, + sctp_hash_params); +} + +struct sctp_transport *sctp_epaddr_lookup_transport( + const struct sctp_endpoint *ep, + const union sctp_addr *paddr) +{ + struct sctp_sockaddr_entry *addr; + struct net *net = sock_net(ep->base.sk); + + addr = list_entry(ep->base.bind_addr.address_list.next, + struct sctp_sockaddr_entry, list); + + return sctp_addrs_lookup_transport(net, &addr->a, paddr); +} + /* Insert association into the hash table. */ static void __sctp_hash_established(struct sctp_association *asoc) { -- cgit v1.2.3-59-g8ed1b From b5eff7128366c4a7a9b502097a968ec9cae2bea2 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 30 Dec 2015 23:50:49 +0800 Subject: sctp: drop the old assoc hashtable of sctp transport hashtable will replace the association hashtable, so association hashtable is not used in sctp any more, so drop the codes about that. Signed-off-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 21 ---------------- include/net/sctp/structs.h | 5 ---- net/sctp/input.c | 61 ---------------------------------------------- net/sctp/protocol.c | 30 ++--------------------- net/sctp/sm_sideeffect.c | 2 -- net/sctp/socket.c | 6 +---- 6 files changed, 3 insertions(+), 122 deletions(-) (limited to 'include') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 7bbdfbab2efa..835aa2ed9870 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -126,8 +126,6 @@ int sctp_primitive_ASCONF(struct net *, struct sctp_association *, void *arg); */ int sctp_rcv(struct sk_buff *skb); void sctp_v4_err(struct sk_buff *skb, u32 info); -void sctp_hash_established(struct sctp_association *); -void sctp_unhash_established(struct sctp_association *); void sctp_hash_endpoint(struct sctp_endpoint *); void sctp_unhash_endpoint(struct sctp_endpoint *); struct sock *sctp_err_lookup(struct net *net, int family, struct sk_buff *, @@ -530,25 +528,6 @@ static inline int sctp_ep_hashfn(struct net *net, __u16 lport) return (net_hash_mix(net) + lport) & (sctp_ep_hashsize - 1); } -/* This is the hash function for the association hash table. */ -static inline int sctp_assoc_hashfn(struct net *net, __u16 lport, __u16 rport) -{ - int h = (lport << 16) + rport + net_hash_mix(net); - h ^= h>>8; - return h & (sctp_assoc_hashsize - 1); -} - -/* This is the hash function for the association hash table. This is - * not used yet, but could be used as a better hash function when - * we have a vtag. - */ -static inline int sctp_vtag_hashfn(__u16 lport, __u16 rport, __u32 vtag) -{ - int h = (lport << 16) + rport; - h ^= vtag; - return h & (sctp_assoc_hashsize - 1); -} - #define sctp_for_each_hentry(epb, head) \ hlist_for_each_entry(epb, head, node) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 4ab87d08e766..20e72129be1c 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -120,8 +120,6 @@ extern struct sctp_globals { /* This is the hash of all endpoints. */ struct sctp_hashbucket *ep_hashtable; - /* This is the hash of all associations. */ - struct sctp_hashbucket *assoc_hashtable; /* This is the sctp port control hash. */ struct sctp_bind_hashbucket *port_hashtable; /* This is the hash of all transports. */ @@ -129,7 +127,6 @@ extern struct sctp_globals { /* Sizes of above hashtables. */ int ep_hashsize; - int assoc_hashsize; int port_hashsize; /* Default initialization values to be applied to new associations. */ @@ -146,8 +143,6 @@ extern struct sctp_globals { #define sctp_address_families (sctp_globals.address_families) #define sctp_ep_hashsize (sctp_globals.ep_hashsize) #define sctp_ep_hashtable (sctp_globals.ep_hashtable) -#define sctp_assoc_hashsize (sctp_globals.assoc_hashsize) -#define sctp_assoc_hashtable (sctp_globals.assoc_hashtable) #define sctp_port_hashsize (sctp_globals.port_hashsize) #define sctp_port_hashtable (sctp_globals.port_hashtable) #define sctp_transport_hashtable (sctp_globals.transport_hashtable) diff --git a/net/sctp/input.c b/net/sctp/input.c index 6f075d835764..d9a6e66c5c8a 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -913,67 +913,6 @@ struct sctp_transport *sctp_epaddr_lookup_transport( return sctp_addrs_lookup_transport(net, &addr->a, paddr); } -/* Insert association into the hash table. */ -static void __sctp_hash_established(struct sctp_association *asoc) -{ - struct net *net = sock_net(asoc->base.sk); - struct sctp_ep_common *epb; - struct sctp_hashbucket *head; - - epb = &asoc->base; - - /* Calculate which chain this entry will belong to. */ - epb->hashent = sctp_assoc_hashfn(net, epb->bind_addr.port, - asoc->peer.port); - - head = &sctp_assoc_hashtable[epb->hashent]; - - write_lock(&head->lock); - hlist_add_head(&epb->node, &head->chain); - write_unlock(&head->lock); -} - -/* Add an association to the hash. Local BH-safe. */ -void sctp_hash_established(struct sctp_association *asoc) -{ - if (asoc->temp) - return; - - local_bh_disable(); - __sctp_hash_established(asoc); - local_bh_enable(); -} - -/* Remove association from the hash table. */ -static void __sctp_unhash_established(struct sctp_association *asoc) -{ - struct net *net = sock_net(asoc->base.sk); - struct sctp_hashbucket *head; - struct sctp_ep_common *epb; - - epb = &asoc->base; - - epb->hashent = sctp_assoc_hashfn(net, epb->bind_addr.port, - asoc->peer.port); - - head = &sctp_assoc_hashtable[epb->hashent]; - - write_lock(&head->lock); - hlist_del_init(&epb->node); - write_unlock(&head->lock); -} - -/* Remove association from the hash table. Local BH-safe. */ -void sctp_unhash_established(struct sctp_association *asoc) -{ - if (asoc->temp) - return; - - local_bh_disable(); - __sctp_unhash_established(asoc); - local_bh_enable(); -} - /* Look up an association. */ static struct sctp_association *__sctp_lookup_association( struct net *net, diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 631cfb380535..ab0d538a74ed 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1416,24 +1416,6 @@ static __init int sctp_init(void) for (order = 0; (1UL << order) < goal; order++) ; - do { - sctp_assoc_hashsize = (1UL << order) * PAGE_SIZE / - sizeof(struct sctp_hashbucket); - if ((sctp_assoc_hashsize > (64 * 1024)) && order > 0) - continue; - sctp_assoc_hashtable = (struct sctp_hashbucket *) - __get_free_pages(GFP_KERNEL | __GFP_NOWARN, order); - } while (!sctp_assoc_hashtable && --order > 0); - if (!sctp_assoc_hashtable) { - pr_err("Failed association hash alloc\n"); - status = -ENOMEM; - goto err_ahash_alloc; - } - for (i = 0; i < sctp_assoc_hashsize; i++) { - rwlock_init(&sctp_assoc_hashtable[i].lock); - INIT_HLIST_HEAD(&sctp_assoc_hashtable[i].chain); - } - /* Allocate and initialize the endpoint hash table. */ sctp_ep_hashsize = 64; sctp_ep_hashtable = @@ -1470,8 +1452,7 @@ static __init int sctp_init(void) if (sctp_transport_hashtable_init()) goto err_thash_alloc; - pr_info("Hash tables configured (established %d bind %d)\n", - sctp_assoc_hashsize, sctp_port_hashsize); + pr_info("Hash tables configured (bind %d)\n", sctp_port_hashsize); sctp_sysctl_register(); @@ -1528,10 +1509,6 @@ err_bhash_alloc: err_thash_alloc: kfree(sctp_ep_hashtable); err_ehash_alloc: - free_pages((unsigned long)sctp_assoc_hashtable, - get_order(sctp_assoc_hashsize * - sizeof(struct sctp_hashbucket))); -err_ahash_alloc: percpu_counter_destroy(&sctp_sockets_allocated); err_percpu_counter_init: kmem_cache_destroy(sctp_chunk_cachep); @@ -1565,13 +1542,10 @@ static __exit void sctp_exit(void) sctp_sysctl_unregister(); - free_pages((unsigned long)sctp_assoc_hashtable, - get_order(sctp_assoc_hashsize * - sizeof(struct sctp_hashbucket))); - kfree(sctp_ep_hashtable); free_pages((unsigned long)sctp_port_hashtable, get_order(sctp_port_hashsize * sizeof(struct sctp_bind_hashbucket))); + kfree(sctp_ep_hashtable); sctp_transport_hashtable_destroy(); percpu_counter_destroy(&sctp_sockets_allocated); diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 05cd16400e0b..4f170ad38ff4 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -866,7 +866,6 @@ static void sctp_cmd_delete_tcb(sctp_cmd_seq_t *cmds, (!asoc->temp) && (sk->sk_shutdown != SHUTDOWN_MASK)) return; - sctp_unhash_established(asoc); sctp_association_free(asoc); } @@ -1269,7 +1268,6 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, asoc = cmd->obj.asoc; BUG_ON(asoc->peer.primary_path == NULL); sctp_endpoint_add_asoc(ep, asoc); - sctp_hash_established(asoc); break; case SCTP_CMD_UPDATE_ASSOC: diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b5f4811cea82..9bb80ec4c08f 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1228,7 +1228,6 @@ out_free: * To the hash table, try to unhash it, just in case, its a noop * if it wasn't hashed so we're safe */ - sctp_unhash_established(asoc); sctp_association_free(asoc); } return err; @@ -1504,7 +1503,6 @@ static void sctp_close(struct sock *sk, long timeout) * ABORT or SHUTDOWN based on the linger options. */ if (sctp_state(asoc, CLOSED)) { - sctp_unhash_established(asoc); sctp_association_free(asoc); continue; } @@ -1986,10 +1984,8 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) goto out_unlock; out_free: - if (new_asoc) { - sctp_unhash_established(asoc); + if (new_asoc) sctp_association_free(asoc); - } out_unlock: release_sock(sk); -- cgit v1.2.3-59-g8ed1b From b0844444590e18704644f707ea88bff1b976b0e7 Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Tue, 29 Dec 2015 14:58:30 +0200 Subject: net/mlx5_core: Introduce access function to read internal timer A preparation step which adds support for reading the hardware internal timer and the hardware timestamping from the CQE. In addition, advertize device_frequency_khz HCA capability. Signed-off-by: Eran Ben Elisha Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 13 +++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h | 1 + include/linux/mlx5/device.h | 20 +++++++++++++++++--- include/linux/mlx5/mlx5_ifc.h | 6 +++--- 4 files changed, 34 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 789882b7b711..67676cf0d507 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -504,6 +504,19 @@ int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id) return mlx5_cmd_status_to_err_v2(out); } +cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev) +{ + u32 timer_h, timer_h1, timer_l; + + timer_h = ioread32be(&dev->iseg->internal_timer_h); + timer_l = ioread32be(&dev->iseg->internal_timer_l); + timer_h1 = ioread32be(&dev->iseg->internal_timer_h); + if (timer_h != timer_h1) /* wrap around */ + timer_l = ioread32be(&dev->iseg->internal_timer_l); + + return (cycle_t)timer_l | (cycle_t)timer_h1 << 32; +} + static int mlx5_irq_set_affinity_hint(struct mlx5_core_dev *mdev, int i) { struct mlx5_priv *priv = &mdev->priv; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index ea6a137fd76c..0336847ec9a1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -98,6 +98,7 @@ int mlx5_core_sriov_configure(struct pci_dev *dev, int num_vfs); int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id); int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id); int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev); +cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev); void mlx5e_init(void); void mlx5e_cleanup(void); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 7d3a85faefb7..df2f79ef3cac 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -443,9 +443,12 @@ struct mlx5_init_seg { __be32 rsvd1[120]; __be32 initializing; struct health_buffer health; - __be32 rsvd2[884]; + __be32 rsvd2[880]; + __be32 internal_timer_h; + __be32 internal_timer_l; + __be32 rsrv3[2]; __be32 health_counter; - __be32 rsvd3[1019]; + __be32 rsvd4[1019]; __be64 ieee1588_clk; __be32 ieee1588_clk_type; __be32 clr_intx; @@ -601,7 +604,8 @@ struct mlx5_cqe64 { __be32 imm_inval_pkey; u8 rsvd40[4]; __be32 byte_cnt; - __be64 timestamp; + __be32 timestamp_h; + __be32 timestamp_l; __be32 sop_drop_qpn; __be16 wqe_counter; u8 signature; @@ -623,6 +627,16 @@ static inline int cqe_has_vlan(struct mlx5_cqe64 *cqe) return !!(cqe->l4_hdr_type_etc & 0x1); } +static inline u64 get_cqe_ts(struct mlx5_cqe64 *cqe) +{ + u32 hi, lo; + + hi = be32_to_cpu(cqe->timestamp_h); + lo = be32_to_cpu(cqe->timestamp_l); + + return (u64)lo | ((u64)hi << 32); +} + enum { CQE_L4_HDR_TYPE_NONE = 0x0, CQE_L4_HDR_TYPE_TCP_NO_ACK = 0x1, diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 131a2737cfa3..1780a85a8797 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -829,9 +829,9 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_66[0x8]; u8 log_uar_page_sz[0x10]; - u8 reserved_67[0xe0]; - - u8 reserved_68[0x1f]; + u8 reserved_67[0x40]; + u8 device_frequency_khz[0x20]; + u8 reserved_68[0x5f]; u8 cqe_zip[0x1]; u8 cqe_zip_timeout[0x10]; -- cgit v1.2.3-59-g8ed1b From 2fbf575867e5a181a3f3e5e29a2f0c205cca5fb3 Mon Sep 17 00:00:00 2001 From: "xypron.glpk@gmx.de" Date: Tue, 5 Jan 2016 10:12:49 +0100 Subject: include/uapi/linux/sockios.h: mark SIOCRTMSG unused IOCTL SIOCRTMSG does nothing but return EINVAL. So comment it as unused. SIOCRTMSG is only used in: * net/ipv4/af_inet.c * include/uapi/linux/sockios.h inet_ioctl calls ip_rt_ioctl. ip_rt_ioctl only handles SIOCADDRT and SIOCDELRT and returns -EINVAL otherwise. Signed-off-by: Heinrich Schuchardt Signed-off-by: David S. Miller --- include/uapi/linux/sockios.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/sockios.h b/include/uapi/linux/sockios.h index e888b1aed69f..8e7890b26d9a 100644 --- a/include/uapi/linux/sockios.h +++ b/include/uapi/linux/sockios.h @@ -27,7 +27,7 @@ /* Routing table calls. */ #define SIOCADDRT 0x890B /* add routing table entry */ #define SIOCDELRT 0x890C /* delete routing table entry */ -#define SIOCRTMSG 0x890D /* call to routing system */ +#define SIOCRTMSG 0x890D /* unused */ /* Socket configuration controls. */ #define SIOCGIFNAME 0x8910 /* get iface name */ -- cgit v1.2.3-59-g8ed1b From a72a5e2d34ec2921c0d9a7545093087e4cb90d0a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 5 Jan 2016 22:17:55 +0100 Subject: inet: kill unused skb_free op The only user was removed in commit 029f7f3b8701cc7a ("netfilter: ipv6: nf_defrag: avoid/free clone operations"). Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/net/inet_frag.h | 1 - net/ieee802154/6lowpan/reassembly.c | 1 - net/ipv4/inet_fragment.c | 10 +--------- net/ipv4/ip_fragment.c | 1 - net/ipv6/reassembly.c | 1 - 5 files changed, 1 insertion(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index ac42bbb37b2d..12aac0fd6ee7 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -99,7 +99,6 @@ struct inet_frags { void (*constructor)(struct inet_frag_queue *q, const void *arg); void (*destructor)(struct inet_frag_queue *); - void (*skb_free)(struct sk_buff *); void (*frag_expire)(unsigned long data); struct kmem_cache *frags_cachep; const char *frags_cache_name; diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 6b437e8760d3..30d875dff6b5 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -624,7 +624,6 @@ int __init lowpan_net_frag_init(void) lowpan_frags.hashfn = lowpan_hashfn; lowpan_frags.constructor = lowpan_frag_init; lowpan_frags.destructor = NULL; - lowpan_frags.skb_free = NULL; lowpan_frags.qsize = sizeof(struct frag_queue); lowpan_frags.match = lowpan_frag_match; lowpan_frags.frag_expire = lowpan_frag_expire; diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index fe144dae7372..3a88b0c73797 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -285,14 +285,6 @@ void inet_frag_kill(struct inet_frag_queue *fq, struct inet_frags *f) } EXPORT_SYMBOL(inet_frag_kill); -static inline void frag_kfree_skb(struct netns_frags *nf, struct inet_frags *f, - struct sk_buff *skb) -{ - if (f->skb_free) - f->skb_free(skb); - kfree_skb(skb); -} - void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f) { struct sk_buff *fp; @@ -309,7 +301,7 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f) struct sk_buff *xp = fp->next; sum_truesize += fp->truesize; - frag_kfree_skb(nf, f, fp); + kfree_skb(fp); fp = xp; } sum = sum_truesize + f->qsize; diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 1fe55ae81781..3f00810b7288 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -891,7 +891,6 @@ void __init ipfrag_init(void) ip4_frags.hashfn = ip4_hashfn; ip4_frags.constructor = ip4_frag_init; ip4_frags.destructor = ip4_frag_free; - ip4_frags.skb_free = NULL; ip4_frags.qsize = sizeof(struct ipq); ip4_frags.match = ip4_frag_match; ip4_frags.frag_expire = ip_expire; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 45f5ae51de65..18f3498a6c80 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -755,7 +755,6 @@ int __init ipv6_frag_init(void) ip6_frags.hashfn = ip6_hashfn; ip6_frags.constructor = ip6_frag_init; ip6_frags.destructor = NULL; - ip6_frags.skb_free = NULL; ip6_frags.qsize = sizeof(struct frag_queue); ip6_frags.match = ip6_frag_match; ip6_frags.frag_expire = ip6_frag_expire; -- cgit v1.2.3-59-g8ed1b From 81435c33e062cbd4508da6f64655cb0967eeb65f Mon Sep 17 00:00:00 2001 From: Elad Raz Date: Wed, 6 Jan 2016 13:01:05 +0100 Subject: switchdev: add bridge vlan_filtering attribute Adding vlan_filtering attribute to allow hardware vendor to support vlan-aware bridges. Vlan_filtering is a per-bridge attribute. Signed-off-by: Elad Raz Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/switchdev.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 6612946167fe..603ae2f88dbb 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -47,6 +47,7 @@ enum switchdev_attr_id { SWITCHDEV_ATTR_ID_PORT_STP_STATE, SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME, + SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING, }; struct switchdev_attr { @@ -58,6 +59,7 @@ struct switchdev_attr { u8 stp_state; /* PORT_STP_STATE */ unsigned long brport_flags; /* PORT_BRIDGE_FLAGS */ u32 ageing_time; /* BRIDGE_AGEING_TIME */ + bool vlan_filtering; /* BRIDGE_VLAN_FILTERING */ } u; }; -- cgit v1.2.3-59-g8ed1b From cdba756f5803a2f0a8bbc6605acc166dd817979e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 6 Jan 2016 06:53:50 -0800 Subject: net: move ndo_features_check() close to ndo_start_xmit() TX fast path uses ndo_start_xmit(), ndo_features_check() and ndo_select_queue(). Move ndo_features_check() close to ndo_start_xmit() to increase data locality. All "struct net_device_ops" should now be using C99 initializers. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c20b814e46a0..8d8e5ca951b4 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -812,6 +812,12 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX) * Required can not be NULL. * + * netdev_features_t (*ndo_fix_features)(struct net_device *dev, + * netdev_features_t features); + * Adjusts the requested feature flags according to device-specific + * constraints, and returns the resulting flags. Must not modify + * the device state. + * * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb, * void *accel_priv, select_queue_fallback_t fallback); * Called to decide which queue to when device supports multiple @@ -959,12 +965,6 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * Called to release previously enslaved netdev. * * Feature/offload setting functions. - * netdev_features_t (*ndo_fix_features)(struct net_device *dev, - * netdev_features_t features); - * Adjusts the requested feature flags according to device-specific - * constraints, and returns the resulting flags. Must not modify - * the device state. - * * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features); * Called to update device configuration to new features. Passed * feature set might be less than what was returned by ndo_fix_features()). @@ -1081,8 +1081,11 @@ struct net_device_ops { void (*ndo_uninit)(struct net_device *dev); int (*ndo_open)(struct net_device *dev); int (*ndo_stop)(struct net_device *dev); - netdev_tx_t (*ndo_start_xmit) (struct sk_buff *skb, - struct net_device *dev); + netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb, + struct net_device *dev); + netdev_features_t (*ndo_features_check)(struct sk_buff *skb, + struct net_device *dev, + netdev_features_t features); u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb, void *accel_priv, @@ -1245,9 +1248,6 @@ struct net_device_ops { struct net_device *dev, void *priv); int (*ndo_get_lock_subclass)(struct net_device *dev); - netdev_features_t (*ndo_features_check) (struct sk_buff *skb, - struct net_device *dev, - netdev_features_t features); int (*ndo_set_tx_maxrate)(struct net_device *dev, int queue_index, u32 maxrate); -- cgit v1.2.3-59-g8ed1b From c7f5d105495a38ed09e70d825f75d9d7d5407264 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 5 Nov 2015 11:34:57 -0500 Subject: net: Add eth_platform_get_mac_address() helper. A repeating pattern in drivers has become to use OF node information and, if not found, platform specific host information to extract the ethernet address for a given device. Currently this is done with a call to of_get_mac_address() and then some ifdef'd stuff for SPARC. Consolidate this into a portable routine, and provide the arch_get_platform_mac_address() weak function hook for all architectures to implement if they want. Signed-off-by: David S. Miller --- arch/sparc/kernel/idprom.c | 7 +++++++ include/linux/etherdevice.h | 3 +++ net/ethernet/eth.c | 31 +++++++++++++++++++++++++++++++ 3 files changed, 41 insertions(+) (limited to 'include') diff --git a/arch/sparc/kernel/idprom.c b/arch/sparc/kernel/idprom.c index 6bd75012109d..f95dd11b75ea 100644 --- a/arch/sparc/kernel/idprom.c +++ b/arch/sparc/kernel/idprom.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -60,6 +61,12 @@ static void __init display_system_type(unsigned char machtype) { } #endif + +unsigned char *arch_get_platform_mac_address(void) +{ + return idprom->id_ethaddr; +} + /* Calculate the IDPROM checksum (xor of the data bytes). */ static unsigned char __init calc_idprom_cksum(struct idprom *idprom) { diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index eb049c622208..37ff4a6faa9a 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -29,6 +29,9 @@ #include #ifdef __KERNEL__ +struct device; +int eth_platform_get_mac_address(struct device *dev, u8 *mac_addr); +unsigned char *arch_get_platform_get_mac_address(void); u32 eth_get_headlen(void *data, unsigned int max_len); __be16 eth_type_trans(struct sk_buff *skb, struct net_device *dev); extern const struct header_ops eth_header_ops; diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 9e63f252a89e..103871784e50 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -52,6 +52,8 @@ #include #include #include +#include +#include #include #include #include @@ -485,3 +487,32 @@ static int __init eth_offload_init(void) } fs_initcall(eth_offload_init); + +unsigned char * __weak arch_get_platform_mac_address(void) +{ + return NULL; +} + +int eth_platform_get_mac_address(struct device *dev, u8 *mac_addr) +{ + const unsigned char *addr; + struct device_node *dp; + + if (dev_is_pci(dev)) + dp = pci_device_to_OF_node(to_pci_dev(dev)); + else + dp = dev->of_node; + + addr = NULL; + if (dp) + addr = of_get_mac_address(dp); + if (!addr) + addr = arch_get_platform_mac_address(); + + if (!addr) + return -ENODEV; + + ether_addr_copy(mac_addr, addr); + return 0; +} +EXPORT_SYMBOL(eth_platform_get_mac_address); -- cgit v1.2.3-59-g8ed1b From ccaa953e9fc7ebb90fba4e4815966683bef4866f Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:06 +0100 Subject: phy: Consistently use addr for address on an MII bus Within phy.h, an address on an MII bus has been called both addr and phy_id. phy_id is particularly confusion, since it also means the ID found in register 3, if the device on the bus is a phy. Consistently use addr. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/phy.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/phy.h b/include/linux/phy.h index a89cb0eef911..77b5e56e2a92 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -158,8 +158,8 @@ struct mii_bus { const char *name; char id[MII_BUS_ID_SIZE]; void *priv; - int (*read)(struct mii_bus *bus, int phy_id, int regnum); - int (*write)(struct mii_bus *bus, int phy_id, int regnum, u16 val); + int (*read)(struct mii_bus *bus, int addr, int regnum); + int (*write)(struct mii_bus *bus, int addr, int regnum, u16 val); int (*reset)(struct mii_bus *bus); /* -- cgit v1.2.3-59-g8ed1b From bac83c653799d7ea3f6cc4d7396d75adc5e0f778 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:07 +0100 Subject: mdio: Move mdiobus_read/write operatings into mdio.h These are logically MDIO operations, not phy operations, so move them into the mdio header. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/mdio.h | 6 ++++++ include/linux/phy.h | 6 +----- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/mdio.h b/include/linux/mdio.h index b42963bc81dd..0d073c23c10d 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -11,6 +11,7 @@ #include +struct mii_bus; static inline bool mdio_phy_id_is_c45(int phy_id) { @@ -173,4 +174,9 @@ static inline u16 ethtool_adv_to_mmd_eee_adv_t(u32 adv) return reg; } +int mdiobus_read(struct mii_bus *bus, int addr, u32 regnum); +int mdiobus_read_nested(struct mii_bus *bus, int addr, u32 regnum); +int mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val); +int mdiobus_write_nested(struct mii_bus *bus, int addr, u32 regnum, u16 val); + #endif /* __LINUX_MDIO_H__ */ diff --git a/include/linux/phy.h b/include/linux/phy.h index 77b5e56e2a92..8ca161a37e8a 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -18,6 +18,7 @@ #include #include +#include #include #include #include @@ -212,11 +213,6 @@ static inline struct mii_bus *devm_mdiobus_alloc(struct device *dev) void devm_mdiobus_free(struct device *dev, struct mii_bus *bus); struct phy_device *mdiobus_scan(struct mii_bus *bus, int addr); -int mdiobus_read(struct mii_bus *bus, int addr, u32 regnum); -int mdiobus_read_nested(struct mii_bus *bus, int addr, u32 regnum); -int mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val); -int mdiobus_write_nested(struct mii_bus *bus, int addr, u32 regnum, u16 val); - #define PHY_INTERRUPT_DISABLED 0x0 #define PHY_INTERRUPT_ENABLED 0x80000000 -- cgit v1.2.3-59-g8ed1b From 72ba48be3ec8e70937ad97d4420ef7144617c64b Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:09 +0100 Subject: phy: Add phydev_err() and phydev_dbg() macros In preparation for moving some of the phy_device structure members, add macros for printing errors and debug information. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/at803x.c | 4 ++-- drivers/net/phy/bcm87xx.c | 5 +++-- drivers/net/phy/micrel.c | 16 +++++++++------- drivers/net/phy/phy.c | 5 +++-- include/linux/phy.h | 6 ++++++ 5 files changed, 23 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c index 2d020a3ec0b5..62361f8af375 100644 --- a/drivers/net/phy/at803x.c +++ b/drivers/net/phy/at803x.c @@ -281,8 +281,8 @@ static void at803x_link_change_notify(struct phy_device *phydev) at803x_context_restore(phydev, &context); - dev_dbg(&phydev->dev, "%s(): phy was reset\n", - __func__); + phydev_dbg(phydev, "%s(): phy was reset\n", + __func__); priv->phy_reset = true; } } else { diff --git a/drivers/net/phy/bcm87xx.c b/drivers/net/phy/bcm87xx.c index 1eca20452f03..71b491c7bf96 100644 --- a/drivers/net/phy/bcm87xx.c +++ b/drivers/net/phy/bcm87xx.c @@ -163,8 +163,9 @@ static int bcm87xx_did_interrupt(struct phy_device *phydev) reg = phy_read(phydev, BCM87XX_LASI_STATUS); if (reg < 0) { - dev_err(&phydev->dev, - "Error: Read of BCM87XX_LASI_STATUS failed: %d\n", reg); + phydev_err(phydev, + "Error: Read of BCM87XX_LASI_STATUS failed: %d\n", + reg); return 0; } return (reg & 1) != 0; diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 1a6048a8c29d..bf72365e90bc 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -224,7 +224,7 @@ static int kszphy_setup_led(struct phy_device *phydev, u32 reg, int val) rc = phy_write(phydev, reg, temp); out: if (rc < 0) - dev_err(&phydev->dev, "failed to set led mode\n"); + phydev_err(phydev, "failed to set led mode\n"); return rc; } @@ -243,7 +243,7 @@ static int kszphy_broadcast_disable(struct phy_device *phydev) ret = phy_write(phydev, MII_KSZPHY_OMSO, ret | KSZPHY_OMSO_B_CAST_OFF); out: if (ret) - dev_err(&phydev->dev, "failed to disable broadcast address\n"); + phydev_err(phydev, "failed to disable broadcast address\n"); return ret; } @@ -263,7 +263,7 @@ static int kszphy_nand_tree_disable(struct phy_device *phydev) ret & ~KSZPHY_OMSO_NAND_TREE_ON); out: if (ret) - dev_err(&phydev->dev, "failed to disable NAND tree mode\n"); + phydev_err(phydev, "failed to disable NAND tree mode\n"); return ret; } @@ -288,7 +288,8 @@ static int kszphy_config_init(struct phy_device *phydev) if (priv->rmii_ref_clk_sel) { ret = kszphy_rmii_clk_sel(phydev, priv->rmii_ref_clk_sel_val); if (ret) { - dev_err(&phydev->dev, "failed to set rmii reference clock\n"); + phydev_err(phydev, + "failed to set rmii reference clock\n"); return ret; } } @@ -649,8 +650,8 @@ static int kszphy_probe(struct phy_device *phydev) priv->led_mode = -1; if (priv->led_mode > 3) { - dev_err(&phydev->dev, "invalid led mode: 0x%02x\n", - priv->led_mode); + phydev_err(phydev, "invalid led mode: 0x%02x\n", + priv->led_mode); priv->led_mode = -1; } } else { @@ -672,7 +673,8 @@ static int kszphy_probe(struct phy_device *phydev) } else if (rate > 49500000 && rate < 50500000) { priv->rmii_ref_clk_sel_val = !rmii_ref_clk_sel_25_mhz; } else { - dev_err(&phydev->dev, "Clock rate out of range: %ld\n", rate); + phydev_err(phydev, "Clock rate out of range: %ld\n", + rate); return -EINVAL; } } diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 47cd306dbb3c..9771941cf0ee 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -995,8 +995,9 @@ void phy_state_machine(struct work_struct *work) if (err < 0) phy_error(phydev); - dev_dbg(&phydev->dev, "PHY state change %s -> %s\n", - phy_state_to_str(old_state), phy_state_to_str(phydev->state)); + phydev_dbg(phydev, "PHY state change %s -> %s\n", + phy_state_to_str(old_state), + phy_state_to_str(phydev->state)); queue_delayed_work(system_power_efficient_wq, &phydev->state_queue, PHY_STATE_TIME * HZ); diff --git a/include/linux/phy.h b/include/linux/phy.h index 8ca161a37e8a..dbcf9fdd960c 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -777,6 +777,12 @@ static inline int phy_read_status(struct phy_device *phydev) return phydev->drv->read_status(phydev); } +#define phydev_err(_phydev, format, args...) \ + dev_err(&_phydev->dev, format, ##args) + +#define phydev_dbg(_phydev, format, args...) \ + dev_dbg(&_phydev->dev, format, ##args) + int genphy_config_init(struct phy_device *phydev); int genphy_setup_forced(struct phy_device *phydev); int genphy_restart_aneg(struct phy_device *phydev); -- cgit v1.2.3-59-g8ed1b From 84eff6d194df442bee62c129f2f47efb0dbd0468 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:10 +0100 Subject: phy: add phydev_name() wrapper Add a phydev_name() function, to help with moving some structure members from phy_device. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/8390/ax88796.c | 2 +- drivers/net/ethernet/adi/bfin_mac.c | 4 ++-- drivers/net/ethernet/agere/et131x.c | 4 ++-- drivers/net/ethernet/amd/au1000_eth.c | 4 ++-- drivers/net/ethernet/broadcom/b44.c | 2 +- drivers/net/ethernet/broadcom/tg3.c | 4 ++-- drivers/net/ethernet/cadence/macb.c | 2 +- drivers/net/ethernet/dnet.c | 6 +++--- drivers/net/ethernet/faraday/ftgmac100.c | 2 +- drivers/net/ethernet/lantiq_etop.c | 4 ++-- drivers/net/ethernet/nxp/lpc_eth.c | 4 ++-- drivers/net/ethernet/rdc/r6040.c | 4 ++-- drivers/net/ethernet/renesas/ravb_main.c | 2 +- drivers/net/ethernet/renesas/sh_eth.c | 2 +- drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c | 2 +- drivers/net/ethernet/smsc/smsc911x.c | 2 +- drivers/net/ethernet/smsc/smsc9420.c | 4 ++-- drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 2 +- drivers/net/ethernet/ti/davinci_emac.c | 2 +- drivers/net/ethernet/ti/davinci_mdio.c | 2 +- drivers/net/ethernet/ti/netcp_ethss.c | 4 ++-- drivers/net/ethernet/toshiba/tc35815.c | 4 ++-- drivers/net/phy/bcm7xxx.c | 2 +- drivers/net/phy/phy_device.c | 2 +- drivers/staging/netlogic/xlr_net.c | 6 +++--- include/linux/phy.h | 5 +++++ 26 files changed, 44 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c index 0443654f0339..90b540a4a561 100644 --- a/drivers/net/ethernet/8390/ax88796.c +++ b/drivers/net/ethernet/8390/ax88796.c @@ -372,7 +372,7 @@ static int ax_mii_probe(struct net_device *dev) ax->phy_dev = phy_dev; netdev_info(dev, "PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phy_dev->drv->name, dev_name(&phy_dev->dev), phy_dev->irq); + phy_dev->drv->name, phydev_name(phy_dev), phy_dev->irq); return 0; } diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index e0e95a15cab0..5f8a5182b8dc 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -419,7 +419,7 @@ static int mii_probe(struct net_device *dev, int phy_mode) return -EINVAL; } - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), &bfin_mac_adjust_link, phy_mode); if (IS_ERR(phydev)) { @@ -446,7 +446,7 @@ static int mii_probe(struct net_device *dev, int phy_mode) pr_info("attached PHY driver [%s] " "(mii_bus:phy_addr=%s, irq=%d, mdc_clk=%dHz(mdc_div=%d)@sclk=%dMHz)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq, + phydev->drv->name, phydev_name(phydev), phydev->irq, MDC_CLK, mdc_div, sclk/1000000); return 0; diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index e0f3d197e7f2..80b706f0fc97 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -3265,7 +3265,7 @@ static int et131x_mii_probe(struct net_device *netdev) return -ENODEV; } - phydev = phy_connect(netdev, dev_name(&phydev->dev), + phydev = phy_connect(netdev, phydev_name(phydev), &et131x_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { @@ -3291,7 +3291,7 @@ static int et131x_mii_probe(struct net_device *netdev) dev_info(&adapter->pdev->dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, dev_name(&phydev->dev)); + phydev->drv->name, phydev_name(phydev)); return 0; } diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 5330bcb8a944..8a8d6f2a0f6f 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -558,7 +558,7 @@ static int au1000_mii_probe(struct net_device *dev) /* now we are supposed to have a proper phydev, to attach to... */ BUG_ON(phydev->attached_dev); - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), &au1000_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { @@ -585,7 +585,7 @@ static int au1000_mii_probe(struct net_device *dev) netdev_info(dev, "attached PHY driver [%s] " "(mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); return 0; } diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index a3b1c07ae0af..928a2210e788 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2316,7 +2316,7 @@ static int b44_register_phy_one(struct b44 *bp) bp->phy_addr = phydev->addr; dev_info(sdev->dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, dev_name(&phydev->dev)); + phydev->drv->name, phydev_name(phydev)); return 0; diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 79789d8e52da..69d84d67f09a 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -2096,7 +2096,7 @@ static int tg3_phy_init(struct tg3 *tp) phydev = tp->mdio_bus->phy_map[tp->phy_addr]; /* Attach the MAC to the PHY. */ - phydev = phy_connect(tp->dev, dev_name(&phydev->dev), + phydev = phy_connect(tp->dev, phydev_name(phydev), tg3_adjust_link, phydev->interface); if (IS_ERR(phydev)) { dev_err(&tp->pdev->dev, "Could not attach to PHY\n"); @@ -17903,7 +17903,7 @@ static int tg3_init_one(struct pci_dev *pdev, phydev = tp->mdio_bus->phy_map[tp->phy_addr]; netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, dev_name(&phydev->dev)); + phydev->drv->name, phydev_name(phydev)); } else { char *ethtype; diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 8b45bc9ac29e..001d60c5521c 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -2951,7 +2951,7 @@ static int macb_probe(struct platform_device *pdev) phydev = bp->phy_dev; netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); return 0; diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 13d00a38a5bd..136b6010f704 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -274,11 +274,11 @@ static int dnet_mii_probe(struct net_device *dev) /* attach the mac to the phy */ if (bp->capabilities & DNET_HAS_RMII) { - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), &dnet_handle_link_change, PHY_INTERFACE_MODE_RMII); } else { - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), &dnet_handle_link_change, PHY_INTERFACE_MODE_MII); } @@ -894,7 +894,7 @@ static int dnet_probe(struct platform_device *pdev) phydev = bp->phy_dev; dev_info(&pdev->dev, "attached PHY driver [%s] " "(mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); return 0; diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 6d0c5d5eea6d..c2e2ac6a0313 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -854,7 +854,7 @@ static int ftgmac100_mii_probe(struct ftgmac100 *priv) return -ENODEV; } - phydev = phy_connect(netdev, dev_name(&phydev->dev), + phydev = phy_connect(netdev, phydev_name(phydev), &ftgmac100_adjust_link, PHY_INTERFACE_MODE_GMII); if (IS_ERR(phydev)) { diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 581928c068f2..274a3cec84c2 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -390,7 +390,7 @@ ltq_etop_mdio_probe(struct net_device *dev) return -ENODEV; } - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), <q_etop_mdio_link, priv->pldata->mii_mode); if (IS_ERR(phydev)) { @@ -410,7 +410,7 @@ ltq_etop_mdio_probe(struct net_device *dev) priv->phydev = phydev; pr_info("%s: attached PHY [%s] (phy_addr=%s, irq=%d)\n", dev->name, phydev->drv->name, - dev_name(&phydev->dev), phydev->irq); + phydev_name(phydev), phydev->irq); return 0; } diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index 057665180f13..5801aa197697 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -797,7 +797,7 @@ static int lpc_mii_probe(struct net_device *ndev) netdev_info(ndev, "using MII interface\n"); else netdev_info(ndev, "using RMII interface\n"); - phydev = phy_connect(ndev, dev_name(&phydev->dev), + phydev = phy_connect(ndev, phydev_name(phydev), &lpc_handle_link_change, lpc_phy_interface_mode(&pldat->pdev->dev)); @@ -818,7 +818,7 @@ static int lpc_mii_probe(struct net_device *ndev) netdev_info(ndev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); return 0; } diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 9a37247cf4b8..86a0887811c7 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -1039,7 +1039,7 @@ static int r6040_mii_probe(struct net_device *dev) return -ENODEV; } - phydev = phy_connect(dev, dev_name(&phydev->dev), &r6040_adjust_link, + phydev = phy_connect(dev, phydev_name(phydev), &r6040_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { @@ -1063,7 +1063,7 @@ static int r6040_mii_probe(struct net_device *dev) dev_info(&lp->pdev->dev, "attached PHY driver [%s] " "(mii_bus:phy_addr=%s)\n", - phydev->drv->name, dev_name(&phydev->dev)); + phydev->drv->name, phydev_name(phydev)); return 0; } diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 0e1ebb39ab46..2f6c974e4a6d 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -928,7 +928,7 @@ static int ravb_phy_init(struct net_device *ndev) phydev->supported &= ~PHY_10BT_FEATURES; netdev_info(ndev, "attached PHY %d (IRQ %d) to driver %s\n", - phydev->addr, phydev->irq, phydev->drv->name); + phydev->addr, phydev->irq, phydev_name(phydev)); priv->phydev = phydev; diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index baa81535a8fc..e14d28474b70 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -1827,7 +1827,7 @@ static int sh_eth_phy_init(struct net_device *ndev) } netdev_info(ndev, "attached PHY %d (IRQ %d) to driver %s\n", - phydev->addr, phydev->irq, phydev->drv->name); + phydev->addr, phydev->irq, phydev_name(phydev)); mdp->phydev = phydev; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c index 43ccb4a6de15..5b13b8c11bef 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c @@ -216,7 +216,7 @@ int sxgbe_mdio_register(struct net_device *ndev) } netdev_info(ndev, "PHY ID %08x at %d IRQ %s (%s)%s\n", phy->phy_id, phy_addr, irq_str, - dev_name(&phy->dev), act ? " active" : ""); + phydev_name(phy), act ? " active" : ""); phy_found = true; } } diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 219a99b7a631..067346d3209d 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1033,7 +1033,7 @@ static int smsc911x_mii_probe(struct net_device *dev) netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); /* mask with MAC supported features */ phydev->supported &= (PHY_BASIC_FEATURES | SUPPORTED_Pause | diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 4a90cdae5444..a02ed6b63064 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -1167,7 +1167,7 @@ static int smsc9420_mii_probe(struct net_device *dev) netif_info(pd, probe, pd->dev, "PHY addr %d, phy_id 0x%08X\n", phydev->addr, phydev->phy_id); - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), smsc9420_phy_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { @@ -1176,7 +1176,7 @@ static int smsc9420_mii_probe(struct net_device *dev) } netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, dev_name(&phydev->dev), phydev->irq); + phydev->drv->name, phydev_name(phydev), phydev->irq); /* mask with MAC supported features */ phydev->supported &= (PHY_BASIC_FEATURES | SUPPORTED_Pause | diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index 16c85ccd1762..05ba84118f37 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -298,7 +298,7 @@ int stmmac_mdio_register(struct net_device *ndev) } pr_info("%s: PHY ID %08x at %d IRQ %s (%s)%s\n", ndev->name, phydev->phy_id, addr, - irq_str, dev_name(&phydev->dev), + irq_str, phydev_name(phydev), act ? " active" : ""); found = 1; } diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index 33bd3b902304..5a40b0256327 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -1646,7 +1646,7 @@ static int emac_dev_open(struct net_device *ndev) dev_info(emac_dev, "attached PHY driver [%s] " "(mii_bus:phy_addr=%s, id=%x)\n", - priv->phydev->drv->name, dev_name(&priv->phydev->dev), + priv->phydev->drv->name, phydev_name(priv->phydev), priv->phydev->phy_id); } diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index c00084d689f3..88e8e6055b9f 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -396,7 +396,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) phy = data->bus->phy_map[addr]; if (phy) { dev_info(dev, "phy[%d]: device %s, driver %s\n", - phy->addr, dev_name(&phy->dev), + phy->addr, phydev_name(phy), phy->drv ? phy->drv->name : "unknown"); } } diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 4e70e7586a09..d543298d6750 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2178,7 +2178,7 @@ static int gbe_slave_open(struct gbe_intf *gbe_intf) return -ENODEV; } dev_dbg(priv->dev, "phy found: id is: 0x%s\n", - dev_name(&slave->phy->dev)); + phydev_name(slave->phy)); phy_start(slave->phy); phy_read_status(slave->phy); } @@ -2681,7 +2681,7 @@ static void init_secondary_ports(struct gbe_priv *gbe_dev, slave->phy = NULL; } else { dev_dbg(dev, "phy found: id is: 0x%s\n", - dev_name(&slave->phy->dev)); + phydev_name(slave->phy)); phy_start(slave->phy); phy_read_status(slave->phy); } diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 45ac38d29ed8..8df6072ac78d 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -631,7 +631,7 @@ static int tc_mii_probe(struct net_device *dev) } /* attach the mac to the phy */ - phydev = phy_connect(dev, dev_name(&phydev->dev), + phydev = phy_connect(dev, phydev_name(phydev), &tc_handle_link_change, lp->chiptype == TC35815_TX4939 ? PHY_INTERFACE_MODE_RMII : PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { @@ -640,7 +640,7 @@ static int tc_mii_probe(struct net_device *dev) } printk(KERN_INFO "%s: attached PHY driver [%s] " "(mii_bus:phy_addr=%s, id=%x)\n", - dev->name, phydev->drv->name, dev_name(&phydev->dev), + dev->name, phydev->drv->name, phydev_name(phydev), phydev->phy_id); /* mask with MAC supported features */ diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c index d4083c381cd1..9f4e6eb886af 100644 --- a/drivers/net/phy/bcm7xxx.c +++ b/drivers/net/phy/bcm7xxx.c @@ -170,7 +170,7 @@ static int bcm7xxx_28nm_config_init(struct phy_device *phydev) int ret = 0; pr_info_once("%s: %s PHY revision: 0x%02x, patch: %d\n", - dev_name(&phydev->dev), phydev->drv->name, rev, patch); + phydev_name(phydev), phydev->drv->name, rev, patch); /* Dummy read to a register to workaround an issue upon reset where the * internal inverter may not allow the first MDIO transaction to pass diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 0bfbabad4431..0f179709a289 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -114,7 +114,7 @@ EXPORT_SYMBOL(phy_register_fixup_for_id); */ static int phy_needs_fixup(struct phy_device *phydev, struct phy_fixup *fixup) { - if (strcmp(fixup->bus_id, dev_name(&phydev->dev)) != 0) + if (strcmp(fixup->bus_id, phydev_name(phydev)) != 0) if (strcmp(fixup->bus_id, PHY_ANY_ID) != 0) return 0; diff --git a/drivers/staging/netlogic/xlr_net.c b/drivers/staging/netlogic/xlr_net.c index 8ae01753b011..b939c4b5f229 100644 --- a/drivers/staging/netlogic/xlr_net.c +++ b/drivers/staging/netlogic/xlr_net.c @@ -838,8 +838,8 @@ static int xlr_mii_probe(struct xlr_net_priv *priv) } /* Attach MAC to PHY */ - phydev = phy_connect(priv->ndev, dev_name(&phydev->dev), - &xlr_gmac_link_adjust, priv->nd->phy_interface); + phydev = phy_connect(priv->ndev, phydev_name(phydev), + &xlr_gmac_link_adjust, priv->nd->phy_interface); if (IS_ERR(phydev)) { pr_err("could not attach PHY\n"); @@ -855,7 +855,7 @@ static int xlr_mii_probe(struct xlr_net_priv *priv) phydev->advertising = phydev->supported; pr_info("attached PHY driver [%s] (mii_bus:phy_addr=%s\n", - phydev->drv->name, dev_name(&phydev->dev)); + phydev->drv->name, phydev_name(phydev)); return 0; } diff --git a/include/linux/phy.h b/include/linux/phy.h index dbcf9fdd960c..5f5cc3424b9e 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -783,6 +783,11 @@ static inline int phy_read_status(struct phy_device *phydev) #define phydev_dbg(_phydev, format, args...) \ dev_dbg(&_phydev->dev, format, ##args) +static inline const char *phydev_name(const struct phy_device *phydev) +{ + return dev_name(&phydev->dev); +} + int genphy_config_init(struct phy_device *phydev); int genphy_setup_forced(struct phy_device *phydev); int genphy_restart_aneg(struct phy_device *phydev); -- cgit v1.2.3-59-g8ed1b From 053e7e169229adebbc27fc176c5369398e9f5eba Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:12 +0100 Subject: phy: phy_{read|write}_mmd_indirect: get addr from phydev The address of the device can be determined from the phydev structure, rather than passing it as a parameter. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm-phy-lib.c | 8 ++++---- drivers/net/phy/dp83867.c | 6 +++--- drivers/net/phy/microchip.c | 5 ++--- drivers/net/phy/phy.c | 36 ++++++++++++++---------------------- include/linux/phy.h | 7 ++----- 5 files changed, 25 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/bcm-phy-lib.c b/drivers/net/phy/bcm-phy-lib.c index ddb377e53633..df0416db0b88 100644 --- a/drivers/net/phy/bcm-phy-lib.c +++ b/drivers/net/phy/bcm-phy-lib.c @@ -184,25 +184,25 @@ int bcm_phy_enable_eee(struct phy_device *phydev) /* Enable EEE at PHY level */ val = phy_read_mmd_indirect(phydev, BRCM_CL45VEN_EEE_CONTROL, - MDIO_MMD_AN, phydev->addr); + MDIO_MMD_AN); if (val < 0) return val; val |= LPI_FEATURE_EN | LPI_FEATURE_EN_DIG1000X; phy_write_mmd_indirect(phydev, BRCM_CL45VEN_EEE_CONTROL, - MDIO_MMD_AN, phydev->addr, (u32)val); + MDIO_MMD_AN, (u32)val); /* Advertise EEE */ val = phy_read_mmd_indirect(phydev, BCM_CL45VEN_EEE_ADV, - MDIO_MMD_AN, phydev->addr); + MDIO_MMD_AN); if (val < 0) return val; val |= (MDIO_AN_EEE_ADV_100TX | MDIO_AN_EEE_ADV_1000T); phy_write_mmd_indirect(phydev, BCM_CL45VEN_EEE_ADV, - MDIO_MMD_AN, phydev->addr, (u32)val); + MDIO_MMD_AN, (u32)val); return 0; } diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index 4ebf601073d9..e4c0b0c0af02 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -160,7 +160,7 @@ static int dp83867_config_init(struct phy_device *phydev) if ((phydev->interface >= PHY_INTERFACE_MODE_RGMII_ID) && (phydev->interface <= PHY_INTERFACE_MODE_RGMII_RXID)) { val = phy_read_mmd_indirect(phydev, DP83867_RGMIICTL, - DP83867_DEVADDR, phydev->addr); + DP83867_DEVADDR); if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) val |= (DP83867_RGMII_TX_CLK_DELAY_EN | DP83867_RGMII_RX_CLK_DELAY_EN); @@ -172,13 +172,13 @@ static int dp83867_config_init(struct phy_device *phydev) val |= DP83867_RGMII_RX_CLK_DELAY_EN; phy_write_mmd_indirect(phydev, DP83867_RGMIICTL, - DP83867_DEVADDR, phydev->addr, val); + DP83867_DEVADDR, val); delay = (dp83867->rx_id_delay | (dp83867->tx_id_delay << DP83867_RGMII_TX_CLK_DELAY_SHIFT)); phy_write_mmd_indirect(phydev, DP83867_RGMIIDCTL, - DP83867_DEVADDR, phydev->addr, delay); + DP83867_DEVADDR, delay); } return 0; diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c index c0a20ebd083b..99df5bc47424 100644 --- a/drivers/net/phy/microchip.c +++ b/drivers/net/phy/microchip.c @@ -78,10 +78,9 @@ static int lan88xx_probe(struct phy_device *phydev) priv->wolopts = 0; /* these values can be used to identify internal PHY */ - priv->chip_id = phy_read_mmd_indirect(phydev, LAN88XX_MMD3_CHIP_ID, - 3, phydev->addr); + priv->chip_id = phy_read_mmd_indirect(phydev, LAN88XX_MMD3_CHIP_ID, 3); priv->chip_rev = phy_read_mmd_indirect(phydev, LAN88XX_MMD3_CHIP_REV, - 3, phydev->addr); + 3); phydev->priv = priv; diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 9771941cf0ee..56c8dd8c0c85 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -1029,7 +1029,6 @@ static inline void mmd_phy_indirect(struct mii_bus *bus, int prtad, int devad, * @phydev: The PHY device bus * @prtad: MMD Address * @devad: MMD DEVAD - * @addr: PHY address on the MII bus * * Description: it reads data from the MMD registers (clause 22 to access to * clause 45) of the specified phy address. @@ -1039,10 +1038,10 @@ static inline void mmd_phy_indirect(struct mii_bus *bus, int prtad, int devad, * 3) Write reg 13 // MMD Data Command for MMD DEVAD * 3) Read reg 14 // Read MMD data */ -int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, - int devad, int addr) +int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, int devad) { struct phy_driver *phydrv = phydev->drv; + int addr = phydev->addr; int value = -1; if (!phydrv->read_mmd_indirect) { @@ -1066,7 +1065,6 @@ EXPORT_SYMBOL(phy_read_mmd_indirect); * @phydev: The PHY device * @prtad: MMD Address * @devad: MMD DEVAD - * @addr: PHY address on the MII bus * @data: data to write in the MMD register * * Description: Write data from the MMD registers of the specified @@ -1078,9 +1076,10 @@ EXPORT_SYMBOL(phy_read_mmd_indirect); * 3) Write reg 14 // Write MMD data */ void phy_write_mmd_indirect(struct phy_device *phydev, int prtad, - int devad, int addr, u32 data) + int devad, u32 data) { struct phy_driver *phydrv = phydev->drv; + int addr = phydev->addr; if (!phydrv->write_mmd_indirect) { struct mii_bus *bus = phydev->bus; @@ -1130,7 +1129,7 @@ int phy_init_eee(struct phy_device *phydev, bool clk_stop_enable) /* First check if the EEE ability is supported */ eee_cap = phy_read_mmd_indirect(phydev, MDIO_PCS_EEE_ABLE, - MDIO_MMD_PCS, phydev->addr); + MDIO_MMD_PCS); if (eee_cap <= 0) goto eee_exit_err; @@ -1142,12 +1141,12 @@ int phy_init_eee(struct phy_device *phydev, bool clk_stop_enable) * the EEE advertising registers. */ eee_lp = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_LPABLE, - MDIO_MMD_AN, phydev->addr); + MDIO_MMD_AN); if (eee_lp <= 0) goto eee_exit_err; eee_adv = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_ADV, - MDIO_MMD_AN, phydev->addr); + MDIO_MMD_AN); if (eee_adv <= 0) goto eee_exit_err; @@ -1161,15 +1160,13 @@ int phy_init_eee(struct phy_device *phydev, bool clk_stop_enable) * clock while it is signaling LPI. */ int val = phy_read_mmd_indirect(phydev, MDIO_CTRL1, - MDIO_MMD_PCS, - phydev->addr); + MDIO_MMD_PCS); if (val < 0) return val; val |= MDIO_PCS_CTRL1_CLKSTOP_EN; phy_write_mmd_indirect(phydev, MDIO_CTRL1, - MDIO_MMD_PCS, phydev->addr, - val); + MDIO_MMD_PCS, val); } return 0; /* EEE supported */ @@ -1188,8 +1185,7 @@ EXPORT_SYMBOL(phy_init_eee); */ int phy_get_eee_err(struct phy_device *phydev) { - return phy_read_mmd_indirect(phydev, MDIO_PCS_EEE_WK_ERR, - MDIO_MMD_PCS, phydev->addr); + return phy_read_mmd_indirect(phydev, MDIO_PCS_EEE_WK_ERR, MDIO_MMD_PCS); } EXPORT_SYMBOL(phy_get_eee_err); @@ -1206,22 +1202,19 @@ int phy_ethtool_get_eee(struct phy_device *phydev, struct ethtool_eee *data) int val; /* Get Supported EEE */ - val = phy_read_mmd_indirect(phydev, MDIO_PCS_EEE_ABLE, - MDIO_MMD_PCS, phydev->addr); + val = phy_read_mmd_indirect(phydev, MDIO_PCS_EEE_ABLE, MDIO_MMD_PCS); if (val < 0) return val; data->supported = mmd_eee_cap_to_ethtool_sup_t(val); /* Get advertisement EEE */ - val = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_ADV, - MDIO_MMD_AN, phydev->addr); + val = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_ADV, MDIO_MMD_AN); if (val < 0) return val; data->advertised = mmd_eee_adv_to_ethtool_adv_t(val); /* Get LP advertisement EEE */ - val = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_LPABLE, - MDIO_MMD_AN, phydev->addr); + val = phy_read_mmd_indirect(phydev, MDIO_AN_EEE_LPABLE, MDIO_MMD_AN); if (val < 0) return val; data->lp_advertised = mmd_eee_adv_to_ethtool_adv_t(val); @@ -1241,8 +1234,7 @@ int phy_ethtool_set_eee(struct phy_device *phydev, struct ethtool_eee *data) { int val = ethtool_adv_to_mmd_eee_adv_t(data->advertised); - phy_write_mmd_indirect(phydev, MDIO_AN_EEE_ADV, MDIO_MMD_AN, - phydev->addr, val); + phy_write_mmd_indirect(phydev, MDIO_AN_EEE_ADV, MDIO_MMD_AN, val); return 0; } diff --git a/include/linux/phy.h b/include/linux/phy.h index 5f5cc3424b9e..08198ce98773 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -629,14 +629,12 @@ static inline int phy_read_mmd(struct phy_device *phydev, int devad, u32 regnum) * phy_read_mmd_indirect - reads data from the MMD registers * @phydev: The PHY device bus * @prtad: MMD Address - * @devad: MMD DEVAD * @addr: PHY address on the MII bus * * Description: it reads data from the MMD registers (clause 22 to access to * clause 45) of the specified phy address. */ -int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, - int devad, int addr); +int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, int devad); /** * phy_read - Convenience function for reading a given PHY register @@ -735,14 +733,13 @@ static inline int phy_write_mmd(struct phy_device *phydev, int devad, * @phydev: The PHY device * @prtad: MMD Address * @devad: MMD DEVAD - * @addr: PHY address on the MII bus * @data: data to write in the MMD register * * Description: Write data from the MMD registers of the specified * phy address. */ void phy_write_mmd_indirect(struct phy_device *phydev, int prtad, - int devad, int addr, u32 data); + int devad, u32 data); struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, bool is_c45, -- cgit v1.2.3-59-g8ed1b From 2220943a21e26d97d7fd8f83c004b947326b469d Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:13 +0100 Subject: phy: Centralise print about attached phy Many Ethernet drivers contain the same netdev_info() print statement about the attached phy. Move it into the phy device code. Additionally add a varargs function which can be used to append additional information. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/adi/bfin_mac.c | 6 ++---- drivers/net/ethernet/agere/et131x.c | 4 +--- drivers/net/ethernet/amd/au1000_eth.c | 4 +--- drivers/net/ethernet/broadcom/b44.c | 3 +-- drivers/net/ethernet/broadcom/bcm63xx_enet.c | 3 +-- drivers/net/ethernet/broadcom/sb1250-mac.c | 7 +++---- drivers/net/ethernet/broadcom/tg3.c | 10 +++------- drivers/net/ethernet/cadence/macb.c | 3 +-- drivers/net/ethernet/dnet.c | 4 +--- drivers/net/ethernet/freescale/fec_main.c | 4 +--- drivers/net/ethernet/lantiq_etop.c | 4 +--- drivers/net/ethernet/nxp/lpc_eth.c | 5 ++--- drivers/net/ethernet/rdc/r6040.c | 4 +--- drivers/net/ethernet/renesas/ravb_main.c | 3 +-- drivers/net/ethernet/renesas/sh_eth.c | 3 +-- drivers/net/ethernet/smsc/smsc911x.c | 4 +--- drivers/net/ethernet/smsc/smsc9420.c | 9 ++------- drivers/net/ethernet/synopsys/dwc_eth_qos.c | 12 +----------- drivers/net/ethernet/ti/cpsw.c | 4 ++-- drivers/net/ethernet/ti/davinci_emac.c | 5 +---- drivers/net/ethernet/toshiba/tc35815.c | 6 ++---- drivers/net/phy/phy_device.c | 27 +++++++++++++++++++++++++++ drivers/staging/netlogic/xlr_net.c | 3 +-- include/linux/phy.h | 4 ++++ net/dsa/slave.c | 5 ++--- 25 files changed, 64 insertions(+), 82 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index 5f8a5182b8dc..ed5c78cb7239 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -444,10 +444,8 @@ static int mii_probe(struct net_device *dev, int phy_mode) lp->old_duplex = -1; lp->phydev = phydev; - pr_info("attached PHY driver [%s] " - "(mii_bus:phy_addr=%s, irq=%d, mdc_clk=%dHz(mdc_div=%d)@sclk=%dMHz)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq, - MDC_CLK, mdc_div, sclk/1000000); + phy_attached_print(phydev, "mdc_clk=%dHz(mdc_div=%d)@sclk=%dMHz)\n", + MDC_CLK, mdc_div, sclk / 1000000); return 0; } diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 80b706f0fc97..825da3af806a 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -3289,9 +3289,7 @@ static int et131x_mii_probe(struct net_device *netdev) phydev->autoneg = AUTONEG_ENABLE; adapter->phydev = phydev; - dev_info(&adapter->pdev->dev, - "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, phydev_name(phydev)); + phy_attached_info(phydev); return 0; } diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 8a8d6f2a0f6f..114618d357d5 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -583,9 +583,7 @@ static int au1000_mii_probe(struct net_device *dev) aup->old_duplex = -1; aup->phy_dev = phydev; - netdev_info(dev, "attached PHY driver [%s] " - "(mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); return 0; } diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 928a2210e788..e7d9308d6760 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2315,8 +2315,7 @@ static int b44_register_phy_one(struct b44 *bp) bp->old_link = 0; bp->phy_addr = phydev->addr; - dev_info(sdev->dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, phydev_name(phydev)); + phy_attached_info(phydev); return 0; diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index a54bafad3538..55f31faa09e6 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -908,8 +908,7 @@ static int bcm_enet_open(struct net_device *dev) else phydev->advertising &= ~SUPPORTED_Pause; - dev_info(kdev, "attached PHY at address %d [%s]\n", - phydev->addr, phydev->drv->name); + phy_attached_info(phydev); priv->old_link = 0; priv->old_duplex = -1; diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c index f557a2aaec23..2470c6084c67 100644 --- a/drivers/net/ethernet/broadcom/sb1250-mac.c +++ b/drivers/net/ethernet/broadcom/sb1250-mac.c @@ -2388,11 +2388,10 @@ static int sbmac_mii_probe(struct net_device *dev) SUPPORTED_MII | SUPPORTED_Pause | SUPPORTED_Asym_Pause; - phy_dev->advertising = phy_dev->supported; - pr_info("%s: attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - dev->name, phy_dev->drv->name, - dev_name(&phy_dev->dev), phy_dev->irq); + phy_attached_info(phydev); + + phy_dev->advertising = phy_dev->supported; sc->phy_dev = phy_dev; diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 69d84d67f09a..07c067590caa 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -2128,6 +2128,8 @@ static int tg3_phy_init(struct tg3 *tp) phydev->advertising = phydev->supported; + phy_attached_info(phydev); + return 0; } @@ -17898,13 +17900,7 @@ static int tg3_init_one(struct pci_dev *pdev, tg3_bus_string(tp, str), dev->dev_addr); - if (tp->phy_flags & TG3_PHYFLG_IS_CONNECTED) { - struct phy_device *phydev; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; - netdev_info(dev, - "attached PHY driver [%s] (mii_bus:phy_addr=%s)\n", - phydev->drv->name, phydev_name(phydev)); - } else { + if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) { char *ethtype; if (tp->phy_flags & TG3_PHYFLG_10_100_ONLY) diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 001d60c5521c..98df33b7a395 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -2950,8 +2950,7 @@ static int macb_probe(struct platform_device *pdev) dev->base_addr, dev->irq, dev->dev_addr); phydev = bp->phy_dev; - netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); return 0; diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 0ec367521354..6557460cf028 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -886,9 +886,7 @@ static int dnet_probe(struct platform_device *pdev) (bp->capabilities & DNET_HAS_GIGABIT) ? "" : "no ", (bp->capabilities & DNET_HAS_DMA) ? "" : "no "); phydev = bp->phy_dev; - dev_info(&pdev->dev, "attached PHY driver [%s] " - "(mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); return 0; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index d2328fc5da57..ceabe21b3b2c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1972,9 +1972,7 @@ static int fec_enet_mii_probe(struct net_device *ndev) fep->link = 0; fep->full_duplex = 0; - netdev_info(ndev, "Freescale FEC PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - fep->phy_dev->drv->name, dev_name(&fep->phy_dev->dev), - fep->phy_dev->irq); + phy_attached_info(phy_dev); return 0; } diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 274a3cec84c2..86238a5eaddf 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -408,9 +408,7 @@ ltq_etop_mdio_probe(struct net_device *dev) phydev->advertising = phydev->supported; priv->phydev = phydev; - pr_info("%s: attached PHY [%s] (phy_addr=%s, irq=%d)\n", - dev->name, phydev->drv->name, - phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); return 0; } diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index 5801aa197697..024bc3675573 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -816,9 +816,8 @@ static int lpc_mii_probe(struct net_device *ndev) pldat->duplex = -1; pldat->phy_dev = phydev; - netdev_info(ndev, - "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); + return 0; } diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 86a0887811c7..174dea787caf 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -1061,9 +1061,7 @@ static int r6040_mii_probe(struct net_device *dev) lp->old_link = 0; lp->old_duplex = -1; - dev_info(&lp->pdev->dev, "attached PHY driver [%s] " - "(mii_bus:phy_addr=%s)\n", - phydev->drv->name, phydev_name(phydev)); + phy_attached_info(phydev); return 0; } diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 2f6c974e4a6d..9e20f37a3b6f 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -927,8 +927,7 @@ static int ravb_phy_init(struct net_device *ndev) /* 10BASE is not supported */ phydev->supported &= ~PHY_10BT_FEATURES; - netdev_info(ndev, "attached PHY %d (IRQ %d) to driver %s\n", - phydev->addr, phydev->irq, phydev_name(phydev)); + phy_attached_info(phydev); priv->phydev = phydev; diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index e14d28474b70..94581be64d65 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -1826,8 +1826,7 @@ static int sh_eth_phy_init(struct net_device *ndev) return PTR_ERR(phydev); } - netdev_info(ndev, "attached PHY %d (IRQ %d) to driver %s\n", - phydev->addr, phydev->irq, phydev_name(phydev)); + phy_attached_info(phydev); mdp->phydev = phydev; diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 067346d3209d..139b99b04099 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1031,9 +1031,7 @@ static int smsc911x_mii_probe(struct net_device *dev) return ret; } - netdev_info(dev, - "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); + phy_attached_info(phydev); /* mask with MAC supported features */ phydev->supported &= (PHY_BASIC_FEATURES | SUPPORTED_Pause | diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index a02ed6b63064..fa8893a804f7 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -1163,10 +1163,6 @@ static int smsc9420_mii_probe(struct net_device *dev) return -ENODEV; } - phydev = pd->mii_bus->phy_map[1]; - netif_info(pd, probe, pd->dev, "PHY addr %d, phy_id 0x%08X\n", - phydev->addr, phydev->phy_id); - phydev = phy_connect(dev, phydev_name(phydev), smsc9420_phy_adjust_link, PHY_INTERFACE_MODE_MII); @@ -1175,14 +1171,13 @@ static int smsc9420_mii_probe(struct net_device *dev) return PTR_ERR(phydev); } - netdev_info(dev, "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", - phydev->drv->name, phydev_name(phydev), phydev->irq); - /* mask with MAC supported features */ phydev->supported &= (PHY_BASIC_FEATURES | SUPPORTED_Pause | SUPPORTED_Asym_Pause); phydev->advertising = phydev->supported; + phy_attached_info(phydev); + pd->phy_dev = phydev; pd->last_duplex = -1; pd->last_carrier = -1; diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index 9066d7a8483c..b25ee370254a 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -972,9 +972,7 @@ static int dwceqos_mii_probe(struct net_device *ndev) } if (netif_msg_probe(lp)) - netdev_dbg(lp->ndev, - "phydev %p, phydev->phy_id 0xa%x, phydev->addr 0x%x\n", - phydev, phydev->phy_id, phydev->addr); + phy_attached_info(phydev); phydev->supported &= PHY_GBIT_FEATURES; @@ -983,14 +981,6 @@ static int dwceqos_mii_probe(struct net_device *ndev) lp->duplex = DUPLEX_UNKNOWN; lp->phy_dev = phydev; - if (netif_msg_probe(lp)) { - netdev_dbg(lp->ndev, "phy_addr 0x%x, phy_id 0x%08x\n", - lp->phy_dev->addr, lp->phy_dev->phy_id); - - netdev_dbg(lp->ndev, "attach [%s] phy driver\n", - lp->phy_dev->drv->name); - } - return 0; } diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 34ce7dce8c9d..49544c0fa6a7 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1159,8 +1159,8 @@ static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv) slave->data->phy_id, slave->slave_num); slave->phy = NULL; } else { - dev_info(priv->dev, "phy found : id is : 0x%x\n", - slave->phy->phy_id); + phy_attached_info(slave->phy); + phy_start(slave->phy); /* Configure GMII_SEL register */ diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index 5a40b0256327..5d9abedd6b75 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -1644,10 +1644,7 @@ static int emac_dev_open(struct net_device *ndev) priv->speed = 0; priv->duplex = ~0; - dev_info(emac_dev, "attached PHY driver [%s] " - "(mii_bus:phy_addr=%s, id=%x)\n", - priv->phydev->drv->name, phydev_name(priv->phydev), - priv->phydev->phy_id); + phy_attached_info(priv->phydev); } if (!priv->phydev) { diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 8df6072ac78d..8fd5e0ba718c 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -638,10 +638,8 @@ static int tc_mii_probe(struct net_device *dev) printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name); return PTR_ERR(phydev); } - printk(KERN_INFO "%s: attached PHY driver [%s] " - "(mii_bus:phy_addr=%s, id=%x)\n", - dev->name, phydev->drv->name, phydev_name(phydev), - phydev->phy_id); + + phy_attached_info(phydev); /* mask with MAC supported features */ phydev->supported &= PHY_BASIC_FEATURES; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 0f179709a289..68fe5738daef 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -607,6 +607,33 @@ int phy_init_hw(struct phy_device *phydev) } EXPORT_SYMBOL(phy_init_hw); +void phy_attached_info(struct phy_device *phydev) +{ + phy_attached_print(phydev, NULL); +} +EXPORT_SYMBOL(phy_attached_info); + +#define ATTACHED_FMT "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)" +void phy_attached_print(struct phy_device *phydev, const char *fmt, ...) +{ + if (!fmt) { + dev_info(&phydev->dev, ATTACHED_FMT "\n", + phydev->drv->name, phydev_name(phydev), + phydev->irq); + } else { + va_list ap; + + dev_info(&phydev->dev, ATTACHED_FMT, + phydev->drv->name, phydev_name(phydev), + phydev->irq); + + va_start(ap, fmt); + vprintk(fmt, ap); + va_end(ap); + } +} +EXPORT_SYMBOL(phy_attached_print); + /** * phy_attach_direct - attach a network device to a given PHY device pointer * @dev: network device to attach diff --git a/drivers/staging/netlogic/xlr_net.c b/drivers/staging/netlogic/xlr_net.c index b939c4b5f229..cbc25b7e70a2 100644 --- a/drivers/staging/netlogic/xlr_net.c +++ b/drivers/staging/netlogic/xlr_net.c @@ -854,8 +854,7 @@ static int xlr_mii_probe(struct xlr_net_priv *priv) | ADVERTISED_MII); phydev->advertising = phydev->supported; - pr_info("attached PHY driver [%s] (mii_bus:phy_addr=%s\n", - phydev->drv->name, phydev_name(phydev)); + phy_attached_info(phydev); return 0; } diff --git a/include/linux/phy.h b/include/linux/phy.h index 08198ce98773..ecbf6382ba29 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -16,6 +16,7 @@ #ifndef __PHY_H #define __PHY_H +#include #include #include #include @@ -785,6 +786,9 @@ static inline const char *phydev_name(const struct phy_device *phydev) return dev_name(&phydev->dev); } +void phy_attached_print(struct phy_device *phydev, const char *fmt, ...) + __printf(2, 3); +void phy_attached_info(struct phy_device *phydev); int genphy_config_init(struct phy_device *phydev); int genphy_setup_forced(struct phy_device *phydev); int genphy_restart_aneg(struct phy_device *phydev); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 1e9e9424a33d..5f45e68b52dc 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1080,11 +1080,10 @@ static int dsa_slave_phy_setup(struct dsa_slave_priv *p, netdev_err(slave_dev, "failed to connect to port %d: %d\n", p->port, ret); return ret; } - } else { - netdev_info(slave_dev, "attached PHY at address %d [%s]\n", - p->phy->addr, p->phy->drv->name); } + phy_attached_info(p->phy); + return 0; } -- cgit v1.2.3-59-g8ed1b From e7f4dc3536a40097f95103ddf98dd55b3a980f5b Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:15 +0100 Subject: mdio: Move allocation of interrupts into core Have mdio_alloc() create the array of interrupt numbers, and initialize it to POLLING. This is what most MDIO drivers want, so allowing code to be removed from the drivers. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- arch/powerpc/platforms/82xx/ep8248e.c | 10 +--------- arch/powerpc/platforms/pasemi/gpio_mdio.c | 3 --- drivers/net/ethernet/8390/ax88796.c | 15 ++------------- drivers/net/ethernet/adi/bfin_mac.c | 11 +---------- drivers/net/ethernet/aeroflex/greth.c | 5 ----- drivers/net/ethernet/aeroflex/greth.h | 1 - drivers/net/ethernet/agere/et131x.c | 13 +------------ drivers/net/ethernet/altera/altera_tse_main.c | 13 +------------ drivers/net/ethernet/amd/au1000_eth.c | 7 ------- drivers/net/ethernet/broadcom/b44.c | 14 +------------- drivers/net/ethernet/broadcom/bcm63xx_enet.c | 9 --------- drivers/net/ethernet/broadcom/bgmac.c | 15 ++------------- drivers/net/ethernet/broadcom/genet/bcmmii.c | 10 ---------- drivers/net/ethernet/broadcom/sb1250-mac.c | 4 ---- drivers/net/ethernet/broadcom/tg3.c | 4 ---- drivers/net/ethernet/broadcom/tg3.h | 1 - drivers/net/ethernet/cadence/macb.c | 14 +------------- drivers/net/ethernet/dnet.c | 12 +----------- drivers/net/ethernet/ethoc.c | 14 +------------- drivers/net/ethernet/faraday/ftgmac100.c | 6 ------ drivers/net/ethernet/freescale/fec_main.c | 16 ++-------------- drivers/net/ethernet/freescale/fec_mpc52xx_phy.c | 4 ---- drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c | 10 +--------- drivers/net/ethernet/freescale/fs_enet/mii-fec.c | 10 +--------- drivers/net/ethernet/freescale/fsl_pq_mdio.c | 2 -- drivers/net/ethernet/hisilicon/hns_mdio.c | 5 ----- drivers/net/ethernet/lantiq_etop.c | 14 +------------- drivers/net/ethernet/marvell/mvmdio.c | 10 +--------- drivers/net/ethernet/nxp/lpc_eth.c | 13 +------------ drivers/net/ethernet/rdc/r6040.c | 14 +------------- drivers/net/ethernet/renesas/sh_eth.c | 12 +----------- drivers/net/ethernet/smsc/smsc911x.c | 6 ++---- drivers/net/ethernet/smsc/smsc9420.c | 6 +----- drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 14 ++++---------- drivers/net/ethernet/synopsys/dwc_eth_qos.c | 15 ++------------- drivers/net/ethernet/ti/cpmac.c | 3 --- drivers/net/ethernet/toshiba/tc35815.c | 14 +------------- drivers/net/ethernet/xilinx/ll_temac.h | 1 - drivers/net/ethernet/xilinx/ll_temac_mdio.c | 2 -- drivers/net/ethernet/xilinx/xilinx_axienet.h | 2 -- drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c | 2 -- drivers/net/ethernet/xilinx/xilinx_emaclite.c | 4 ---- drivers/net/phy/fixed_phy.c | 4 +--- drivers/net/phy/mdio-bcm-unimac.c | 11 +---------- drivers/net/phy/mdio-gpio.c | 2 +- drivers/net/phy/mdio-moxart.c | 7 ------- drivers/net/phy/mdio-mux.c | 3 +-- drivers/net/phy/mdio-octeon.c | 2 -- drivers/net/phy/mdio-sun4i.c | 12 +----------- drivers/net/phy/mdio_bus.c | 5 +++++ drivers/net/usb/ax88172a.c | 14 ++------------ drivers/net/usb/lan78xx.c | 11 +---------- drivers/of/of_mdio.c | 7 +------ include/linux/phy.h | 6 +++--- 54 files changed, 48 insertions(+), 396 deletions(-) (limited to 'include') diff --git a/arch/powerpc/platforms/82xx/ep8248e.c b/arch/powerpc/platforms/82xx/ep8248e.c index a0cb8bd41958..6781bda117be 100644 --- a/arch/powerpc/platforms/82xx/ep8248e.c +++ b/arch/powerpc/platforms/82xx/ep8248e.c @@ -131,23 +131,15 @@ static int ep8248e_mdio_probe(struct platform_device *ofdev) if (!bus) return -ENOMEM; - bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (bus->irq == NULL) { - ret = -ENOMEM; - goto err_free_bus; - } - bus->name = "ep8248e-mdio-bitbang"; bus->parent = &ofdev->dev; snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start); ret = of_mdiobus_register(bus, ofdev->dev.of_node); if (ret) - goto err_free_irq; + goto err_free_bus; return 0; -err_free_irq: - kfree(bus->irq); err_free_bus: free_mdio_bitbang(bus); return ret; diff --git a/arch/powerpc/platforms/pasemi/gpio_mdio.c b/arch/powerpc/platforms/pasemi/gpio_mdio.c index ae3f47b25b18..ddf635000c6b 100644 --- a/arch/powerpc/platforms/pasemi/gpio_mdio.c +++ b/arch/powerpc/platforms/pasemi/gpio_mdio.c @@ -41,7 +41,6 @@ static void __iomem *gpio_regs; struct gpio_priv { int mdc_pin; int mdio_pin; - int mdio_irqs[PHY_MAX_ADDR]; }; #define MDC_PIN(bus) (((struct gpio_priv *)bus->priv)->mdc_pin) @@ -245,8 +244,6 @@ static int gpio_mdio_probe(struct platform_device *ofdev) snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", *prop); new_bus->priv = priv; - new_bus->irq = priv->mdio_irqs; - prop = of_get_property(np, "mdc-pin", NULL); priv->mdc_pin = *prop; diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c index 90b540a4a561..c89b9aeeceb6 100644 --- a/drivers/net/ethernet/8390/ax88796.c +++ b/drivers/net/ethernet/8390/ax88796.c @@ -627,7 +627,7 @@ static int ax_mii_init(struct net_device *dev) struct platform_device *pdev = to_platform_device(dev->dev.parent); struct ei_device *ei_local = netdev_priv(dev); struct ax_device *ax = to_ax_dev(dev); - int err, i; + int err; ax->bb_ctrl.ops = &bb_ops; ax->addr_memr = ei_local->mem + AX_MEMR; @@ -642,23 +642,12 @@ static int ax_mii_init(struct net_device *dev) snprintf(ax->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x", pdev->name, pdev->id); - ax->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!ax->mii_bus->irq) { - err = -ENOMEM; - goto out_free_mdio_bitbang; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - ax->mii_bus->irq[i] = PHY_POLL; - err = mdiobus_register(ax->mii_bus); if (err) - goto out_free_irq; + goto out_free_mdio_bitbang; return 0; - out_free_irq: - kfree(ax->mii_bus->irq); out_free_mdio_bitbang: free_mdio_bitbang(ax->mii_bus); out: diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index ed5c78cb7239..62862744c870 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -1838,12 +1838,6 @@ static int bfin_mii_bus_probe(struct platform_device *pdev) snprintf(miibus->id, MII_BUS_ID_SIZE, "%s-%x", pdev->name, pdev->id); - miibus->irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); - if (!miibus->irq) - goto out_err_irq_alloc; - - for (i = rc; i < PHY_MAX_ADDR; ++i) - miibus->irq[i] = PHY_POLL; rc = clamp(mii_bus_pd->phydev_number, 0, PHY_MAX_ADDR); if (rc != mii_bus_pd->phydev_number) @@ -1862,14 +1856,12 @@ static int bfin_mii_bus_probe(struct platform_device *pdev) rc = mdiobus_register(miibus); if (rc) { dev_err(&pdev->dev, "Cannot register MDIO bus!\n"); - goto out_err_mdiobus_register; + goto out_err_alloc; } platform_set_drvdata(pdev, miibus); return 0; -out_err_mdiobus_register: - kfree(miibus->irq); out_err_irq_alloc: mdiobus_free(miibus); out_err_alloc: @@ -1885,7 +1877,6 @@ static int bfin_mii_bus_remove(struct platform_device *pdev) dev_get_platdata(&pdev->dev); mdiobus_unregister(miibus); - kfree(miibus->irq); mdiobus_free(miibus); peripheral_free_list(mii_bus_pd->mac_peripherals); diff --git a/drivers/net/ethernet/aeroflex/greth.c b/drivers/net/ethernet/aeroflex/greth.c index 20bf55dbd76f..b873531c5575 100644 --- a/drivers/net/ethernet/aeroflex/greth.c +++ b/drivers/net/ethernet/aeroflex/greth.c @@ -1337,11 +1337,6 @@ static int greth_mdio_init(struct greth_private *greth) greth->mdio->write = greth_mdio_write; greth->mdio->priv = greth; - greth->mdio->irq = greth->mdio_irqs; - - for (phy = 0; phy < PHY_MAX_ADDR; phy++) - greth->mdio->irq[phy] = PHY_POLL; - ret = mdiobus_register(greth->mdio); if (ret) { goto error; diff --git a/drivers/net/ethernet/aeroflex/greth.h b/drivers/net/ethernet/aeroflex/greth.h index ae16ac94daf8..92dd918e4a83 100644 --- a/drivers/net/ethernet/aeroflex/greth.h +++ b/drivers/net/ethernet/aeroflex/greth.h @@ -125,7 +125,6 @@ struct greth_private { struct phy_device *phy; struct mii_bus *mdio; - int mdio_irqs[PHY_MAX_ADDR]; unsigned int link; unsigned int speed; unsigned int duplex; diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 825da3af806a..f29d45eea1d9 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -3325,7 +3325,6 @@ static void et131x_pci_remove(struct pci_dev *pdev) netif_napi_del(&adapter->napi); phy_disconnect(adapter->phydev); mdiobus_unregister(adapter->mii_bus); - kfree(adapter->mii_bus->irq); mdiobus_free(adapter->mii_bus); et131x_adapter_memory_free(adapter); @@ -3944,7 +3943,6 @@ static int et131x_pci_setup(struct pci_dev *pdev, struct net_device *netdev; struct et131x_adapter *adapter; int rc; - int ii; rc = pci_enable_device(pdev); if (rc < 0) { @@ -4034,18 +4032,11 @@ static int et131x_pci_setup(struct pci_dev *pdev, adapter->mii_bus->priv = netdev; adapter->mii_bus->read = et131x_mdio_read; adapter->mii_bus->write = et131x_mdio_write; - adapter->mii_bus->irq = kmalloc_array(PHY_MAX_ADDR, sizeof(int), - GFP_KERNEL); - if (!adapter->mii_bus->irq) - goto err_mdio_free; - - for (ii = 0; ii < PHY_MAX_ADDR; ii++) - adapter->mii_bus->irq[ii] = PHY_POLL; rc = mdiobus_register(adapter->mii_bus); if (rc < 0) { dev_err(&pdev->dev, "failed to register MII bus\n"); - goto err_mdio_free_irq; + goto err_mdio_free; } rc = et131x_mii_probe(netdev); @@ -4085,8 +4076,6 @@ err_phy_disconnect: phy_disconnect(adapter->phydev); err_mdio_unregister: mdiobus_unregister(adapter->mii_bus); -err_mdio_free_irq: - kfree(adapter->mii_bus->irq); err_mdio_free: mdiobus_free(adapter->mii_bus); err_mem_free: diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index fe644823ceaf..10d51e8aefe0 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -131,7 +131,6 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id) { struct altera_tse_private *priv = netdev_priv(dev); int ret; - int i; struct device_node *mdio_node = NULL; struct mii_bus *mdio = NULL; struct device_node *child_node = NULL; @@ -161,14 +160,6 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id) mdio->write = &altera_tse_mdio_write; snprintf(mdio->id, MII_BUS_ID_SIZE, "%s-%u", mdio->name, id); - mdio->irq = kcalloc(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL); - if (mdio->irq == NULL) { - ret = -ENOMEM; - goto out_free_mdio; - } - for (i = 0; i < PHY_MAX_ADDR; i++) - mdio->irq[i] = PHY_POLL; - mdio->priv = dev; mdio->parent = priv->device; @@ -176,7 +167,7 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id) if (ret != 0) { netdev_err(dev, "Cannot register MDIO bus %s\n", mdio->id); - goto out_free_mdio_irq; + goto out_free_mdio; } if (netif_msg_drv(priv)) @@ -184,8 +175,6 @@ static int altera_tse_mdio_create(struct net_device *dev, unsigned int id) priv->mdio = mdio; return 0; -out_free_mdio_irq: - kfree(mdio->irq); out_free_mdio: mdiobus_free(mdio); mdio = NULL; diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 114618d357d5..982b581d3484 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -1291,14 +1291,7 @@ static int au1000_probe(struct platform_device *pdev) aup->mii_bus->name = "au1000_eth_mii"; snprintf(aup->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x", pdev->name, aup->mac_id); - aup->mii_bus->irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); - if (aup->mii_bus->irq == NULL) { - err = -ENOMEM; - goto err_out; - } - for (i = 0; i < PHY_MAX_ADDR; ++i) - aup->mii_bus->irq[i] = PHY_POLL; /* if known, set corresponding PHY IRQs */ if (aup->phy_static_config) if (aup->phy_irq && aup->phy_busid == aup->mac_id) diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index e7d9308d6760..4d08bc02c7a8 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2263,21 +2263,13 @@ static int b44_register_phy_one(struct b44 *bp) mii_bus->parent = sdev->dev; mii_bus->phy_mask = ~(1 << bp->phy_addr); snprintf(mii_bus->id, MII_BUS_ID_SIZE, "%x", instance); - mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!mii_bus->irq) { - dev_err(sdev->dev, "mii_bus irq allocation failed\n"); - err = -ENOMEM; - goto err_out_mdiobus; - } - - memset(mii_bus->irq, PHY_POLL, sizeof(int) * PHY_MAX_ADDR); bp->mii_bus = mii_bus; err = mdiobus_register(mii_bus); if (err) { dev_err(sdev->dev, "failed to register MII bus\n"); - goto err_out_mdiobus_irq; + goto err_out_mdiobus; } if (!bp->mii_bus->phy_map[bp->phy_addr] && @@ -2322,9 +2314,6 @@ static int b44_register_phy_one(struct b44 *bp) err_out_mdiobus_unregister: mdiobus_unregister(mii_bus); -err_out_mdiobus_irq: - kfree(mii_bus->irq); - err_out_mdiobus: mdiobus_free(mii_bus); @@ -2338,7 +2327,6 @@ static void b44_unregister_phy_one(struct b44 *bp) phy_disconnect(bp->phydev); mdiobus_unregister(mii_bus); - kfree(mii_bus->irq); mdiobus_free(mii_bus); } diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index 55f31faa09e6..87c6b5bdd616 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -1848,17 +1848,8 @@ static int bcm_enet_probe(struct platform_device *pdev) * if a slave is not present on hw */ bus->phy_mask = ~(1 << priv->phy_id); - bus->irq = devm_kzalloc(&pdev->dev, sizeof(int) * PHY_MAX_ADDR, - GFP_KERNEL); - if (!bus->irq) { - ret = -ENOMEM; - goto out_free_mdio; - } - if (priv->has_phy_interrupt) bus->irq[priv->phy_id] = priv->phy_interrupt; - else - bus->irq[priv->phy_id] = PHY_POLL; ret = mdiobus_register(bus); if (ret) { diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 28f7610b03fe..c7798d360512 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1471,7 +1471,7 @@ static int bgmac_mii_register(struct bgmac *bgmac) struct mii_bus *mii_bus; struct phy_device *phy_dev; char bus_id[MII_BUS_ID_SIZE + 3]; - int i, err = 0; + int err = 0; if (ci->id == BCMA_CHIP_ID_BCM4707 || ci->id == BCMA_CHIP_ID_BCM53018) @@ -1490,18 +1490,10 @@ static int bgmac_mii_register(struct bgmac *bgmac) mii_bus->parent = &bgmac->core->dev; mii_bus->phy_mask = ~(1 << bgmac->phyaddr); - mii_bus->irq = kmalloc_array(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL); - if (!mii_bus->irq) { - err = -ENOMEM; - goto err_free_bus; - } - for (i = 0; i < PHY_MAX_ADDR; i++) - mii_bus->irq[i] = PHY_POLL; - err = mdiobus_register(mii_bus); if (err) { bgmac_err(bgmac, "Registration of mii bus failed\n"); - goto err_free_irq; + goto err_free_bus; } bgmac->mii_bus = mii_bus; @@ -1522,8 +1514,6 @@ static int bgmac_mii_register(struct bgmac *bgmac) err_unregister_bus: mdiobus_unregister(mii_bus); -err_free_irq: - kfree(mii_bus->irq); err_free_bus: mdiobus_free(mii_bus); return err; @@ -1534,7 +1524,6 @@ static void bgmac_mii_unregister(struct bgmac *bgmac) struct mii_bus *mii_bus = bgmac->mii_bus; mdiobus_unregister(mii_bus); - kfree(mii_bus->irq); mdiobus_free(mii_bus); } diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 8bdfe53754ba..4523acd8c1c2 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -402,8 +402,6 @@ int bcmgenet_mii_probe(struct net_device *dev) */ if (priv->internal_phy) priv->mii_bus->irq[phydev->addr] = PHY_IGNORE_INTERRUPT; - else - priv->mii_bus->irq[phydev->addr] = PHY_POLL; return 0; } @@ -477,12 +475,6 @@ static int bcmgenet_mii_alloc(struct bcmgenet_priv *priv) snprintf(bus->id, MII_BUS_ID_SIZE, "%s-%d", priv->pdev->name, priv->pdev->id); - bus->irq = kcalloc(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL); - if (!bus->irq) { - mdiobus_free(priv->mii_bus); - return -ENOMEM; - } - return 0; } @@ -648,7 +640,6 @@ int bcmgenet_mii_init(struct net_device *dev) out: of_node_put(priv->phy_dn); mdiobus_unregister(priv->mii_bus); - kfree(priv->mii_bus->irq); mdiobus_free(priv->mii_bus); return ret; } @@ -659,6 +650,5 @@ void bcmgenet_mii_exit(struct net_device *dev) of_node_put(priv->phy_dn); mdiobus_unregister(priv->mii_bus); - kfree(priv->mii_bus->irq); mdiobus_free(priv->mii_bus); } diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c index 2470c6084c67..68a363708d27 100644 --- a/drivers/net/ethernet/broadcom/sb1250-mac.c +++ b/drivers/net/ethernet/broadcom/sb1250-mac.c @@ -238,7 +238,6 @@ struct sbmac_softc { struct napi_struct napi; struct phy_device *phy_dev; /* the associated PHY device */ struct mii_bus *mii_bus; /* the MII bus */ - int phy_irq[PHY_MAX_ADDR]; spinlock_t sbm_lock; /* spin lock */ int sbm_devflags; /* current device flags */ @@ -2250,9 +2249,6 @@ static int sbmac_init(struct platform_device *pldev, long long base) sc->mii_bus->priv = sc; sc->mii_bus->read = sbmac_mii_read; sc->mii_bus->write = sbmac_mii_write; - sc->mii_bus->irq = sc->phy_irq; - for (i = 0; i < PHY_MAX_ADDR; ++i) - sc->mii_bus->irq[i] = SBMAC_PHY_INT; sc->mii_bus->parent = &pldev->dev; /* diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 07c067590caa..04e7d0d0e5b1 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -1538,10 +1538,6 @@ static int tg3_mdio_init(struct tg3 *tp) tp->mdio_bus->read = &tg3_mdio_read; tp->mdio_bus->write = &tg3_mdio_write; tp->mdio_bus->phy_mask = ~(1 << tp->phy_addr); - tp->mdio_bus->irq = &tp->mdio_irq[0]; - - for (i = 0; i < PHY_MAX_ADDR; i++) - tp->mdio_bus->irq[i] = PHY_POLL; /* The bus registration will look for all the PHYs on the mdio bus. * Unfortunately, it does not ensure the PHY is powered up before diff --git a/drivers/net/ethernet/broadcom/tg3.h b/drivers/net/ethernet/broadcom/tg3.h index 31c9f8295953..3b5e98ecba00 100644 --- a/drivers/net/ethernet/broadcom/tg3.h +++ b/drivers/net/ethernet/broadcom/tg3.h @@ -3254,7 +3254,6 @@ struct tg3 { int pcie_readrq; struct mii_bus *mdio_bus; - int mdio_irq[PHY_MAX_ADDR]; int old_link; u8 phy_addr; diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 98df33b7a395..eb1397484eef 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -441,12 +441,6 @@ static int macb_mii_init(struct macb *bp) bp->mii_bus->parent = &bp->dev->dev; pdata = dev_get_platdata(&bp->pdev->dev); - bp->mii_bus->irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); - if (!bp->mii_bus->irq) { - err = -ENOMEM; - goto err_out_free_mdiobus; - } - dev_set_drvdata(&bp->dev->dev, bp->mii_bus); np = bp->pdev->dev.of_node; @@ -471,9 +465,6 @@ static int macb_mii_init(struct macb *bp) goto err_out_unregister_bus; } } else { - for (i = 0; i < PHY_MAX_ADDR; i++) - bp->mii_bus->irq[i] = PHY_POLL; - if (pdata) bp->mii_bus->phy_mask = pdata->phy_mask; @@ -481,7 +472,7 @@ static int macb_mii_init(struct macb *bp) } if (err) - goto err_out_free_mdio_irq; + goto err_out_free_mdiobus; err = macb_mii_probe(bp->dev); if (err) @@ -491,8 +482,6 @@ static int macb_mii_init(struct macb *bp) err_out_unregister_bus: mdiobus_unregister(bp->mii_bus); -err_out_free_mdio_irq: - kfree(bp->mii_bus->irq); err_out_free_mdiobus: mdiobus_free(bp->mii_bus); err_out: @@ -2980,7 +2969,6 @@ static int macb_remove(struct platform_device *pdev) if (bp->phy_dev) phy_disconnect(bp->phy_dev); mdiobus_unregister(bp->mii_bus); - kfree(bp->mii_bus->irq); mdiobus_free(bp->mii_bus); /* Shutdown the PHY if there is a GPIO reset */ diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 6557460cf028..b69a9eacc531 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -302,7 +302,7 @@ static int dnet_mii_probe(struct net_device *dev) static int dnet_mii_init(struct dnet *bp) { - int err, i; + int err; bp->mii_bus = mdiobus_alloc(); if (bp->mii_bus == NULL) @@ -317,16 +317,6 @@ static int dnet_mii_init(struct dnet *bp) bp->mii_bus->priv = bp; - bp->mii_bus->irq = devm_kmalloc(&bp->pdev->dev, - sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!bp->mii_bus->irq) { - err = -ENOMEM; - goto err_out; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - bp->mii_bus->irq[i] = PHY_POLL; - if (mdiobus_register(bp->mii_bus)) { err = -ENXIO; goto err_out; diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index ff665493ca97..c028b299ab3f 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -1015,7 +1015,6 @@ static int ethoc_probe(struct platform_device *pdev) struct resource *mmio = NULL; struct resource *mem = NULL; struct ethoc *priv = NULL; - unsigned int phy; int num_bd; int ret = 0; bool random_mac = false; @@ -1206,19 +1205,10 @@ static int ethoc_probe(struct platform_device *pdev) priv->mdio->write = ethoc_mdio_write; priv->mdio->priv = priv; - priv->mdio->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!priv->mdio->irq) { - ret = -ENOMEM; - goto free_mdio; - } - - for (phy = 0; phy < PHY_MAX_ADDR; phy++) - priv->mdio->irq[phy] = PHY_POLL; - ret = mdiobus_register(priv->mdio); if (ret) { dev_err(&netdev->dev, "failed to register MDIO bus\n"); - goto free_mdio; + goto free; } ret = ethoc_mdio_probe(netdev); @@ -1250,8 +1240,6 @@ error2: netif_napi_del(&priv->napi); error: mdiobus_unregister(priv->mdio); -free_mdio: - kfree(priv->mdio->irq); mdiobus_free(priv->mdio); free: if (priv->clk) diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index c2e2ac6a0313..8f3f2cf0dcbf 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -71,7 +71,6 @@ struct ftgmac100 { struct napi_struct napi; struct mii_bus *mii_bus; - int phy_irq[PHY_MAX_ADDR]; struct phy_device *phydev; int old_speed; }; @@ -1188,7 +1187,6 @@ static int ftgmac100_probe(struct platform_device *pdev) struct net_device *netdev; struct ftgmac100 *priv; int err; - int i; if (!pdev) return -ENODEV; @@ -1257,10 +1255,6 @@ static int ftgmac100_probe(struct platform_device *pdev) priv->mii_bus->priv = netdev; priv->mii_bus->read = ftgmac100_mdiobus_read; priv->mii_bus->write = ftgmac100_mdiobus_write; - priv->mii_bus->irq = priv->phy_irq; - - for (i = 0; i < PHY_MAX_ADDR; i++) - priv->mii_bus->irq[i] = PHY_POLL; err = mdiobus_register(priv->mii_bus); if (err) { diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index ceabe21b3b2c..da255fb4f1d5 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1983,7 +1983,7 @@ static int fec_enet_mii_init(struct platform_device *pdev) struct net_device *ndev = platform_get_drvdata(pdev); struct fec_enet_private *fep = netdev_priv(ndev); struct device_node *node; - int err = -ENXIO, i; + int err = -ENXIO; u32 mii_speed, holdtime; /* @@ -2065,15 +2065,6 @@ static int fec_enet_mii_init(struct platform_device *pdev) fep->mii_bus->priv = fep; fep->mii_bus->parent = &pdev->dev; - fep->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!fep->mii_bus->irq) { - err = -ENOMEM; - goto err_out_free_mdiobus; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - fep->mii_bus->irq[i] = PHY_POLL; - node = of_get_child_by_name(pdev->dev.of_node, "mdio"); if (node) { err = of_mdiobus_register(fep->mii_bus, node); @@ -2083,7 +2074,7 @@ static int fec_enet_mii_init(struct platform_device *pdev) } if (err) - goto err_out_free_mdio_irq; + goto err_out_free_mdiobus; mii_cnt++; @@ -2093,8 +2084,6 @@ static int fec_enet_mii_init(struct platform_device *pdev) return 0; -err_out_free_mdio_irq: - kfree(fep->mii_bus->irq); err_out_free_mdiobus: mdiobus_free(fep->mii_bus); err_out: @@ -2105,7 +2094,6 @@ static void fec_enet_mii_remove(struct fec_enet_private *fep) { if (--mii_cnt == 0) { mdiobus_unregister(fep->mii_bus); - kfree(fep->mii_bus->irq); mdiobus_free(fep->mii_bus); } } diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c b/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c index 1e647beaf989..b5497e308302 100644 --- a/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c +++ b/drivers/net/ethernet/freescale/fec_mpc52xx_phy.c @@ -22,7 +22,6 @@ struct mpc52xx_fec_mdio_priv { struct mpc52xx_fec __iomem *regs; - int mdio_irqs[PHY_MAX_ADDR]; }; static int mpc52xx_fec_mdio_transfer(struct mii_bus *bus, int phy_id, @@ -83,9 +82,6 @@ static int mpc52xx_fec_mdio_probe(struct platform_device *of) bus->read = mpc52xx_fec_mdio_read; bus->write = mpc52xx_fec_mdio_write; - /* setup irqs */ - bus->irq = priv->mdio_irqs; - /* setup registers */ err = of_address_to_resource(np, 0, &res); if (err) diff --git a/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c b/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c index 68a428de0bc0..1f015edcca22 100644 --- a/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c +++ b/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c @@ -172,23 +172,16 @@ static int fs_enet_mdio_probe(struct platform_device *ofdev) goto out_free_bus; new_bus->phy_mask = ~0; - new_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!new_bus->irq) { - ret = -ENOMEM; - goto out_unmap_regs; - } new_bus->parent = &ofdev->dev; platform_set_drvdata(ofdev, new_bus); ret = of_mdiobus_register(new_bus, ofdev->dev.of_node); if (ret) - goto out_free_irqs; + goto out_unmap_regs; return 0; -out_free_irqs: - kfree(new_bus->irq); out_unmap_regs: iounmap(bitbang->dir); out_free_bus: @@ -205,7 +198,6 @@ static int fs_enet_mdio_remove(struct platform_device *ofdev) struct bb_info *bitbang = bus->priv; mdiobus_unregister(bus); - kfree(bus->irq); free_mdio_bitbang(bus); iounmap(bitbang->dir); kfree(bitbang); diff --git a/drivers/net/ethernet/freescale/fs_enet/mii-fec.c b/drivers/net/ethernet/freescale/fs_enet/mii-fec.c index 2be383e6d258..a89267b94352 100644 --- a/drivers/net/ethernet/freescale/fs_enet/mii-fec.c +++ b/drivers/net/ethernet/freescale/fs_enet/mii-fec.c @@ -166,23 +166,16 @@ static int fs_enet_mdio_probe(struct platform_device *ofdev) clrsetbits_be32(&fec->fecp->fec_mii_speed, 0x7E, fec->mii_speed); new_bus->phy_mask = ~0; - new_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!new_bus->irq) { - ret = -ENOMEM; - goto out_unmap_regs; - } new_bus->parent = &ofdev->dev; platform_set_drvdata(ofdev, new_bus); ret = of_mdiobus_register(new_bus, ofdev->dev.of_node); if (ret) - goto out_free_irqs; + goto out_unmap_regs; return 0; -out_free_irqs: - kfree(new_bus->irq); out_unmap_regs: iounmap(fec->fecp); out_res: @@ -200,7 +193,6 @@ static int fs_enet_mdio_remove(struct platform_device *ofdev) struct fec_info *fec = bus->priv; mdiobus_unregister(bus); - kfree(bus->irq); iounmap(fec->fecp); kfree(fec); mdiobus_free(bus); diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c index 40071dad1c57..622005abf859 100644 --- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c +++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c @@ -69,7 +69,6 @@ struct fsl_pq_mdio { struct fsl_pq_mdio_priv { void __iomem *map; struct fsl_pq_mii __iomem *regs; - int irqs[PHY_MAX_ADDR]; }; /* @@ -401,7 +400,6 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev) new_bus->read = &fsl_pq_mdio_read; new_bus->write = &fsl_pq_mdio_write; new_bus->reset = &fsl_pq_mdio_reset; - new_bus->irq = priv->irqs; err = of_address_to_resource(np, 0, &res); if (err < 0) { diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 37491c85bc42..58c96c412fe8 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -463,11 +463,6 @@ static int hns_mdio_probe(struct platform_device *pdev) dev_warn(&pdev->dev, "no syscon hisilicon,peri-c-subctrl\n"); mdio_dev->subctrl_vbase = NULL; } - new_bus->irq = devm_kcalloc(&pdev->dev, PHY_MAX_ADDR, - sizeof(int), GFP_KERNEL); - if (!new_bus->irq) - return -ENOMEM; - new_bus->parent = &pdev->dev; platform_set_drvdata(pdev, new_bus); diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 86238a5eaddf..fb61f7f96bb4 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -433,18 +433,9 @@ ltq_etop_mdio_init(struct net_device *dev) priv->mii_bus->name = "ltq_mii"; snprintf(priv->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x", priv->pdev->name, priv->pdev->id); - priv->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!priv->mii_bus->irq) { - err = -ENOMEM; - goto err_out_free_mdiobus; - } - - for (i = 0; i < PHY_MAX_ADDR; ++i) - priv->mii_bus->irq[i] = PHY_POLL; - if (mdiobus_register(priv->mii_bus)) { err = -ENXIO; - goto err_out_free_mdio_irq; + goto err_out_free_mdiobus; } if (ltq_etop_mdio_probe(dev)) { @@ -455,8 +446,6 @@ ltq_etop_mdio_init(struct net_device *dev) err_out_unregister_bus: mdiobus_unregister(priv->mii_bus); -err_out_free_mdio_irq: - kfree(priv->mii_bus->irq); err_out_free_mdiobus: mdiobus_free(priv->mii_bus); err_out: @@ -470,7 +459,6 @@ ltq_etop_mdio_cleanup(struct net_device *dev) phy_disconnect(priv->phydev); mdiobus_unregister(priv->mii_bus); - kfree(priv->mii_bus->irq); mdiobus_free(priv->mii_bus); } diff --git a/drivers/net/ethernet/marvell/mvmdio.c b/drivers/net/ethernet/marvell/mvmdio.c index fc2fb25343f4..8982c882af1b 100644 --- a/drivers/net/ethernet/marvell/mvmdio.c +++ b/drivers/net/ethernet/marvell/mvmdio.c @@ -187,7 +187,7 @@ static int orion_mdio_probe(struct platform_device *pdev) struct resource *r; struct mii_bus *bus; struct orion_mdio_dev *dev; - int i, ret; + int ret; r = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!r) { @@ -207,14 +207,6 @@ static int orion_mdio_probe(struct platform_device *pdev) dev_name(&pdev->dev)); bus->parent = &pdev->dev; - bus->irq = devm_kmalloc_array(&pdev->dev, PHY_MAX_ADDR, sizeof(int), - GFP_KERNEL); - if (!bus->irq) - return -ENOMEM; - - for (i = 0; i < PHY_MAX_ADDR; i++) - bus->irq[i] = PHY_POLL; - dev = bus->priv; dev->regs = devm_ioremap(&pdev->dev, r->start, resource_size(r)); if (!dev->regs) { diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index 024bc3675573..a9ce37f9bb2e 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -850,19 +850,10 @@ static int lpc_mii_init(struct netdata_local *pldat) pldat->mii_bus->priv = pldat; pldat->mii_bus->parent = &pldat->pdev->dev; - pldat->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!pldat->mii_bus->irq) { - err = -ENOMEM; - goto err_out_1; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - pldat->mii_bus->irq[i] = PHY_POLL; - platform_set_drvdata(pldat->pdev, pldat->mii_bus); if (mdiobus_register(pldat->mii_bus)) - goto err_out_free_mdio_irq; + goto err_out_unregister_bus; if (lpc_mii_probe(pldat->ndev) != 0) goto err_out_unregister_bus; @@ -871,8 +862,6 @@ static int lpc_mii_init(struct netdata_local *pldat) err_out_unregister_bus: mdiobus_unregister(pldat->mii_bus); -err_out_free_mdio_irq: - kfree(pldat->mii_bus->irq); err_out_1: mdiobus_free(pldat->mii_bus); err_out: diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 174dea787caf..6b541e57c96a 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -1075,7 +1075,6 @@ static int r6040_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) static int card_idx = -1; int bar = 0; u16 *adrp; - int i; pr_info("%s\n", version); @@ -1187,19 +1186,11 @@ static int r6040_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) lp->mii_bus->name = "r6040_eth_mii"; snprintf(lp->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x", dev_name(&pdev->dev), card_idx); - lp->mii_bus->irq = kmalloc_array(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL); - if (!lp->mii_bus->irq) { - err = -ENOMEM; - goto err_out_mdio; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - lp->mii_bus->irq[i] = PHY_POLL; err = mdiobus_register(lp->mii_bus); if (err) { dev_err(&pdev->dev, "failed to register MII bus\n"); - goto err_out_mdio_irq; + goto err_out_mdio; } err = r6040_mii_probe(dev); @@ -1218,8 +1209,6 @@ static int r6040_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) err_out_mdio_unregister: mdiobus_unregister(lp->mii_bus); -err_out_mdio_irq: - kfree(lp->mii_bus->irq); err_out_mdio: mdiobus_free(lp->mii_bus); err_out_unmap: @@ -1242,7 +1231,6 @@ static void r6040_remove_one(struct pci_dev *pdev) unregister_netdev(dev); mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); netif_napi_del(&lp->napi); pci_iounmap(pdev, lp->base); diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 94581be64d65..c5ec57ce3ddc 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -2859,7 +2859,7 @@ static int sh_mdio_release(struct sh_eth_private *mdp) static int sh_mdio_init(struct sh_eth_private *mdp, struct sh_eth_plat_data *pd) { - int ret, i; + int ret; struct bb_info *bitbang; struct platform_device *pdev = mdp->pdev; struct device *dev = &mdp->pdev->dev; @@ -2885,20 +2885,10 @@ static int sh_mdio_init(struct sh_eth_private *mdp, snprintf(mdp->mii_bus->id, MII_BUS_ID_SIZE, "%s-%x", pdev->name, pdev->id); - /* PHY IRQ */ - mdp->mii_bus->irq = devm_kmalloc_array(dev, PHY_MAX_ADDR, sizeof(int), - GFP_KERNEL); - if (!mdp->mii_bus->irq) { - ret = -ENOMEM; - goto out_free_bus; - } - /* register MDIO bus */ if (dev->of_node) { ret = of_mdiobus_register(mdp->mii_bus, dev->of_node); } else { - for (i = 0; i < PHY_MAX_ADDR; i++) - mdp->mii_bus->irq[i] = PHY_POLL; if (pd->phy_irq > 0) mdp->mii_bus->irq[pd->phy] = pd->phy_irq; diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 139b99b04099..c74e78dd989a 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1059,7 +1059,7 @@ static int smsc911x_mii_init(struct platform_device *pdev, struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); - int err = -ENXIO, i; + int err = -ENXIO; pdata->mii_bus = mdiobus_alloc(); if (!pdata->mii_bus) { @@ -1073,9 +1073,7 @@ static int smsc911x_mii_init(struct platform_device *pdev, pdata->mii_bus->priv = pdata; pdata->mii_bus->read = smsc911x_mii_read; pdata->mii_bus->write = smsc911x_mii_write; - pdata->mii_bus->irq = pdata->phy_irq; - for (i = 0; i < PHY_MAX_ADDR; ++i) - pdata->mii_bus->irq[i] = PHY_POLL; + memcpy(pdata->mii_bus->irq, pdata->phy_irq, sizeof(pdata->mii_bus)); pdata->mii_bus->parent = &pdev->dev; diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index fa8893a804f7..59bf4c353d50 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -78,7 +78,6 @@ struct smsc9420_pdata { struct phy_device *phy_dev; struct mii_bus *mii_bus; - int phy_irq[PHY_MAX_ADDR]; int last_duplex; int last_carrier; }; @@ -1188,7 +1187,7 @@ static int smsc9420_mii_probe(struct net_device *dev) static int smsc9420_mii_init(struct net_device *dev) { struct smsc9420_pdata *pd = netdev_priv(dev); - int err = -ENXIO, i; + int err = -ENXIO; pd->mii_bus = mdiobus_alloc(); if (!pd->mii_bus) { @@ -1201,9 +1200,6 @@ static int smsc9420_mii_init(struct net_device *dev) pd->mii_bus->priv = pd; pd->mii_bus->read = smsc9420_mii_read; pd->mii_bus->write = smsc9420_mii_write; - pd->mii_bus->irq = pd->phy_irq; - for (i = 0; i < PHY_MAX_ADDR; ++i) - pd->mii_bus->irq[i] = PHY_POLL; /* Mask all PHYs except ID 1 (internal) */ pd->mii_bus->phy_mask = ~(1 << 1); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index 05ba84118f37..f0990eb9460f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -196,7 +196,6 @@ int stmmac_mdio_register(struct net_device *ndev) { int err = 0; struct mii_bus *new_bus; - int *irqlist; struct stmmac_priv *priv = netdev_priv(ndev); struct stmmac_mdio_bus_data *mdio_bus_data = priv->plat->mdio_bus_data; int addr, found; @@ -227,13 +226,8 @@ int stmmac_mdio_register(struct net_device *ndev) if (new_bus == NULL) return -ENOMEM; - if (mdio_bus_data->irqs) { - irqlist = mdio_bus_data->irqs; - } else { - for (addr = 0; addr < PHY_MAX_ADDR; addr++) - priv->mii_irq[addr] = PHY_POLL; - irqlist = priv->mii_irq; - } + if (mdio_bus_data->irqs) + memcpy(new_bus->irq, mdio_bus_data, sizeof(new_bus->irq)); #ifdef CONFIG_OF if (priv->device->of_node) @@ -247,7 +241,6 @@ int stmmac_mdio_register(struct net_device *ndev) snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%x", new_bus->name, priv->plat->bus_id); new_bus->priv = ndev; - new_bus->irq = irqlist; new_bus->phy_mask = mdio_bus_data->phy_mask; new_bus->parent = priv->device; @@ -271,7 +264,8 @@ int stmmac_mdio_register(struct net_device *ndev) */ if ((mdio_bus_data->irqs == NULL) && (mdio_bus_data->probed_phy_irq > 0)) { - irqlist[addr] = mdio_bus_data->probed_phy_irq; + new_bus->irq[addr] = + mdio_bus_data->probed_phy_irq; phydev->irq = mdio_bus_data->probed_phy_irq; } diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index b25ee370254a..6dbb681912f2 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -1220,7 +1220,7 @@ static void dwceqos_enable_mmc_interrupt(struct net_local *lp) static int dwceqos_mii_init(struct net_local *lp) { - int ret = -ENXIO, i; + int ret = -ENXIO; struct resource res; struct device_node *mdionode; @@ -1241,24 +1241,14 @@ static int dwceqos_mii_init(struct net_local *lp) lp->mii_bus->priv = lp; lp->mii_bus->parent = &lp->ndev->dev; - lp->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!lp->mii_bus->irq) { - ret = -ENOMEM; - goto err_out_free_mdiobus; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - lp->mii_bus->irq[i] = PHY_POLL; of_address_to_resource(lp->pdev->dev.of_node, 0, &res); snprintf(lp->mii_bus->id, MII_BUS_ID_SIZE, "%.8llx", (unsigned long long)res.start); if (of_mdiobus_register(lp->mii_bus, mdionode)) - goto err_out_free_mdio_irq; + goto err_out_free_mdiobus; return 0; -err_out_free_mdio_irq: - kfree(lp->mii_bus->irq); err_out_free_mdiobus: mdiobus_free(lp->mii_bus); err_out: @@ -2977,7 +2967,6 @@ static int dwceqos_remove(struct platform_device *pdev) if (lp->phy_dev) phy_disconnect(lp->phy_dev); mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); unregister_netdev(ndev); diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index 77d26fe286c0..0b483301767d 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -316,8 +316,6 @@ static int cpmac_mdio_reset(struct mii_bus *bus) return 0; } -static int mii_irqs[PHY_MAX_ADDR] = { PHY_POLL, }; - static struct mii_bus *cpmac_mii; static void cpmac_set_multicast_list(struct net_device *dev) @@ -1226,7 +1224,6 @@ int cpmac_init(void) cpmac_mii->read = cpmac_mdio_read; cpmac_mii->write = cpmac_mdio_write; cpmac_mii->reset = cpmac_mdio_reset; - cpmac_mii->irq = mii_irqs; cpmac_mii->priv = ioremap(AR7_REGS_MDIO, 256); diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 8fd5e0ba718c..fed5e3dfbc8f 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -682,18 +682,9 @@ static int tc_mii_init(struct net_device *dev) (lp->pci_dev->bus->number << 8) | lp->pci_dev->devfn); lp->mii_bus->priv = dev; lp->mii_bus->parent = &lp->pci_dev->dev; - lp->mii_bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!lp->mii_bus->irq) { - err = -ENOMEM; - goto err_out_free_mii_bus; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - lp->mii_bus->irq[i] = PHY_POLL; - err = mdiobus_register(lp->mii_bus); if (err) - goto err_out_free_mdio_irq; + goto err_out_free_mii_bus; err = tc_mii_probe(dev); if (err) goto err_out_unregister_bus; @@ -701,8 +692,6 @@ static int tc_mii_init(struct net_device *dev) err_out_unregister_bus: mdiobus_unregister(lp->mii_bus); -err_out_free_mdio_irq: - kfree(lp->mii_bus->irq); err_out_free_mii_bus: mdiobus_free(lp->mii_bus); err_out: @@ -880,7 +869,6 @@ static void tc35815_remove_one(struct pci_dev *pdev) phy_disconnect(lp->phy_dev); mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); unregister_netdev(dev); free_netdev(dev); diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 522abe2ff25a..902457e43628 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -337,7 +337,6 @@ struct temac_local { /* MDIO bus data */ struct mii_bus *mii_bus; /* MII bus reference */ - int mdio_irqs[PHY_MAX_ADDR]; /* IRQs table for MDIO bus */ /* IO registers, dma functions and IRQs */ void __iomem *regs; diff --git a/drivers/net/ethernet/xilinx/ll_temac_mdio.c b/drivers/net/ethernet/xilinx/ll_temac_mdio.c index 415de1eaf641..7714aff78b7d 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_mdio.c +++ b/drivers/net/ethernet/xilinx/ll_temac_mdio.c @@ -92,7 +92,6 @@ int temac_mdio_setup(struct temac_local *lp, struct device_node *np) bus->read = temac_mdio_read; bus->write = temac_mdio_write; bus->parent = lp->dev; - bus->irq = lp->mdio_irqs; /* preallocated IRQ table */ lp->mii_bus = bus; @@ -114,7 +113,6 @@ int temac_mdio_setup(struct temac_local *lp, struct device_node *np) void temac_mdio_teardown(struct temac_local *lp) { mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); lp->mii_bus = NULL; } diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h index 7cb9abac95c8..9ead4e269409 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h @@ -385,7 +385,6 @@ struct axidma_bd { * @phy_dev: Pointer to PHY device structure attached to the axienet_local * @phy_node: Pointer to device node structure * @mii_bus: Pointer to MII bus structure - * @mdio_irqs: IRQs table for MDIO bus required in mii_bus structure * @regs: Base address for the axienet_local device address space * @dma_regs: Base address for the axidma device address space * @dma_err_tasklet: Tasklet structure to process Axi DMA errors @@ -426,7 +425,6 @@ struct axienet_local { /* MDIO bus data */ struct mii_bus *mii_bus; /* MII bus reference */ - int mdio_irqs[PHY_MAX_ADDR]; /* IRQs table for MDIO bus */ /* IO registers, dma functions and IRQs */ void __iomem *regs; diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c index 507bbb0355c2..63307ea97846 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c @@ -212,7 +212,6 @@ issue: bus->read = axienet_mdio_read; bus->write = axienet_mdio_write; bus->parent = lp->dev; - bus->irq = lp->mdio_irqs; /* preallocated IRQ table */ lp->mii_bus = bus; ret = of_mdiobus_register(bus, np1); @@ -232,7 +231,6 @@ issue: void axienet_mdio_teardown(struct axienet_local *lp) { mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); lp->mii_bus = NULL; } diff --git a/drivers/net/ethernet/xilinx/xilinx_emaclite.c b/drivers/net/ethernet/xilinx/xilinx_emaclite.c index cf468c87ce57..d1a0167dff84 100644 --- a/drivers/net/ethernet/xilinx/xilinx_emaclite.c +++ b/drivers/net/ethernet/xilinx/xilinx_emaclite.c @@ -114,7 +114,6 @@ * @phy_dev: pointer to the PHY device * @phy_node: pointer to the PHY device node * @mii_bus: pointer to the MII bus - * @mdio_irqs: IRQs table for MDIO bus * @last_link: last link status * @has_mdio: indicates whether MDIO is included in the HW */ @@ -135,7 +134,6 @@ struct net_local { struct device_node *phy_node; struct mii_bus *mii_bus; - int mdio_irqs[PHY_MAX_ADDR]; int last_link; bool has_mdio; @@ -852,7 +850,6 @@ static int xemaclite_mdio_setup(struct net_local *lp, struct device *dev) bus->read = xemaclite_mdio_read; bus->write = xemaclite_mdio_write; bus->parent = dev; - bus->irq = lp->mdio_irqs; /* preallocated IRQ table */ lp->mii_bus = bus; @@ -1196,7 +1193,6 @@ static int xemaclite_of_remove(struct platform_device *of_dev) /* Un-register the mii_bus, if configured */ if (lp->has_mdio) { mdiobus_unregister(lp->mii_bus); - kfree(lp->mii_bus->irq); mdiobus_free(lp->mii_bus); lp->mii_bus = NULL; } diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index e23bf5b90e17..0a1cde6803b0 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -27,7 +27,6 @@ #define MII_REGS_NUM 29 struct fixed_mdio_bus { - int irqs[PHY_MAX_ADDR]; struct mii_bus *mii_bus; struct list_head phys; }; @@ -256,7 +255,7 @@ int fixed_phy_add(unsigned int irq, int phy_addr, memset(fp->regs, 0xFF, sizeof(fp->regs[0]) * MII_REGS_NUM); - fmb->irqs[phy_addr] = irq; + fmb->mii_bus->irq[phy_addr] = irq; fp->addr = phy_addr; fp->status = *status; @@ -395,7 +394,6 @@ static int __init fixed_mdio_bus_init(void) fmb->mii_bus->parent = &pdev->dev; fmb->mii_bus->read = &fixed_mdio_read; fmb->mii_bus->write = &fixed_mdio_write; - fmb->mii_bus->irq = fmb->irqs; ret = mdiobus_register(fmb->mii_bus); if (ret) diff --git a/drivers/net/phy/mdio-bcm-unimac.c b/drivers/net/phy/mdio-bcm-unimac.c index 4bde5e728fe0..8c73b2e771dd 100644 --- a/drivers/net/phy/mdio-bcm-unimac.c +++ b/drivers/net/phy/mdio-bcm-unimac.c @@ -200,16 +200,10 @@ static int unimac_mdio_probe(struct platform_device *pdev) bus->reset = unimac_mdio_reset; snprintf(bus->id, MII_BUS_ID_SIZE, "%s", pdev->name); - bus->irq = kcalloc(PHY_MAX_ADDR, sizeof(int), GFP_KERNEL); - if (!bus->irq) { - ret = -ENOMEM; - goto out_mdio_free; - } - ret = of_mdiobus_register(bus, np); if (ret) { dev_err(&pdev->dev, "MDIO bus registration failed\n"); - goto out_mdio_irq; + goto out_mdio_free; } platform_set_drvdata(pdev, priv); @@ -218,8 +212,6 @@ static int unimac_mdio_probe(struct platform_device *pdev) return 0; -out_mdio_irq: - kfree(bus->irq); out_mdio_free: mdiobus_free(bus); return ret; @@ -230,7 +222,6 @@ static int unimac_mdio_remove(struct platform_device *pdev) struct unimac_mdio_priv *priv = platform_get_drvdata(pdev); mdiobus_unregister(priv->mii_bus); - kfree(priv->mii_bus->irq); mdiobus_free(priv->mii_bus); return 0; diff --git a/drivers/net/phy/mdio-gpio.c b/drivers/net/phy/mdio-gpio.c index 95f51d7267b3..27ab63064f95 100644 --- a/drivers/net/phy/mdio-gpio.c +++ b/drivers/net/phy/mdio-gpio.c @@ -159,7 +159,7 @@ static struct mii_bus *mdio_gpio_bus_init(struct device *dev, new_bus->phy_mask = pdata->phy_mask; new_bus->phy_ignore_ta_mask = pdata->phy_ignore_ta_mask; - new_bus->irq = pdata->irqs; + memcpy(new_bus->irq, pdata->irqs, sizeof(new_bus->irq)); new_bus->parent = dev; if (new_bus->phy_mask == ~0) diff --git a/drivers/net/phy/mdio-moxart.c b/drivers/net/phy/mdio-moxart.c index f1fc51f655d9..5bb56d126693 100644 --- a/drivers/net/phy/mdio-moxart.c +++ b/drivers/net/phy/mdio-moxart.c @@ -130,13 +130,6 @@ static int moxart_mdio_probe(struct platform_device *pdev) snprintf(bus->id, MII_BUS_ID_SIZE, "%s-%d-mii", pdev->name, pdev->id); bus->parent = &pdev->dev; - bus->irq = devm_kzalloc(&pdev->dev, sizeof(int) * PHY_MAX_ADDR, - GFP_KERNEL); - if (!bus->irq) { - ret = -ENOMEM; - goto err_out_free_mdiobus; - } - /* Setting PHY_IGNORE_INTERRUPT here even if it has no effect, * of_mdiobus_register() sets these PHY_POLL. * Ideally, the interrupt from MAC controller could be used to diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c index 7f8e7662e28c..308ade0eb1b6 100644 --- a/drivers/net/phy/mdio-mux.c +++ b/drivers/net/phy/mdio-mux.c @@ -34,7 +34,6 @@ struct mdio_mux_child_bus { struct mdio_mux_parent_bus *parent; struct mdio_mux_child_bus *next; int bus_number; - int phy_irq[PHY_MAX_ADDR]; }; /* @@ -157,7 +156,7 @@ int mdio_mux_init(struct device *dev, break; } cb->mii_bus->priv = cb; - cb->mii_bus->irq = cb->phy_irq; + cb->mii_bus->name = "mdio_mux"; snprintf(cb->mii_bus->id, MII_BUS_ID_SIZE, "%x.%x", pb->parent_id, v); diff --git a/drivers/net/phy/mdio-octeon.c b/drivers/net/phy/mdio-octeon.c index 0d5da1312dd3..47d4f2f263d1 100644 --- a/drivers/net/phy/mdio-octeon.c +++ b/drivers/net/phy/mdio-octeon.c @@ -113,7 +113,6 @@ struct octeon_mdiobus { resource_size_t mdio_phys; resource_size_t regsize; enum octeon_mdiobus_mode mode; - int phy_irq[PHY_MAX_ADDR]; }; #ifdef CONFIG_CAVIUM_OCTEON_SOC @@ -306,7 +305,6 @@ static int octeon_mdiobus_probe(struct platform_device *pdev) oct_mdio_writeq(smi_en.u64, bus->register_base + SMI_EN); bus->mii_bus->priv = bus; - bus->mii_bus->irq = bus->phy_irq; bus->mii_bus->name = "mdio-octeon"; snprintf(bus->mii_bus->id, MII_BUS_ID_SIZE, "%llx", bus->register_base); bus->mii_bus->parent = &pdev->dev; diff --git a/drivers/net/phy/mdio-sun4i.c b/drivers/net/phy/mdio-sun4i.c index 15bc7f9ea224..f70522c35163 100644 --- a/drivers/net/phy/mdio-sun4i.c +++ b/drivers/net/phy/mdio-sun4i.c @@ -96,7 +96,7 @@ static int sun4i_mdio_probe(struct platform_device *pdev) struct mii_bus *bus; struct sun4i_mdio_data *data; struct resource *res; - int ret, i; + int ret; bus = mdiobus_alloc_size(sizeof(*data)); if (!bus) @@ -108,16 +108,6 @@ static int sun4i_mdio_probe(struct platform_device *pdev) snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii", dev_name(&pdev->dev)); bus->parent = &pdev->dev; - bus->irq = devm_kzalloc(&pdev->dev, sizeof(int) * PHY_MAX_ADDR, - GFP_KERNEL); - if (!bus->irq) { - ret = -ENOMEM; - goto err_out_free_mdiobus; - } - - for (i = 0; i < PHY_MAX_ADDR; i++) - bus->irq[i] = PHY_POLL; - data = bus->priv; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); data->membase = devm_ioremap_resource(&pdev->dev, res); diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 88cb4592b6fb..05381d0f559c 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -51,6 +51,7 @@ struct mii_bus *mdiobus_alloc_size(size_t size) struct mii_bus *bus; size_t aligned_size = ALIGN(sizeof(*bus), NETDEV_ALIGN); size_t alloc_size; + int i; /* If we alloc extra space, it should be aligned */ if (size) @@ -65,6 +66,10 @@ struct mii_bus *mdiobus_alloc_size(size_t size) bus->priv = (void *)bus + aligned_size; } + /* Initialise the interrupts to polling */ + for (i = 0; i < PHY_MAX_ADDR; i++) + bus->irq[i] = PHY_POLL; + return bus; } EXPORT_SYMBOL(mdiobus_alloc_size); diff --git a/drivers/net/usb/ax88172a.c b/drivers/net/usb/ax88172a.c index 5f18fcb8dcc7..224e7d82de6d 100644 --- a/drivers/net/usb/ax88172a.c +++ b/drivers/net/usb/ax88172a.c @@ -98,7 +98,7 @@ static void ax88172a_status(struct usbnet *dev, struct urb *urb) static int ax88172a_init_mdio(struct usbnet *dev) { struct ax88172a_private *priv = dev->driver_priv; - int ret, i; + int ret; priv->mdio = mdiobus_alloc(); if (!priv->mdio) { @@ -114,25 +114,15 @@ static int ax88172a_init_mdio(struct usbnet *dev) snprintf(priv->mdio->id, MII_BUS_ID_SIZE, "usb-%03d:%03d", dev->udev->bus->busnum, dev->udev->devnum); - priv->mdio->irq = kzalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!priv->mdio->irq) { - ret = -ENOMEM; - goto mfree; - } - for (i = 0; i < PHY_MAX_ADDR; i++) - priv->mdio->irq[i] = PHY_POLL; - ret = mdiobus_register(priv->mdio); if (ret) { netdev_err(dev->net, "Could not register MDIO bus\n"); - goto ifree; + goto mfree; } netdev_info(dev->net, "registered mdio bus %s\n", priv->mdio->id); return 0; -ifree: - kfree(priv->mdio->irq); mfree: mdiobus_free(priv->mdio); return ret; diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index 226668ead0d8..1662b7b144a8 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -1458,12 +1458,6 @@ static int lan78xx_mdio_init(struct lan78xx_net *dev) snprintf(dev->mdiobus->id, MII_BUS_ID_SIZE, "usb-%03d:%03d", dev->udev->bus->busnum, dev->udev->devnum); - dev->mdiobus->irq = kzalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL); - if (!dev->mdiobus->irq) { - ret = -ENOMEM; - goto exit1; - } - /* handle our own interrupt */ for (i = 0; i < PHY_MAX_ADDR; i++) dev->mdiobus->irq[i] = PHY_IGNORE_INTERRUPT; @@ -1479,13 +1473,11 @@ static int lan78xx_mdio_init(struct lan78xx_net *dev) ret = mdiobus_register(dev->mdiobus); if (ret) { netdev_err(dev->net, "can't register MDIO bus\n"); - goto exit2; + goto exit1; } netdev_dbg(dev->net, "registered mdiobus bus %s\n", dev->mdiobus->id); return 0; -exit2: - kfree(dev->mdiobus->irq); exit1: mdiobus_free(dev->mdiobus); return ret; @@ -1494,7 +1486,6 @@ exit1: static void lan78xx_remove_mdio(struct lan78xx_net *dev) { mdiobus_unregister(dev->mdiobus); - kfree(dev->mdiobus->irq); mdiobus_free(dev->mdiobus); } diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index a87a868fed64..2f88ff4654da 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -127,17 +127,12 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) struct device_node *child; const __be32 *paddr; bool scanphys = false; - int addr, rc, i; + int addr, rc; /* Mask out all PHYs from auto probing. Instead the PHYs listed in * the device tree are populated after the bus has been registered */ mdio->phy_mask = ~0; - /* Clear all the IRQ properties */ - if (mdio->irq) - for (i=0; iirq[i] = PHY_POLL; - mdio->dev.of_node = np; /* Register the MDIO bus */ diff --git a/include/linux/phy.h b/include/linux/phy.h index ecbf6382ba29..a5473c9e19de 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -189,10 +189,10 @@ struct mii_bus { u32 phy_ignore_ta_mask; /* - * Pointer to an array of interrupts, each PHY's - * interrupt at the index matching its address + * An array of interrupts, each PHY's interrupt at the index + * matching its address */ - int *irq; + int irq[PHY_MAX_ADDR]; }; #define to_mii_bus(d) container_of(d, struct mii_bus, dev) -- cgit v1.2.3-59-g8ed1b From e5a03bfd873c29eb786655ef2e95e53ed242b404 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:16 +0100 Subject: phy: Add an mdio_device structure Not all devices attached to an MDIO bus are phys. So add an mdio_device structure to represent the generic parts of an mdio device, and place this structure into the phy_device. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/agere/et131x.c | 30 ++++++------ drivers/net/ethernet/altera/altera_tse_main.c | 2 +- drivers/net/ethernet/broadcom/b44.c | 2 +- drivers/net/ethernet/broadcom/genet/bcmmii.c | 2 +- drivers/net/ethernet/broadcom/sb1250-mac.c | 4 +- drivers/net/ethernet/freescale/fman/fman_dtsec.c | 6 +-- drivers/net/ethernet/freescale/fman/fman_memac.c | 6 +-- drivers/net/ethernet/freescale/fs_enet/mac-fec.c | 2 +- drivers/net/ethernet/freescale/gianfar.c | 4 +- drivers/net/ethernet/freescale/ucc_geth.c | 4 +- drivers/net/ethernet/hisilicon/hns/hns_ethtool.c | 2 +- drivers/net/ethernet/marvell/mv643xx_eth.c | 2 +- drivers/net/ethernet/marvell/mvneta.c | 2 +- drivers/net/ethernet/smsc/smsc911x.c | 11 +++-- drivers/net/ethernet/smsc/smsc9420.c | 3 +- drivers/net/ethernet/ti/cpsw.c | 3 +- drivers/net/ethernet/ti/davinci_mdio.c | 2 +- drivers/net/ethernet/xilinx/xilinx_emaclite.c | 2 +- drivers/net/phy/at803x.c | 2 +- drivers/net/phy/bcm87xx.c | 4 +- drivers/net/phy/dp83640.c | 22 +++++---- drivers/net/phy/dp83867.c | 4 +- drivers/net/phy/fixed_phy.c | 10 ++-- drivers/net/phy/icplus.c | 18 +++---- drivers/net/phy/marvell.c | 7 +-- drivers/net/phy/mdio_bus.c | 12 ++--- drivers/net/phy/micrel.c | 12 ++--- drivers/net/phy/microchip.c | 4 +- drivers/net/phy/phy.c | 25 +++++----- drivers/net/phy/phy_device.c | 62 ++++++++++++------------ drivers/net/phy/smsc.c | 2 +- drivers/of/of_mdio.c | 6 +-- include/linux/mdio.h | 9 ++++ include/linux/phy.h | 26 +++++----- net/dsa/dsa.c | 2 +- 35 files changed, 165 insertions(+), 151 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index f29d45eea1d9..3f3bcbea15bd 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -1235,7 +1235,7 @@ static int et131x_mii_read(struct et131x_adapter *adapter, u8 reg, u16 *value) if (!phydev) return -EIO; - return et131x_phy_mii_read(adapter, phydev->addr, reg, value); + return et131x_phy_mii_read(adapter, phydev->mdio.addr, reg, value); } static int et131x_mii_write(struct et131x_adapter *adapter, u8 addr, u8 reg, @@ -1462,7 +1462,7 @@ static void et1310_phy_power_switch(struct et131x_adapter *adapter, bool down) data &= ~BMCR_PDOWN; if (down) data |= BMCR_PDOWN; - et131x_mii_write(adapter, phydev->addr, MII_BMCR, data); + et131x_mii_write(adapter, phydev->mdio.addr, MII_BMCR, data); } /* et131x_xcvr_init - Init the phy if we are setting it into force mode */ @@ -1490,7 +1490,7 @@ static void et131x_xcvr_init(struct et131x_adapter *adapter) else lcr2 |= (LED_VAL_LINKON << LED_TXRX_SHIFT); - et131x_mii_write(adapter, phydev->addr, PHY_LED_2, lcr2); + et131x_mii_write(adapter, phydev->mdio.addr, PHY_LED_2, lcr2); } } @@ -3192,14 +3192,14 @@ static void et131x_adjust_link(struct net_device *netdev) et131x_mii_read(adapter, PHY_MPHY_CONTROL_REG, ®ister18); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, PHY_MPHY_CONTROL_REG, register18 | 0x4); - et131x_mii_write(adapter, phydev->addr, PHY_INDEX_REG, - register18 | 0x8402); - et131x_mii_write(adapter, phydev->addr, PHY_DATA_REG, - register18 | 511); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, + PHY_INDEX_REG, register18 | 0x8402); + et131x_mii_write(adapter, phydev->mdio.addr, + PHY_DATA_REG, register18 | 511); + et131x_mii_write(adapter, phydev->mdio.addr, PHY_MPHY_CONTROL_REG, register18); } @@ -3212,8 +3212,8 @@ static void et131x_adjust_link(struct net_device *netdev) et131x_mii_read(adapter, PHY_CONFIG, ®); reg &= ~ET_PHY_CONFIG_TX_FIFO_DEPTH; reg |= ET_PHY_CONFIG_FIFO_DEPTH_32; - et131x_mii_write(adapter, phydev->addr, PHY_CONFIG, - reg); + et131x_mii_write(adapter, phydev->mdio.addr, + PHY_CONFIG, reg); } et131x_set_rx_dma_timer(adapter); @@ -3226,14 +3226,14 @@ static void et131x_adjust_link(struct net_device *netdev) et131x_mii_read(adapter, PHY_MPHY_CONTROL_REG, ®ister18); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, PHY_MPHY_CONTROL_REG, register18 | 0x4); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, PHY_INDEX_REG, register18 | 0x8402); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, PHY_DATA_REG, register18 | 511); - et131x_mii_write(adapter, phydev->addr, + et131x_mii_write(adapter, phydev->mdio.addr, PHY_MPHY_CONTROL_REG, register18); } diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 10d51e8aefe0..17472851674f 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -844,7 +844,7 @@ static int init_phy(struct net_device *dev) } netdev_dbg(dev, "attached to PHY %d UID 0x%08x Link = %d\n", - phydev->addr, phydev->phy_id, phydev->link); + phydev->mdio.addr, phydev->phy_id, phydev->link); priv->phydev = phydev; return 0; diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 4d08bc02c7a8..843a4a5864fc 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2305,7 +2305,7 @@ static int b44_register_phy_one(struct b44 *bp) bp->phydev = phydev; bp->old_link = 0; - bp->phy_addr = phydev->addr; + bp->phy_addr = phydev->mdio.addr; phy_attached_info(phydev); diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 4523acd8c1c2..633b59db813a 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -401,7 +401,7 @@ int bcmgenet_mii_probe(struct net_device *dev) * Ethernet MAC ISRs */ if (priv->internal_phy) - priv->mii_bus->irq[phydev->addr] = PHY_IGNORE_INTERRUPT; + priv->mii_bus->irq[phydev->mdio.addr] = PHY_IGNORE_INTERRUPT; return 0; } diff --git a/drivers/net/ethernet/broadcom/sb1250-mac.c b/drivers/net/ethernet/broadcom/sb1250-mac.c index 68a363708d27..768c18da510c 100644 --- a/drivers/net/ethernet/broadcom/sb1250-mac.c +++ b/drivers/net/ethernet/broadcom/sb1250-mac.c @@ -2366,8 +2366,8 @@ static int sbmac_mii_probe(struct net_device *dev) return -ENXIO; } - phy_dev = phy_connect(dev, dev_name(&phy_dev->dev), &sbmac_mii_poll, - PHY_INTERFACE_MODE_GMII); + phy_dev = phy_connect(dev, dev_name(&phy_dev->mdio.dev), + &sbmac_mii_poll, PHY_INTERFACE_MODE_GMII); if (IS_ERR(phy_dev)) { printk(KERN_ERR "%s: could not attach to PHY\n", dev->name); return PTR_ERR(phy_dev); diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c index 587f9b40cfaa..6b1261c0b1c2 100644 --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c @@ -1295,7 +1295,7 @@ int dtsec_init(struct fman_mac *dtsec) err = init(dtsec->regs, dtsec_drv_param, dtsec->phy_if, dtsec->max_speed, (u8 *)eth_addr, dtsec->exceptions, - dtsec->tbiphy->addr); + dtsec->tbiphy->mdio.addr); if (err) { free_init_resources(dtsec); pr_err("DTSEC version doesn't support this i/f mode\n"); @@ -1434,11 +1434,11 @@ struct fman_mac *dtsec_config(struct fman_mac_params *params) dtsec->tbiphy = of_phy_find_device(params->internal_phy_node); if (!dtsec->tbiphy) { pr_err("of_phy_find_device (TBI PHY) failed\n"); - put_device(&dtsec->tbiphy->dev); + put_device(&dtsec->tbiphy->mdio.dev); goto err_dtsec_drv_param; } - put_device(&dtsec->tbiphy->dev); + put_device(&dtsec->tbiphy->mdio.dev); /* Save FMan revision */ fman_get_revision(dtsec->fm, &dtsec->fm_rev_info); diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index 58bb72071c14..45e98fd8b79e 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -1054,15 +1054,15 @@ int memac_init(struct fman_mac *memac) * register address space and access each one of 4 * ports inside QSGMII. */ - phy_addr = memac->pcsphy->addr; + phy_addr = memac->pcsphy->mdio.addr; qsmgii_phy_addr = (u8)((phy_addr << 2) | i); - memac->pcsphy->addr = qsmgii_phy_addr; + memac->pcsphy->mdio.addr = qsmgii_phy_addr; if (memac->basex_if) setup_sgmii_internal_phy_base_x(memac); else setup_sgmii_internal_phy(memac, fixed_link); - memac->pcsphy->addr = phy_addr; + memac->pcsphy->mdio.addr = phy_addr; } } diff --git a/drivers/net/ethernet/freescale/fs_enet/mac-fec.c b/drivers/net/ethernet/freescale/fs_enet/mac-fec.c index 016743e355de..c158d409f6af 100644 --- a/drivers/net/ethernet/freescale/fs_enet/mac-fec.c +++ b/drivers/net/ethernet/freescale/fs_enet/mac-fec.c @@ -254,7 +254,7 @@ static void restart(struct net_device *dev) int r; u32 addrhi, addrlo; - struct mii_bus* mii = fep->phydev->bus; + struct mii_bus *mii = fep->phydev->mdio.bus; struct fec_info* fec_inf = mii->priv; r = whack_reset(fep->fec.fecp); diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 4e394f75261e..2aa7b401cc3b 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1834,7 +1834,7 @@ static void gfar_configure_serdes(struct net_device *dev) * several seconds for it to come back. */ if (phy_read(tbiphy, MII_BMSR) & BMSR_LSTATUS) { - put_device(&tbiphy->dev); + put_device(&tbiphy->mdio.dev); return; } @@ -1849,7 +1849,7 @@ static void gfar_configure_serdes(struct net_device *dev) BMCR_ANENABLE | BMCR_ANRESTART | BMCR_FULLDPLX | BMCR_SPEED1000); - put_device(&tbiphy->dev); + put_device(&tbiphy->mdio.dev); } static int __gfar_is_rx_idle(struct gfar_private *priv) diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 650f7888e32b..0e7f24ec3239 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -1385,7 +1385,7 @@ static int adjust_enet_interface(struct ucc_geth_private *ugeth) value &= ~0x1000; /* Turn off autonegotiation */ phy_write(tbiphy, ENET_TBI_MII_CR, value); - put_device(&tbiphy->dev); + put_device(&tbiphy->mdio.dev); } init_check_frame_length_mode(ug_info->lengthCheckRx, &ug_regs->maccfg2); @@ -1705,7 +1705,7 @@ static void uec_configure_serdes(struct net_device *dev) * several seconds for it to come back. */ if (phy_read(tbiphy, ENET_TBI_MII_SR) & TBISR_LSTATUS) { - put_device(&tbiphy->dev); + put_device(&tbiphy->mdio.dev); return; } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 4eddbeb19307..3df22840fcd1 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -71,7 +71,7 @@ static void hns_get_mdix_mode(struct net_device *net_dev, struct hns_nic_priv *priv = netdev_priv(net_dev); struct phy_device *phy_dev = priv->phy; - if (!phy_dev || !phy_dev->bus) { + if (!phy_dev || !phy_dev->mdio.bus) { cmd->eth_tp_mdix_ctrl = ETH_TP_MDI_INVALID; cmd->eth_tp_mdix = ETH_TP_MDI_INVALID; return; diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 4eba2ed53052..a0c03834a2f7 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -3133,7 +3133,7 @@ static int mv643xx_eth_probe(struct platform_device *pdev) if (!mp->phy) err = -ENODEV; else - phy_addr_set(mp, mp->phy->addr); + phy_addr_set(mp, mp->phy->mdio.addr); } else if (pd->phy_addr != MV643XX_ETH_PHY_NONE) { mp->phy = phy_scan(mp, pd->phy_addr); diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 15b1f6bbd92d..fabc8df40392 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3714,7 +3714,7 @@ static int mvneta_probe(struct platform_device *pdev) mvneta_fixed_link_update(pp, phy); - put_device(&phy->dev); + put_device(&phy->mdio.dev); } return 0; diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index c74e78dd989a..8af25563f627 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -864,8 +864,8 @@ static int smsc911x_phy_loopbacktest(struct net_device *dev) for (i = 0; i < 10; i++) { /* Set PHY to 10/FD, no ANEG, and loopback mode */ - smsc911x_mii_write(phy_dev->bus, phy_dev->addr, MII_BMCR, - BMCR_LOOPBACK | BMCR_FULLDPLX); + smsc911x_mii_write(phy_dev->mdio.bus, phy_dev->mdio.addr, + MII_BMCR, BMCR_LOOPBACK | BMCR_FULLDPLX); /* Enable MAC tx/rx, FD */ spin_lock_irqsave(&pdata->mac_lock, flags); @@ -893,7 +893,7 @@ static int smsc911x_phy_loopbacktest(struct net_device *dev) spin_unlock_irqrestore(&pdata->mac_lock, flags); /* Cancel PHY loopback mode */ - smsc911x_mii_write(phy_dev->bus, phy_dev->addr, MII_BMCR, 0); + smsc911x_mii_write(phy_dev->mdio.bus, phy_dev->mdio.addr, MII_BMCR, 0); smsc911x_reg_write(pdata, TX_CFG, 0); smsc911x_reg_write(pdata, RX_CFG, 0); @@ -1021,7 +1021,7 @@ static int smsc911x_mii_probe(struct net_device *dev) } SMSC_TRACE(pdata, probe, "PHY: addr %d, phy_id 0x%08X", - phydev->addr, phydev->phy_id); + phydev->mdio.addr, phydev->phy_id); ret = phy_connect_direct(dev, phydev, &smsc911x_phy_adjust_link, pdata->config.phy_interface); @@ -1988,7 +1988,8 @@ smsc911x_ethtool_getregs(struct net_device *dev, struct ethtool_regs *regs, } for (i = 0; i <= 31; i++) - data[j++] = smsc911x_mii_read(phy_dev->bus, phy_dev->addr, i); + data[j++] = smsc911x_mii_read(phy_dev->mdio.bus, + phy_dev->mdio.addr, i); } static void smsc911x_eeprom_enable_access(struct smsc911x_data *pdata) diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 59bf4c353d50..53355c323f54 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -315,7 +315,8 @@ smsc9420_ethtool_getregs(struct net_device *dev, struct ethtool_regs *regs, return; for (i = 0; i <= 31; i++) - data[j++] = smsc9420_mii_read(phy_dev->bus, phy_dev->addr, i); + data[j++] = smsc9420_mii_read(phy_dev->mdio.bus, + phy_dev->mdio.addr, i); } static void smsc9420_eeprom_enable_access(struct smsc9420_pdata *pd) diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 49544c0fa6a7..42fdfd4d9d4f 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -2050,7 +2050,8 @@ static int cpsw_probe_dt(struct cpsw_priv *priv, if (!phy_dev) return -ENODEV; snprintf(slave_data->phy_id, sizeof(slave_data->phy_id), - PHY_ID_FMT, phy_dev->bus->id, phy_dev->addr); + PHY_ID_FMT, phy_dev->mdio.bus->id, + phy_dev->mdio.addr); } else if (parp) { u32 phyid; struct device_node *mdio_node; diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 88e8e6055b9f..78299c1592c1 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -396,7 +396,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) phy = data->bus->phy_map[addr]; if (phy) { dev_info(dev, "phy[%d]: device %s, driver %s\n", - phy->addr, phydev_name(phy), + phy->mdio.addr, phydev_name(phy), phy->drv ? phy->drv->name : "unknown"); } } diff --git a/drivers/net/ethernet/xilinx/xilinx_emaclite.c b/drivers/net/ethernet/xilinx/xilinx_emaclite.c index d1a0167dff84..e324b3092380 100644 --- a/drivers/net/ethernet/xilinx/xilinx_emaclite.c +++ b/drivers/net/ethernet/xilinx/xilinx_emaclite.c @@ -827,7 +827,7 @@ static int xemaclite_mdio_setup(struct net_local *lp, struct device *dev) dev_info(dev, "MDIO of the phy is not registered yet\n"); else - put_device(&phydev->dev); + put_device(&phydev->mdio.dev); return 0; } diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c index 62361f8af375..b76ac09a554f 100644 --- a/drivers/net/phy/at803x.c +++ b/drivers/net/phy/at803x.c @@ -190,7 +190,7 @@ static int at803x_resume(struct phy_device *phydev) static int at803x_probe(struct phy_device *phydev) { - struct device *dev = &phydev->dev; + struct device *dev = &phydev->mdio.dev; struct at803x_priv *priv; struct gpio_desc *gpiod_reset; diff --git a/drivers/net/phy/bcm87xx.c b/drivers/net/phy/bcm87xx.c index 71b491c7bf96..e536e30d1643 100644 --- a/drivers/net/phy/bcm87xx.c +++ b/drivers/net/phy/bcm87xx.c @@ -40,10 +40,10 @@ static int bcm87xx_of_reg_init(struct phy_device *phydev) const __be32 *paddr_end; int len, ret; - if (!phydev->dev.of_node) + if (!phydev->mdio.dev.of_node) return 0; - paddr = of_get_property(phydev->dev.of_node, + paddr = of_get_property(phydev->mdio.dev.of_node, "broadcom,c45-reg-init", &len); if (!paddr) return 0; diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 47b711739ba9..39da6fc6a85e 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -220,9 +220,10 @@ static void rx_timestamp_work(struct work_struct *work); #define BROADCAST_ADDR 31 -static inline int broadcast_write(struct mii_bus *bus, u32 regnum, u16 val) +static inline int broadcast_write(struct phy_device *phydev, u32 regnum, + u16 val) { - return mdiobus_write(bus, BROADCAST_ADDR, regnum, val); + return mdiobus_write(phydev->mdio.bus, BROADCAST_ADDR, regnum, val); } /* Caller must hold extreg_lock. */ @@ -232,7 +233,7 @@ static int ext_read(struct phy_device *phydev, int page, u32 regnum) int val; if (dp83640->clock->page != page) { - broadcast_write(phydev->bus, PAGESEL, page); + broadcast_write(phydev, PAGESEL, page); dp83640->clock->page = page; } val = phy_read(phydev, regnum); @@ -247,11 +248,11 @@ static void ext_write(int broadcast, struct phy_device *phydev, struct dp83640_private *dp83640 = phydev->priv; if (dp83640->clock->page != page) { - broadcast_write(phydev->bus, PAGESEL, page); + broadcast_write(phydev, PAGESEL, page); dp83640->clock->page = page; } if (broadcast) - broadcast_write(phydev->bus, regnum, val); + broadcast_write(phydev, regnum, val); else phy_write(phydev, regnum, val); } @@ -1039,7 +1040,7 @@ static int choose_this_phy(struct dp83640_clock *clock, if (chosen_phy == -1 && !clock->chosen) return 1; - if (chosen_phy == phydev->addr) + if (chosen_phy == phydev->mdio.addr) return 1; return 0; @@ -1103,10 +1104,10 @@ static int dp83640_probe(struct phy_device *phydev) struct dp83640_private *dp83640; int err = -ENOMEM, i; - if (phydev->addr == BROADCAST_ADDR) + if (phydev->mdio.addr == BROADCAST_ADDR) return 0; - clock = dp83640_clock_get_bus(phydev->bus); + clock = dp83640_clock_get_bus(phydev->mdio.bus); if (!clock) goto no_clock; @@ -1132,7 +1133,8 @@ static int dp83640_probe(struct phy_device *phydev) if (choose_this_phy(clock, phydev)) { clock->chosen = dp83640; - clock->ptp_clock = ptp_clock_register(&clock->caps, &phydev->dev); + clock->ptp_clock = ptp_clock_register(&clock->caps, + &phydev->mdio.dev); if (IS_ERR(clock->ptp_clock)) { err = PTR_ERR(clock->ptp_clock); goto no_register; @@ -1158,7 +1160,7 @@ static void dp83640_remove(struct phy_device *phydev) struct list_head *this, *next; struct dp83640_private *tmp, *dp83640 = phydev->priv; - if (phydev->addr == BROADCAST_ADDR) + if (phydev->mdio.addr == BROADCAST_ADDR) return; enable_status_frames(phydev, false); diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index e4c0b0c0af02..74e4521bd2d3 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -103,7 +103,7 @@ static int dp83867_config_intr(struct phy_device *phydev) static int dp83867_of_init(struct phy_device *phydev) { struct dp83867_private *dp83867 = phydev->priv; - struct device *dev = &phydev->dev; + struct device *dev = &phydev->mdio.dev; struct device_node *of_node = dev->of_node; int ret; @@ -137,7 +137,7 @@ static int dp83867_config_init(struct phy_device *phydev) u16 val, delay; if (!phydev->priv) { - dp83867 = devm_kzalloc(&phydev->dev, sizeof(*dp83867), + dp83867 = devm_kzalloc(&phydev->mdio.dev, sizeof(*dp83867), GFP_KERNEL); if (!dp83867) return -ENOMEM; diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 0a1cde6803b0..ab9c473d75ea 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -197,11 +197,11 @@ int fixed_phy_set_link_update(struct phy_device *phydev, struct fixed_mdio_bus *fmb = &platform_fmb; struct fixed_phy *fp; - if (!phydev || !phydev->bus) + if (!phydev || !phydev->mdio.bus) return -EINVAL; list_for_each_entry(fp, &fmb->phys, node) { - if (fp->addr == phydev->addr) { + if (fp->addr == phydev->mdio.addr) { fp->link_update = link_update; fp->phydev = phydev; return 0; @@ -219,11 +219,11 @@ int fixed_phy_update_state(struct phy_device *phydev, struct fixed_mdio_bus *fmb = &platform_fmb; struct fixed_phy *fp; - if (!phydev || phydev->bus != fmb->mii_bus) + if (!phydev || phydev->mdio.bus != fmb->mii_bus) return -EINVAL; list_for_each_entry(fp, &fmb->phys, node) { - if (fp->addr == phydev->addr) { + if (fp->addr == phydev->mdio.addr) { #define _UPD(x) if (changed->x) \ fp->status.x = status->x _UPD(link); @@ -344,7 +344,7 @@ struct phy_device *fixed_phy_register(unsigned int irq, } of_node_get(np); - phy->dev.of_node = np; + phy->mdio.dev.of_node = np; phy->is_pseudo_fixed_link = true; switch (status->speed) { diff --git a/drivers/net/phy/icplus.c b/drivers/net/phy/icplus.c index 0dbc445a5fa0..c12170d07b62 100644 --- a/drivers/net/phy/icplus.c +++ b/drivers/net/phy/icplus.c @@ -53,43 +53,43 @@ static int ip175c_config_init(struct phy_device *phydev) if (full_reset_performed == 0) { /* master reset */ - err = mdiobus_write(phydev->bus, 30, 0, 0x175c); + err = mdiobus_write(phydev->mdio.bus, 30, 0, 0x175c); if (err < 0) return err; /* ensure no bus delays overlap reset period */ - err = mdiobus_read(phydev->bus, 30, 0); + err = mdiobus_read(phydev->mdio.bus, 30, 0); /* data sheet specifies reset period is 2 msec */ mdelay(2); /* enable IP175C mode */ - err = mdiobus_write(phydev->bus, 29, 31, 0x175c); + err = mdiobus_write(phydev->mdio.bus, 29, 31, 0x175c); if (err < 0) return err; /* Set MII0 speed and duplex (in PHY mode) */ - err = mdiobus_write(phydev->bus, 29, 22, 0x420); + err = mdiobus_write(phydev->mdio.bus, 29, 22, 0x420); if (err < 0) return err; /* reset switch ports */ for (i = 0; i < 5; i++) { - err = mdiobus_write(phydev->bus, i, + err = mdiobus_write(phydev->mdio.bus, i, MII_BMCR, BMCR_RESET); if (err < 0) return err; } for (i = 0; i < 5; i++) - err = mdiobus_read(phydev->bus, i, MII_BMCR); + err = mdiobus_read(phydev->mdio.bus, i, MII_BMCR); mdelay(2); full_reset_performed = 1; } - if (phydev->addr != 4) { + if (phydev->mdio.addr != 4) { phydev->state = PHY_RUNNING; phydev->speed = SPEED_100; phydev->duplex = DUPLEX_FULL; @@ -184,7 +184,7 @@ static int ip101a_g_config_init(struct phy_device *phydev) static int ip175c_read_status(struct phy_device *phydev) { - if (phydev->addr == 4) /* WAN port */ + if (phydev->mdio.addr == 4) /* WAN port */ genphy_read_status(phydev); else /* Don't need to read status for switch ports */ @@ -195,7 +195,7 @@ static int ip175c_read_status(struct phy_device *phydev) static int ip175c_config_aneg(struct phy_device *phydev) { - if (phydev->addr == 4) /* WAN port */ + if (phydev->mdio.addr == 4) /* WAN port */ genphy_config_aneg(phydev); return 0; diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 50b5eac75854..f96c93c9819a 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -300,10 +300,11 @@ static int marvell_of_reg_init(struct phy_device *phydev) const __be32 *paddr; int len, i, saved_page, current_page, page_changed, ret; - if (!phydev->dev.of_node) + if (!phydev->mdio.dev.of_node) return 0; - paddr = of_get_property(phydev->dev.of_node, "marvell,reg-init", &len); + paddr = of_get_property(phydev->mdio.dev.of_node, + "marvell,reg-init", &len); if (!paddr || len < (4 * sizeof(*paddr))) return 0; @@ -1060,7 +1061,7 @@ static int marvell_probe(struct phy_device *phydev) { struct marvell_priv *priv; - priv = devm_kzalloc(&phydev->dev, sizeof(*priv), GFP_KERNEL); + priv = devm_kzalloc(&phydev->mdio.dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 05381d0f559c..e5b1ccde835b 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -200,16 +200,16 @@ EXPORT_SYMBOL(of_mdio_find_bus); * the phy. This allows auto-probed pyh devices to be supplied with information * passed in via DT. */ -static void of_mdiobus_link_phydev(struct mii_bus *mdio, +static void of_mdiobus_link_phydev(struct mii_bus *bus, struct phy_device *phydev) { - struct device *dev = &phydev->dev; + struct device *dev = &phydev->mdio.dev; struct device_node *child; - if (dev->of_node || !mdio->dev.of_node) + if (dev->of_node || !bus->dev.of_node) return; - for_each_available_child_of_node(mdio->dev.of_node, child) { + for_each_available_child_of_node(bus->dev.of_node, child) { int addr; int ret; @@ -227,7 +227,7 @@ static void of_mdiobus_link_phydev(struct mii_bus *mdio, continue; } - if (addr == phydev->addr) { + if (addr == phydev->mdio.addr) { dev->of_node = child; return; } @@ -522,7 +522,7 @@ static int mdio_bus_match(struct device *dev, struct device_driver *drv) static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) { - struct device_driver *drv = phydev->dev.driver; + struct device_driver *drv = phydev->mdio.dev.driver; struct phy_driver *phydrv = to_phy_driver(drv); struct net_device *netdev = phydev->attached_dev; diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index bf72365e90bc..b51505be1fa9 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -350,7 +350,7 @@ static int ksz9021_load_values_from_of(struct phy_device *phydev, static int ksz9021_config_init(struct phy_device *phydev) { - const struct device *dev = &phydev->dev; + const struct device *dev = &phydev->mdio.dev; const struct device_node *of_node = dev->of_node; const struct device *dev_walker; @@ -358,7 +358,7 @@ static int ksz9021_config_init(struct phy_device *phydev) * properties in the MAC node. Walk up the tree of devices to * find a device with an OF node. */ - dev_walker = &phydev->dev; + dev_walker = &phydev->mdio.dev; do { of_node = dev_walker->of_node; dev_walker = dev_walker->parent; @@ -471,7 +471,7 @@ static int ksz9031_center_flp_timing(struct phy_device *phydev) static int ksz9031_config_init(struct phy_device *phydev) { - const struct device *dev = &phydev->dev; + const struct device *dev = &phydev->mdio.dev; const struct device_node *of_node = dev->of_node; static const char *clk_skews[2] = {"rxc-skew-ps", "txc-skew-ps"}; static const char *rx_data_skews[4] = { @@ -630,12 +630,12 @@ static void kszphy_get_stats(struct phy_device *phydev, static int kszphy_probe(struct phy_device *phydev) { const struct kszphy_type *type = phydev->drv->driver_data; - const struct device_node *np = phydev->dev.of_node; + const struct device_node *np = phydev->mdio.dev.of_node; struct kszphy_priv *priv; struct clk *clk; int ret; - priv = devm_kzalloc(&phydev->dev, sizeof(*priv), GFP_KERNEL); + priv = devm_kzalloc(&phydev->mdio.dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; @@ -658,7 +658,7 @@ static int kszphy_probe(struct phy_device *phydev) priv->led_mode = -1; } - clk = devm_clk_get(&phydev->dev, "rmii-ref"); + clk = devm_clk_get(&phydev->mdio.dev, "rmii-ref"); /* NOTE: clk may be NULL if building without CONFIG_HAVE_CLK */ if (!IS_ERR_OR_NULL(clk)) { unsigned long rate = clk_get_rate(clk); diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c index 99df5bc47424..5e34b49be0b3 100644 --- a/drivers/net/phy/microchip.c +++ b/drivers/net/phy/microchip.c @@ -68,7 +68,7 @@ int lan88xx_suspend(struct phy_device *phydev) static int lan88xx_probe(struct phy_device *phydev) { - struct device *dev = &phydev->dev; + struct device *dev = &phydev->mdio.dev; struct lan88xx_priv *priv; priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); @@ -89,7 +89,7 @@ static int lan88xx_probe(struct phy_device *phydev) static void lan88xx_remove(struct phy_device *phydev) { - struct device *dev = &phydev->dev; + struct device *dev = &phydev->mdio.dev; struct lan88xx_priv *priv = phydev->priv; if (priv) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 56c8dd8c0c85..8763bb20988a 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -319,7 +319,7 @@ int phy_ethtool_sset(struct phy_device *phydev, struct ethtool_cmd *cmd) { u32 speed = ethtool_cmd_speed(cmd); - if (cmd->phy_address != phydev->addr) + if (cmd->phy_address != phydev->mdio.addr) return -EINVAL; /* We make sure that we don't pass unsupported values in to the PHY */ @@ -375,7 +375,7 @@ int phy_ethtool_gset(struct phy_device *phydev, struct ethtool_cmd *cmd) cmd->port = PORT_BNC; else cmd->port = PORT_MII; - cmd->phy_address = phydev->addr; + cmd->phy_address = phydev->mdio.addr; cmd->transceiver = phy_is_internal(phydev) ? XCVR_INTERNAL : XCVR_EXTERNAL; cmd->autoneg = phydev->autoneg; @@ -403,16 +403,17 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd) switch (cmd) { case SIOCGMIIPHY: - mii_data->phy_id = phydev->addr; + mii_data->phy_id = phydev->mdio.addr; /* fall through */ case SIOCGMIIREG: - mii_data->val_out = mdiobus_read(phydev->bus, mii_data->phy_id, + mii_data->val_out = mdiobus_read(phydev->mdio.bus, + mii_data->phy_id, mii_data->reg_num); return 0; case SIOCSMIIREG: - if (mii_data->phy_id == phydev->addr) { + if (mii_data->phy_id == phydev->mdio.addr) { switch (mii_data->reg_num) { case MII_BMCR: if ((val & (BMCR_RESET | BMCR_ANENABLE)) == 0) { @@ -445,10 +446,10 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd) } } - mdiobus_write(phydev->bus, mii_data->phy_id, + mdiobus_write(phydev->mdio.bus, mii_data->phy_id, mii_data->reg_num, val); - if (mii_data->phy_id == phydev->addr && + if (mii_data->phy_id == phydev->mdio.addr && mii_data->reg_num == MII_BMCR && val & BMCR_RESET) return phy_init_hw(phydev); @@ -643,7 +644,7 @@ int phy_start_interrupts(struct phy_device *phydev) if (request_irq(phydev->irq, phy_interrupt, 0, "phy_interrupt", phydev) < 0) { pr_warn("%s: Can't get IRQ %d (PHY)\n", - phydev->bus->name, phydev->irq); + phydev->mdio.bus->name, phydev->irq); phydev->irq = PHY_POLL; return 0; } @@ -1041,11 +1042,11 @@ static inline void mmd_phy_indirect(struct mii_bus *bus, int prtad, int devad, int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, int devad) { struct phy_driver *phydrv = phydev->drv; - int addr = phydev->addr; + int addr = phydev->mdio.addr; int value = -1; if (!phydrv->read_mmd_indirect) { - struct mii_bus *bus = phydev->bus; + struct mii_bus *bus = phydev->mdio.bus; mutex_lock(&bus->mdio_lock); mmd_phy_indirect(bus, prtad, devad, addr); @@ -1079,10 +1080,10 @@ void phy_write_mmd_indirect(struct phy_device *phydev, int prtad, int devad, u32 data) { struct phy_driver *phydrv = phydev->drv; - int addr = phydev->addr; + int addr = phydev->mdio.addr; if (!phydrv->write_mmd_indirect) { - struct mii_bus *bus = phydev->bus; + struct mii_bus *bus = phydev->mdio.bus; mutex_lock(&bus->mdio_lock); mmd_phy_indirect(bus, prtad, devad, addr); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 68fe5738daef..01e5d52dc37c 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -43,7 +43,7 @@ MODULE_LICENSE("GPL"); void phy_device_free(struct phy_device *phydev) { - put_device(&phydev->dev); + put_device(&phydev->mdio.dev); } EXPORT_SYMBOL(phy_device_free); @@ -65,7 +65,7 @@ static DEFINE_MUTEX(phy_fixup_lock); /** * phy_register_fixup - creates a new phy_fixup and adds it to the list - * @bus_id: A string which matches phydev->dev.bus_id (or PHY_ANY_ID) + * @bus_id: A string which matches phydev->mdio.dev.bus_id (or PHY_ANY_ID) * @phy_uid: Used to match against phydev->phy_id (the UID of the PHY) * It can also be PHY_ANY_UID * @phy_uid_mask: Applied to phydev->phy_id and fixup->phy_uid before @@ -153,13 +153,19 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, struct phy_c45_device_ids *c45_ids) { struct phy_device *dev; + struct mdio_device *mdiodev; /* We allocate the device, and initialize the default values */ dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return ERR_PTR(-ENOMEM); - dev->dev.release = phy_device_release; + mdiodev = &dev->mdio; + mdiodev->dev.release = phy_device_release; + mdiodev->dev.parent = &bus->dev; + mdiodev->dev.bus = &mdio_bus_type; + mdiodev->bus = bus; + mdiodev->addr = addr; dev->speed = 0; dev->duplex = -1; @@ -171,15 +177,11 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, dev->autoneg = AUTONEG_ENABLE; dev->is_c45 = is_c45; - dev->addr = addr; dev->phy_id = phy_id; if (c45_ids) dev->c45_ids = *c45_ids; - dev->bus = bus; - dev->dev.parent = &bus->dev; - dev->dev.bus = &mdio_bus_type; dev->irq = bus->irq ? bus->irq[addr] : PHY_POLL; - dev_set_name(&dev->dev, PHY_ID_FMT, bus->id, addr); + dev_set_name(&mdiodev->dev, PHY_ID_FMT, bus->id, addr); dev->state = PHY_DOWN; @@ -199,7 +201,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, */ request_module(MDIO_MODULE_PREFIX MDIO_ID_FMT, MDIO_ID_ARGS(phy_id)); - device_initialize(&dev->dev); + device_initialize(&mdiodev->dev); return dev; } @@ -382,27 +384,27 @@ int phy_device_register(struct phy_device *phydev) int err; /* Don't register a phy if one is already registered at this address */ - if (phydev->bus->phy_map[phydev->addr]) + if (phydev->mdio.bus->phy_map[phydev->mdio.addr]) return -EINVAL; - phydev->bus->phy_map[phydev->addr] = phydev; + phydev->mdio.bus->phy_map[phydev->mdio.addr] = phydev; /* Run all of the fixups for this PHY */ err = phy_scan_fixups(phydev); if (err) { - pr_err("PHY %d failed to initialize\n", phydev->addr); + pr_err("PHY %d failed to initialize\n", phydev->mdio.addr); goto out; } - err = device_add(&phydev->dev); + err = device_add(&phydev->mdio.dev); if (err) { - pr_err("PHY %d failed to add\n", phydev->addr); + pr_err("PHY %d failed to add\n", phydev->mdio.addr); goto out; } return 0; out: - phydev->bus->phy_map[phydev->addr] = NULL; + phydev->mdio.bus->phy_map[phydev->mdio.addr] = NULL; return err; } EXPORT_SYMBOL(phy_device_register); @@ -417,10 +419,10 @@ EXPORT_SYMBOL(phy_device_register); */ void phy_device_remove(struct phy_device *phydev) { - struct mii_bus *bus = phydev->bus; - int addr = phydev->addr; + struct mii_bus *bus = phydev->mdio.bus; + int addr = phydev->mdio.addr; - device_del(&phydev->dev); + device_del(&phydev->mdio.dev); bus->phy_map[addr] = NULL; } EXPORT_SYMBOL(phy_device_remove); @@ -617,13 +619,13 @@ EXPORT_SYMBOL(phy_attached_info); void phy_attached_print(struct phy_device *phydev, const char *fmt, ...) { if (!fmt) { - dev_info(&phydev->dev, ATTACHED_FMT "\n", + dev_info(&phydev->mdio.dev, ATTACHED_FMT "\n", phydev->drv->name, phydev_name(phydev), phydev->irq); } else { va_list ap; - dev_info(&phydev->dev, ATTACHED_FMT, + dev_info(&phydev->mdio.dev, ATTACHED_FMT, phydev->drv->name, phydev_name(phydev), phydev->irq); @@ -652,8 +654,8 @@ EXPORT_SYMBOL(phy_attached_print); int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface) { - struct mii_bus *bus = phydev->bus; - struct device *d = &phydev->dev; + struct mii_bus *bus = phydev->mdio.bus; + struct device *d = &phydev->mdio.dev; int err; if (!try_module_get(bus->owner)) { @@ -771,8 +773,8 @@ void phy_detach(struct phy_device *phydev) * real driver could be loaded */ for (i = 0; i < ARRAY_SIZE(genphy_driver); i++) { - if (phydev->dev.driver == &genphy_driver[i].driver) { - device_release_driver(&phydev->dev); + if (phydev->mdio.dev.driver == &genphy_driver[i].driver) { + device_release_driver(&phydev->mdio.dev); break; } } @@ -781,16 +783,16 @@ void phy_detach(struct phy_device *phydev) * The phydev might go away on the put_device() below, so avoid * a use-after-free bug by reading the underlying bus first. */ - bus = phydev->bus; + bus = phydev->mdio.bus; - put_device(&phydev->dev); + put_device(&phydev->mdio.dev); module_put(bus->owner); } EXPORT_SYMBOL(phy_detach); int phy_suspend(struct phy_device *phydev) { - struct phy_driver *phydrv = to_phy_driver(phydev->dev.driver); + struct phy_driver *phydrv = to_phy_driver(phydev->mdio.dev.driver); struct ethtool_wolinfo wol = { .cmd = ETHTOOL_GWOL }; int ret = 0; @@ -813,7 +815,7 @@ EXPORT_SYMBOL(phy_suspend); int phy_resume(struct phy_device *phydev) { - struct phy_driver *phydrv = to_phy_driver(phydev->dev.driver); + struct phy_driver *phydrv = to_phy_driver(phydev->mdio.dev.driver); int ret = 0; if (phydrv->resume) @@ -1330,7 +1332,7 @@ EXPORT_SYMBOL(phy_set_max_speed); static void of_set_phy_supported(struct phy_device *phydev) { - struct device_node *node = phydev->dev.of_node; + struct device_node *node = phydev->mdio.dev.of_node; u32 max_speed; if (!IS_ENABLED(CONFIG_OF_MDIO)) @@ -1354,7 +1356,7 @@ static void of_set_phy_supported(struct phy_device *phydev) static int phy_probe(struct device *dev) { struct phy_device *phydev = to_phy_device(dev); - struct device_driver *drv = phydev->dev.driver; + struct device_driver *drv = phydev->mdio.dev.driver; struct phy_driver *phydrv = to_phy_driver(drv); int err = 0; diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c index dc2da8770918..18c981b95910 100644 --- a/drivers/net/phy/smsc.c +++ b/drivers/net/phy/smsc.c @@ -44,7 +44,7 @@ static int smsc_phy_ack_interrupt(struct phy_device *phydev) static int smsc_phy_config_init(struct phy_device *phydev) { int __maybe_unused len; - struct device *dev __maybe_unused = &phydev->dev; + struct device *dev __maybe_unused = &phydev->mdio.dev; struct device_node *of_node __maybe_unused = dev->of_node; int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS); int enable_energy = 1; diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 2f88ff4654da..bc9d76329435 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -75,7 +75,7 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi /* Associate the OF node with the device structure so it * can be looked up later */ of_node_get(child); - phy->dev.of_node = child; + phy->mdio.dev.of_node = child; /* All data is now stored in the phy struct; * register it */ @@ -233,7 +233,7 @@ struct phy_device *of_phy_connect(struct net_device *dev, ret = phy_connect_direct(dev, phy, hndlr, iface); /* refcount is held by phy_connect_direct() on success */ - put_device(&phy->dev); + put_device(&phy->mdio.dev); return ret ? NULL : phy; } @@ -263,7 +263,7 @@ struct phy_device *of_phy_attach(struct net_device *dev, ret = phy_attach_direct(dev, phy, flags, iface); /* refcount is held by phy_attach_direct() on success */ - put_device(&phy->dev); + put_device(&phy->mdio.dev); return ret ? NULL : phy; } diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 0d073c23c10d..94f9f1491cde 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -13,6 +13,15 @@ struct mii_bus; +struct mdio_device { + struct device dev; + + struct mii_bus *bus; + /* Bus address of the MDIO device (0-31) */ + int addr; +}; +#define to_mdio_device(d) container_of(d, struct mdio_device, dev) + static inline bool mdio_phy_id_is_c45(int phy_id) { return (phy_id & MDIO_PHY_ID_C45) && !(phy_id & ~MDIO_PHY_ID_C45_MASK); diff --git a/include/linux/phy.h b/include/linux/phy.h index a5473c9e19de..239a0c2bc49d 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -358,14 +358,12 @@ struct phy_c45_device_ids { * handling, as well as handling shifts in PHY hardware state */ struct phy_device { + struct mdio_device mdio; + /* Information about the PHY type */ /* And management functions */ struct phy_driver *drv; - struct mii_bus *bus; - - struct device dev; - u32 phy_id; struct phy_c45_device_ids c45_ids; @@ -381,9 +379,6 @@ struct phy_device { phy_interface_t interface; - /* Bus address of the PHY (0-31) */ - int addr; - /* * forced speed & duplex (no autoneg) * partner speed & duplex & pause (autoneg) @@ -432,7 +427,8 @@ struct phy_device { void (*adjust_link)(struct net_device *dev); }; -#define to_phy_device(d) container_of(d, struct phy_device, dev) +#define to_phy_device(d) container_of(to_mdio_device(d), \ + struct phy_device, mdio) /* struct phy_driver: Driver structure for a particular PHY type * @@ -622,7 +618,7 @@ static inline int phy_read_mmd(struct phy_device *phydev, int devad, u32 regnum) if (!phydev->is_c45) return -EOPNOTSUPP; - return mdiobus_read(phydev->bus, phydev->addr, + return mdiobus_read(phydev->mdio.bus, phydev->mdio.addr, MII_ADDR_C45 | (devad << 16) | (regnum & 0xffff)); } @@ -648,7 +644,7 @@ int phy_read_mmd_indirect(struct phy_device *phydev, int prtad, int devad); */ static inline int phy_read(struct phy_device *phydev, u32 regnum) { - return mdiobus_read(phydev->bus, phydev->addr, regnum); + return mdiobus_read(phydev->mdio.bus, phydev->mdio.addr, regnum); } /** @@ -663,7 +659,7 @@ static inline int phy_read(struct phy_device *phydev, u32 regnum) */ static inline int phy_write(struct phy_device *phydev, u32 regnum, u16 val) { - return mdiobus_write(phydev->bus, phydev->addr, regnum, val); + return mdiobus_write(phydev->mdio.bus, phydev->mdio.addr, regnum, val); } /** @@ -726,7 +722,7 @@ static inline int phy_write_mmd(struct phy_device *phydev, int devad, regnum = MII_ADDR_C45 | ((devad & 0x1f) << 16) | (regnum & 0xffff); - return mdiobus_write(phydev->bus, phydev->addr, regnum, val); + return mdiobus_write(phydev->mdio.bus, phydev->mdio.addr, regnum, val); } /** @@ -776,14 +772,14 @@ static inline int phy_read_status(struct phy_device *phydev) } #define phydev_err(_phydev, format, args...) \ - dev_err(&_phydev->dev, format, ##args) + dev_err(&_phydev->mdio.dev, format, ##args) #define phydev_dbg(_phydev, format, args...) \ - dev_dbg(&_phydev->dev, format, ##args) + dev_dbg(&_phydev->mdio.dev, format, ##args); static inline const char *phydev_name(const struct phy_device *phydev) { - return dev_name(&phydev->dev); + return dev_name(&phydev->mdio.dev); } void phy_attached_print(struct phy_device *phydev, const char *fmt, ...) diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 208d1b257194..fa4daba8db55 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -439,7 +439,7 @@ static void dsa_switch_destroy(struct dsa_switch *ds) if (of_phy_is_fixed_link(port_dn)) { phydev = of_phy_find_device(port_dn); if (phydev) { - int addr = phydev->addr; + int addr = phydev->mdio.addr; phy_device_free(phydev); of_node_put(port_dn); -- cgit v1.2.3-59-g8ed1b From 7f854420fbfe9d49afe2ffb1df052cfe8e215541 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:18 +0100 Subject: phy: Add API for {un}registering an mdio device to a bus. Rather than have drivers directly manipulate the mii_bus structure, provide and API for registering and unregistering devices on an MDIO bus, and performing lookups. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/au1000_eth.c | 9 +++-- drivers/net/ethernet/broadcom/b44.c | 2 +- drivers/net/ethernet/broadcom/genet/bcmmii.c | 2 +- drivers/net/ethernet/broadcom/tg3.c | 30 +++++++-------- drivers/net/ethernet/ethoc.c | 4 +- drivers/net/ethernet/faraday/ftgmac100.c | 2 +- drivers/net/ethernet/freescale/fec_main.c | 7 +--- drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c | 2 +- drivers/net/ethernet/smsc/smsc9420.c | 3 +- drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 2 +- drivers/net/ethernet/ti/davinci_mdio.c | 2 +- drivers/net/phy/mdio_bus.c | 46 ++++++++++++++++++++++- drivers/net/phy/phy_device.c | 21 +++++------ drivers/of/of_mdio.c | 2 +- include/linux/mdio.h | 8 ++++ include/linux/phy.h | 2 +- net/dsa/slave.c | 3 +- 17 files changed, 98 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 982b581d3484..c8640418fc37 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -502,7 +502,7 @@ static int au1000_mii_probe(struct net_device *dev) BUG_ON(aup->mac_id < 0 || aup->mac_id > 1); if (aup->phy_addr) - phydev = aup->mii_bus->phy_map[aup->phy_addr]; + phydev = mdiobus_get_phy(aup->mii_bus, aup->phy_addr); else netdev_info(dev, "using PHY-less setup\n"); return 0; @@ -512,8 +512,8 @@ static int au1000_mii_probe(struct net_device *dev) * on the current MAC's MII bus */ for (phy_addr = 0; phy_addr < PHY_MAX_ADDR; phy_addr++) - if (aup->mii_bus->phy_map[phy_addr]) { - phydev = aup->mii_bus->phy_map[phy_addr]; + if (mdiobus_get_phy(aup->mii_bus, aup->phy_addr)) { + phydev = mdiobus_get_phy(aup->mii_bus, aup->phy_addr); if (!aup->phy_search_highest_addr) /* break out with first one found */ break; @@ -531,7 +531,8 @@ static int au1000_mii_probe(struct net_device *dev) */ for (phy_addr = 0; phy_addr < PHY_MAX_ADDR; phy_addr++) { struct phy_device *const tmp_phydev = - aup->mii_bus->phy_map[phy_addr]; + mdiobus_get_phy(aup->mii_bus, + phy_addr); if (aup->mac_id == 1) break; diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 843a4a5864fc..74f0a37c4eb6 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2272,7 +2272,7 @@ static int b44_register_phy_one(struct b44 *bp) goto err_out_mdiobus; } - if (!bp->mii_bus->phy_map[bp->phy_addr] && + if (!mdiobus_is_registered_device(bp->mii_bus, bp->phy_addr) && (sprom->boardflags_lo & (B44_BOARDFLAG_ROBO | B44_BOARDFLAG_ADM))) { dev_info(sdev->dev, diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 633b59db813a..0d775964b060 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -573,7 +573,7 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv) } if (pd->phy_address >= 0 && pd->phy_address < PHY_MAX_ADDR) - phydev = mdio->phy_map[pd->phy_address]; + phydev = mdiobus_get_phy(mdio, pd->phy_address); else phydev = phy_find_first(mdio); diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 04e7d0d0e5b1..9293675df7ba 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -1406,7 +1406,7 @@ static void tg3_mdio_config_5785(struct tg3 *tp) u32 val; struct phy_device *phydev; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); switch (phydev->drv->phy_id & phydev->drv->phy_id_mask) { case PHY_ID_BCM50610: case PHY_ID_BCM50610M: @@ -1554,7 +1554,7 @@ static int tg3_mdio_init(struct tg3 *tp) return i; } - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); if (!phydev || !phydev->drv) { dev_warn(&tp->pdev->dev, "No PHY devices\n"); @@ -1964,7 +1964,7 @@ static void tg3_setup_flow_control(struct tg3 *tp, u32 lcladv, u32 rmtadv) u32 old_tx_mode = tp->tx_mode; if (tg3_flag(tp, USE_PHYLIB)) - autoneg = tp->mdio_bus->phy_map[tp->phy_addr]->autoneg; + autoneg = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr)->autoneg; else autoneg = tp->link_config.autoneg; @@ -2000,7 +2000,7 @@ static void tg3_adjust_link(struct net_device *dev) u8 oldflowctrl, linkmesg = 0; u32 mac_mode, lcl_adv, rmt_adv; struct tg3 *tp = netdev_priv(dev); - struct phy_device *phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + struct phy_device *phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); spin_lock_bh(&tp->lock); @@ -2089,7 +2089,7 @@ static int tg3_phy_init(struct tg3 *tp) /* Bring the PHY back to a known state. */ tg3_bmcr_reset(tp); - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); /* Attach the MAC to the PHY. */ phydev = phy_connect(tp->dev, phydev_name(phydev), @@ -2116,7 +2116,7 @@ static int tg3_phy_init(struct tg3 *tp) SUPPORTED_Asym_Pause); break; default: - phy_disconnect(tp->mdio_bus->phy_map[tp->phy_addr]); + phy_disconnect(mdiobus_get_phy(tp->mdio_bus, tp->phy_addr)); return -EINVAL; } @@ -2136,7 +2136,7 @@ static void tg3_phy_start(struct tg3 *tp) if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); if (tp->phy_flags & TG3_PHYFLG_IS_LOW_POWER) { tp->phy_flags &= ~TG3_PHYFLG_IS_LOW_POWER; @@ -2156,13 +2156,13 @@ static void tg3_phy_stop(struct tg3 *tp) if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return; - phy_stop(tp->mdio_bus->phy_map[tp->phy_addr]); + phy_stop(mdiobus_get_phy(tp->mdio_bus, tp->phy_addr)); } static void tg3_phy_fini(struct tg3 *tp) { if (tp->phy_flags & TG3_PHYFLG_IS_CONNECTED) { - phy_disconnect(tp->mdio_bus->phy_map[tp->phy_addr]); + phy_disconnect(mdiobus_get_phy(tp->mdio_bus, tp->phy_addr)); tp->phy_flags &= ~TG3_PHYFLG_IS_CONNECTED; } } @@ -4046,7 +4046,7 @@ static int tg3_power_down_prepare(struct tg3 *tp) struct phy_device *phydev; u32 phyid, advertising; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); tp->phy_flags |= TG3_PHYFLG_IS_LOW_POWER; @@ -12074,7 +12074,7 @@ static int tg3_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) struct phy_device *phydev; if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return -EAGAIN; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); return phy_ethtool_gset(phydev, cmd); } @@ -12141,7 +12141,7 @@ static int tg3_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) struct phy_device *phydev; if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return -EAGAIN; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); return phy_ethtool_sset(phydev, cmd); } @@ -12296,7 +12296,7 @@ static int tg3_nway_reset(struct net_device *dev) if (tg3_flag(tp, USE_PHYLIB)) { if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return -EAGAIN; - r = phy_start_aneg(tp->mdio_bus->phy_map[tp->phy_addr]); + r = phy_start_aneg(mdiobus_get_phy(tp->mdio_bus, tp->phy_addr)); } else { u32 bmcr; @@ -12414,7 +12414,7 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam u32 newadv; struct phy_device *phydev; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); if (!(phydev->supported & SUPPORTED_Pause) || (!(phydev->supported & SUPPORTED_Asym_Pause) && @@ -13924,7 +13924,7 @@ static int tg3_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) struct phy_device *phydev; if (!(tp->phy_flags & TG3_PHYFLG_IS_CONNECTED)) return -EAGAIN; - phydev = tp->mdio_bus->phy_map[tp->phy_addr]; + phydev = mdiobus_get_phy(tp->mdio_bus, tp->phy_addr); return phy_mii_ioctl(phydev, ifr, cmd); } diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index c028b299ab3f..62fa136554ac 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -678,7 +678,7 @@ static int ethoc_mdio_probe(struct net_device *dev) int err; if (priv->phy_id != -1) - phy = priv->mdio->phy_map[priv->phy_id]; + phy = mdiobus_get_phy(priv->mdio, priv->phy_id); else phy = phy_find_first(priv->mdio); @@ -766,7 +766,7 @@ static int ethoc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) if (mdio->phy_id >= PHY_MAX_ADDR) return -ERANGE; - phy = priv->mdio->phy_map[mdio->phy_id]; + phy = mdiobus_get_phy(priv->mdio, mdio->phy_id); if (!phy) return -ENODEV; } else { diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 8f3f2cf0dcbf..bb116ad646f6 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -839,7 +839,7 @@ static int ftgmac100_mii_probe(struct ftgmac100 *priv) /* search for connect PHY device */ for (i = 0; i < PHY_MAX_ADDR; i++) { - struct phy_device *tmp = priv->mii_bus->phy_map[i]; + struct phy_device *tmp = mdiobus_get_phy(priv->mii_bus, i); if (tmp) { phydev = tmp; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index da255fb4f1d5..502da6f48f95 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include #include @@ -1926,11 +1927,7 @@ static int fec_enet_mii_probe(struct net_device *ndev) } else { /* check for attached phy */ for (phy_id = 0; (phy_id < PHY_MAX_ADDR); phy_id++) { - if ((fep->mii_bus->phy_mask & (1 << phy_id))) - continue; - if (fep->mii_bus->phy_map[phy_id] == NULL) - continue; - if (fep->mii_bus->phy_map[phy_id]->phy_id == 0) + if (!mdiobus_is_registered_device(fep->mii_bus, phy_id)) continue; if (dev_id--) continue; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c index 5b13b8c11bef..467ff7033606 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_mdio.c @@ -180,7 +180,7 @@ int sxgbe_mdio_register(struct net_device *ndev) } for (phy_addr = 0; phy_addr < PHY_MAX_ADDR; phy_addr++) { - struct phy_device *phy = mdio_bus->phy_map[phy_addr]; + struct phy_device *phy = mdiobus_get_phy(mdio_bus, phy_addr); if (phy) { char irq_num[4]; diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 53355c323f54..8594b9e8b28b 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -1158,7 +1158,8 @@ static int smsc9420_mii_probe(struct net_device *dev) BUG_ON(pd->phy_dev); /* Device only supports internal PHY at address 1 */ - if (!pd->mii_bus->phy_map[1]) { + phydev = mdiobus_get_phy(pd->mii_bus, 1); + if (!phydev) { netdev_err(dev, "no PHY found at address 1\n"); return -ENODEV; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index f0990eb9460f..bff28595b427 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -252,7 +252,7 @@ int stmmac_mdio_register(struct net_device *ndev) found = 0; for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - struct phy_device *phydev = new_bus->phy_map[addr]; + struct phy_device *phydev = mdiobus_get_phy(new_bus, addr); if (phydev) { int act = 0; char irq_num[4]; diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 78299c1592c1..4e7c9b9b042a 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -393,7 +393,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) /* scan and dump the bus */ for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - phy = data->bus->phy_map[addr]; + phy = mdiobus_get_phy(data->bus, addr); if (phy) { dev_info(dev, "phy[%d]: device %s, driver %s\n", phy->mdio.addr, phydev_name(phy), diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index e5b1ccde835b..f28f89e109ba 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -38,6 +38,48 @@ #include +int mdiobus_register_device(struct mdio_device *mdiodev) +{ + if (mdiodev->bus->mdio_map[mdiodev->addr]) + return -EBUSY; + + mdiodev->bus->mdio_map[mdiodev->addr] = mdiodev; + + return 0; +} +EXPORT_SYMBOL(mdiobus_register_device); + +int mdiobus_unregister_device(struct mdio_device *mdiodev) +{ + if (mdiodev->bus->mdio_map[mdiodev->addr] != mdiodev) + return -EINVAL; + + mdiodev->bus->mdio_map[mdiodev->addr] = NULL; + + return 0; +} +EXPORT_SYMBOL(mdiobus_unregister_device); + +struct phy_device *mdiobus_get_phy(struct mii_bus *bus, int addr) +{ + struct mdio_device *mdiodev = bus->mdio_map[addr]; + + if (!mdiodev) + return NULL; + + if (!(mdiodev->flags & MDIO_DEVICE_FLAG_PHY)) + return NULL; + + return container_of(mdiodev, struct phy_device, mdio); +} +EXPORT_SYMBOL(mdiobus_get_phy); + +bool mdiobus_is_registered_device(struct mii_bus *bus, int addr) +{ + return bus->mdio_map[addr]; +} +EXPORT_SYMBOL(mdiobus_is_registered_device); + /** * mdiobus_alloc_size - allocate a mii_bus structure * @size: extra amount of memory to allocate for private storage. @@ -299,7 +341,7 @@ int __mdiobus_register(struct mii_bus *bus, struct module *owner) error: while (--i >= 0) { - struct phy_device *phydev = bus->phy_map[i]; + struct phy_device *phydev = mdiobus_get_phy(bus, i); if (phydev) { phy_device_remove(phydev); phy_device_free(phydev); @@ -318,7 +360,7 @@ void mdiobus_unregister(struct mii_bus *bus) bus->state = MDIOBUS_UNREGISTERED; for (i = 0; i < PHY_MAX_ADDR; i++) { - struct phy_device *phydev = bus->phy_map[i]; + struct phy_device *phydev = mdiobus_get_phy(bus, i); if (phydev) { phy_device_remove(phydev); phy_device_free(phydev); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 01e5d52dc37c..e0d5dbb96700 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -166,6 +166,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, mdiodev->dev.bus = &mdio_bus_type; mdiodev->bus = bus; mdiodev->addr = addr; + mdiodev->flags = MDIO_DEVICE_FLAG_PHY; dev->speed = 0; dev->duplex = -1; @@ -383,10 +384,9 @@ int phy_device_register(struct phy_device *phydev) { int err; - /* Don't register a phy if one is already registered at this address */ - if (phydev->mdio.bus->phy_map[phydev->mdio.addr]) - return -EINVAL; - phydev->mdio.bus->phy_map[phydev->mdio.addr] = phydev; + err = mdiobus_register_device(&phydev->mdio); + if (err) + return err; /* Run all of the fixups for this PHY */ err = phy_scan_fixups(phydev); @@ -404,7 +404,7 @@ int phy_device_register(struct phy_device *phydev) return 0; out: - phydev->mdio.bus->phy_map[phydev->mdio.addr] = NULL; + mdiobus_unregister_device(&phydev->mdio); return err; } EXPORT_SYMBOL(phy_device_register); @@ -419,11 +419,8 @@ EXPORT_SYMBOL(phy_device_register); */ void phy_device_remove(struct phy_device *phydev) { - struct mii_bus *bus = phydev->mdio.bus; - int addr = phydev->mdio.addr; - device_del(&phydev->mdio.dev); - bus->phy_map[addr] = NULL; + mdiobus_unregister_device(&phydev->mdio); } EXPORT_SYMBOL(phy_device_remove); @@ -433,11 +430,13 @@ EXPORT_SYMBOL(phy_device_remove); */ struct phy_device *phy_find_first(struct mii_bus *bus) { + struct phy_device *phydev; int addr; for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - if (bus->phy_map[addr]) - return bus->phy_map[addr]; + phydev = mdiobus_get_phy(bus, addr); + if (phydev) + return phydev; } return NULL; } diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index c0292051392e..6febe2df76f9 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -193,7 +193,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) for (addr = 0; addr < PHY_MAX_ADDR; addr++) { /* skip already registered PHYs */ - if (mdio->phy_map[addr]) + if (mdiobus_is_registered_device(mdio, addr)) continue; /* be noisy to encourage people to set reg property */ diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 94f9f1491cde..8cd9579e18ea 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -19,9 +19,12 @@ struct mdio_device { struct mii_bus *bus; /* Bus address of the MDIO device (0-31) */ int addr; + int flags; }; #define to_mdio_device(d) container_of(d, struct mdio_device, dev) +#define MDIO_DEVICE_FLAG_PHY 1 + static inline bool mdio_phy_id_is_c45(int phy_id) { return (phy_id & MDIO_PHY_ID_C45) && !(phy_id & ~MDIO_PHY_ID_C45_MASK); @@ -188,4 +191,9 @@ int mdiobus_read_nested(struct mii_bus *bus, int addr, u32 regnum); int mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val); int mdiobus_write_nested(struct mii_bus *bus, int addr, u32 regnum, u16 val); +int mdiobus_register_device(struct mdio_device *mdiodev); +int mdiobus_unregister_device(struct mdio_device *mdiodev); +bool mdiobus_is_registered_device(struct mii_bus *bus, int addr); +struct phy_device *mdiobus_get_phy(struct mii_bus *bus, int addr); + #endif /* __LINUX_MDIO_H__ */ diff --git a/include/linux/phy.h b/include/linux/phy.h index 239a0c2bc49d..2d7beef20825 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -180,7 +180,7 @@ struct mii_bus { struct device dev; /* list of all PHYs on bus */ - struct phy_device *phy_map[PHY_MAX_ADDR]; + struct mdio_device *mdio_map[PHY_MAX_ADDR]; /* PHY addresses to be ignored when probing */ u32 phy_mask; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 5f45e68b52dc..2771713714f1 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -997,7 +998,7 @@ static int dsa_slave_phy_connect(struct dsa_slave_priv *p, { struct dsa_switch *ds = p->parent; - p->phy = ds->slave_mii_bus->phy_map[addr]; + p->phy = mdiobus_get_phy(ds->slave_mii_bus, addr); if (!p->phy) { netdev_err(slave_dev, "no phy at %d\n", addr); return -ENODEV; -- cgit v1.2.3-59-g8ed1b From bc87922ff59d364a33e9bce0febdef21a7fbd2af Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:21 +0100 Subject: phy: Move PHY PM operations into phy_device The MDIO PM operations are really PHY device PM operations. So move them into phy_device. This will be needed when we support devices on the mdio bus which are not PHYs. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 83 ++++---------------------------- drivers/net/phy/phy_device.c | 110 +++++++++++++++++++++++++++++++++++++++++++ include/linux/mdio.h | 2 +- 3 files changed, 121 insertions(+), 74 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index e6dddb086265..65ff8199bd09 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -561,95 +561,32 @@ static int mdio_bus_match(struct device *dev, struct device_driver *drv) } #ifdef CONFIG_PM - -static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) -{ - struct device_driver *drv = phydev->mdio.dev.driver; - struct phy_driver *phydrv = to_phy_driver(drv); - struct net_device *netdev = phydev->attached_dev; - - if (!drv || !phydrv->suspend) - return false; - - /* PHY not attached? May suspend if the PHY has not already been - * suspended as part of a prior call to phy_disconnect() -> - * phy_detach() -> phy_suspend() because the parent netdev might be the - * MDIO bus driver and clock gated at this point. - */ - if (!netdev) - return !phydev->suspended; - - /* Don't suspend PHY if the attched netdev parent may wakeup. - * The parent may point to a PCI device, as in tg3 driver. - */ - if (netdev->dev.parent && device_may_wakeup(netdev->dev.parent)) - return false; - - /* Also don't suspend PHY if the netdev itself may wakeup. This - * is the case for devices w/o underlaying pwr. mgmt. aware bus, - * e.g. SoC devices. - */ - if (device_may_wakeup(&netdev->dev)) - return false; - - return true; -} - static int mdio_bus_suspend(struct device *dev) { - struct phy_device *phydev = to_phy_device(dev); + struct mdio_device *mdio = to_mdio_device(dev); - /* We must stop the state machine manually, otherwise it stops out of - * control, possibly with the phydev->lock held. Upon resume, netdev - * may call phy routines that try to grab the same lock, and that may - * lead to a deadlock. - */ - if (phydev->attached_dev && phydev->adjust_link) - phy_stop_machine(phydev); - - if (!mdio_bus_phy_may_suspend(phydev)) - return 0; + if (mdio->pm_ops && mdio->pm_ops->suspend) + return mdio->pm_ops->suspend(dev); - return phy_suspend(phydev); + return 0; } static int mdio_bus_resume(struct device *dev) { - struct phy_device *phydev = to_phy_device(dev); - int ret; - - if (!mdio_bus_phy_may_suspend(phydev)) - goto no_resume; + struct mdio_device *mdio = to_mdio_device(dev); - ret = phy_resume(phydev); - if (ret < 0) - return ret; - -no_resume: - if (phydev->attached_dev && phydev->adjust_link) - phy_start_machine(phydev); + if (mdio->pm_ops && mdio->pm_ops->resume) + return mdio->pm_ops->resume(dev); return 0; } static int mdio_bus_restore(struct device *dev) { - struct phy_device *phydev = to_phy_device(dev); - struct net_device *netdev = phydev->attached_dev; - int ret; - - if (!netdev) - return 0; - - ret = phy_init_hw(phydev); - if (ret < 0) - return ret; - - /* The PHY needs to renegotiate. */ - phydev->link = 0; - phydev->state = PHY_UP; + struct mdio_device *mdio = to_mdio_device(dev); - phy_start_machine(phydev); + if (mdio->pm_ops && mdio->pm_ops->restore) + return mdio->pm_ops->restore(dev); return 0; } diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 7a5222daff93..eb0b0ed32662 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -63,6 +63,115 @@ static struct phy_driver genphy_driver[GENPHY_DRV_MAX]; static LIST_HEAD(phy_fixup_list); static DEFINE_MUTEX(phy_fixup_lock); +#ifdef CONFIG_PM +static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) +{ + struct device_driver *drv = phydev->mdio.dev.driver; + struct phy_driver *phydrv = to_phy_driver(drv); + struct net_device *netdev = phydev->attached_dev; + + if (!drv || !phydrv->suspend) + return false; + + /* PHY not attached? May suspend if the PHY has not already been + * suspended as part of a prior call to phy_disconnect() -> + * phy_detach() -> phy_suspend() because the parent netdev might be the + * MDIO bus driver and clock gated at this point. + */ + if (!netdev) + return !phydev->suspended; + + /* Don't suspend PHY if the attached netdev parent may wakeup. + * The parent may point to a PCI device, as in tg3 driver. + */ + if (netdev->dev.parent && device_may_wakeup(netdev->dev.parent)) + return false; + + /* Also don't suspend PHY if the netdev itself may wakeup. This + * is the case for devices w/o underlaying pwr. mgmt. aware bus, + * e.g. SoC devices. + */ + if (device_may_wakeup(&netdev->dev)) + return false; + + return true; +} + +static int mdio_bus_phy_suspend(struct device *dev) +{ + struct phy_device *phydev = to_phy_device(dev); + + /* We must stop the state machine manually, otherwise it stops out of + * control, possibly with the phydev->lock held. Upon resume, netdev + * may call phy routines that try to grab the same lock, and that may + * lead to a deadlock. + */ + if (phydev->attached_dev && phydev->adjust_link) + phy_stop_machine(phydev); + + if (!mdio_bus_phy_may_suspend(phydev)) + return 0; + + return phy_suspend(phydev); +} + +static int mdio_bus_phy_resume(struct device *dev) +{ + struct phy_device *phydev = to_phy_device(dev); + int ret; + + if (!mdio_bus_phy_may_suspend(phydev)) + goto no_resume; + + ret = phy_resume(phydev); + if (ret < 0) + return ret; + +no_resume: + if (phydev->attached_dev && phydev->adjust_link) + phy_start_machine(phydev); + + return 0; +} + +static int mdio_bus_phy_restore(struct device *dev) +{ + struct phy_device *phydev = to_phy_device(dev); + struct net_device *netdev = phydev->attached_dev; + int ret; + + if (!netdev) + return 0; + + ret = phy_init_hw(phydev); + if (ret < 0) + return ret; + + /* The PHY needs to renegotiate. */ + phydev->link = 0; + phydev->state = PHY_UP; + + phy_start_machine(phydev); + + return 0; +} + +static const struct dev_pm_ops mdio_bus_phy_pm_ops = { + .suspend = mdio_bus_phy_suspend, + .resume = mdio_bus_phy_resume, + .freeze = mdio_bus_phy_suspend, + .thaw = mdio_bus_phy_resume, + .restore = mdio_bus_phy_restore, +}; + +#define MDIO_BUS_PHY_PM_OPS (&mdio_bus_phy_pm_ops) + +#else + +#define MDIO_BUS_PHY_PM_OPS NULL + +#endif /* CONFIG_PM */ + /** * phy_register_fixup - creates a new phy_fixup and adds it to the list * @bus_id: A string which matches phydev->mdio.dev.bus_id (or PHY_ANY_ID) @@ -165,6 +274,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, mdiodev->dev.parent = &bus->dev; mdiodev->dev.bus = &mdio_bus_type; mdiodev->bus = bus; + mdiodev->pm_ops = MDIO_BUS_PHY_PM_OPS; mdiodev->addr = addr; mdiodev->flags = MDIO_DEVICE_FLAG_PHY; diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 8cd9579e18ea..9f844d372ed5 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -15,7 +15,7 @@ struct mii_bus; struct mdio_device { struct device dev; - + const struct dev_pm_ops *pm_ops; struct mii_bus *bus; /* Bus address of the MDIO device (0-31) */ int addr; -- cgit v1.2.3-59-g8ed1b From be01da72b1b832b89fbdf59ae6f1b60e53ca2987 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:22 +0100 Subject: phy: Centralize setting driver module owner Rather than have each driver set the driver owner field, do it once in the core code. This will also help with later changes, when the device structure will move. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/amd.c | 1 - drivers/net/phy/aquantia.c | 4 ---- drivers/net/phy/at803x.c | 9 --------- drivers/net/phy/bcm63xx.c | 2 -- drivers/net/phy/bcm7xxx.c | 6 ------ drivers/net/phy/bcm87xx.c | 2 -- drivers/net/phy/broadcom.c | 12 ------------ drivers/net/phy/cicada.c | 2 -- drivers/net/phy/davicom.c | 4 ---- drivers/net/phy/dp83640.c | 3 +-- drivers/net/phy/dp83848.c | 2 -- drivers/net/phy/dp83867.c | 2 -- drivers/net/phy/et1011c.c | 1 - drivers/net/phy/icplus.c | 3 --- drivers/net/phy/lxt.c | 4 ---- drivers/net/phy/marvell.c | 13 ------------- drivers/net/phy/micrel.c | 13 ------------- drivers/net/phy/microchip.c | 2 -- drivers/net/phy/national.c | 1 - drivers/net/phy/phy_device.c | 13 +++++++------ drivers/net/phy/qsemi.c | 1 - drivers/net/phy/realtek.c | 5 ----- drivers/net/phy/smsc.c | 10 ---------- drivers/net/phy/ste10Xp.c | 2 -- drivers/net/phy/teranetics.c | 1 - drivers/net/phy/vitesse.c | 8 -------- include/linux/phy.h | 7 ++++--- 27 files changed, 12 insertions(+), 121 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/amd.c b/drivers/net/phy/amd.c index 65a488f82eb8..18141c022b13 100644 --- a/drivers/net/phy/amd.c +++ b/drivers/net/phy/amd.c @@ -72,7 +72,6 @@ static struct phy_driver am79c_driver[] = { { .read_status = genphy_read_status, .ack_interrupt = am79c_ack_interrupt, .config_intr = am79c_config_intr, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(am79c_driver); diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aquantia.c index f1936b7a7af6..09b0b0aa8d68 100644 --- a/drivers/net/phy/aquantia.c +++ b/drivers/net/phy/aquantia.c @@ -128,7 +128,6 @@ static struct phy_driver aquantia_driver[] = { .config_intr = aquantia_config_intr, .ack_interrupt = aquantia_ack_interrupt, .read_status = aquantia_read_status, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_AQ2104, @@ -141,7 +140,6 @@ static struct phy_driver aquantia_driver[] = { .config_intr = aquantia_config_intr, .ack_interrupt = aquantia_ack_interrupt, .read_status = aquantia_read_status, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_AQR105, @@ -154,7 +152,6 @@ static struct phy_driver aquantia_driver[] = { .config_intr = aquantia_config_intr, .ack_interrupt = aquantia_ack_interrupt, .read_status = aquantia_read_status, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_AQR405, @@ -167,7 +164,6 @@ static struct phy_driver aquantia_driver[] = { .config_intr = aquantia_config_intr, .ack_interrupt = aquantia_ack_interrupt, .read_status = aquantia_read_status, - .driver = { .owner = THIS_MODULE,}, }, }; diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c index b76ac09a554f..8a8f6fb2880d 100644 --- a/drivers/net/phy/at803x.c +++ b/drivers/net/phy/at803x.c @@ -310,9 +310,6 @@ static struct phy_driver at803x_driver[] = { .read_status = genphy_read_status, .ack_interrupt = at803x_ack_interrupt, .config_intr = at803x_config_intr, - .driver = { - .owner = THIS_MODULE, - }, }, { /* ATHEROS 8030 */ .phy_id = ATH8030_PHY_ID, @@ -331,9 +328,6 @@ static struct phy_driver at803x_driver[] = { .read_status = genphy_read_status, .ack_interrupt = at803x_ack_interrupt, .config_intr = at803x_config_intr, - .driver = { - .owner = THIS_MODULE, - }, }, { /* ATHEROS 8031 */ .phy_id = ATH8031_PHY_ID, @@ -352,9 +346,6 @@ static struct phy_driver at803x_driver[] = { .read_status = genphy_read_status, .ack_interrupt = &at803x_ack_interrupt, .config_intr = &at803x_config_intr, - .driver = { - .owner = THIS_MODULE, - }, } }; module_phy_driver(at803x_driver); diff --git a/drivers/net/phy/bcm63xx.c b/drivers/net/phy/bcm63xx.c index 86b28052bf06..e741bf614c4e 100644 --- a/drivers/net/phy/bcm63xx.c +++ b/drivers/net/phy/bcm63xx.c @@ -56,7 +56,6 @@ static struct phy_driver bcm63xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { /* same phy as above, with just a different OUI */ .phy_id = 0x002bdc00, @@ -69,7 +68,6 @@ static struct phy_driver bcm63xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, } }; module_phy_driver(bcm63xx_driver); diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c index 9f4e6eb886af..bf241a3ec5e5 100644 --- a/drivers/net/phy/bcm7xxx.c +++ b/drivers/net/phy/bcm7xxx.c @@ -324,7 +324,6 @@ static int bcm7xxx_dummy_config_init(struct phy_device *phydev) .config_aneg = genphy_config_aneg, \ .read_status = genphy_read_status, \ .resume = bcm7xxx_28nm_resume, \ - .driver = { .owner = THIS_MODULE }, \ } static struct phy_driver bcm7xxx_driver[] = { @@ -346,7 +345,6 @@ static struct phy_driver bcm7xxx_driver[] = { .read_status = genphy_read_status, .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM7429, .phy_id_mask = 0xfffffff0, @@ -359,7 +357,6 @@ static struct phy_driver bcm7xxx_driver[] = { .read_status = genphy_read_status, .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM7435, .phy_id_mask = 0xfffffff0, @@ -372,7 +369,6 @@ static struct phy_driver bcm7xxx_driver[] = { .read_status = genphy_read_status, .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_BCM_OUI_4, .phy_id_mask = 0xffff0000, @@ -385,7 +381,6 @@ static struct phy_driver bcm7xxx_driver[] = { .read_status = genphy_read_status, .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_BCM_OUI_5, .phy_id_mask = 0xffffff00, @@ -398,7 +393,6 @@ static struct phy_driver bcm7xxx_driver[] = { .read_status = genphy_read_status, .suspend = bcm7xxx_suspend, .resume = bcm7xxx_config_init, - .driver = { .owner = THIS_MODULE }, } }; static struct mdio_device_id __maybe_unused bcm7xxx_tbl[] = { diff --git a/drivers/net/phy/bcm87xx.c b/drivers/net/phy/bcm87xx.c index e536e30d1643..f7ebdcff53e4 100644 --- a/drivers/net/phy/bcm87xx.c +++ b/drivers/net/phy/bcm87xx.c @@ -201,7 +201,6 @@ static struct phy_driver bcm87xx_driver[] = { .config_intr = bcm87xx_config_intr, .did_interrupt = bcm87xx_did_interrupt, .match_phy_device = bcm8706_match_phy_device, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM8727, .phy_id_mask = 0xffffffff, @@ -214,7 +213,6 @@ static struct phy_driver bcm87xx_driver[] = { .config_intr = bcm87xx_config_intr, .did_interrupt = bcm87xx_did_interrupt, .match_phy_device = bcm8727_match_phy_device, - .driver = { .owner = THIS_MODULE }, } }; module_phy_driver(bcm87xx_driver); diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index 3ce5d9514623..870327efccf7 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -460,7 +460,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5421, .phy_id_mask = 0xfffffff0, @@ -473,7 +472,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5461, .phy_id_mask = 0xfffffff0, @@ -486,7 +484,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM54616S, .phy_id_mask = 0xfffffff0, @@ -499,7 +496,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5464, .phy_id_mask = 0xfffffff0, @@ -512,7 +508,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5481, .phy_id_mask = 0xfffffff0, @@ -525,7 +520,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5482, .phy_id_mask = 0xfffffff0, @@ -538,7 +532,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = bcm5482_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM50610, .phy_id_mask = 0xfffffff0, @@ -551,7 +544,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM50610M, .phy_id_mask = 0xfffffff0, @@ -564,7 +556,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM57780, .phy_id_mask = 0xfffffff0, @@ -577,7 +568,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = bcm_phy_ack_intr, .config_intr = bcm_phy_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCMAC131, .phy_id_mask = 0xfffffff0, @@ -590,7 +580,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = brcm_fet_ack_interrupt, .config_intr = brcm_fet_config_intr, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = PHY_ID_BCM5241, .phy_id_mask = 0xfffffff0, @@ -603,7 +592,6 @@ static struct phy_driver broadcom_drivers[] = { .read_status = genphy_read_status, .ack_interrupt = brcm_fet_ack_interrupt, .config_intr = brcm_fet_config_intr, - .driver = { .owner = THIS_MODULE }, } }; module_phy_driver(broadcom_drivers); diff --git a/drivers/net/phy/cicada.c b/drivers/net/phy/cicada.c index 27f5464899d4..d339c1afea77 100644 --- a/drivers/net/phy/cicada.c +++ b/drivers/net/phy/cicada.c @@ -114,7 +114,6 @@ static struct phy_driver cis820x_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &cis820x_ack_interrupt, .config_intr = &cis820x_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x000fc440, .name = "Cicada Cis8204", @@ -126,7 +125,6 @@ static struct phy_driver cis820x_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &cis820x_ack_interrupt, .config_intr = &cis820x_config_intr, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(cis820x_driver); diff --git a/drivers/net/phy/davicom.c b/drivers/net/phy/davicom.c index 2a328703b4ae..36e3e2033eca 100644 --- a/drivers/net/phy/davicom.c +++ b/drivers/net/phy/davicom.c @@ -156,7 +156,6 @@ static struct phy_driver dm91xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = dm9161_ack_interrupt, .config_intr = dm9161_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x0181b8b0, .name = "Davicom DM9161B/C", @@ -168,7 +167,6 @@ static struct phy_driver dm91xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = dm9161_ack_interrupt, .config_intr = dm9161_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x0181b8a0, .name = "Davicom DM9161A", @@ -180,7 +178,6 @@ static struct phy_driver dm91xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = dm9161_ack_interrupt, .config_intr = dm9161_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x00181b80, .name = "Davicom DM9131", @@ -191,7 +188,6 @@ static struct phy_driver dm91xx_driver[] = { .read_status = genphy_read_status, .ack_interrupt = dm9161_ack_interrupt, .config_intr = dm9161_config_intr, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(dm91xx_driver); diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 39da6fc6a85e..180f69952779 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -1492,12 +1492,11 @@ static struct phy_driver dp83640_driver = { .hwtstamp = dp83640_hwtstamp, .rxtstamp = dp83640_rxtstamp, .txtstamp = dp83640_txtstamp, - .driver = {.owner = THIS_MODULE,} }; static int __init dp83640_init(void) { - return phy_driver_register(&dp83640_driver); + return phy_driver_register(&dp83640_driver, THIS_MODULE); } static void __exit dp83640_exit(void) diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c index 5ce9bef54468..5e14e629c597 100644 --- a/drivers/net/phy/dp83848.c +++ b/drivers/net/phy/dp83848.c @@ -88,8 +88,6 @@ static struct phy_driver dp83848_driver[] = { /* IRQ related */ .ack_interrupt = dp83848_ack_interrupt, .config_intr = dp83848_config_intr, - - .driver = { .owner = THIS_MODULE, }, }, }; module_phy_driver(dp83848_driver); diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index 74e4521bd2d3..2afa61b51d41 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -214,8 +214,6 @@ static struct phy_driver dp83867_driver[] = { .read_status = genphy_read_status, .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = {.owner = THIS_MODULE,} }, }; module_phy_driver(dp83867_driver); diff --git a/drivers/net/phy/et1011c.c b/drivers/net/phy/et1011c.c index a907743816a8..a9a4edfa23c8 100644 --- a/drivers/net/phy/et1011c.c +++ b/drivers/net/phy/et1011c.c @@ -95,7 +95,6 @@ static struct phy_driver et1011c_driver[] = { { .flags = PHY_POLL, .config_aneg = et1011c_config_aneg, .read_status = et1011c_read_status, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(et1011c_driver); diff --git a/drivers/net/phy/icplus.c b/drivers/net/phy/icplus.c index c12170d07b62..e5f251b91578 100644 --- a/drivers/net/phy/icplus.c +++ b/drivers/net/phy/icplus.c @@ -221,7 +221,6 @@ static struct phy_driver icplus_driver[] = { .read_status = &ip175c_read_status, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x02430d90, .name = "ICPlus IP1001", @@ -233,7 +232,6 @@ static struct phy_driver icplus_driver[] = { .read_status = &genphy_read_status, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x02430c54, .name = "ICPlus IP101A/G", @@ -247,7 +245,6 @@ static struct phy_driver icplus_driver[] = { .read_status = &genphy_read_status, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(icplus_driver); diff --git a/drivers/net/phy/lxt.c b/drivers/net/phy/lxt.c index a3a5a703635b..f6078376ef50 100644 --- a/drivers/net/phy/lxt.c +++ b/drivers/net/phy/lxt.c @@ -278,7 +278,6 @@ static struct phy_driver lxt97x_driver[] = { .read_status = genphy_read_status, .ack_interrupt = lxt970_ack_interrupt, .config_intr = lxt970_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x001378e0, .name = "LXT971", @@ -289,7 +288,6 @@ static struct phy_driver lxt97x_driver[] = { .read_status = genphy_read_status, .ack_interrupt = lxt971_ack_interrupt, .config_intr = lxt971_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x00137a10, .name = "LXT973-A2", @@ -299,7 +297,6 @@ static struct phy_driver lxt97x_driver[] = { .probe = lxt973_probe, .config_aneg = lxt973_config_aneg, .read_status = lxt973a2_read_status, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x00137a10, .name = "LXT973", @@ -309,7 +306,6 @@ static struct phy_driver lxt97x_driver[] = { .probe = lxt973_probe, .config_aneg = lxt973_config_aneg, .read_status = genphy_read_status, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(lxt97x_driver); diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index f96c93c9819a..e3eb96443c97 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -1087,7 +1087,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1112, @@ -1106,7 +1105,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1111, @@ -1125,7 +1123,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1118, @@ -1144,7 +1141,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = {.owner = THIS_MODULE,}, }, { .phy_id = MARVELL_PHY_ID_88E1121R, @@ -1163,7 +1159,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1318S, @@ -1184,7 +1179,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1145, @@ -1203,7 +1197,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1149R, @@ -1222,7 +1215,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1240, @@ -1241,7 +1233,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1116R, @@ -1260,7 +1251,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1510, @@ -1279,7 +1269,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E1540, @@ -1298,7 +1287,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, { .phy_id = MARVELL_PHY_ID_88E3016, @@ -1319,7 +1307,6 @@ static struct phy_driver marvell_drivers[] = { .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, - .driver = { .owner = THIS_MODULE }, }, }; diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index b51505be1fa9..0dbc6496b6f6 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -706,7 +706,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8021, .phy_id_mask = 0x00ffffff, @@ -726,7 +725,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8031, .phy_id_mask = 0x00ffffff, @@ -746,7 +744,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8041, .phy_id_mask = 0x00fffff0, @@ -766,7 +763,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8041RNLI, .phy_id_mask = 0x00fffff0, @@ -786,7 +782,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8051, .phy_id_mask = 0x00fffff0, @@ -806,7 +801,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8001, .name = "Micrel KSZ8001 or KS8721", @@ -825,7 +819,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8081, .name = "Micrel KSZ8081 or KSZ8091", @@ -844,7 +837,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8061, .name = "Micrel KSZ8061", @@ -861,7 +853,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ9021, .phy_id_mask = 0x000ffffe, @@ -881,7 +872,6 @@ static struct phy_driver ksphy_driver[] = { .resume = genphy_resume, .read_mmd_indirect = ksz9021_rd_mmd_phyreg, .write_mmd_indirect = ksz9021_wr_mmd_phyreg, - .driver = { .owner = THIS_MODULE, }, }, { .phy_id = PHY_ID_KSZ9031, .phy_id_mask = 0x00fffff0, @@ -899,7 +889,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE, }, }, { .phy_id = PHY_ID_KSZ8873MLL, .phy_id_mask = 0x00fffff0, @@ -914,7 +903,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE, }, }, { .phy_id = PHY_ID_KSZ886X, .phy_id_mask = 0x00fffff0, @@ -929,7 +917,6 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE, }, } }; module_phy_driver(ksphy_driver); diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c index 5e34b49be0b3..15f820648f82 100644 --- a/drivers/net/phy/microchip.c +++ b/drivers/net/phy/microchip.c @@ -129,8 +129,6 @@ static struct phy_driver microchip_phy_driver[] = { .suspend = lan88xx_suspend, .resume = genphy_resume, .set_wol = lan88xx_set_wol, - - .driver = { .owner = THIS_MODULE, } } }; module_phy_driver(microchip_phy_driver); diff --git a/drivers/net/phy/national.c b/drivers/net/phy/national.c index 0a7b9c7f09a2..2a1b490bc587 100644 --- a/drivers/net/phy/national.c +++ b/drivers/net/phy/national.c @@ -140,7 +140,6 @@ static struct phy_driver dp83865_driver[] = { { .read_status = genphy_read_status, .ack_interrupt = ns_ack_interrupt, .config_intr = ns_config_intr, - .driver = {.owner = THIS_MODULE,} } }; module_phy_driver(dp83865_driver); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index eb0b0ed32662..a1b833cd4183 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -1564,8 +1564,9 @@ static int phy_remove(struct device *dev) /** * phy_driver_register - register a phy_driver with the PHY layer * @new_driver: new phy_driver to register + * @owner: module owning this PHY */ -int phy_driver_register(struct phy_driver *new_driver) +int phy_driver_register(struct phy_driver *new_driver, struct module *owner) { int retval; @@ -1573,6 +1574,7 @@ int phy_driver_register(struct phy_driver *new_driver) new_driver->driver.bus = &mdio_bus_type; new_driver->driver.probe = phy_probe; new_driver->driver.remove = phy_remove; + new_driver->driver.owner = owner; retval = driver_register(&new_driver->driver); if (retval) { @@ -1588,12 +1590,13 @@ int phy_driver_register(struct phy_driver *new_driver) } EXPORT_SYMBOL(phy_driver_register); -int phy_drivers_register(struct phy_driver *new_driver, int n) +int phy_drivers_register(struct phy_driver *new_driver, int n, + struct module *owner) { int i, ret = 0; for (i = 0; i < n; i++) { - ret = phy_driver_register(new_driver + i); + ret = phy_driver_register(new_driver + i, owner); if (ret) { while (i-- > 0) phy_driver_unregister(new_driver + i); @@ -1634,7 +1637,6 @@ static struct phy_driver genphy_driver[] = { .read_status = genphy_read_status, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE, }, }, { .phy_id = 0xffffffff, .phy_id_mask = 0xffffffff, @@ -1646,7 +1648,6 @@ static struct phy_driver genphy_driver[] = { .read_status = gen10g_read_status, .suspend = gen10g_suspend, .resume = gen10g_resume, - .driver = {.owner = THIS_MODULE, }, } }; static int __init phy_init(void) @@ -1658,7 +1659,7 @@ static int __init phy_init(void) return rc; rc = phy_drivers_register(genphy_driver, - ARRAY_SIZE(genphy_driver)); + ARRAY_SIZE(genphy_driver), THIS_MODULE); if (rc) mdio_bus_exit(); diff --git a/drivers/net/phy/qsemi.c b/drivers/net/phy/qsemi.c index be4c6f7c3645..d470db89e8dd 100644 --- a/drivers/net/phy/qsemi.c +++ b/drivers/net/phy/qsemi.c @@ -122,7 +122,6 @@ static struct phy_driver qs6612_driver[] = { { .read_status = genphy_read_status, .ack_interrupt = qs6612_ack_interrupt, .config_intr = qs6612_config_intr, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(qs6612_driver); diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c index 43ab691362d4..aadd6e9f54ad 100644 --- a/drivers/net/phy/realtek.c +++ b/drivers/net/phy/realtek.c @@ -124,7 +124,6 @@ static struct phy_driver realtek_drvs[] = { .flags = PHY_HAS_INTERRUPT, .config_aneg = &genphy_config_aneg, .read_status = &genphy_read_status, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x001cc912, .name = "RTL8211B Gigabit Ethernet", @@ -135,7 +134,6 @@ static struct phy_driver realtek_drvs[] = { .read_status = &genphy_read_status, .ack_interrupt = &rtl821x_ack_interrupt, .config_intr = &rtl8211b_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x001cc914, .name = "RTL8211DN Gigabit Ethernet", @@ -148,7 +146,6 @@ static struct phy_driver realtek_drvs[] = { .config_intr = rtl8211e_config_intr, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x001cc915, .name = "RTL8211E Gigabit Ethernet", @@ -161,7 +158,6 @@ static struct phy_driver realtek_drvs[] = { .config_intr = &rtl8211e_config_intr, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = 0x001cc916, .name = "RTL8211F Gigabit Ethernet", @@ -175,7 +171,6 @@ static struct phy_driver realtek_drvs[] = { .config_intr = &rtl8211f_config_intr, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = { .owner = THIS_MODULE }, }, }; diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c index 18c981b95910..ddb06135c21d 100644 --- a/drivers/net/phy/smsc.c +++ b/drivers/net/phy/smsc.c @@ -171,8 +171,6 @@ static struct phy_driver smsc_phy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = { .owner = THIS_MODULE, } }, { .phy_id = 0x0007c0b0, /* OUI=0x00800f, Model#=0x0b */ .phy_id_mask = 0xfffffff0, @@ -194,8 +192,6 @@ static struct phy_driver smsc_phy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = { .owner = THIS_MODULE, } }, { .phy_id = 0x0007c0c0, /* OUI=0x00800f, Model#=0x0c */ .phy_id_mask = 0xfffffff0, @@ -217,8 +213,6 @@ static struct phy_driver smsc_phy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = { .owner = THIS_MODULE, } }, { .phy_id = 0x0007c0d0, /* OUI=0x00800f, Model#=0x0d */ .phy_id_mask = 0xfffffff0, @@ -239,8 +233,6 @@ static struct phy_driver smsc_phy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = { .owner = THIS_MODULE, } }, { .phy_id = 0x0007c0f0, /* OUI=0x00800f, Model#=0x0f */ .phy_id_mask = 0xfffffff0, @@ -262,8 +254,6 @@ static struct phy_driver smsc_phy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, - - .driver = { .owner = THIS_MODULE, } } }; module_phy_driver(smsc_phy_driver); diff --git a/drivers/net/phy/ste10Xp.c b/drivers/net/phy/ste10Xp.c index 3fc199b773e6..d00cfb64529e 100644 --- a/drivers/net/phy/ste10Xp.c +++ b/drivers/net/phy/ste10Xp.c @@ -95,7 +95,6 @@ static struct phy_driver ste10xp_pdriver[] = { .config_intr = ste10Xp_config_intr, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = {.owner = THIS_MODULE,} }, { .phy_id = STE100P_PHY_ID, .phy_id_mask = 0xffffffff, @@ -109,7 +108,6 @@ static struct phy_driver ste10xp_pdriver[] = { .config_intr = ste10Xp_config_intr, .suspend = genphy_suspend, .resume = genphy_resume, - .driver = {.owner = THIS_MODULE,} } }; module_phy_driver(ste10xp_pdriver); diff --git a/drivers/net/phy/teranetics.c b/drivers/net/phy/teranetics.c index 07463fcca212..fb2cef764e9a 100644 --- a/drivers/net/phy/teranetics.c +++ b/drivers/net/phy/teranetics.c @@ -108,7 +108,6 @@ static struct phy_driver teranetics_driver[] = { .config_aneg = teranetics_config_aneg, .read_status = teranetics_read_status, .match_phy_device = teranetics_match_phy_device, - .driver = { .owner = THIS_MODULE,}, }, }; diff --git a/drivers/net/phy/vitesse.c b/drivers/net/phy/vitesse.c index dd295dbaa074..2e37eb337d48 100644 --- a/drivers/net/phy/vitesse.c +++ b/drivers/net/phy/vitesse.c @@ -236,7 +236,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8244, .name = "Vitesse VSC8244", @@ -248,7 +247,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8514, .name = "Vitesse VSC8514", @@ -260,7 +258,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8574, .name = "Vitesse VSC8574", @@ -272,7 +269,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8601, .name = "Vitesse VSC8601", @@ -284,7 +280,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_VSC8662, .name = "Vitesse VSC8662", @@ -296,7 +291,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { /* Vitesse 8221 */ .phy_id = PHY_ID_VSC8221, @@ -309,7 +303,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, }, { /* Vitesse 8211 */ .phy_id = PHY_ID_VSC8211, @@ -322,7 +315,6 @@ static struct phy_driver vsc82xx_driver[] = { .read_status = &genphy_read_status, .ack_interrupt = &vsc824x_ack_interrupt, .config_intr = &vsc82xx_config_intr, - .driver = { .owner = THIS_MODULE,}, } }; module_phy_driver(vsc82xx_driver); diff --git a/include/linux/phy.h b/include/linux/phy.h index 2d7beef20825..49e4418822b3 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -797,8 +797,9 @@ int genphy_resume(struct phy_device *phydev); int genphy_soft_reset(struct phy_device *phydev); void phy_driver_unregister(struct phy_driver *drv); void phy_drivers_unregister(struct phy_driver *drv, int n); -int phy_driver_register(struct phy_driver *new_driver); -int phy_drivers_register(struct phy_driver *new_driver, int n); +int phy_driver_register(struct phy_driver *new_driver, struct module *owner); +int phy_drivers_register(struct phy_driver *new_driver, int n, + struct module *owner); void phy_state_machine(struct work_struct *work); void phy_change(struct work_struct *work); void phy_mac_interrupt(struct phy_device *phydev, int new_link); @@ -843,7 +844,7 @@ extern struct bus_type mdio_bus_type; #define phy_module_driver(__phy_drivers, __count) \ static int __init phy_module_init(void) \ { \ - return phy_drivers_register(__phy_drivers, __count); \ + return phy_drivers_register(__phy_drivers, __count, THIS_MODULE); \ } \ module_init(phy_module_init); \ static void __exit phy_module_exit(void) \ -- cgit v1.2.3-59-g8ed1b From e76a4957c5ee5cf69cea89d450c29c536e77ce9e Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:23 +0100 Subject: phy: Move phy specific bus match into phy_device Matching a driver to a device has both generic parts, and parts which are specific to PHY devices. Move the PHY specific parts into phy_device. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 38 ++++++++++++-------------------------- drivers/net/phy/phy_device.c | 28 ++++++++++++++++++++++++++++ include/linux/mdio.h | 1 + 3 files changed, 41 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 65ff8199bd09..bd523b2c6331 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -523,41 +523,27 @@ int mdiobus_write(struct mii_bus *bus, int addr, u32 regnum, u16 val) EXPORT_SYMBOL(mdiobus_write); /** - * mdio_bus_match - determine if given PHY driver supports the given PHY device - * @dev: target PHY device - * @drv: given PHY driver + * mdio_bus_match - determine if given MDIO driver supports the given + * MDIO device + * @dev: target MDIO device + * @drv: given MDIO driver * - * Description: Given a PHY device, and a PHY driver, return 1 if - * the driver supports the device. Otherwise, return 0. + * Description: Given a MDIO device, and a MDIO driver, return 1 if + * the driver supports the device. Otherwise, return 0. This may + * require calling the devices own match function, since different classes + * of MDIO devices have different match criteria. */ static int mdio_bus_match(struct device *dev, struct device_driver *drv) { - struct phy_device *phydev = to_phy_device(dev); - struct phy_driver *phydrv = to_phy_driver(drv); - const int num_ids = ARRAY_SIZE(phydev->c45_ids.device_ids); - int i; + struct mdio_device *mdio = to_mdio_device(dev); if (of_driver_match_device(dev, drv)) return 1; - if (phydrv->match_phy_device) - return phydrv->match_phy_device(phydev); + if (mdio->bus_match) + return mdio->bus_match(dev, drv); - if (phydev->is_c45) { - for (i = 1; i < num_ids; i++) { - if (!(phydev->c45_ids.devices_in_package & (1 << i))) - continue; - - if ((phydrv->phy_id & phydrv->phy_id_mask) == - (phydev->c45_ids.device_ids[i] & - phydrv->phy_id_mask)) - return 1; - } - return 0; - } else { - return (phydrv->phy_id & phydrv->phy_id_mask) == - (phydev->phy_id & phydrv->phy_id_mask); - } + return 0; } #ifdef CONFIG_PM diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index a1b833cd4183..78628428ee28 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -257,6 +257,33 @@ static int phy_scan_fixups(struct phy_device *phydev) return 0; } +static int phy_bus_match(struct device *dev, struct device_driver *drv) +{ + struct phy_device *phydev = to_phy_device(dev); + struct phy_driver *phydrv = to_phy_driver(drv); + const int num_ids = ARRAY_SIZE(phydev->c45_ids.device_ids); + int i; + + if (phydrv->match_phy_device) + return phydrv->match_phy_device(phydev); + + if (phydev->is_c45) { + for (i = 1; i < num_ids; i++) { + if (!(phydev->c45_ids.devices_in_package & (1 << i))) + continue; + + if ((phydrv->phy_id & phydrv->phy_id_mask) == + (phydev->c45_ids.device_ids[i] & + phydrv->phy_id_mask)) + return 1; + } + return 0; + } else { + return (phydrv->phy_id & phydrv->phy_id_mask) == + (phydev->phy_id & phydrv->phy_id_mask); + } +} + struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, bool is_c45, struct phy_c45_device_ids *c45_ids) @@ -275,6 +302,7 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, mdiodev->dev.bus = &mdio_bus_type; mdiodev->bus = bus; mdiodev->pm_ops = MDIO_BUS_PHY_PM_OPS; + mdiodev->bus_match = phy_bus_match; mdiodev->addr = addr; mdiodev->flags = MDIO_DEVICE_FLAG_PHY; diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 9f844d372ed5..0690359e55a5 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -17,6 +17,7 @@ struct mdio_device { struct device dev; const struct dev_pm_ops *pm_ops; struct mii_bus *bus; + int (*bus_match)(struct device *dev, struct device_driver *drv); /* Bus address of the MDIO device (0-31) */ int addr; int flags; -- cgit v1.2.3-59-g8ed1b From a9049e0c513c4521dbfaa302af8ed08b3366b41f Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:26 +0100 Subject: mdio: Add support for mdio drivers. Not all devices on an MDIO bus are PHYs. Meaning not all MDIO drivers are PHY drivers. Add support for generic MDIO drivers. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/Makefile | 2 +- drivers/net/phy/mdio_bus.c | 13 +++- drivers/net/phy/mdio_device.c | 169 ++++++++++++++++++++++++++++++++++++++++++ drivers/net/phy/phy_device.c | 27 ++++--- drivers/of/of_mdio.c | 33 +++++++++ include/linux/mdio.h | 50 +++++++++++++ include/linux/phy.h | 9 ++- 7 files changed, 286 insertions(+), 17 deletions(-) create mode 100644 drivers/net/phy/mdio_device.c (limited to 'include') diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index f31a4e25cf15..680e88f9915a 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -1,6 +1,6 @@ # Makefile for Linux PHY drivers -libphy-objs := phy.o phy_device.o mdio_bus.o +libphy-objs := phy.o phy_device.o mdio_bus.o mdio_device.o obj-$(CONFIG_PHYLIB) += libphy.o obj-$(CONFIG_AQUANTIA_PHY) += aquantia.o diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 6f9ca51446db..0573cfd2116f 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -357,16 +357,25 @@ EXPORT_SYMBOL(__mdiobus_register); void mdiobus_unregister(struct mii_bus *bus) { + struct mdio_device *mdiodev; + struct phy_device *phydev; int i; BUG_ON(bus->state != MDIOBUS_REGISTERED); bus->state = MDIOBUS_UNREGISTERED; for (i = 0; i < PHY_MAX_ADDR; i++) { - struct phy_device *phydev = mdiobus_get_phy(bus, i); - if (phydev) { + mdiodev = bus->mdio_map[i]; + if (!mdiodev) + continue; + + if (!(mdiodev->flags & MDIO_DEVICE_FLAG_PHY)) { + phydev = container_of(mdiodev, struct phy_device, mdio); phy_device_remove(phydev); phy_device_free(phydev); + } else { + mdio_device_remove(mdiodev); + mdio_device_free(mdiodev); } } device_del(&bus->dev); diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c new file mode 100644 index 000000000000..64e3777c85b4 --- /dev/null +++ b/drivers/net/phy/mdio_device.c @@ -0,0 +1,169 @@ +/* Framework for MDIO devices, other than PHYs. + * + * Copyright (c) 2016 Andrew Lunn + * + * 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. + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +void mdio_device_free(struct mdio_device *mdiodev) +{ + put_device(&mdiodev->dev); +} +EXPORT_SYMBOL(mdio_device_free); + +static void mdio_device_release(struct device *dev) +{ + kfree(to_mdio_device(dev)); +} + +struct mdio_device *mdio_device_create(struct mii_bus *bus, int addr) +{ + struct mdio_device *mdiodev; + + /* We allocate the device, and initialize the default values */ + mdiodev = kzalloc(sizeof(*mdiodev), GFP_KERNEL); + if (!mdiodev) + return ERR_PTR(-ENOMEM); + + mdiodev->dev.release = mdio_device_release; + mdiodev->dev.parent = &bus->dev; + mdiodev->dev.bus = &mdio_bus_type; + mdiodev->bus = bus; + mdiodev->addr = addr; + + dev_set_name(&mdiodev->dev, PHY_ID_FMT, bus->id, addr); + + device_initialize(&mdiodev->dev); + + return mdiodev; +} +EXPORT_SYMBOL(mdio_device_create); + +/** + * mdio_device_register - Register the mdio device on the MDIO bus + * @mdiodev: mdio_device structure to be added to the MDIO bus + */ +int mdio_device_register(struct mdio_device *mdiodev) +{ + int err; + + dev_info(&mdiodev->dev, "mdio_device_register\n"); + + err = mdiobus_register_device(mdiodev); + if (err) + return err; + + err = device_add(&mdiodev->dev); + if (err) { + pr_err("MDIO %d failed to add\n", mdiodev->addr); + goto out; + } + + return 0; + + out: + mdiobus_unregister_device(mdiodev); + return err; +} +EXPORT_SYMBOL(mdio_device_register); + +/** + * mdio_device_remove - Remove a previously registered mdio device from the + * MDIO bus + * @mdiodev: mdio_device structure to remove + * + * This doesn't free the mdio_device itself, it merely reverses the effects + * of mdio_device_register(). Use mdio_device_free() to free the device + * after calling this function. + */ +void mdio_device_remove(struct mdio_device *mdiodev) +{ + device_del(&mdiodev->dev); + mdiobus_unregister_device(mdiodev); +} +EXPORT_SYMBOL(mdio_device_remove); + +/** + * mdio_probe - probe an MDIO device + * @dev: device to probe + * + * Description: Take care of setting up the mdio_device structure + * and calling the driver to probe the device. + */ +static int mdio_probe(struct device *dev) +{ + struct mdio_device *mdiodev = to_mdio_device(dev); + struct device_driver *drv = mdiodev->dev.driver; + struct mdio_driver *mdiodrv = to_mdio_driver(drv); + int err = 0; + + if (mdiodrv->probe) + err = mdiodrv->probe(mdiodev); + + return err; +} + +static int mdio_remove(struct device *dev) +{ + struct mdio_device *mdiodev = to_mdio_device(dev); + struct device_driver *drv = mdiodev->dev.driver; + struct mdio_driver *mdiodrv = to_mdio_driver(drv); + + if (mdiodrv->remove) + mdiodrv->remove(mdiodev); + + return 0; +} + +/** + * mdio_driver_register - register an mdio_driver with the MDIO layer + * @new_driver: new mdio_driver to register + */ +int mdio_driver_register(struct mdio_driver *drv) +{ + struct mdio_driver_common *mdiodrv = &drv->mdiodrv; + int retval; + + pr_info("mdio_driver_register: %s\n", mdiodrv->driver.name); + + mdiodrv->driver.bus = &mdio_bus_type; + mdiodrv->driver.probe = mdio_probe; + mdiodrv->driver.remove = mdio_remove; + + retval = driver_register(&mdiodrv->driver); + if (retval) { + pr_err("%s: Error %d in registering driver\n", + mdiodrv->driver.name, retval); + + return retval; + } + + return 0; +} +EXPORT_SYMBOL(mdio_driver_register); + +void mdio_driver_unregister(struct mdio_driver *drv) +{ + struct mdio_driver_common *mdiodrv = &drv->mdiodrv; + + driver_unregister(&mdiodrv->driver); +} +EXPORT_SYMBOL(mdio_driver_unregister); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 78628428ee28..af6cb6556cf9 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -264,6 +264,9 @@ static int phy_bus_match(struct device *dev, struct device_driver *drv) const int num_ids = ARRAY_SIZE(phydev->c45_ids.device_ids); int i; + if (!(phydrv->mdiodrv.flags & MDIO_DEVICE_IS_PHY)) + return 0; + if (phydrv->match_phy_device) return phydrv->match_phy_device(phydev); @@ -851,9 +854,11 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, */ if (!d->driver) { if (phydev->is_c45) - d->driver = &genphy_driver[GENPHY_DRV_10G].driver; + d->driver = + &genphy_driver[GENPHY_DRV_10G].mdiodrv.driver; else - d->driver = &genphy_driver[GENPHY_DRV_1G].driver; + d->driver = + &genphy_driver[GENPHY_DRV_1G].mdiodrv.driver; err = d->driver->probe(d); if (err >= 0) @@ -954,7 +959,8 @@ void phy_detach(struct phy_device *phydev) * real driver could be loaded */ for (i = 0; i < ARRAY_SIZE(genphy_driver); i++) { - if (phydev->mdio.dev.driver == &genphy_driver[i].driver) { + if (phydev->mdio.dev.driver == + &genphy_driver[i].mdiodrv.driver) { device_release_driver(&phydev->mdio.dev); break; } @@ -1598,13 +1604,14 @@ int phy_driver_register(struct phy_driver *new_driver, struct module *owner) { int retval; - new_driver->driver.name = new_driver->name; - new_driver->driver.bus = &mdio_bus_type; - new_driver->driver.probe = phy_probe; - new_driver->driver.remove = phy_remove; - new_driver->driver.owner = owner; + new_driver->mdiodrv.flags |= MDIO_DEVICE_IS_PHY; + new_driver->mdiodrv.driver.name = new_driver->name; + new_driver->mdiodrv.driver.bus = &mdio_bus_type; + new_driver->mdiodrv.driver.probe = phy_probe; + new_driver->mdiodrv.driver.remove = phy_remove; + new_driver->mdiodrv.driver.owner = owner; - retval = driver_register(&new_driver->driver); + retval = driver_register(&new_driver->mdiodrv.driver); if (retval) { pr_err("%s: Error %d in registering driver\n", new_driver->name, retval); @@ -1637,7 +1644,7 @@ EXPORT_SYMBOL(phy_drivers_register); void phy_driver_unregister(struct phy_driver *drv) { - driver_unregister(&drv->driver); + driver_unregister(&drv->mdiodrv.driver); } EXPORT_SYMBOL(phy_driver_unregister); diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 6febe2df76f9..c0a8f84d92db 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -92,6 +92,37 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi return 0; } +static int of_mdiobus_register_device(struct mii_bus *mdio, + struct device_node *child, + u32 addr) +{ + struct mdio_device *mdiodev; + int rc; + + mdiodev = mdio_device_create(mdio, addr); + if (!mdiodev || IS_ERR(mdiodev)) + return 1; + + /* Associate the OF node with the device structure so it + * can be looked up later. + */ + of_node_get(child); + mdiodev->dev.of_node = child; + + /* All data is now stored in the mdiodev struct; register it. */ + rc = mdio_device_register(mdiodev); + if (rc) { + mdio_device_free(mdiodev); + of_node_put(child); + return 1; + } + + dev_dbg(&mdio->dev, "registered mdio device %s at address %i\n", + child->name, addr); + + return 0; +} + int of_mdio_parse_addr(struct device *dev, const struct device_node *np) { u32 addr; @@ -179,6 +210,8 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) if (of_mdiobus_child_is_phy(child)) of_mdiobus_register_phy(mdio, child, addr); + else + of_mdiobus_register_device(mdio, child, addr); } if (!scanphys) diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 0690359e55a5..75f7fad0af4f 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -15,6 +15,7 @@ struct mii_bus; struct mdio_device { struct device dev; + const struct dev_pm_ops *pm_ops; struct mii_bus *bus; int (*bus_match)(struct device *dev, struct device_driver *drv); @@ -24,7 +25,37 @@ struct mdio_device { }; #define to_mdio_device(d) container_of(d, struct mdio_device, dev) +/* struct mdio_driver_common: Common to all MDIO drivers */ +struct mdio_driver_common { + struct device_driver driver; + int flags; +}; #define MDIO_DEVICE_FLAG_PHY 1 +#define to_mdio_common_driver(d) \ + container_of(d, struct mdio_driver_common, driver) + +/* struct mdio_driver: Generic MDIO driver */ +struct mdio_driver { + struct mdio_driver_common mdiodrv; + + /* + * Called during discovery. Used to set + * up device-specific structures, if any + */ + int (*probe)(struct mdio_device *mdiodev); + + /* Clears up any memory if needed */ + void (*remove)(struct mdio_device *mdiodev); +}; +#define to_mdio_driver(d) \ + container_of(to_mdio_common_driver(d), struct mdio_driver, mdiodrv) + +void mdio_device_free(struct mdio_device *mdiodev); +struct mdio_device *mdio_device_create(struct mii_bus *bus, int addr); +int mdio_device_register(struct mdio_device *mdiodev); +void mdio_device_remove(struct mdio_device *mdiodev); +int mdio_driver_register(struct mdio_driver *drv); +void mdio_driver_unregister(struct mdio_driver *drv); static inline bool mdio_phy_id_is_c45(int phy_id) { @@ -197,4 +228,23 @@ int mdiobus_unregister_device(struct mdio_device *mdiodev); bool mdiobus_is_registered_device(struct mii_bus *bus, int addr); struct phy_device *mdiobus_get_phy(struct mii_bus *bus, int addr); +/** + * module_mdio_driver() - Helper macro for registering mdio drivers + * + * Helper macro for MDIO drivers which do not do anything special in module + * init/exit. Each module may only use this macro once, and calling it + * replaces module_init() and module_exit(). + */ +#define mdio_module_driver(_mdio_driver) \ +static int __init mdio_module_init(void) \ +{ \ + return mdio_driver_register(&_mdio_driver); \ +} \ +module_init(mdio_module_init); \ +static void __exit mdio_module_exit(void) \ +{ \ + mdio_driver_unregister(&_mdio_driver); \ +} \ +module_exit(mdio_module_exit) + #endif /* __LINUX_MDIO_H__ */ diff --git a/include/linux/phy.h b/include/linux/phy.h index 49e4418822b3..d6f3641e7933 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -60,6 +60,7 @@ #define PHY_HAS_INTERRUPT 0x00000001 #define PHY_HAS_MAGICANEG 0x00000002 #define PHY_IS_INTERNAL 0x00000004 +#define MDIO_DEVICE_IS_PHY 0x80000000 /* Interface Mode definitions */ typedef enum { @@ -432,6 +433,7 @@ struct phy_device { /* struct phy_driver: Driver structure for a particular PHY type * + * driver_data: static driver data * phy_id: The result of reading the UID registers of this PHY * type, and ANDing them with the phy_id_mask. This driver * only works for PHYs with IDs which match this field @@ -441,7 +443,6 @@ struct phy_device { * by this PHY * flags: A bitfield defining certain other features this PHY * supports (like interrupts) - * driver_data: static driver data * * The drivers must implement config_aneg and read_status. All * other functions are optional. Note that none of these @@ -452,6 +453,7 @@ struct phy_device { * supported in the driver). */ struct phy_driver { + struct mdio_driver_common mdiodrv; u32 phy_id; char *name; unsigned int phy_id_mask; @@ -587,10 +589,9 @@ struct phy_driver { void (*get_strings)(struct phy_device *dev, u8 *data); void (*get_stats)(struct phy_device *dev, struct ethtool_stats *stats, u64 *data); - - struct device_driver driver; }; -#define to_phy_driver(d) container_of(d, struct phy_driver, driver) +#define to_phy_driver(d) container_of(to_mdio_common_driver(d), \ + struct phy_driver, mdiodrv) #define PHY_ANY_ID "MATCH ANY PHY" #define PHY_ANY_UID 0xffffffff -- cgit v1.2.3-59-g8ed1b From 711fdba37a3dd7ee487e28767f9f0e67144cbf80 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 6 Jan 2016 20:11:27 +0100 Subject: mdio: Abstract device_remove() and device_free() Make device_free and device_remove operations in the mdio device structure, so the core code does not need to differentiate between phy devices and generic mdio devices. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 23 +++++++++-------------- drivers/net/phy/mdio_device.c | 2 ++ drivers/net/phy/phy_device.c | 18 ++++++++++++++++++ include/linux/mdio.h | 4 ++++ 4 files changed, 33 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 0573cfd2116f..0be7b3d65f0f 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -299,6 +299,7 @@ static inline void of_mdiobus_link_mdiodev(struct mii_bus *mdio, */ int __mdiobus_register(struct mii_bus *bus, struct module *owner) { + struct mdio_device *mdiodev; int i, err; if (NULL == bus || NULL == bus->name || @@ -344,11 +345,12 @@ int __mdiobus_register(struct mii_bus *bus, struct module *owner) error: while (--i >= 0) { - struct phy_device *phydev = mdiobus_get_phy(bus, i); - if (phydev) { - phy_device_remove(phydev); - phy_device_free(phydev); - } + mdiodev = bus->mdio_map[i]; + if (!mdiodev) + continue; + + mdiodev->device_remove(mdiodev); + mdiodev->device_free(mdiodev); } device_del(&bus->dev); return err; @@ -358,7 +360,6 @@ EXPORT_SYMBOL(__mdiobus_register); void mdiobus_unregister(struct mii_bus *bus) { struct mdio_device *mdiodev; - struct phy_device *phydev; int i; BUG_ON(bus->state != MDIOBUS_REGISTERED); @@ -369,14 +370,8 @@ void mdiobus_unregister(struct mii_bus *bus) if (!mdiodev) continue; - if (!(mdiodev->flags & MDIO_DEVICE_FLAG_PHY)) { - phydev = container_of(mdiodev, struct phy_device, mdio); - phy_device_remove(phydev); - phy_device_free(phydev); - } else { - mdio_device_remove(mdiodev); - mdio_device_free(mdiodev); - } + mdiodev->device_remove(mdiodev); + mdiodev->device_free(mdiodev); } device_del(&bus->dev); } diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c index 64e3777c85b4..9c88e6749b9a 100644 --- a/drivers/net/phy/mdio_device.c +++ b/drivers/net/phy/mdio_device.c @@ -46,6 +46,8 @@ struct mdio_device *mdio_device_create(struct mii_bus *bus, int addr) mdiodev->dev.release = mdio_device_release; mdiodev->dev.parent = &bus->dev; mdiodev->dev.bus = &mdio_bus_type; + mdiodev->device_free = mdio_device_free; + mdiodev->device_remove = mdio_device_remove; mdiodev->bus = bus; mdiodev->addr = addr; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index af6cb6556cf9..319300627c0b 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -47,11 +47,27 @@ void phy_device_free(struct phy_device *phydev) } EXPORT_SYMBOL(phy_device_free); +static void phy_mdio_device_free(struct mdio_device *mdiodev) +{ + struct phy_device *phydev; + + phydev = container_of(mdiodev, struct phy_device, mdio); + phy_device_free(phydev); +} + static void phy_device_release(struct device *dev) { kfree(to_phy_device(dev)); } +static void phy_mdio_device_remove(struct mdio_device *mdiodev) +{ + struct phy_device *phydev; + + phydev = container_of(mdiodev, struct phy_device, mdio); + phy_device_remove(phydev); +} + enum genphy_driver { GENPHY_DRV_1G, GENPHY_DRV_10G, @@ -308,6 +324,8 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, int phy_id, mdiodev->bus_match = phy_bus_match; mdiodev->addr = addr; mdiodev->flags = MDIO_DEVICE_FLAG_PHY; + mdiodev->device_free = phy_mdio_device_free; + mdiodev->device_remove = phy_mdio_device_remove; dev->speed = 0; dev->duplex = -1; diff --git a/include/linux/mdio.h b/include/linux/mdio.h index 75f7fad0af4f..5bfd99d1a40a 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -18,7 +18,11 @@ struct mdio_device { const struct dev_pm_ops *pm_ops; struct mii_bus *bus; + int (*bus_match)(struct device *dev, struct device_driver *drv); + void (*device_free)(struct mdio_device *mdiodev); + void (*device_remove)(struct mdio_device *mdiodev); + /* Bus address of the MDIO device (0-31) */ int addr; int flags; -- cgit v1.2.3-59-g8ed1b From e6d8ecac9e68265aee9be711c5bd29406129666f Mon Sep 17 00:00:00 2001 From: Carlos Falgueras García Date: Tue, 5 Jan 2016 14:03:32 +0100 Subject: netfilter: nf_tables: Add new attributes into nft_set to store user data. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit User data is stored at after 'nft_set_ops' private data into 'data[]' flexible array. The field 'udata' points to user data and 'udlen' stores its length. Add new flag NFTA_SET_USERDATA. Signed-off-by: Carlos Falgueras García Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 4 ++++ include/uapi/linux/netfilter/nf_tables.h | 2 ++ net/netfilter/nf_tables_api.c | 21 ++++++++++++++++++++- 3 files changed, 26 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 0191fbb33a2f..f6b1daf2e698 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -291,6 +291,8 @@ void nft_unregister_set(struct nft_set_ops *ops); * @timeout: default timeout value in msecs * @gc_int: garbage collection interval in msecs * @policy: set parameterization (see enum nft_set_policies) + * @udlen: user data length + * @udata: user data * @ops: set ops * @pnet: network namespace * @flags: set flags @@ -310,6 +312,8 @@ struct nft_set { u64 timeout; u32 gc_int; u16 policy; + u16 udlen; + unsigned char *udata; /* runtime data below here */ const struct nft_set_ops *ops ____cacheline_aligned; possible_net_t pnet; diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 731288a039f6..03c28a402c63 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -291,6 +291,7 @@ enum nft_set_desc_attributes { * @NFTA_SET_ID: uniquely identifies a set in a transaction (NLA_U32) * @NFTA_SET_TIMEOUT: default timeout value (NLA_U64) * @NFTA_SET_GC_INTERVAL: garbage collection interval (NLA_U32) + * @NFTA_SET_USERDATA: user data (NLA_BINARY) */ enum nft_set_attributes { NFTA_SET_UNSPEC, @@ -306,6 +307,7 @@ enum nft_set_attributes { NFTA_SET_ID, NFTA_SET_TIMEOUT, NFTA_SET_GC_INTERVAL, + NFTA_SET_USERDATA, __NFTA_SET_MAX }; #define NFTA_SET_MAX (__NFTA_SET_MAX - 1) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index f5c397158e29..2011977cd79d 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2323,6 +2323,8 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { [NFTA_SET_ID] = { .type = NLA_U32 }, [NFTA_SET_TIMEOUT] = { .type = NLA_U64 }, [NFTA_SET_GC_INTERVAL] = { .type = NLA_U32 }, + [NFTA_SET_USERDATA] = { .type = NLA_BINARY, + .len = NFT_USERDATA_MAXLEN }, }; static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { @@ -2482,6 +2484,9 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, goto nla_put_failure; } + if (nla_put(skb, NFTA_SET_USERDATA, set->udlen, set->udata)) + goto nla_put_failure; + desc = nla_nest_start(skb, NFTA_SET_DESC); if (desc == NULL) goto nla_put_failure; @@ -2691,6 +2696,8 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, u64 timeout; u32 ktype, dtype, flags, policy, gc_int; struct nft_set_desc desc; + unsigned char *udata; + u16 udlen; int err; if (nla[NFTA_SET_TABLE] == NULL || @@ -2803,12 +2810,16 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, if (IS_ERR(ops)) return PTR_ERR(ops); + udlen = 0; + if (nla[NFTA_SET_USERDATA]) + udlen = nla_len(nla[NFTA_SET_USERDATA]); + size = 0; if (ops->privsize != NULL) size = ops->privsize(nla); err = -ENOMEM; - set = kzalloc(sizeof(*set) + size, GFP_KERNEL); + set = kzalloc(sizeof(*set) + size + udlen, GFP_KERNEL); if (set == NULL) goto err1; @@ -2817,6 +2828,12 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, if (err < 0) goto err2; + udata = NULL; + if (udlen) { + udata = set->data + size; + nla_memcpy(udata, nla[NFTA_SET_USERDATA], udlen); + } + INIT_LIST_HEAD(&set->bindings); write_pnet(&set->pnet, net); set->ops = ops; @@ -2827,6 +2844,8 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, set->flags = flags; set->size = desc.size; set->policy = policy; + set->udlen = udlen; + set->udata = udata; set->timeout = timeout; set->gc_int = gc_int; -- cgit v1.2.3-59-g8ed1b From 48f66c905a976bf0ff092fc24f08d9addd82a245 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 7 Jan 2016 21:34:24 +0100 Subject: netfilter: nft_ct: add byte/packet counter support If the accounting extension isn't present, we'll return a counter value of 0. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 2 ++ net/netfilter/nft_ct.c | 38 ++++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) (limited to 'include') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 03c28a402c63..be41ffc128b8 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -757,6 +757,8 @@ enum nft_ct_keys { NFT_CT_PROTO_SRC, NFT_CT_PROTO_DST, NFT_CT_LABELS, + NFT_CT_PKTS, + NFT_CT_BYTES, }; /** diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 8cbca3432f90..6f74109a7ed3 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -30,6 +31,18 @@ struct nft_ct { }; }; +static u64 nft_ct_get_eval_counter(const struct nf_conn_counter *c, + enum nft_ct_keys k, + enum ip_conntrack_dir d) +{ + if (d < IP_CT_DIR_MAX) + return k == NFT_CT_BYTES ? atomic64_read(&c[d].bytes) : + atomic64_read(&c[d].packets); + + return nft_ct_get_eval_counter(c, k, IP_CT_DIR_ORIGINAL) + + nft_ct_get_eval_counter(c, k, IP_CT_DIR_REPLY); +} + static void nft_ct_get_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) @@ -114,6 +127,17 @@ static void nft_ct_get_eval(const struct nft_expr *expr, NF_CT_LABELS_MAX_SIZE - size); return; } + case NFT_CT_BYTES: /* fallthrough */ + case NFT_CT_PKTS: { + const struct nf_conn_acct *acct = nf_conn_acct_find(ct); + u64 count = 0; + + if (acct) + count = nft_ct_get_eval_counter(acct->counter, + priv->key, priv->dir); + memcpy(dest, &count, sizeof(count)); + return; + } #endif default: break; @@ -291,6 +315,13 @@ static int nft_ct_get_init(const struct nft_ctx *ctx, return -EINVAL; len = FIELD_SIZEOF(struct nf_conntrack_tuple, src.u.all); break; + case NFT_CT_BYTES: + case NFT_CT_PKTS: + /* no direction? return sum of original + reply */ + if (tb[NFTA_CT_DIRECTION] == NULL) + priv->dir = IP_CT_DIR_MAX; + len = sizeof(u64); + break; default: return -EOPNOTSUPP; } @@ -373,6 +404,13 @@ static int nft_ct_get_dump(struct sk_buff *skb, const struct nft_expr *expr) case NFT_CT_PROTO_DST: if (nla_put_u8(skb, NFTA_CT_DIRECTION, priv->dir)) goto nla_put_failure; + break; + case NFT_CT_BYTES: + case NFT_CT_PKTS: + if (priv->dir < IP_CT_DIR_MAX && + nla_put_u8(skb, NFTA_CT_DIRECTION, priv->dir)) + goto nla_put_failure; + break; default: break; } -- cgit v1.2.3-59-g8ed1b From 0797cbd8e2741c69a4d416d8f669639a064db8d1 Mon Sep 17 00:00:00 2001 From: Lance Richardson Date: Wed, 6 Jan 2016 17:22:46 -0500 Subject: ipv4: eliminate endianness warnings in ip_fib.h fib_multipath_hash() computes a hash using __be32 values, force cast these to u32 to pacify sparse. Signed-off-by: Lance Richardson Signed-off-by: David S. Miller --- include/net/ip_fib.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 9f4df68105ab..7029527725dd 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -325,7 +325,8 @@ extern u32 fib_multipath_secret __read_mostly; static inline int fib_multipath_hash(__be32 saddr, __be32 daddr) { - return jhash_2words(saddr, daddr, fib_multipath_secret) >> 1; + return jhash_2words((__force u32)saddr, (__force u32)daddr, + fib_multipath_secret) >> 1; } void fib_select_multipath(struct fib_result *res, int hash); -- cgit v1.2.3-59-g8ed1b From 01dd194c387af5b3c4c1f6459d30f596565e466c Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 6 Jan 2016 22:32:16 +0100 Subject: bpf: cleanup bpf_prog_run_{save,clear}_cb helpers Move the details behind the cb[] access into a small helper to decouple and make them generic for bpf_prog_run_save_cb()/bpf_prog_run_clear_cb() that was introduced via commit ff936a04e5f2 ("bpf: fix cb access in socket filter programs"). Also add a comment to better clarify what is done in bpf_skb_cb(). Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/filter.h | 39 +++++++++++++++++++++++++++++---------- 1 file changed, 29 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/filter.h b/include/linux/filter.h index f5b5891ed1ba..43aa1f8855c7 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -350,25 +350,43 @@ struct sk_filter { #define BPF_PROG_RUN(filter, ctx) (*filter->bpf_func)(ctx, filter->insnsi) +#define BPF_SKB_CB_LEN QDISC_CB_PRIV_LEN + +static inline u8 *bpf_skb_cb(struct sk_buff *skb) +{ + /* eBPF programs may read/write skb->cb[] area to transfer meta + * data between tail calls. Since this also needs to work with + * tc, that scratch memory is mapped to qdisc_skb_cb's data area. + * + * In some socket filter cases, the cb unfortunately needs to be + * saved/restored so that protocol specific skb->cb[] data won't + * be lost. In any case, due to unpriviledged eBPF programs + * attached to sockets, we need to clear the bpf_skb_cb() area + * to not leak previous contents to user space. + */ + BUILD_BUG_ON(FIELD_SIZEOF(struct __sk_buff, cb) != BPF_SKB_CB_LEN); + BUILD_BUG_ON(FIELD_SIZEOF(struct __sk_buff, cb) != + FIELD_SIZEOF(struct qdisc_skb_cb, data)); + + return qdisc_skb_cb(skb)->data; +} + static inline u32 bpf_prog_run_save_cb(const struct bpf_prog *prog, struct sk_buff *skb) { - u8 *cb_data = qdisc_skb_cb(skb)->data; - u8 saved_cb[QDISC_CB_PRIV_LEN]; + u8 *cb_data = bpf_skb_cb(skb); + u8 cb_saved[BPF_SKB_CB_LEN]; u32 res; - BUILD_BUG_ON(FIELD_SIZEOF(struct __sk_buff, cb) != - QDISC_CB_PRIV_LEN); - if (unlikely(prog->cb_access)) { - memcpy(saved_cb, cb_data, sizeof(saved_cb)); - memset(cb_data, 0, sizeof(saved_cb)); + memcpy(cb_saved, cb_data, sizeof(cb_saved)); + memset(cb_data, 0, sizeof(cb_saved)); } res = BPF_PROG_RUN(prog, skb); if (unlikely(prog->cb_access)) - memcpy(cb_data, saved_cb, sizeof(saved_cb)); + memcpy(cb_data, cb_saved, sizeof(cb_saved)); return res; } @@ -376,10 +394,11 @@ static inline u32 bpf_prog_run_save_cb(const struct bpf_prog *prog, static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog, struct sk_buff *skb) { - u8 *cb_data = qdisc_skb_cb(skb)->data; + u8 *cb_data = bpf_skb_cb(skb); if (unlikely(prog->cb_access)) - memset(cb_data, 0, QDISC_CB_PRIV_LEN); + memset(cb_data, 0, BPF_SKB_CB_LEN); + return BPF_PROG_RUN(prog, skb); } -- cgit v1.2.3-59-g8ed1b From 4d41e12593a9a6c4aaf113d44c8c619067b2b0aa Mon Sep 17 00:00:00 2001 From: Elad Raz Date: Sun, 10 Jan 2016 21:06:22 +0100 Subject: switchdev: Adding MDB entry offload Define HW multicast entry: MAC and VID. Using a MAC address simplifies support for both IPV4 and IPv6. Signed-off-by: Elad Raz Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/switchdev.h | 11 +++++++++++ net/switchdev/switchdev.c | 2 ++ 2 files changed, 13 insertions(+) (limited to 'include') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 603ae2f88dbb..d451122e8404 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -68,6 +68,7 @@ enum switchdev_obj_id { SWITCHDEV_OBJ_ID_PORT_VLAN, SWITCHDEV_OBJ_ID_IPV4_FIB, SWITCHDEV_OBJ_ID_PORT_FDB, + SWITCHDEV_OBJ_ID_PORT_MDB, }; struct switchdev_obj { @@ -113,6 +114,16 @@ struct switchdev_obj_port_fdb { #define SWITCHDEV_OBJ_PORT_FDB(obj) \ container_of(obj, struct switchdev_obj_port_fdb, obj) +/* SWITCHDEV_OBJ_ID_PORT_MDB */ +struct switchdev_obj_port_mdb { + struct switchdev_obj obj; + unsigned char addr[ETH_ALEN]; + u16 vid; +}; + +#define SWITCHDEV_OBJ_PORT_MDB(obj) \ + container_of(obj, struct switchdev_obj_port_mdb, obj) + void switchdev_trans_item_enqueue(struct switchdev_trans *trans, void *data, void (*destructor)(void const *), struct switchdev_trans_item *tritem); diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index df790d3385a2..ebc661d3b6e3 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -345,6 +345,8 @@ static size_t switchdev_obj_size(const struct switchdev_obj *obj) return sizeof(struct switchdev_obj_ipv4_fib); 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: BUG(); } -- cgit v1.2.3-59-g8ed1b From 787d7ac308ff2279e4b2ea393ad4d990de486ef2 Mon Sep 17 00:00:00 2001 From: Hannes Frederic Sowa Date: Thu, 7 Jan 2016 14:28:39 +0100 Subject: udp: restrict offloads to one namespace udp tunnel offloads tend to aggregate datagrams based on inner headers. gro engine gets notified by tunnel implementations about possible offloads. The match is solely based on the port number. Imagine a tunnel bound to port 53, the offloading will look into all DNS packets and tries to aggregate them based on the inner data found within. This could lead to data corruption and malformed DNS packets. While this patch minimizes the problem and helps an administrator to find the issue by querying ip tunnel/fou, a better way would be to match on the specific destination ip address so if a user space socket is bound to the same address it will conflict. Cc: Tom Herbert Cc: Eric Dumazet Signed-off-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- drivers/net/geneve.c | 2 +- drivers/net/vxlan.c | 2 +- include/net/protocol.h | 2 +- net/ipv4/fou.c | 2 +- net/ipv4/udp_offload.c | 10 +++++++--- 5 files changed, 11 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 58efdec12f30..db96b0cbb8ba 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -376,7 +376,7 @@ static void geneve_notify_add_rx_port(struct geneve_sock *gs) int err; if (sa_family == AF_INET) { - err = udp_add_offload(&gs->udp_offloads); + err = udp_add_offload(sock_net(sk), &gs->udp_offloads); if (err) pr_warn("geneve: udp_add_offload failed with status %d\n", err); diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 405a7b6cca25..e1e147f2d6ce 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -621,7 +621,7 @@ static void vxlan_notify_add_rx_port(struct vxlan_sock *vs) int err; if (sa_family == AF_INET) { - err = udp_add_offload(&vs->udp_offloads); + err = udp_add_offload(net, &vs->udp_offloads); if (err) pr_warn("vxlan: udp_add_offload failed with status %d\n", err); } diff --git a/include/net/protocol.h b/include/net/protocol.h index d6fcc1fcdb5b..da689f5432de 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -107,7 +107,7 @@ int inet_del_offload(const struct net_offload *prot, unsigned char num); void inet_register_protosw(struct inet_protosw *p); void inet_unregister_protosw(struct inet_protosw *p); -int udp_add_offload(struct udp_offload *prot); +int udp_add_offload(struct net *net, struct udp_offload *prot); void udp_del_offload(struct udp_offload *prot); #if IS_ENABLED(CONFIG_IPV6) diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index bd903fe0f750..976f0dcf6991 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -498,7 +498,7 @@ static int fou_create(struct net *net, struct fou_cfg *cfg, sk->sk_allocation = GFP_ATOMIC; if (cfg->udp_config.family == AF_INET) { - err = udp_add_offload(&fou->udp_offloads); + err = udp_add_offload(net, &fou->udp_offloads); if (err) goto error; } diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index f9386160cbee..5d396b96ae8b 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -21,6 +21,7 @@ static struct udp_offload_priv __rcu *udp_offload_base __read_mostly; struct udp_offload_priv { struct udp_offload *offload; + possible_net_t net; struct rcu_head rcu; struct udp_offload_priv __rcu *next; }; @@ -241,13 +242,14 @@ out: return segs; } -int udp_add_offload(struct udp_offload *uo) +int udp_add_offload(struct net *net, struct udp_offload *uo) { struct udp_offload_priv *new_offload = kzalloc(sizeof(*new_offload), GFP_ATOMIC); if (!new_offload) return -ENOMEM; + write_pnet(&new_offload->net, net); new_offload->offload = uo; spin_lock(&udp_offload_lock); @@ -311,7 +313,8 @@ struct sk_buff **udp_gro_receive(struct sk_buff **head, struct sk_buff *skb, rcu_read_lock(); uo_priv = rcu_dereference(udp_offload_base); for (; uo_priv != NULL; uo_priv = rcu_dereference(uo_priv->next)) { - if (uo_priv->offload->port == uh->dest && + if (net_eq(read_pnet(&uo_priv->net), dev_net(skb->dev)) && + uo_priv->offload->port == uh->dest && uo_priv->offload->callbacks.gro_receive) goto unflush; } @@ -389,7 +392,8 @@ int udp_gro_complete(struct sk_buff *skb, int nhoff) uo_priv = rcu_dereference(udp_offload_base); for (; uo_priv != NULL; uo_priv = rcu_dereference(uo_priv->next)) { - if (uo_priv->offload->port == uh->dest && + if (net_eq(read_pnet(&uo_priv->net), dev_net(skb->dev)) && + uo_priv->offload->port == uh->dest && uo_priv->offload->callbacks.gro_complete) break; } -- cgit v1.2.3-59-g8ed1b From 13b287e8d1cad951634389f85b8c9b816bd3bb1e Mon Sep 17 00:00:00 2001 From: Nikolay Borisov Date: Thu, 7 Jan 2016 16:38:43 +0200 Subject: ipv4: Namespaceify tcp_keepalive_time sysctl knob Different net namespaces might have different requirements as to the keepalive time of tcp sockets. This might be required in cases where different firewall rules are in place which require tcp timeout sockets to be increased/decreased independently of the host. Signed-off-by: Nikolay Borisov Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 2 ++ include/net/tcp.h | 5 +++-- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 2 ++ net/ipv4/tcp_timer.c | 1 - 5 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index d75be32650ba..9e9bbebaebd1 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -94,6 +94,8 @@ struct netns_ipv4 { int sysctl_tcp_probe_threshold; u32 sysctl_tcp_probe_interval; + int sysctl_tcp_keepalive_time; + struct ping_group_range ping_group_range; atomic_t dev_addr_genid; diff --git a/include/net/tcp.h b/include/net/tcp.h index f33fecf4e282..cb4d4cf25744 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps; extern int sysctl_tcp_window_scaling; extern int sysctl_tcp_sack; extern int sysctl_tcp_fin_timeout; -extern int sysctl_tcp_keepalive_time; extern int sysctl_tcp_keepalive_probes; extern int sysctl_tcp_keepalive_intvl; extern int sysctl_tcp_syn_retries; @@ -1230,7 +1229,9 @@ static inline int keepalive_intvl_when(const struct tcp_sock *tp) static inline int keepalive_time_when(const struct tcp_sock *tp) { - return tp->keepalive_time ? : sysctl_tcp_keepalive_time; + struct net *net = sock_net((struct sock *)tp); + + return tp->keepalive_time ? : net->ipv4.sysctl_tcp_keepalive_time; } static inline int keepalive_probes(const struct tcp_sock *tp) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 41ff1f87dfd7..1886cc842871 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -336,13 +336,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_keepalive_time", - .data = &sysctl_tcp_keepalive_time, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_jiffies, - }, { .procname = "tcp_keepalive_probes", .data = &sysctl_tcp_keepalive_probes, @@ -961,6 +954,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_keepalive_time", + .data = &init_net.ipv4.sysctl_tcp_keepalive_time, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index fc4f72686705..6e14ff9a8580 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2385,6 +2385,8 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_probe_threshold = TCP_PROBE_THRESHOLD; net->ipv4.sysctl_tcp_probe_interval = TCP_PROBE_INTERVAL; + net->ipv4.sysctl_tcp_keepalive_time = TCP_KEEPALIVE_TIME; + return 0; fail: tcp_sk_exit(net); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 193ba1fa8a9a..166f27b43cc0 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -24,7 +24,6 @@ int sysctl_tcp_syn_retries __read_mostly = TCP_SYN_RETRIES; int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES; -int sysctl_tcp_keepalive_time __read_mostly = TCP_KEEPALIVE_TIME; int sysctl_tcp_keepalive_probes __read_mostly = TCP_KEEPALIVE_PROBES; int sysctl_tcp_keepalive_intvl __read_mostly = TCP_KEEPALIVE_INTVL; int sysctl_tcp_retries1 __read_mostly = TCP_RETR1; -- cgit v1.2.3-59-g8ed1b From 9bd6861bd4326e3afd3f14a9ec8a723771fb20bb Mon Sep 17 00:00:00 2001 From: Nikolay Borisov Date: Thu, 7 Jan 2016 16:38:44 +0200 Subject: ipv4: Namespecify tcp_keepalive_probes sysctl knob This is required to have full tcp keepalive mechanism namespace support. Signed-off-by: Nikolay Borisov Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 5 +++-- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_timer.c | 1 - 5 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 9e9bbebaebd1..6e26ea2d0374 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -95,6 +95,7 @@ struct netns_ipv4 { u32 sysctl_tcp_probe_interval; int sysctl_tcp_keepalive_time; + int sysctl_tcp_keepalive_probes; struct ping_group_range ping_group_range; diff --git a/include/net/tcp.h b/include/net/tcp.h index cb4d4cf25744..0646521400bf 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps; extern int sysctl_tcp_window_scaling; extern int sysctl_tcp_sack; extern int sysctl_tcp_fin_timeout; -extern int sysctl_tcp_keepalive_probes; extern int sysctl_tcp_keepalive_intvl; extern int sysctl_tcp_syn_retries; extern int sysctl_tcp_synack_retries; @@ -1236,7 +1235,9 @@ static inline int keepalive_time_when(const struct tcp_sock *tp) static inline int keepalive_probes(const struct tcp_sock *tp) { - return tp->keepalive_probes ? : sysctl_tcp_keepalive_probes; + struct net *net = sock_net((struct sock *)tp); + + return tp->keepalive_probes ? : net->ipv4.sysctl_tcp_keepalive_probes; } static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 1886cc842871..e99fbb77dba7 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -336,13 +336,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_keepalive_probes", - .data = &sysctl_tcp_keepalive_probes, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_keepalive_intvl", .data = &sysctl_tcp_keepalive_intvl, @@ -961,6 +954,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, + { + .procname = "tcp_keepalive_probes", + .data = &init_net.ipv4.sysctl_tcp_keepalive_probes, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 6e14ff9a8580..ed98de85871e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2386,6 +2386,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_probe_interval = TCP_PROBE_INTERVAL; net->ipv4.sysctl_tcp_keepalive_time = TCP_KEEPALIVE_TIME; + net->ipv4.sysctl_tcp_keepalive_probes = TCP_KEEPALIVE_PROBES; return 0; fail: diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 166f27b43cc0..0ccb120d591a 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -24,7 +24,6 @@ int sysctl_tcp_syn_retries __read_mostly = TCP_SYN_RETRIES; int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES; -int sysctl_tcp_keepalive_probes __read_mostly = TCP_KEEPALIVE_PROBES; int sysctl_tcp_keepalive_intvl __read_mostly = TCP_KEEPALIVE_INTVL; int sysctl_tcp_retries1 __read_mostly = TCP_RETR1; int sysctl_tcp_retries2 __read_mostly = TCP_RETR2; -- cgit v1.2.3-59-g8ed1b From b840d15d39128d08ed4486085e5507d2617b9ae1 Mon Sep 17 00:00:00 2001 From: Nikolay Borisov Date: Thu, 7 Jan 2016 16:38:45 +0200 Subject: ipv4: Namespecify the tcp_keepalive_intvl sysctl knob This is the final part required to namespaceify the tcp keep alive mechanism. Signed-off-by: Nikolay Borisov Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 5 +++-- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_timer.c | 1 - 5 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 6e26ea2d0374..2b7907a35568 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -96,6 +96,7 @@ struct netns_ipv4 { int sysctl_tcp_keepalive_time; int sysctl_tcp_keepalive_probes; + int sysctl_tcp_keepalive_intvl; struct ping_group_range ping_group_range; diff --git a/include/net/tcp.h b/include/net/tcp.h index 0646521400bf..a80255f4ca33 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps; extern int sysctl_tcp_window_scaling; extern int sysctl_tcp_sack; extern int sysctl_tcp_fin_timeout; -extern int sysctl_tcp_keepalive_intvl; extern int sysctl_tcp_syn_retries; extern int sysctl_tcp_synack_retries; extern int sysctl_tcp_retries1; @@ -1223,7 +1222,9 @@ void tcp_enter_memory_pressure(struct sock *sk); static inline int keepalive_intvl_when(const struct tcp_sock *tp) { - return tp->keepalive_intvl ? : sysctl_tcp_keepalive_intvl; + struct net *net = sock_net((struct sock *)tp); + + return tp->keepalive_intvl ? : net->ipv4.sysctl_tcp_keepalive_intvl; } static inline int keepalive_time_when(const struct tcp_sock *tp) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e99fbb77dba7..46ce410703b1 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -336,13 +336,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_keepalive_intvl", - .data = &sysctl_tcp_keepalive_intvl, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_jiffies, - }, { .procname = "tcp_retries1", .data = &sysctl_tcp_retries1, @@ -961,6 +954,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_keepalive_intvl", + .data = &init_net.ipv4.sysctl_tcp_keepalive_intvl, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ed98de85871e..65947c1f4733 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2387,6 +2387,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_keepalive_time = TCP_KEEPALIVE_TIME; net->ipv4.sysctl_tcp_keepalive_probes = TCP_KEEPALIVE_PROBES; + net->ipv4.sysctl_tcp_keepalive_intvl = TCP_KEEPALIVE_INTVL; return 0; fail: diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 0ccb120d591a..a4730a28b220 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -24,7 +24,6 @@ int sysctl_tcp_syn_retries __read_mostly = TCP_SYN_RETRIES; int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES; -int sysctl_tcp_keepalive_intvl __read_mostly = TCP_KEEPALIVE_INTVL; int sysctl_tcp_retries1 __read_mostly = TCP_RETR1; int sysctl_tcp_retries2 __read_mostly = TCP_RETR2; int sysctl_tcp_orphan_retries __read_mostly; -- cgit v1.2.3-59-g8ed1b From fdc5432a7b44ab7de17141beec19d946b9344e91 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 7 Jan 2016 15:50:22 +0100 Subject: net, sched: add skb_at_tc_ingress helper Add a skb_at_tc_ingress() as this will be needed elsewhere as well and can hide the ugly ifdef. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/net/sch_generic.h | 9 +++++++++ net/sched/cls_bpf.c | 6 +----- 2 files changed, 10 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index b2a8e6338576..636a362a0e03 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -407,6 +407,15 @@ bool tcf_destroy(struct tcf_proto *tp, bool force); void tcf_destroy_chain(struct tcf_proto __rcu **fl); int skb_do_redirect(struct sk_buff *); +static inline bool skb_at_tc_ingress(const struct sk_buff *skb) +{ +#ifdef CONFIG_NET_CLS_ACT + return G_TC_AT(skb->tc_verd) & AT_INGRESS; +#else + return false; +#endif +} + /* Reset all TX qdiscs greater then index of a device. */ static inline void qdisc_reset_all_tx_gt(struct net_device *dev, unsigned int i) { diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 5faaa5425f7b..b3c8bb4aeef5 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -79,12 +79,8 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res) { struct cls_bpf_head *head = rcu_dereference_bh(tp->root); + bool at_ingress = skb_at_tc_ingress(skb); struct cls_bpf_prog *prog; -#ifdef CONFIG_NET_CLS_ACT - bool at_ingress = G_TC_AT(skb->tc_verd) & AT_INGRESS; -#else - bool at_ingress = false; -#endif int ret = -1; if (unlikely(!skb_mac_header_was_set(skb))) -- cgit v1.2.3-59-g8ed1b From f8ffad69c9f8b8dfb0b633425d4ef4d2493ba61a Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 7 Jan 2016 15:50:23 +0100 Subject: bpf: add skb_postpush_rcsum and fix dev_forward_skb occasions Add a small helper skb_postpush_rcsum() and fix up redirect locations that need CHECKSUM_COMPLETE fixups on ingress. dev_forward_skb() expects a proper csum that covers also Ethernet header, f.e. since 2c26d34bbcc0 ("net/core: Handle csum for CHECKSUM_COMPLETE VXLAN forwarding"), we also do skb_postpull_rcsum() after pulling Ethernet header off via eth_type_trans(). When using eBPF in a netns setup f.e. with vxlan in collect metadata mode, I can trigger the following csum issue with an IPv6 setup: [ 505.144065] dummy1: hw csum failure [...] [ 505.144108] Call Trace: [ 505.144112] [] dump_stack+0x44/0x5c [ 505.144134] [] netdev_rx_csum_fault+0x3a/0x40 [ 505.144142] [] __skb_checksum_complete+0xcf/0xe0 [ 505.144149] [] nf_ip6_checksum+0xb2/0x120 [ 505.144161] [] icmpv6_error+0x17e/0x328 [nf_conntrack_ipv6] [ 505.144170] [] ? ip6t_do_table+0x2fa/0x645 [ip6_tables] [ 505.144177] [] ? ipv6_get_l4proto+0x65/0xd0 [nf_conntrack_ipv6] [ 505.144189] [] nf_conntrack_in+0xc2/0x5a0 [nf_conntrack] [ 505.144196] [] ipv6_conntrack_in+0x1c/0x20 [nf_conntrack_ipv6] [ 505.144204] [] nf_iterate+0x5d/0x70 [ 505.144210] [] nf_hook_slow+0x66/0xc0 [ 505.144218] [] ipv6_rcv+0x3f2/0x4f0 [ 505.144225] [] ? ip6_make_skb+0x1b0/0x1b0 [ 505.144232] [] __netif_receive_skb_core+0x36b/0x9a0 [ 505.144239] [] ? __netif_receive_skb+0x18/0x60 [ 505.144245] [] __netif_receive_skb+0x18/0x60 [ 505.144252] [] process_backlog+0x9f/0x140 [ 505.144259] [] net_rx_action+0x145/0x320 [...] What happens is that on ingress, we push Ethernet header back in, either from cls_bpf or right before skb_do_redirect(), but without updating csum. The "hw csum failure" can be fixed by using the new skb_postpush_rcsum() helper for the dev_forward_skb() case to correct the csum diff again. Thanks to Hannes Frederic Sowa for the csum_partial() idea! Fixes: 3896d655f4d4 ("bpf: introduce bpf_clone_redirect() helper") Fixes: 27b29f63058d ("bpf: add bpf_redirect() helper") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/skbuff.h | 17 +++++++++++++++++ net/core/filter.c | 17 +++++++++++++---- 2 files changed, 30 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 6b6bd42d6134..07f9ccd28654 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2805,6 +2805,23 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len); +static inline void skb_postpush_rcsum(struct sk_buff *skb, + const void *start, unsigned int len) +{ + /* For performing the reverse operation to skb_postpull_rcsum(), + * we can instead of ... + * + * skb->csum = csum_add(skb->csum, csum_partial(start, len, 0)); + * + * ... just use this equivalent version here to save a few + * instructions. Feeding csum of 0 in csum_partial() and later + * on adding skb->csum is equivalent to feed skb->csum in the + * first place. + */ + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->csum = csum_partial(start, len, skb->csum); +} + /** * pskb_trim_rcsum - trim received skb and update checksum * @skb: buffer to trim diff --git a/net/core/filter.c b/net/core/filter.c index 35e6fed28709..0db92b5e2cbf 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1368,8 +1368,9 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) /* skb_store_bits cannot return -EFAULT here */ skb_store_bits(skb, offset, ptr, len); - if (BPF_RECOMPUTE_CSUM(flags) && skb->ip_summed == CHECKSUM_COMPLETE) - skb->csum = csum_add(skb->csum, csum_partial(ptr, len, 0)); + if (BPF_RECOMPUTE_CSUM(flags)) + skb_postpush_rcsum(skb, ptr, len); + return 0; } @@ -1525,8 +1526,12 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) if (unlikely(!skb2)) return -ENOMEM; - if (BPF_IS_REDIRECT_INGRESS(flags)) + if (BPF_IS_REDIRECT_INGRESS(flags)) { + if (skb_at_tc_ingress(skb2)) + skb_postpush_rcsum(skb2, skb_mac_header(skb2), + skb2->mac_len); return dev_forward_skb(dev, skb2); + } skb2->dev = dev; skb_sender_cpu_clear(skb2); @@ -1569,8 +1574,12 @@ int skb_do_redirect(struct sk_buff *skb) return -EINVAL; } - if (BPF_IS_REDIRECT_INGRESS(ri->flags)) + if (BPF_IS_REDIRECT_INGRESS(ri->flags)) { + if (skb_at_tc_ingress(skb)) + skb_postpush_rcsum(skb, skb_mac_header(skb), + skb->mac_len); return dev_forward_skb(dev, skb); + } skb->dev = dev; skb_sender_cpu_clear(skb); -- cgit v1.2.3-59-g8ed1b From 1f211a1b929c804100e138c5d3d656992cfd5622 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 7 Jan 2016 22:29:47 +0100 Subject: net, sched: add clsact qdisc This work adds a generalization of the ingress qdisc as a qdisc holding only classifiers. The clsact qdisc works on ingress, but also on egress. In both cases, it's execution happens without taking the qdisc lock, and the main difference for the egress part compared to prior version of [1] is that this can be applied with _any_ underlying real egress qdisc (also classless ones). Besides solving the use-case of [1], that is, allowing for more programmability on assigning skb->priority for the mqprio case that is supported by most popular 10G+ NICs, it also opens up a lot more flexibility for other tc applications. The main work on classification can already be done at clsact egress time if the use-case allows and state stored for later retrieval f.e. again in skb->priority with major/minors (which is checked by most classful qdiscs before consulting tc_classify()) and/or in other skb fields like skb->tc_index for some light-weight post-processing to get to the eventual classid in case of a classful qdisc. Another use case is that the clsact egress part allows to have a central egress counterpart to the ingress classifiers, so that classifiers can easily share state (e.g. in cls_bpf via eBPF maps) for ingress and egress. Currently, default setups like mq + pfifo_fast would require for this to use, for example, prio qdisc instead (to get a tc_classify() run) and to duplicate the egress classifier for each queue. With clsact, it allows for leaving the setup as is, it can additionally assign skb->priority to put the skb in one of pfifo_fast's bands and it can share state with maps. Moreover, we can access the skb's dst entry (f.e. to retrieve tclassid) w/o the need to perform a skb_dst_force() to hold on to it any longer. In lwt case, we can also use this facility to setup dst metadata via cls_bpf (bpf_skb_set_tunnel_key()) without needing a real egress qdisc just for that (case of IFF_NO_QUEUE devices, for example). The realization can be done without any changes to the scheduler core framework. All it takes is that we have two a-priori defined minors/child classes, where we can mux between ingress and egress classifier list (dev->ingress_cl_list and dev->egress_cl_list, latter stored close to dev->_tx to avoid extra cacheline miss for moderate loads). The egress part is a bit similar modelled to handle_ing() and patched to a noop in case the functionality is not used. Both handlers are now called sch_handle_ingress() and sch_handle_egress(), code sharing among the two doesn't seem practical as there are various minor differences in both paths, so that making them conditional in a single handler would rather slow things down. Full compatibility to ingress qdisc is provided as well. Since both piggyback on TC_H_CLSACT, only one of them (ingress/clsact) can exist per netdevice, and thus ingress qdisc specific behaviour can be retained for user space. This means, either a user does 'tc qdisc add dev foo ingress' and configures ingress qdisc as usual, or the 'tc qdisc add dev foo clsact' alternative, where both, ingress and egress classifier can be configured as in the below example. ingress qdisc supports attaching classifier to any minor number whereas clsact has two fixed minors for muxing between the lists, therefore to not break user space setups, they are better done as two separate qdiscs. I decided to extend the sch_ingress module with clsact functionality so that commonly used code can be reused, the module is being aliased with sch_clsact so that it can be auto-loaded properly. Alternative would have been to add a flag when initializing ingress to alter its behaviour plus aliasing to a different name (as it's more than just ingress). However, the first would end up, based on the flag, choosing the new/old behaviour by calling different function implementations to handle each anyway, the latter would require to register ingress qdisc once again under different alias. So, this really begs to provide a minimal, cleaner approach to have Qdisc_ops and Qdisc_class_ops by its own that share callbacks used by both. Example, adding qdisc: # tc qdisc add dev foo clsact # tc qdisc show dev foo qdisc mq 0: root qdisc pfifo_fast 0: parent :1 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 qdisc pfifo_fast 0: parent :2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 qdisc pfifo_fast 0: parent :3 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 qdisc pfifo_fast 0: parent :4 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 qdisc clsact ffff: parent ffff:fff1 Adding filters (deleting, etc works analogous by specifying ingress/egress): # tc filter add dev foo ingress bpf da obj bar.o sec ingress # tc filter add dev foo egress bpf da obj bar.o sec egress # tc filter show dev foo ingress filter protocol all pref 49152 bpf filter protocol all pref 49152 bpf handle 0x1 bar.o:[ingress] direct-action # tc filter show dev foo egress filter protocol all pref 49152 bpf filter protocol all pref 49152 bpf handle 0x1 bar.o:[egress] direct-action A 'tc filter show dev foo' or 'tc filter show dev foo parent ffff:' will show an empty list for clsact. Either using the parent names (ingress/egress) or specifying the full major/minor will then show the related filter lists. Prior work on a mqprio prequeue() facility [1] was done mainly by John Fastabend. [1] http://patchwork.ozlabs.org/patch/512949/ Signed-off-by: Daniel Borkmann Acked-by: John Fastabend Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 +- include/linux/rtnetlink.h | 5 +++ include/uapi/linux/pkt_sched.h | 4 ++ net/Kconfig | 3 ++ net/core/dev.c | 82 +++++++++++++++++++++++++++++++++++---- net/sched/Kconfig | 14 +++++-- net/sched/cls_bpf.c | 2 +- net/sched/sch_ingress.c | 88 +++++++++++++++++++++++++++++++++++++++++- 8 files changed, 186 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 8d8e5ca951b4..2285596e7045 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1739,7 +1739,9 @@ struct net_device { #ifdef CONFIG_XPS struct xps_dev_maps __rcu *xps_maps; #endif - +#ifdef CONFIG_NET_CLS_ACT + struct tcf_proto __rcu *egress_cl_list; +#endif #ifdef CONFIG_NET_SWITCHDEV u32 offload_fwd_mark; #endif diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 4be5048b1fbe..c006cc900c44 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -84,6 +84,11 @@ void net_inc_ingress_queue(void); void net_dec_ingress_queue(void); #endif +#ifdef CONFIG_NET_EGRESS +void net_inc_egress_queue(void); +void net_dec_egress_queue(void); +#endif + extern void rtnetlink_init(void); extern void __rtnl_unlock(void); diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h index 8d2530daca9f..8cb18b44968e 100644 --- a/include/uapi/linux/pkt_sched.h +++ b/include/uapi/linux/pkt_sched.h @@ -72,6 +72,10 @@ struct tc_estimator { #define TC_H_UNSPEC (0U) #define TC_H_ROOT (0xFFFFFFFFU) #define TC_H_INGRESS (0xFFFFFFF1U) +#define TC_H_CLSACT TC_H_INGRESS + +#define TC_H_MIN_INGRESS 0xFFF2U +#define TC_H_MIN_EGRESS 0xFFF3U /* Need to corrospond to iproute2 tc/tc_core.h "enum link_layer" */ enum tc_link_layer { diff --git a/net/Kconfig b/net/Kconfig index 11f8c22af34d..174354618f8a 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -48,6 +48,9 @@ config COMPAT_NETLINK_MESSAGES config NET_INGRESS bool +config NET_EGRESS + bool + menu "Networking options" source "net/packet/Kconfig" diff --git a/net/core/dev.c b/net/core/dev.c index 914b4a24c654..0ca95d5d7af0 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1676,6 +1676,22 @@ void net_dec_ingress_queue(void) EXPORT_SYMBOL_GPL(net_dec_ingress_queue); #endif +#ifdef CONFIG_NET_EGRESS +static struct static_key egress_needed __read_mostly; + +void net_inc_egress_queue(void) +{ + static_key_slow_inc(&egress_needed); +} +EXPORT_SYMBOL_GPL(net_inc_egress_queue); + +void net_dec_egress_queue(void) +{ + static_key_slow_dec(&egress_needed); +} +EXPORT_SYMBOL_GPL(net_dec_egress_queue); +#endif + static struct static_key netstamp_needed __read_mostly; #ifdef HAVE_JUMP_LABEL /* We are not allowed to call static_key_slow_dec() from irq context @@ -3007,7 +3023,6 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, bool contended; int rc; - qdisc_pkt_len_init(skb); qdisc_calculate_pkt_len(skb, q); /* * Heuristic to force contended enqueues to serialize on a @@ -3100,6 +3115,49 @@ int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *skb) } EXPORT_SYMBOL(dev_loopback_xmit); +#ifdef CONFIG_NET_EGRESS +static struct sk_buff * +sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) +{ + struct tcf_proto *cl = rcu_dereference_bh(dev->egress_cl_list); + struct tcf_result cl_res; + + if (!cl) + return skb; + + /* skb->tc_verd and qdisc_skb_cb(skb)->pkt_len were already set + * earlier by the caller. + */ + qdisc_bstats_cpu_update(cl->q, skb); + + switch (tc_classify(skb, cl, &cl_res, false)) { + case TC_ACT_OK: + case TC_ACT_RECLASSIFY: + skb->tc_index = TC_H_MIN(cl_res.classid); + break; + case TC_ACT_SHOT: + qdisc_qstats_cpu_drop(cl->q); + *ret = NET_XMIT_DROP; + goto drop; + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + *ret = NET_XMIT_SUCCESS; +drop: + kfree_skb(skb); + return NULL; + case TC_ACT_REDIRECT: + /* No need to push/pop skb's mac_header here on egress! */ + skb_do_redirect(skb); + *ret = NET_XMIT_SUCCESS; + return NULL; + default: + break; + } + + return skb; +} +#endif /* CONFIG_NET_EGRESS */ + static inline int get_xps_queue(struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_XPS @@ -3226,6 +3284,17 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) skb_update_prio(skb); + qdisc_pkt_len_init(skb); +#ifdef CONFIG_NET_CLS_ACT + skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS); +# ifdef CONFIG_NET_EGRESS + if (static_key_false(&egress_needed)) { + skb = sch_handle_egress(skb, &rc, dev); + if (!skb) + goto out; + } +# endif +#endif /* If device/qdisc don't need skb->dst, release it right now while * its hot in this cpu cache. */ @@ -3247,9 +3316,6 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) txq = netdev_pick_tx(dev, skb, accel_priv); q = rcu_dereference_bh(txq->qdisc); -#ifdef CONFIG_NET_CLS_ACT - skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS); -#endif trace_net_dev_queue(skb); if (q->enqueue) { rc = __dev_xmit_skb(skb, q, dev, txq); @@ -3806,9 +3872,9 @@ int (*br_fdb_test_addr_hook)(struct net_device *dev, EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook); #endif -static inline struct sk_buff *handle_ing(struct sk_buff *skb, - struct packet_type **pt_prev, - int *ret, struct net_device *orig_dev) +static inline struct sk_buff * +sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, + struct net_device *orig_dev) { #ifdef CONFIG_NET_CLS_ACT struct tcf_proto *cl = rcu_dereference_bh(skb->dev->ingress_cl_list); @@ -4002,7 +4068,7 @@ another_round: skip_taps: #ifdef CONFIG_NET_INGRESS if (static_key_false(&ingress_needed)) { - skb = handle_ing(skb, &pt_prev, &ret, orig_dev); + skb = sch_handle_ingress(skb, &pt_prev, &ret, orig_dev); if (!skb) goto out; diff --git a/net/sched/Kconfig b/net/sched/Kconfig index daa33432b716..82830824fb1f 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -310,15 +310,21 @@ config NET_SCH_PIE If unsure, say N. config NET_SCH_INGRESS - tristate "Ingress Qdisc" + tristate "Ingress/classifier-action Qdisc" depends on NET_CLS_ACT select NET_INGRESS + select NET_EGRESS ---help--- - Say Y here if you want to use classifiers for incoming packets. + Say Y here if you want to use classifiers for incoming and/or outgoing + packets. This qdisc doesn't do anything else besides running classifiers, + which can also have actions attached to them. In case of outgoing packets, + classifiers that this qdisc holds are executed in the transmit path + before real enqueuing to an egress qdisc happens. + If unsure, say Y. - To compile this code as a module, choose M here: the - module will be called sch_ingress. + To compile this code as a module, choose M here: the module will be + called sch_ingress with alias of sch_clsact. config NET_SCH_PLUG tristate "Plug network traffic until release (PLUG)" diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index b3c8bb4aeef5..8dc84300ee79 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -291,7 +291,7 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog, prog->bpf_name = name; prog->filter = fp; - if (fp->dst_needed) + if (fp->dst_needed && !(tp->q->flags & TCQ_F_INGRESS)) netif_keep_dst(qdisc_dev(tp->q)); return 0; diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index e7c648fa9dc3..10adbc617905 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -1,4 +1,5 @@ -/* net/sched/sch_ingress.c - Ingress qdisc +/* net/sched/sch_ingress.c - Ingress and clsact qdisc + * * 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 @@ -98,17 +99,100 @@ static struct Qdisc_ops ingress_qdisc_ops __read_mostly = { .owner = THIS_MODULE, }; +static unsigned long clsact_get(struct Qdisc *sch, u32 classid) +{ + switch (TC_H_MIN(classid)) { + case TC_H_MIN(TC_H_MIN_INGRESS): + case TC_H_MIN(TC_H_MIN_EGRESS): + return TC_H_MIN(classid); + default: + return 0; + } +} + +static unsigned long clsact_bind_filter(struct Qdisc *sch, + unsigned long parent, u32 classid) +{ + return clsact_get(sch, classid); +} + +static struct tcf_proto __rcu **clsact_find_tcf(struct Qdisc *sch, + unsigned long cl) +{ + struct net_device *dev = qdisc_dev(sch); + + switch (cl) { + case TC_H_MIN(TC_H_MIN_INGRESS): + return &dev->ingress_cl_list; + case TC_H_MIN(TC_H_MIN_EGRESS): + return &dev->egress_cl_list; + default: + return NULL; + } +} + +static int clsact_init(struct Qdisc *sch, struct nlattr *opt) +{ + net_inc_ingress_queue(); + net_inc_egress_queue(); + + sch->flags |= TCQ_F_CPUSTATS; + + return 0; +} + +static void clsact_destroy(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + + tcf_destroy_chain(&dev->ingress_cl_list); + tcf_destroy_chain(&dev->egress_cl_list); + + net_dec_ingress_queue(); + net_dec_egress_queue(); +} + +static const struct Qdisc_class_ops clsact_class_ops = { + .leaf = ingress_leaf, + .get = clsact_get, + .put = ingress_put, + .walk = ingress_walk, + .tcf_chain = clsact_find_tcf, + .bind_tcf = clsact_bind_filter, + .unbind_tcf = ingress_put, +}; + +static struct Qdisc_ops clsact_qdisc_ops __read_mostly = { + .cl_ops = &clsact_class_ops, + .id = "clsact", + .init = clsact_init, + .destroy = clsact_destroy, + .dump = ingress_dump, + .owner = THIS_MODULE, +}; + static int __init ingress_module_init(void) { - return register_qdisc(&ingress_qdisc_ops); + int ret; + + ret = register_qdisc(&ingress_qdisc_ops); + if (!ret) { + ret = register_qdisc(&clsact_qdisc_ops); + if (ret) + unregister_qdisc(&ingress_qdisc_ops); + } + + return ret; } static void __exit ingress_module_exit(void) { unregister_qdisc(&ingress_qdisc_ops); + unregister_qdisc(&clsact_qdisc_ops); } module_init(ingress_module_init); module_exit(ingress_module_exit); +MODULE_ALIAS("sch_clsact"); MODULE_LICENSE("GPL"); -- cgit v1.2.3-59-g8ed1b From 712f4aad406bb1ed67f3f98d04c044191f0ff593 Mon Sep 17 00:00:00 2001 From: willy tarreau Date: Sun, 10 Jan 2016 07:54:56 +0100 Subject: unix: properly account for FDs passed over unix sockets It is possible for a process to allocate and accumulate far more FDs than the process' limit by sending them over a unix socket then closing them to keep the process' fd count low. This change addresses this problem by keeping track of the number of FDs in flight per user and preventing non-privileged processes from having more FDs in flight than their configured FD limit. Reported-by: socketpair@gmail.com Reported-by: Tetsuo Handa Mitigates: CVE-2013-4312 (Linux 2.0+) Suggested-by: Linus Torvalds Acked-by: Hannes Frederic Sowa Signed-off-by: Willy Tarreau Signed-off-by: David S. Miller --- include/linux/sched.h | 1 + net/unix/af_unix.c | 24 ++++++++++++++++++++---- net/unix/garbage.c | 13 ++++++++----- 3 files changed, 29 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index edad7a43edea..fbf25f19b3b5 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -830,6 +830,7 @@ struct user_struct { unsigned long mq_bytes; /* How many bytes can be allocated to mqueue? */ #endif unsigned long locked_shm; /* How many pages of mlocked shm ? */ + unsigned long unix_inflight; /* How many files in flight in unix sockets */ #ifdef CONFIG_KEYS struct key *uid_keyring; /* UID specific keyring */ diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index ef05cd9403d4..e3f85bc8b135 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1513,6 +1513,21 @@ static void unix_destruct_scm(struct sk_buff *skb) sock_wfree(skb); } +/* + * The "user->unix_inflight" variable is protected by the garbage + * collection lock, and we just read it locklessly here. If you go + * over the limit, there might be a tiny race in actually noticing + * it across threads. Tough. + */ +static inline bool too_many_unix_fds(struct task_struct *p) +{ + struct user_struct *user = current_user(); + + if (unlikely(user->unix_inflight > task_rlimit(p, RLIMIT_NOFILE))) + return !capable(CAP_SYS_RESOURCE) && !capable(CAP_SYS_ADMIN); + return false; +} + #define MAX_RECURSION_LEVEL 4 static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) @@ -1521,6 +1536,9 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) unsigned char max_level = 0; int unix_sock_count = 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]); @@ -1542,10 +1560,8 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) if (!UNIXCB(skb).fp) return -ENOMEM; - if (unix_sock_count) { - for (i = scm->fp->count - 1; i >= 0; i--) - unix_inflight(scm->fp->fp[i]); - } + for (i = scm->fp->count - 1; i >= 0; i--) + unix_inflight(scm->fp->fp[i]); return max_level; } diff --git a/net/unix/garbage.c b/net/unix/garbage.c index a73a226f2d33..8fcdc2283af5 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -120,11 +120,11 @@ void unix_inflight(struct file *fp) { struct sock *s = unix_get_socket(fp); + spin_lock(&unix_gc_lock); + if (s) { struct unix_sock *u = unix_sk(s); - spin_lock(&unix_gc_lock); - if (atomic_long_inc_return(&u->inflight) == 1) { BUG_ON(!list_empty(&u->link)); list_add_tail(&u->link, &gc_inflight_list); @@ -132,25 +132,28 @@ void unix_inflight(struct file *fp) BUG_ON(list_empty(&u->link)); } unix_tot_inflight++; - spin_unlock(&unix_gc_lock); } + fp->f_cred->user->unix_inflight++; + spin_unlock(&unix_gc_lock); } void unix_notinflight(struct file *fp) { struct sock *s = unix_get_socket(fp); + spin_lock(&unix_gc_lock); + if (s) { struct unix_sock *u = unix_sk(s); - spin_lock(&unix_gc_lock); BUG_ON(list_empty(&u->link)); if (atomic_long_dec_and_test(&u->inflight)) list_del_init(&u->link); unix_tot_inflight--; - spin_unlock(&unix_gc_lock); } + fp->f_cred->user->unix_inflight--; + spin_unlock(&unix_gc_lock); } static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *), -- cgit v1.2.3-59-g8ed1b From 781c53bc5d5628065a46c70f02f5a0450f5842f4 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 11 Jan 2016 01:16:38 +0100 Subject: bpf: export helper function flags and reject invalid ones Export flags used by eBPF helper functions through UAPI, so they can be used by programs (instead of them redefining all flags each time or just using the hard-coded values). It also gives a better overview what flags are used where and we can further get rid of the extra macros defined in filter.c. Moreover, reject invalid flags. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 16 ++++++++++++++++ net/core/filter.c | 37 +++++++++++++++++++++++-------------- 2 files changed, 39 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 8bed7f1176b8..d94797ce9a5a 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -273,6 +273,22 @@ enum bpf_func_id { __BPF_FUNC_MAX_ID, }; +/* All flags used by eBPF helper functions, placed here. */ + +/* BPF_FUNC_skb_store_bytes flags. */ +#define BPF_F_RECOMPUTE_CSUM (1ULL << 0) + +/* BPF_FUNC_l3_csum_replace and BPF_FUNC_l4_csum_replace flags. + * First 4 bits are for passing the header field size. + */ +#define BPF_F_HDR_FIELD_MASK 0xfULL + +/* BPF_FUNC_l4_csum_replace flags. */ +#define BPF_F_PSEUDO_HDR (1ULL << 4) + +/* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */ +#define BPF_F_INGRESS (1ULL << 0) + /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure */ diff --git a/net/core/filter.c b/net/core/filter.c index 0db92b5e2cbf..7c55cadc0f38 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1328,8 +1328,7 @@ int sk_reuseport_attach_bpf(u32 ufd, struct sock *sk) return 0; } -#define BPF_RECOMPUTE_CSUM(flags) ((flags) & 1) -#define BPF_LDST_LEN 16U +#define BPF_LDST_LEN 16U static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) { @@ -1340,6 +1339,9 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) char buf[BPF_LDST_LEN]; void *ptr; + if (unlikely(flags & ~(BPF_F_RECOMPUTE_CSUM))) + return -EINVAL; + /* bpf verifier guarantees that: * 'from' pointer points to bpf program stack * 'len' bytes of it were initialized @@ -1359,7 +1361,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) if (unlikely(!ptr)) return -EFAULT; - if (BPF_RECOMPUTE_CSUM(flags)) + if (flags & BPF_F_RECOMPUTE_CSUM) skb_postpull_rcsum(skb, ptr, len); memcpy(ptr, from, len); @@ -1368,7 +1370,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags) /* skb_store_bits cannot return -EFAULT here */ skb_store_bits(skb, offset, ptr, len); - if (BPF_RECOMPUTE_CSUM(flags)) + if (flags & BPF_F_RECOMPUTE_CSUM) skb_postpush_rcsum(skb, ptr, len); return 0; @@ -1415,15 +1417,14 @@ const struct bpf_func_proto bpf_skb_load_bytes_proto = { .arg4_type = ARG_CONST_STACK_SIZE, }; -#define BPF_HEADER_FIELD_SIZE(flags) ((flags) & 0x0f) -#define BPF_IS_PSEUDO_HEADER(flags) ((flags) & 0x10) - static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) { struct sk_buff *skb = (struct sk_buff *) (long) r1; int offset = (int) r2; __sum16 sum, *ptr; + if (unlikely(flags & ~(BPF_F_HDR_FIELD_MASK))) + return -EINVAL; if (unlikely((u32) offset > 0xffff)) return -EFAULT; @@ -1435,7 +1436,7 @@ static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) if (unlikely(!ptr)) return -EFAULT; - switch (BPF_HEADER_FIELD_SIZE(flags)) { + switch (flags & BPF_F_HDR_FIELD_MASK) { case 2: csum_replace2(ptr, from, to); break; @@ -1467,10 +1468,12 @@ const struct bpf_func_proto bpf_l3_csum_replace_proto = { static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) { struct sk_buff *skb = (struct sk_buff *) (long) r1; - bool is_pseudo = !!BPF_IS_PSEUDO_HEADER(flags); + bool is_pseudo = flags & BPF_F_PSEUDO_HDR; int offset = (int) r2; __sum16 sum, *ptr; + if (unlikely(flags & ~(BPF_F_PSEUDO_HDR | BPF_F_HDR_FIELD_MASK))) + return -EINVAL; if (unlikely((u32) offset > 0xffff)) return -EFAULT; @@ -1482,7 +1485,7 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags) if (unlikely(!ptr)) return -EFAULT; - switch (BPF_HEADER_FIELD_SIZE(flags)) { + switch (flags & BPF_F_HDR_FIELD_MASK) { case 2: inet_proto_csum_replace2(ptr, skb, from, to, is_pseudo); break; @@ -1511,13 +1514,14 @@ const struct bpf_func_proto bpf_l4_csum_replace_proto = { .arg5_type = ARG_ANYTHING, }; -#define BPF_IS_REDIRECT_INGRESS(flags) ((flags) & 1) - static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) { struct sk_buff *skb = (struct sk_buff *) (long) r1, *skb2; struct net_device *dev; + if (unlikely(flags & ~(BPF_F_INGRESS))) + return -EINVAL; + dev = dev_get_by_index_rcu(dev_net(skb->dev), ifindex); if (unlikely(!dev)) return -EINVAL; @@ -1526,7 +1530,7 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) if (unlikely(!skb2)) return -ENOMEM; - if (BPF_IS_REDIRECT_INGRESS(flags)) { + if (flags & BPF_F_INGRESS) { if (skb_at_tc_ingress(skb2)) skb_postpush_rcsum(skb2, skb_mac_header(skb2), skb2->mac_len); @@ -1553,12 +1557,17 @@ struct redirect_info { }; static DEFINE_PER_CPU(struct redirect_info, redirect_info); + static u64 bpf_redirect(u64 ifindex, u64 flags, u64 r3, u64 r4, u64 r5) { struct redirect_info *ri = this_cpu_ptr(&redirect_info); + if (unlikely(flags & ~(BPF_F_INGRESS))) + return TC_ACT_SHOT; + ri->ifindex = ifindex; ri->flags = flags; + return TC_ACT_REDIRECT; } @@ -1574,7 +1583,7 @@ int skb_do_redirect(struct sk_buff *skb) return -EINVAL; } - if (BPF_IS_REDIRECT_INGRESS(ri->flags)) { + if (ri->flags & BPF_F_INGRESS) { if (skb_at_tc_ingress(skb)) skb_postpush_rcsum(skb, skb_mac_header(skb), skb->mac_len); -- cgit v1.2.3-59-g8ed1b From c6c33454072fc9fe961e2b25f22a619e4fa98838 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 11 Jan 2016 01:16:39 +0100 Subject: bpf: support ipv6 for bpf_skb_{set,get}_tunnel_key After IPv6 support has recently been added to metadata dst and related encaps, add support for populating/reading it from an eBPF program. Commit d3aa45ce6b ("bpf: add helpers to access tunnel metadata") started with initial IPv4-only support back then (due to IPv6 metadata support not being available yet). To stay compatible with older programs, we need to test for the passed structure size. Also TOS and TTL support from the ip_tunnel_info key has been added. Tested with vxlan devs in collect meta data mode with IPv4, IPv6 and in compat mode over different network namespaces. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 10 ++++++- net/core/filter.c | 69 +++++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 71 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index d94797ce9a5a..aa6f8571de13 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -289,6 +289,9 @@ enum bpf_func_id { /* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */ #define BPF_F_INGRESS (1ULL << 0) +/* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */ +#define BPF_F_TUNINFO_IPV6 (1ULL << 0) + /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure */ @@ -312,7 +315,12 @@ struct __sk_buff { struct bpf_tunnel_key { __u32 tunnel_id; - __u32 remote_ipv4; + union { + __u32 remote_ipv4; + __u32 remote_ipv6[4]; + }; + __u8 tunnel_tos; + __u8 tunnel_ttl; }; #endif /* _UAPI__LINUX_BPF_H__ */ diff --git a/net/core/filter.c b/net/core/filter.c index 7c55cadc0f38..77cdfb455e7f 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1680,19 +1680,49 @@ bool bpf_helper_changes_skb_data(void *func) return false; } +static unsigned short bpf_tunnel_key_af(u64 flags) +{ + return flags & BPF_F_TUNINFO_IPV6 ? AF_INET6 : AF_INET; +} + static u64 bpf_skb_get_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5) { struct sk_buff *skb = (struct sk_buff *) (long) r1; struct bpf_tunnel_key *to = (struct bpf_tunnel_key *) (long) r2; - struct ip_tunnel_info *info = skb_tunnel_info(skb); + const struct ip_tunnel_info *info = skb_tunnel_info(skb); + u8 compat[sizeof(struct bpf_tunnel_key)]; - if (unlikely(size != sizeof(struct bpf_tunnel_key) || flags || !info)) - return -EINVAL; - if (ip_tunnel_info_af(info) != AF_INET) + if (unlikely(!info || (flags & ~(BPF_F_TUNINFO_IPV6)))) return -EINVAL; + if (ip_tunnel_info_af(info) != bpf_tunnel_key_af(flags)) + return -EPROTO; + if (unlikely(size != sizeof(struct bpf_tunnel_key))) { + switch (size) { + case offsetof(struct bpf_tunnel_key, remote_ipv6[1]): + /* Fixup deprecated structure layouts here, so we have + * a common path later on. + */ + if (ip_tunnel_info_af(info) != AF_INET) + return -EINVAL; + to = (struct bpf_tunnel_key *)compat; + break; + default: + return -EINVAL; + } + } to->tunnel_id = be64_to_cpu(info->key.tun_id); - to->remote_ipv4 = be32_to_cpu(info->key.u.ipv4.src); + to->tunnel_tos = info->key.tos; + to->tunnel_ttl = info->key.ttl; + + if (flags & BPF_F_TUNINFO_IPV6) + memcpy(to->remote_ipv6, &info->key.u.ipv6.src, + sizeof(to->remote_ipv6)); + else + to->remote_ipv4 = be32_to_cpu(info->key.u.ipv4.src); + + if (unlikely(size != sizeof(struct bpf_tunnel_key))) + memcpy((void *)(long) r2, to, size); return 0; } @@ -1714,10 +1744,25 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5) struct sk_buff *skb = (struct sk_buff *) (long) r1; struct bpf_tunnel_key *from = (struct bpf_tunnel_key *) (long) r2; struct metadata_dst *md = this_cpu_ptr(md_dst); + u8 compat[sizeof(struct bpf_tunnel_key)]; struct ip_tunnel_info *info; - if (unlikely(size != sizeof(struct bpf_tunnel_key) || flags)) + if (unlikely(flags & ~(BPF_F_TUNINFO_IPV6))) return -EINVAL; + if (unlikely(size != sizeof(struct bpf_tunnel_key))) { + switch (size) { + case offsetof(struct bpf_tunnel_key, remote_ipv6[1]): + /* Fixup deprecated structure layouts here, so we have + * a common path later on. + */ + memcpy(compat, from, size); + memset(compat + size, 0, sizeof(compat) - size); + from = (struct bpf_tunnel_key *)compat; + break; + default: + return -EINVAL; + } + } skb_dst_drop(skb); dst_hold((struct dst_entry *) md); @@ -1725,9 +1770,19 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5) info = &md->u.tun_info; info->mode = IP_TUNNEL_INFO_TX; + info->key.tun_flags = TUNNEL_KEY; info->key.tun_id = cpu_to_be64(from->tunnel_id); - info->key.u.ipv4.dst = cpu_to_be32(from->remote_ipv4); + info->key.tos = from->tunnel_tos; + info->key.ttl = from->tunnel_ttl; + + if (flags & BPF_F_TUNINFO_IPV6) { + info->mode |= IP_TUNNEL_INFO_IPV6; + memcpy(&info->key.u.ipv6.dst, from->remote_ipv6, + sizeof(from->remote_ipv6)); + } else { + info->key.u.ipv4.dst = cpu_to_be32(from->remote_ipv4); + } return 0; } -- cgit v1.2.3-59-g8ed1b From f0d22d1874730530a2ac304fd0888cb8a6864527 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:25:57 +0200 Subject: net/mlx5_core: Introduce flow steering autogrouped flow table When user add rule to autogrouped flow table, we search for flow group with the same match criteria, if we don't find such group then we create new flow group with the required match criteria and insert the rule to this group. We divide the flow table into required_groups + 1, in order to reserve a part of the flow table for rules which don't match any existing group. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 166 +++++++++++++++++++--- drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 5 + include/linux/mlx5/fs.h | 6 + 3 files changed, 158 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index f7d62fe595f6..7d24bbba58ba 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -85,6 +85,12 @@ static struct init_tree_node { } }; +enum fs_i_mutex_lock_class { + FS_MUTEX_GRANDPARENT, + FS_MUTEX_PARENT, + FS_MUTEX_CHILD +}; + static void del_rule(struct fs_node *node); static void del_flow_table(struct fs_node *node); static void del_flow_group(struct fs_node *node); @@ -119,10 +125,11 @@ static void tree_get_node(struct fs_node *node) atomic_inc(&node->refcount); } -static void nested_lock_ref_node(struct fs_node *node) +static void nested_lock_ref_node(struct fs_node *node, + enum fs_i_mutex_lock_class class) { if (node) { - mutex_lock_nested(&node->lock, SINGLE_DEPTH_NESTING); + mutex_lock_nested(&node->lock, class); atomic_inc(&node->refcount); } } @@ -481,9 +488,7 @@ struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, list_add_tail(&ft->node.list, &fs_prio->node.children); fs_prio->num_ft++; unlock_ref_node(&fs_prio->node); - return ft; - free_ft: kfree(ft); unlock_prio: @@ -491,8 +496,32 @@ unlock_prio: return ERR_PTR(err); } -struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, - u32 *fg_in) +struct mlx5_flow_table *mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns, + int prio, + int num_flow_table_entries, + int max_num_groups) +{ + struct mlx5_flow_table *ft; + + if (max_num_groups > num_flow_table_entries) + return ERR_PTR(-EINVAL); + + ft = mlx5_create_flow_table(ns, prio, num_flow_table_entries); + if (IS_ERR(ft)) + return ft; + + ft->autogroup.active = true; + ft->autogroup.required_groups = max_num_groups; + + return ft; +} + +/* Flow table should be locked */ +static struct mlx5_flow_group *create_flow_group_common(struct mlx5_flow_table *ft, + u32 *fg_in, + struct list_head + *prev_fg, + bool is_auto_fg) { struct mlx5_flow_group *fg; struct mlx5_core_dev *dev = get_dev(&ft->node); @@ -505,18 +534,33 @@ struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, if (IS_ERR(fg)) return fg; - lock_ref_node(&ft->node); err = mlx5_cmd_create_flow_group(dev, ft, fg_in, &fg->id); if (err) { kfree(fg); - unlock_ref_node(&ft->node); return ERR_PTR(err); } + + if (ft->autogroup.active) + ft->autogroup.num_groups++; /* Add node to tree */ - tree_init_node(&fg->node, 1, del_flow_group); + tree_init_node(&fg->node, !is_auto_fg, del_flow_group); tree_add_node(&fg->node, &ft->node); /* Add node to group list */ list_add(&fg->node.list, ft->node.children.prev); + + return fg; +} + +struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, + u32 *fg_in) +{ + struct mlx5_flow_group *fg; + + if (ft->autogroup.active) + return ERR_PTR(-EPERM); + + lock_ref_node(&ft->node); + fg = create_flow_group_common(ft, fg_in, &ft->node.children, false); unlock_ref_node(&ft->node); return fg; @@ -614,7 +658,63 @@ static struct fs_fte *create_fte(struct mlx5_flow_group *fg, return fte; } -/* Assuming parent fg(flow table) is locked */ +static struct mlx5_flow_group *create_autogroup(struct mlx5_flow_table *ft, + u8 match_criteria_enable, + u32 *match_criteria) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct list_head *prev = &ft->node.children; + unsigned int candidate_index = 0; + struct mlx5_flow_group *fg; + void *match_criteria_addr; + unsigned int group_size = 0; + u32 *in; + + if (!ft->autogroup.active) + return ERR_PTR(-ENOENT); + + in = mlx5_vzalloc(inlen); + if (!in) + return ERR_PTR(-ENOMEM); + + if (ft->autogroup.num_groups < ft->autogroup.required_groups) + /* We save place for flow groups in addition to max types */ + group_size = ft->max_fte / (ft->autogroup.required_groups + 1); + + /* ft->max_fte == ft->autogroup.max_types */ + if (group_size == 0) + group_size = 1; + + /* sorted by start_index */ + fs_for_each_fg(fg, ft) { + if (candidate_index + group_size > fg->start_index) + candidate_index = fg->start_index + fg->max_ftes; + else + break; + prev = &fg->node.list; + } + + if (candidate_index + group_size > ft->max_fte) { + fg = ERR_PTR(-ENOSPC); + goto out; + } + + MLX5_SET(create_flow_group_in, in, match_criteria_enable, + match_criteria_enable); + MLX5_SET(create_flow_group_in, in, start_flow_index, candidate_index); + MLX5_SET(create_flow_group_in, in, end_flow_index, candidate_index + + group_size - 1); + match_criteria_addr = MLX5_ADDR_OF(create_flow_group_in, + in, match_criteria); + memcpy(match_criteria_addr, match_criteria, + MLX5_ST_SZ_BYTES(fte_match_param)); + + fg = create_flow_group_common(ft, in, prev, true); +out: + kvfree(in); + return fg; +} + static struct mlx5_flow_rule *add_rule_fg(struct mlx5_flow_group *fg, u32 *match_value, u8 action, @@ -626,9 +726,9 @@ static struct mlx5_flow_rule *add_rule_fg(struct mlx5_flow_group *fg, struct mlx5_flow_table *ft; struct list_head *prev; - lock_ref_node(&fg->node); + nested_lock_ref_node(&fg->node, FS_MUTEX_PARENT); fs_for_each_fte(fte, fg) { - nested_lock_ref_node(&fte->node); + nested_lock_ref_node(&fte->node, FS_MUTEX_CHILD); if (compare_match_value(&fg->mask, match_value, &fte->val) && action == fte->action && flow_tag == fte->flow_tag) { rule = add_rule_fte(fte, fg, dest); @@ -669,6 +769,33 @@ unlock_fg: return rule; } +static struct mlx5_flow_rule *add_rule_to_auto_fg(struct mlx5_flow_table *ft, + u8 match_criteria_enable, + u32 *match_criteria, + u32 *match_value, + u8 action, + u32 flow_tag, + struct mlx5_flow_destination *dest) +{ + struct mlx5_flow_rule *rule; + struct mlx5_flow_group *g; + + g = create_autogroup(ft, match_criteria_enable, match_criteria); + if (IS_ERR(g)) + return (void *)g; + + rule = add_rule_fg(g, match_value, + action, flow_tag, dest); + if (IS_ERR(rule)) { + /* Remove assumes refcount > 0 and autogroup creates a group + * with a refcount = 0. + */ + tree_get_node(&g->node); + tree_remove_node(&g->node); + } + return rule; +} + struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, u8 match_criteria_enable, @@ -679,23 +806,24 @@ mlx5_add_flow_rule(struct mlx5_flow_table *ft, struct mlx5_flow_destination *dest) { struct mlx5_flow_group *g; - struct mlx5_flow_rule *rule = ERR_PTR(-EINVAL); + struct mlx5_flow_rule *rule; - tree_get_node(&ft->node); - lock_ref_node(&ft->node); + nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT); fs_for_each_fg(g, ft) if (compare_match_criteria(g->mask.match_criteria_enable, match_criteria_enable, g->mask.match_criteria, match_criteria)) { - unlock_ref_node(&ft->node); rule = add_rule_fg(g, match_value, action, flow_tag, dest); - goto put; + if (!IS_ERR(rule) || PTR_ERR(rule) != -ENOSPC) + goto unlock; } + + rule = add_rule_to_auto_fg(ft, match_criteria_enable, match_criteria, + match_value, action, flow_tag, dest); +unlock: unlock_ref_node(&ft->node); -put: - tree_put_node(&ft->node); return rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 4ebb97fd5544..0f98257c0d31 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -77,6 +77,11 @@ struct mlx5_flow_table { unsigned int max_fte; unsigned int level; enum fs_flow_table_type type; + struct { + bool active; + unsigned int required_groups; + unsigned int num_groups; + } autogroup; }; /* Type of children is mlx5_flow_rule */ diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index bc7ad019afde..06ac6e8fccfa 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -61,6 +61,12 @@ struct mlx5_flow_namespace * mlx5_get_flow_namespace(struct mlx5_core_dev *dev, enum mlx5_flow_namespace_type type); +struct mlx5_flow_table * +mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns, + int prio, + int num_flow_table_entries, + int max_num_groups); + struct mlx5_flow_table * mlx5_create_flow_table(struct mlx5_flow_namespace *ns, int prio, -- cgit v1.2.3-59-g8ed1b From 2cc43b494a6c30ec0e554ea91ce763c97069e8cc Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:25:59 +0200 Subject: net/mlx5_core: Managing root flow table The root Flow Table for each Flow Table Type is defined, by default, as the Flow Table with level 0. In order not to use an empty flow tables and introduce new hops, but still preserve space for flow-tables that have a priority greater(lower number) than the current flow table, we introduce this new set root flow table command. This command tells the HW to start matching packets from the assigned root flow table. This command is used when we create new flow table with level lower than the current lowest flow table or it is the first flow table. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 18 +++++ drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h | 2 + drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 97 ++++++++++++++++++++--- drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 6 ++ include/linux/mlx5/mlx5_ifc.h | 31 +++++++- 5 files changed, 144 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index 5096f4f336bd..d8b1195fba3d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -38,6 +38,24 @@ #include "fs_cmd.h" #include "mlx5_core.h" +int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft) +{ + u32 in[MLX5_ST_SZ_DW(set_flow_table_root_in)]; + u32 out[MLX5_ST_SZ_DW(set_flow_table_root_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(set_flow_table_root_in, in, opcode, + MLX5_CMD_OP_SET_FLOW_TABLE_ROOT); + MLX5_SET(set_flow_table_root_in, in, table_type, ft->type); + MLX5_SET(set_flow_table_root_in, in, table_id, ft->id); + + memset(out, 0, sizeof(out)); + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, + sizeof(out)); +} + int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, enum fs_flow_table_type type, unsigned int level, unsigned int log_size, unsigned int *table_id) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h index f39304ede186..70d18ec145c2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -62,4 +62,6 @@ int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, struct mlx5_flow_table *ft, unsigned int index); +int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft); #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index c5a96e6abe0d..64bdb54041d1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -510,6 +510,29 @@ static struct mlx5_flow_table *find_prev_chained_ft(struct fs_prio *prio) return find_closest_ft(prio, true); } +static int update_root_ft_create(struct mlx5_flow_table *ft, struct fs_prio + *prio) +{ + struct mlx5_flow_root_namespace *root = find_root(&prio->node); + int min_level = INT_MAX; + int err; + + if (root->root_ft) + min_level = root->root_ft->level; + + if (ft->level >= min_level) + return 0; + + err = mlx5_cmd_update_root_ft(root->dev, ft); + if (err) + mlx5_core_warn(root->dev, "Update root flow table of id=%u failed\n", + ft->id); + else + root->root_ft = ft; + + return err; +} + struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, int prio, int max_fte) @@ -526,14 +549,15 @@ struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, return ERR_PTR(-ENODEV); } + mutex_lock(&root->chain_lock); fs_prio = find_prio(ns, prio); - if (!fs_prio) - return ERR_PTR(-EINVAL); - - lock_ref_node(&fs_prio->node); + if (!fs_prio) { + err = -EINVAL; + goto unlock_root; + } if (fs_prio->num_ft == fs_prio->max_ft) { err = -ENOSPC; - goto unlock_prio; + goto unlock_root; } ft = alloc_flow_table(find_next_free_level(fs_prio), @@ -541,7 +565,7 @@ struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, root->table_type); if (!ft) { err = -ENOMEM; - goto unlock_prio; + goto unlock_root; } tree_init_node(&ft->node, 1, del_flow_table); @@ -551,15 +575,25 @@ struct mlx5_flow_table *mlx5_create_flow_table(struct mlx5_flow_namespace *ns, if (err) goto free_ft; + if (MLX5_CAP_FLOWTABLE(root->dev, + flow_table_properties_nic_receive.modify_root)) { + err = update_root_ft_create(ft, fs_prio); + if (err) + goto destroy_ft; + } + lock_ref_node(&fs_prio->node); tree_add_node(&ft->node, &fs_prio->node); list_add_tail(&ft->node.list, &fs_prio->node.children); fs_prio->num_ft++; unlock_ref_node(&fs_prio->node); + mutex_unlock(&root->chain_lock); return ft; +destroy_ft: + mlx5_cmd_destroy_flow_table(root->dev, ft); free_ft: kfree(ft); -unlock_prio: - unlock_ref_node(&fs_prio->node); +unlock_root: + mutex_unlock(&root->chain_lock); return ERR_PTR(err); } @@ -899,13 +933,57 @@ void mlx5_del_flow_rule(struct mlx5_flow_rule *rule) tree_remove_node(&rule->node); } +/* Assuming prio->node.children(flow tables) is sorted by level */ +static struct mlx5_flow_table *find_next_ft(struct mlx5_flow_table *ft) +{ + struct fs_prio *prio; + + fs_get_obj(prio, ft->node.parent); + + if (!list_is_last(&ft->node.list, &prio->node.children)) + return list_next_entry(ft, node.list); + return find_next_chained_ft(prio); +} + +static int update_root_ft_destroy(struct mlx5_flow_table *ft) +{ + struct mlx5_flow_root_namespace *root = find_root(&ft->node); + struct mlx5_flow_table *new_root_ft = NULL; + + if (root->root_ft != ft) + return 0; + + new_root_ft = find_next_ft(ft); + if (new_root_ft) { + int err = mlx5_cmd_update_root_ft(root->dev, new_root_ft); + + if (err) { + mlx5_core_warn(root->dev, "Update root flow table of id=%u failed\n", + ft->id); + return err; + } + root->root_ft = new_root_ft; + } + return 0; +} + int mlx5_destroy_flow_table(struct mlx5_flow_table *ft) { + struct mlx5_flow_root_namespace *root = find_root(&ft->node); + int err = 0; + + mutex_lock(&root->chain_lock); + err = update_root_ft_destroy(ft); + if (err) { + mutex_unlock(&root->chain_lock); + return err; + } if (tree_remove_node(&ft->node)) mlx5_core_warn(get_dev(&ft->node), "Flow table %d wasn't destroyed, refcount > 1\n", ft->id); + mutex_unlock(&root->chain_lock); - return 0; + return err; } void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) @@ -1072,6 +1150,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev ns = &root_ns->ns; fs_init_namespace(ns); + mutex_init(&root_ns->chain_lock); tree_init_node(&ns->node, 1, NULL); tree_add_node(&ns->node, NULL); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 0f98257c0d31..1a2e08bad529 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -129,6 +129,9 @@ struct mlx5_flow_root_namespace { struct mlx5_flow_namespace ns; enum fs_flow_table_type table_type; struct mlx5_core_dev *dev; + struct mlx5_flow_table *root_ft; + /* Should be held when chaining flow tables */ + struct mutex chain_lock; }; int mlx5_init_fs(struct mlx5_core_dev *dev); @@ -148,6 +151,9 @@ void mlx5_cleanup_fs(struct mlx5_core_dev *dev); #define fs_for_each_prio(pos, ns) \ fs_list_for_each_entry(pos, &(ns)->node.children) +#define fs_for_each_ft(pos, prio) \ + fs_list_for_each_entry(pos, &(prio)->node.children) + #define fs_for_each_fg(pos, ft) \ fs_list_for_each_entry(pos, &(ft)->node.children) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 1780a85a8797..323e713c44ba 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -185,6 +185,7 @@ enum { MLX5_CMD_OP_MODIFY_RQT = 0x917, MLX5_CMD_OP_DESTROY_RQT = 0x918, MLX5_CMD_OP_QUERY_RQT = 0x919, + MLX5_CMD_OP_SET_FLOW_TABLE_ROOT = 0x92f, MLX5_CMD_OP_CREATE_FLOW_TABLE = 0x930, MLX5_CMD_OP_DESTROY_FLOW_TABLE = 0x931, MLX5_CMD_OP_QUERY_FLOW_TABLE = 0x932, @@ -258,7 +259,8 @@ struct mlx5_ifc_flow_table_prop_layout_bits { u8 ft_support[0x1]; u8 reserved_0[0x2]; u8 flow_modify_en[0x1]; - u8 reserved_1[0x1c]; + u8 modify_root[0x1]; + u8 reserved_1[0x1b]; u8 reserved_2[0x2]; u8 log_max_ft_size[0x6]; @@ -6946,4 +6948,31 @@ union mlx5_ifc_uplink_pci_interface_document_bits { u8 reserved_0[0x20060]; }; +struct mlx5_ifc_set_flow_table_root_out_bits { + u8 status[0x8]; + u8 reserved_0[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_1[0x40]; +}; + +struct mlx5_ifc_set_flow_table_root_in_bits { + u8 opcode[0x10]; + u8 reserved_0[0x10]; + + u8 reserved_1[0x10]; + u8 op_mod[0x10]; + + u8 reserved_2[0x40]; + + u8 table_type[0x8]; + u8 reserved_3[0x18]; + + u8 reserved_4[0x8]; + u8 table_id[0x18]; + + u8 reserved_5[0x140]; +}; + #endif /* MLX5_IFC_H */ -- cgit v1.2.3-59-g8ed1b From 34a40e689393a6b13673ab395a9a4d063d249fe9 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:26:00 +0200 Subject: net/mlx5_core: Introduce modify flow table command Introduce the modify flow table command. This command is used when we want to change the next flow table of an existing flow table. The next flow table is defined as the table we search (in order to find a match), if we couldn't find a match in any of the flow table entries in the current flow table. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 27 ++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h | 4 ++ include/linux/mlx5/mlx5_ifc.h | 56 ++++++++++++++++++++++-- 3 files changed, 83 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index d8b1195fba3d..2b5562553f2d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -101,6 +101,33 @@ int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, sizeof(out)); } +int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft) +{ + u32 in[MLX5_ST_SZ_DW(modify_flow_table_in)]; + u32 out[MLX5_ST_SZ_DW(modify_flow_table_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(modify_flow_table_in, in, opcode, + MLX5_CMD_OP_MODIFY_FLOW_TABLE); + MLX5_SET(modify_flow_table_in, in, table_type, ft->type); + MLX5_SET(modify_flow_table_in, in, table_id, ft->id); + MLX5_SET(modify_flow_table_in, in, modify_field_select, + MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID); + if (next_ft) { + MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1); + MLX5_SET(modify_flow_table_in, in, table_miss_id, next_ft->id); + } else { + MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0); + } + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, + sizeof(out)); +} + int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, struct mlx5_flow_table *ft, u32 *in, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h index 70d18ec145c2..1ae9b685c783 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -40,6 +40,10 @@ int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, struct mlx5_flow_table *ft); +int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft); + int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, struct mlx5_flow_table *ft, u32 *in, unsigned int *group_id); diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 323e713c44ba..7f166955d4c9 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -194,7 +194,8 @@ enum { MLX5_CMD_OP_QUERY_FLOW_GROUP = 0x935, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY = 0x936, MLX5_CMD_OP_QUERY_FLOW_TABLE_ENTRY = 0x937, - MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY = 0x938 + MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY = 0x938, + MLX5_CMD_OP_MODIFY_FLOW_TABLE = 0x93c }; struct mlx5_ifc_flow_table_fields_supported_bits { @@ -260,7 +261,9 @@ struct mlx5_ifc_flow_table_prop_layout_bits { u8 reserved_0[0x2]; u8 flow_modify_en[0x1]; u8 modify_root[0x1]; - u8 reserved_1[0x1b]; + u8 identified_miss_table_mode[0x1]; + u8 flow_table_modify[0x1]; + u8 reserved_1[0x19]; u8 reserved_2[0x2]; u8 log_max_ft_size[0x6]; @@ -5669,12 +5672,16 @@ struct mlx5_ifc_create_flow_table_in_bits { u8 reserved_4[0x20]; - u8 reserved_5[0x8]; + u8 reserved_5[0x4]; + u8 table_miss_mode[0x4]; u8 level[0x8]; u8 reserved_6[0x8]; u8 log_size[0x8]; - u8 reserved_7[0x120]; + u8 reserved_7[0x8]; + u8 table_miss_id[0x18]; + + u8 reserved_8[0x100]; }; struct mlx5_ifc_create_flow_group_out_bits { @@ -6975,4 +6982,45 @@ struct mlx5_ifc_set_flow_table_root_in_bits { u8 reserved_5[0x140]; }; +enum { + MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID = 0x1, +}; + +struct mlx5_ifc_modify_flow_table_out_bits { + u8 status[0x8]; + u8 reserved_0[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_1[0x40]; +}; + +struct mlx5_ifc_modify_flow_table_in_bits { + u8 opcode[0x10]; + u8 reserved_0[0x10]; + + u8 reserved_1[0x10]; + u8 op_mod[0x10]; + + u8 reserved_2[0x20]; + + u8 reserved_3[0x10]; + u8 modify_field_select[0x10]; + + u8 table_type[0x8]; + u8 reserved_4[0x18]; + + u8 reserved_5[0x8]; + u8 table_id[0x18]; + + u8 reserved_6[0x4]; + u8 table_miss_mode[0x4]; + u8 reserved_7[0x18]; + + u8 reserved_8[0x8]; + u8 table_miss_id[0x18]; + + u8 reserved_9[0x100]; +}; + #endif /* MLX5_IFC_H */ -- cgit v1.2.3-59-g8ed1b From 4cbdd30ed5c8bc5cf40813b025b4fb57b376a592 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:26:04 +0200 Subject: net/mlx5_core: Enable flow steering support for the IB driver When the driver is loaded, we create flow steering namespace for kernel bypass with nine priorities and another namespace for leftovers(in order to catch packets that weren't matched). Verbs applications will use these priorities. we found nine as a number that balances the requirements from the user and retains performance. The bypass namespace is used by verbs applications that want to bypass the kernel networking stack. The leftovers namespace is used by verbs applications and the sniffer in order to catch packets that weren't handled by any preceding rules. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 55 +++++++++++++++++++---- include/linux/mlx5/device.h | 2 + include/linux/mlx5/fs.h | 2 + 3 files changed, 51 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 96e287a326ae..757725bf48a8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -40,18 +40,19 @@ #define INIT_TREE_NODE_ARRAY_SIZE(...) (sizeof((struct init_tree_node[]){__VA_ARGS__}) /\ sizeof(struct init_tree_node)) -#define ADD_PRIO(min_level_val, max_ft_val, caps_val,\ +#define ADD_PRIO(num_prios_val, min_level_val, max_ft_val, caps_val,\ ...) {.type = FS_TYPE_PRIO,\ .min_ft_level = min_level_val,\ .max_ft = max_ft_val,\ + .num_leaf_prios = num_prios_val,\ .caps = caps_val,\ .children = (struct init_tree_node[]) {__VA_ARGS__},\ .ar_size = INIT_TREE_NODE_ARRAY_SIZE(__VA_ARGS__) \ } -#define ADD_FT_PRIO(max_ft_val, ...)\ - ADD_PRIO(0, max_ft_val, {},\ - __VA_ARGS__)\ +#define ADD_MULTIPLE_PRIO(num_prios_val, max_ft_val, ...)\ + ADD_PRIO(num_prios_val, 0, max_ft_val, {},\ + __VA_ARGS__)\ #define ADD_NS(...) {.type = FS_TYPE_NAMESPACE,\ .children = (struct init_tree_node[]) {__VA_ARGS__},\ @@ -66,7 +67,14 @@ #define FS_REQUIRED_CAPS(...) {.arr_sz = INIT_CAPS_ARRAY_SIZE(__VA_ARGS__), \ .caps = (long[]) {__VA_ARGS__} } +#define LEFTOVERS_MAX_FT 1 +#define LEFTOVERS_NUM_PRIOS 1 +#define BY_PASS_PRIO_MAX_FT 1 +#define BY_PASS_MIN_LEVEL (KENREL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ + LEFTOVERS_MAX_FT) + #define KERNEL_MAX_FT 2 +#define KERNEL_NUM_PRIOS 1 #define KENREL_MIN_LEVEL 2 struct node_caps { @@ -79,14 +87,27 @@ static struct init_tree_node { int ar_size; struct node_caps caps; int min_ft_level; + int num_leaf_prios; int prio; int max_ft; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 1, + .ar_size = 3, .children = (struct init_tree_node[]) { - ADD_PRIO(KENREL_MIN_LEVEL, 0, {}, - ADD_NS(ADD_FT_PRIO(KERNEL_MAX_FT))), + ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, + FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), + FS_CAP(flow_table_properties_nic_receive.modify_root), + FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), + FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_MAX_FT))), + ADD_PRIO(0, KENREL_MIN_LEVEL, 0, {}, + ADD_NS(ADD_MULTIPLE_PRIO(KERNEL_NUM_PRIOS, KERNEL_MAX_FT))), + ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, + FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), + FS_CAP(flow_table_properties_nic_receive.modify_root), + FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), + FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + ADD_NS(ADD_MULTIPLE_PRIO(LEFTOVERS_NUM_PRIOS, LEFTOVERS_MAX_FT))), } }; @@ -1098,8 +1119,10 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, return NULL; switch (type) { + case MLX5_FLOW_NAMESPACE_BYPASS: case MLX5_FLOW_NAMESPACE_KERNEL: - prio = 0; + case MLX5_FLOW_NAMESPACE_LEFTOVERS: + prio = type; break; case MLX5_FLOW_NAMESPACE_FDB: if (dev->priv.fdb_root_ns) @@ -1164,6 +1187,20 @@ static struct mlx5_flow_namespace *fs_create_namespace(struct fs_prio *prio) return ns; } +static int create_leaf_prios(struct mlx5_flow_namespace *ns, struct init_tree_node + *prio_metadata) +{ + struct fs_prio *fs_prio; + int i; + + for (i = 0; i < prio_metadata->num_leaf_prios; i++) { + fs_prio = fs_create_prio(ns, i, prio_metadata->max_ft); + if (IS_ERR(fs_prio)) + return PTR_ERR(fs_prio); + } + return 0; +} + #define FLOW_TABLE_BIT_SZ 1 #define GET_FLOW_TABLE_CAP(dev, offset) \ ((be32_to_cpu(*((__be32 *)(dev->hca_caps_cur[MLX5_CAP_FLOW_TABLE]) + \ @@ -1201,6 +1238,8 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, return 0; fs_get_obj(fs_ns, fs_parent_node); + if (init_node->num_leaf_prios) + return create_leaf_prios(fs_ns, init_node); fs_prio = fs_create_prio(fs_ns, index, init_node->max_ft); if (IS_ERR(fs_prio)) return PTR_ERR(fs_prio); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index df2f79ef3cac..7be845e30689 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1258,4 +1258,6 @@ static inline u16 mlx5_to_sw_pkey_sz(int pkey_sz) return MLX5_MIN_PKEY_TABLE_SIZE << pkey_sz; } +#define MLX5_BY_PASS_NUM_PRIOS 9 + #endif /* MLX5_DEVICE_H */ diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 06ac6e8fccfa..a94341271e3f 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -39,7 +39,9 @@ #define MLX5_FS_DEFAULT_FLOW_TAG 0x0 enum mlx5_flow_namespace_type { + MLX5_FLOW_NAMESPACE_BYPASS, MLX5_FLOW_NAMESPACE_KERNEL, + MLX5_FLOW_NAMESPACE_LEFTOVERS, MLX5_FLOW_NAMESPACE_FDB, }; -- cgit v1.2.3-59-g8ed1b From b4d1f032d75b2efb73304e8c12faa7149ad700c7 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:26:05 +0200 Subject: net/mlx5_core: Make ipv4/ipv6 location more clear Change the mlx5 firmware interface header to make it more clear which bytes should be used by IPv4 or IPv6 addresses. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- include/linux/mlx5/mlx5_ifc.h | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 7f166955d4c9..68d73f82e009 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -298,6 +298,22 @@ struct mlx5_ifc_odp_per_transport_service_cap_bits { u8 reserved_1[0x1a]; }; +struct mlx5_ifc_ipv4_layout_bits { + u8 reserved_0[0x60]; + + u8 ipv4[0x20]; +}; + +struct mlx5_ifc_ipv6_layout_bits { + u8 ipv6[16][0x8]; +}; + +union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits { + struct mlx5_ifc_ipv6_layout_bits ipv6_layout; + struct mlx5_ifc_ipv4_layout_bits ipv4_layout; + u8 reserved_0[0x80]; +}; + struct mlx5_ifc_fte_match_set_lyr_2_4_bits { u8 smac_47_16[0x20]; @@ -328,9 +344,9 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits { u8 udp_sport[0x10]; u8 udp_dport[0x10]; - u8 src_ip[4][0x20]; + union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits src_ipv4_src_ipv6; - u8 dst_ip[4][0x20]; + union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits dst_ipv4_dst_ipv6; }; struct mlx5_ifc_fte_match_set_misc_bits { -- cgit v1.2.3-59-g8ed1b From 038d2ef87572757861a177b19f9d489def2c48b8 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 11 Jan 2016 10:26:07 +0200 Subject: IB/mlx5: Add flow steering support Adding flow steering support by creating a flow-table per priority (if rules exist in the priority). mlx5_ib uses autogrouping and thus only creates the required destinations. Also includes adding of these flow steering utilities 1. Parsing verbs flow attributes hardware steering specs. 2. Check if flow is multicast - this is required in order to decide to which flow table will we add the steering rule. 3. Set outer headers in flow match criteria to zeros. Signed-off-by: Maor Gottlieb Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx5/main.c | 463 +++++++++++++++++++++++++++++++++++ drivers/infiniband/hw/mlx5/mlx5_ib.h | 45 +++- include/linux/mlx5/fs.h | 10 + 3 files changed, 517 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 7e97cb55a6bf..b0ec175cc6ba 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -43,6 +43,9 @@ #include #include #include +#include +#include +#include #include "user.h" #include "mlx5_ib.h" @@ -835,6 +838,457 @@ static int mlx5_ib_dealloc_pd(struct ib_pd *pd) return 0; } +static bool outer_header_zero(u32 *match_criteria) +{ + int size = MLX5_ST_SZ_BYTES(fte_match_param); + char *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers); + + return outer_headers_c[0] == 0 && !memcmp(outer_headers_c, + outer_headers_c + 1, + size - 1); +} + +static int parse_flow_attr(u32 *match_c, u32 *match_v, + union ib_flow_spec *ib_spec) +{ + void *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers); + switch (ib_spec->type) { + case IB_FLOW_SPEC_ETH: + if (ib_spec->size != sizeof(ib_spec->eth)) + return -EINVAL; + + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + dmac_47_16), + ib_spec->eth.mask.dst_mac); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + dmac_47_16), + ib_spec->eth.val.dst_mac); + + if (ib_spec->eth.mask.vlan_tag) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + vlan_tag, 1); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_vid, ntohs(ib_spec->eth.mask.vlan_tag)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_vid, ntohs(ib_spec->eth.val.vlan_tag)); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_cfi, + ntohs(ib_spec->eth.mask.vlan_tag) >> 12); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_cfi, + ntohs(ib_spec->eth.val.vlan_tag) >> 12); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_prio, + ntohs(ib_spec->eth.mask.vlan_tag) >> 13); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_prio, + ntohs(ib_spec->eth.val.vlan_tag) >> 13); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + ethertype, ntohs(ib_spec->eth.mask.ether_type)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + ethertype, ntohs(ib_spec->eth.val.ether_type)); + break; + case IB_FLOW_SPEC_IPV4: + if (ib_spec->size != sizeof(ib_spec->ipv4)) + return -EINVAL; + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + ethertype, 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + ethertype, ETH_P_IP); + + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &ib_spec->ipv4.mask.src_ip, + sizeof(ib_spec->ipv4.mask.src_ip)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &ib_spec->ipv4.val.src_ip, + sizeof(ib_spec->ipv4.val.src_ip)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &ib_spec->ipv4.mask.dst_ip, + sizeof(ib_spec->ipv4.mask.dst_ip)); + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &ib_spec->ipv4.val.dst_ip, + sizeof(ib_spec->ipv4.val.dst_ip)); + break; + case IB_FLOW_SPEC_TCP: + if (ib_spec->size != sizeof(ib_spec->tcp_udp)) + return -EINVAL; + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_TCP); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_sport, + ntohs(ib_spec->tcp_udp.mask.src_port)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_sport, + ntohs(ib_spec->tcp_udp.val.src_port)); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_dport, + ntohs(ib_spec->tcp_udp.mask.dst_port)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_dport, + ntohs(ib_spec->tcp_udp.val.dst_port)); + break; + case IB_FLOW_SPEC_UDP: + if (ib_spec->size != sizeof(ib_spec->tcp_udp)) + return -EINVAL; + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_UDP); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_sport, + ntohs(ib_spec->tcp_udp.mask.src_port)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_sport, + ntohs(ib_spec->tcp_udp.val.src_port)); + + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_dport, + ntohs(ib_spec->tcp_udp.mask.dst_port)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_dport, + ntohs(ib_spec->tcp_udp.val.dst_port)); + break; + default: + return -EINVAL; + } + + return 0; +} + +/* If a flow could catch both multicast and unicast packets, + * it won't fall into the multicast flow steering table and this rule + * could steal other multicast packets. + */ +static bool flow_is_multicast_only(struct ib_flow_attr *ib_attr) +{ + struct ib_flow_spec_eth *eth_spec; + + if (ib_attr->type != IB_FLOW_ATTR_NORMAL || + ib_attr->size < sizeof(struct ib_flow_attr) + + sizeof(struct ib_flow_spec_eth) || + ib_attr->num_of_specs < 1) + return false; + + eth_spec = (struct ib_flow_spec_eth *)(ib_attr + 1); + if (eth_spec->type != IB_FLOW_SPEC_ETH || + eth_spec->size != sizeof(*eth_spec)) + return false; + + return is_multicast_ether_addr(eth_spec->mask.dst_mac) && + is_multicast_ether_addr(eth_spec->val.dst_mac); +} + +static bool is_valid_attr(struct ib_flow_attr *flow_attr) +{ + union ib_flow_spec *ib_spec = (union ib_flow_spec *)(flow_attr + 1); + bool has_ipv4_spec = false; + bool eth_type_ipv4 = true; + unsigned int spec_index; + + /* Validate that ethertype is correct */ + for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { + if (ib_spec->type == IB_FLOW_SPEC_ETH && + ib_spec->eth.mask.ether_type) { + if (!((ib_spec->eth.mask.ether_type == htons(0xffff)) && + ib_spec->eth.val.ether_type == htons(ETH_P_IP))) + eth_type_ipv4 = false; + } else if (ib_spec->type == IB_FLOW_SPEC_IPV4) { + has_ipv4_spec = true; + } + ib_spec = (void *)ib_spec + ib_spec->size; + } + return !has_ipv4_spec || eth_type_ipv4; +} + +static void put_flow_table(struct mlx5_ib_dev *dev, + struct mlx5_ib_flow_prio *prio, bool ft_added) +{ + prio->refcount -= !!ft_added; + if (!prio->refcount) { + mlx5_destroy_flow_table(prio->flow_table); + prio->flow_table = NULL; + } +} + +static int mlx5_ib_destroy_flow(struct ib_flow *flow_id) +{ + struct mlx5_ib_dev *dev = to_mdev(flow_id->qp->device); + struct mlx5_ib_flow_handler *handler = container_of(flow_id, + struct mlx5_ib_flow_handler, + ibflow); + struct mlx5_ib_flow_handler *iter, *tmp; + + mutex_lock(&dev->flow_db.lock); + + list_for_each_entry_safe(iter, tmp, &handler->list, list) { + mlx5_del_flow_rule(iter->rule); + list_del(&iter->list); + kfree(iter); + } + + mlx5_del_flow_rule(handler->rule); + put_flow_table(dev, &dev->flow_db.prios[handler->prio], true); + mutex_unlock(&dev->flow_db.lock); + + kfree(handler); + + return 0; +} + +#define MLX5_FS_MAX_TYPES 10 +#define MLX5_FS_MAX_ENTRIES 32000UL +static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, + struct ib_flow_attr *flow_attr) +{ + struct mlx5_flow_namespace *ns = NULL; + struct mlx5_ib_flow_prio *prio; + struct mlx5_flow_table *ft; + int num_entries; + int num_groups; + int priority; + int err = 0; + + if (flow_attr->type == IB_FLOW_ATTR_NORMAL) { + if (flow_is_multicast_only(flow_attr)) + priority = MLX5_IB_FLOW_MCAST_PRIO; + else + priority = flow_attr->priority; + ns = mlx5_get_flow_namespace(dev->mdev, + MLX5_FLOW_NAMESPACE_BYPASS); + num_entries = MLX5_FS_MAX_ENTRIES; + num_groups = MLX5_FS_MAX_TYPES; + prio = &dev->flow_db.prios[priority]; + } else if (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT || + flow_attr->type == IB_FLOW_ATTR_MC_DEFAULT) { + ns = mlx5_get_flow_namespace(dev->mdev, + MLX5_FLOW_NAMESPACE_LEFTOVERS); + build_leftovers_ft_param(&priority, + &num_entries, + &num_groups); + prio = &dev->flow_db.prios[MLX5_IB_FLOW_LEFTOVERS_PRIO]; + } + + if (!ns) + return ERR_PTR(-ENOTSUPP); + + ft = prio->flow_table; + if (!ft) { + ft = mlx5_create_auto_grouped_flow_table(ns, priority, + num_entries, + num_groups); + + if (!IS_ERR(ft)) { + prio->refcount = 0; + prio->flow_table = ft; + } else { + err = PTR_ERR(ft); + } + } + + return err ? ERR_PTR(err) : prio; +} + +static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, + struct mlx5_ib_flow_prio *ft_prio, + struct ib_flow_attr *flow_attr, + struct mlx5_flow_destination *dst) +{ + struct mlx5_flow_table *ft = ft_prio->flow_table; + struct mlx5_ib_flow_handler *handler; + void *ib_flow = flow_attr + 1; + u8 match_criteria_enable = 0; + unsigned int spec_index; + u32 *match_c; + u32 *match_v; + int err = 0; + + if (!is_valid_attr(flow_attr)) + return ERR_PTR(-EINVAL); + + match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + handler = kzalloc(sizeof(*handler), GFP_KERNEL); + if (!handler || !match_c || !match_v) { + err = -ENOMEM; + goto free; + } + + INIT_LIST_HEAD(&handler->list); + + for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { + err = parse_flow_attr(match_c, match_v, ib_flow); + if (err < 0) + goto free; + + ib_flow += ((union ib_flow_spec *)ib_flow)->size; + } + + /* Outer header support only */ + match_criteria_enable = (!outer_header_zero(match_c)) << 0; + handler->rule = mlx5_add_flow_rule(ft, match_criteria_enable, + match_c, match_v, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + MLX5_FS_DEFAULT_FLOW_TAG, + dst); + + if (IS_ERR(handler->rule)) { + err = PTR_ERR(handler->rule); + goto free; + } + + handler->prio = ft_prio - dev->flow_db.prios; + + ft_prio->flow_table = ft; +free: + if (err) + kfree(handler); + kfree(match_c); + kfree(match_v); + return err ? ERR_PTR(err) : handler; +} + +enum { + LEFTOVERS_MC, + LEFTOVERS_UC, +}; + +static struct mlx5_ib_flow_handler *create_leftovers_rule(struct mlx5_ib_dev *dev, + struct mlx5_ib_flow_prio *ft_prio, + struct ib_flow_attr *flow_attr, + struct mlx5_flow_destination *dst) +{ + struct mlx5_ib_flow_handler *handler_ucast = NULL; + struct mlx5_ib_flow_handler *handler = NULL; + + static struct { + struct ib_flow_attr flow_attr; + struct ib_flow_spec_eth eth_flow; + } leftovers_specs[] = { + [LEFTOVERS_MC] = { + .flow_attr = { + .num_of_specs = 1, + .size = sizeof(leftovers_specs[0]) + }, + .eth_flow = { + .type = IB_FLOW_SPEC_ETH, + .size = sizeof(struct ib_flow_spec_eth), + .mask = {.dst_mac = {0x1} }, + .val = {.dst_mac = {0x1} } + } + }, + [LEFTOVERS_UC] = { + .flow_attr = { + .num_of_specs = 1, + .size = sizeof(leftovers_specs[0]) + }, + .eth_flow = { + .type = IB_FLOW_SPEC_ETH, + .size = sizeof(struct ib_flow_spec_eth), + .mask = {.dst_mac = {0x1} }, + .val = {.dst_mac = {} } + } + } + }; + + handler = create_flow_rule(dev, ft_prio, + &leftovers_specs[LEFTOVERS_MC].flow_attr, + dst); + if (!IS_ERR(handler) && + flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT) { + handler_ucast = create_flow_rule(dev, ft_prio, + &leftovers_specs[LEFTOVERS_UC].flow_attr, + dst); + if (IS_ERR(handler_ucast)) { + kfree(handler); + handler = handler_ucast; + } else { + list_add(&handler_ucast->list, &handler->list); + } + } + + return handler; +} + +static struct ib_flow *mlx5_ib_create_flow(struct ib_qp *qp, + struct ib_flow_attr *flow_attr, + int domain) +{ + struct mlx5_ib_dev *dev = to_mdev(qp->device); + struct mlx5_ib_flow_handler *handler = NULL; + struct mlx5_flow_destination *dst = NULL; + struct mlx5_ib_flow_prio *ft_prio; + int err; + + if (flow_attr->priority > MLX5_IB_FLOW_LAST_PRIO) + return ERR_PTR(-ENOSPC); + + if (domain != IB_FLOW_DOMAIN_USER || + flow_attr->port > MLX5_CAP_GEN(dev->mdev, num_ports) || + flow_attr->flags) + return ERR_PTR(-EINVAL); + + dst = kzalloc(sizeof(*dst), GFP_KERNEL); + if (!dst) + return ERR_PTR(-ENOMEM); + + mutex_lock(&dev->flow_db.lock); + + ft_prio = get_flow_table(dev, flow_attr); + if (IS_ERR(ft_prio)) { + err = PTR_ERR(ft_prio); + goto unlock; + } + + dst->type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dst->tir_num = to_mqp(qp)->raw_packet_qp.rq.tirn; + + if (flow_attr->type == IB_FLOW_ATTR_NORMAL) { + handler = create_flow_rule(dev, ft_prio, flow_attr, + dst); + } else if (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT || + flow_attr->type == IB_FLOW_ATTR_MC_DEFAULT) { + handler = create_leftovers_rule(dev, ft_prio, flow_attr, + dst); + } else { + err = -EINVAL; + goto destroy_ft; + } + + if (IS_ERR(handler)) { + err = PTR_ERR(handler); + handler = NULL; + goto destroy_ft; + } + + ft_prio->refcount++; + mutex_unlock(&dev->flow_db.lock); + kfree(dst); + + return &handler->ibflow; + +destroy_ft: + put_flow_table(dev, ft_prio, false); +unlock: + mutex_unlock(&dev->flow_db.lock); + kfree(dst); + kfree(handler); + return ERR_PTR(err); +} + static int mlx5_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) { struct mlx5_ib_dev *dev = to_mdev(ibqp->device); @@ -1439,10 +1893,19 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) (1ull << IB_USER_VERBS_CMD_CLOSE_XRCD); } + if (mlx5_ib_port_link_layer(&dev->ib_dev) == + IB_LINK_LAYER_ETHERNET) { + dev->ib_dev.create_flow = mlx5_ib_create_flow; + dev->ib_dev.destroy_flow = mlx5_ib_destroy_flow; + dev->ib_dev.uverbs_ex_cmd_mask |= + (1ull << IB_USER_VERBS_EX_CMD_CREATE_FLOW) | + (1ull << IB_USER_VERBS_EX_CMD_DESTROY_FLOW); + } err = init_node_data(dev); if (err) goto err_dealloc; + mutex_init(&dev->flow_db.lock); mutex_init(&dev->cap_mask_mutex); err = create_dev_resources(&dev->devr); diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index 633347260b79..1474cccd1e0f 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -105,6 +105,36 @@ struct mlx5_ib_pd { u32 pdn; }; +#define MLX5_IB_FLOW_MCAST_PRIO (MLX5_BY_PASS_NUM_PRIOS - 1) +#define MLX5_IB_FLOW_LAST_PRIO (MLX5_IB_FLOW_MCAST_PRIO - 1) +#if (MLX5_IB_FLOW_LAST_PRIO <= 0) +#error "Invalid number of bypass priorities" +#endif +#define MLX5_IB_FLOW_LEFTOVERS_PRIO (MLX5_IB_FLOW_MCAST_PRIO + 1) + +#define MLX5_IB_NUM_FLOW_FT (MLX5_IB_FLOW_LEFTOVERS_PRIO + 1) +struct mlx5_ib_flow_prio { + struct mlx5_flow_table *flow_table; + unsigned int refcount; +}; + +struct mlx5_ib_flow_handler { + struct list_head list; + struct ib_flow ibflow; + unsigned int prio; + struct mlx5_flow_rule *rule; +}; + +struct mlx5_ib_flow_db { + struct mlx5_ib_flow_prio prios[MLX5_IB_NUM_FLOW_FT]; + /* Protect flow steering bypass flow tables + * when add/del flow rules. + * only single add/removal of flow steering rule could be done + * simultaneously. + */ + struct mutex lock; +}; + /* Use macros here so that don't have to duplicate * enum ib_send_flags and enum ib_qp_type for low-level driver */ @@ -171,9 +201,21 @@ struct mlx5_ib_pfault { struct mlx5_pagefault mpfault; }; +struct mlx5_ib_rq { + u32 tirn; +}; + +struct mlx5_ib_raw_packet_qp { + struct mlx5_ib_rq rq; +}; + struct mlx5_ib_qp { struct ib_qp ibqp; - struct mlx5_core_qp mqp; + union { + struct mlx5_core_qp mqp; + struct mlx5_ib_raw_packet_qp raw_packet_qp; + }; + struct mlx5_buf buf; struct mlx5_db db; @@ -431,6 +473,7 @@ struct mlx5_ib_dev { */ struct srcu_struct mr_srcu; #endif + struct mlx5_ib_flow_db flow_db; }; static inline struct mlx5_ib_cq *to_mibcq(struct mlx5_core_cq *mcq) diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index a94341271e3f..8230caa3fb6e 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -38,6 +38,16 @@ #define MLX5_FS_DEFAULT_FLOW_TAG 0x0 +#define LEFTOVERS_RULE_NUM 2 +static inline void build_leftovers_ft_param(int *priority, + int *n_ent, + int *n_grp) +{ + *priority = 0; /* Priority of leftovers_prio-0 */ + *n_ent = LEFTOVERS_RULE_NUM; + *n_grp = LEFTOVERS_RULE_NUM; +} + enum mlx5_flow_namespace_type { MLX5_FLOW_NAMESPACE_BYPASS, MLX5_FLOW_NAMESPACE_KERNEL, -- cgit v1.2.3-59-g8ed1b From b6a0e72ad3cffabaf30b856deb58fbe64a0f36a8 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 11 Jan 2016 10:19:10 -0800 Subject: net: Fix typo in netdev_intersect_features Obviously need to 'or in NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM. Fixes: c8cd0989bd151f ("net: Eliminate NETIF_F_GEN_CSUM and NETIF_F_V[46]_CSUM") Reported-by: Jack Morgenstein Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2285596e7045..5ac140dcb789 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3873,9 +3873,9 @@ static inline netdev_features_t netdev_intersect_features(netdev_features_t f1, { if ((f1 ^ f2) & NETIF_F_HW_CSUM) { if (f1 & NETIF_F_HW_CSUM) - f1 |= (NETIF_F_IP_CSUM|NETIF_F_IP_CSUM); + f1 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM); else - f2 |= (NETIF_F_IP_CSUM|NETIF_F_IP_CSUM); + f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM); } return f1 & f2; -- cgit v1.2.3-59-g8ed1b