From bb4cf02d4c74f0db60f58c406ddfdfed16d14f84 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 2 Mar 2020 05:59:33 -0600 Subject: netdevice: Replace zero-length array with flexible-array member The current codebase makes use of the zero-length array language extension to the C90 standard, but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on. Also, notice that, dynamic memory allocations won't be affected by this change: "Flexible array members have incomplete type, and so the sizeof operator may not be applied. As a quirk of the original implementation of zero-length arrays, sizeof evaluates to zero."[1] This issue was found with the help of Coccinelle. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by: Gustavo A. R. Silva Signed-off-by: David S. Miller --- include/linux/netdevice.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 6c3f7032e8d9..b6fedd54cd8e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -664,7 +664,7 @@ static inline void netdev_queue_numa_node_write(struct netdev_queue *q, int node struct rps_map { unsigned int len; struct rcu_head rcu; - u16 cpus[0]; + u16 cpus[]; }; #define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16))) @@ -686,7 +686,7 @@ struct rps_dev_flow { struct rps_dev_flow_table { unsigned int mask; struct rcu_head rcu; - struct rps_dev_flow flows[0]; + struct rps_dev_flow flows[]; }; #define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \ ((_num) * sizeof(struct rps_dev_flow))) @@ -704,7 +704,7 @@ struct rps_dev_flow_table { struct rps_sock_flow_table { u32 mask; - u32 ents[0] ____cacheline_aligned_in_smp; + u32 ents[] ____cacheline_aligned_in_smp; }; #define RPS_SOCK_FLOW_TABLE_SIZE(_num) (offsetof(struct rps_sock_flow_table, ents[_num])) @@ -767,7 +767,7 @@ struct xps_map { unsigned int len; unsigned int alloc_len; struct rcu_head rcu; - u16 queues[0]; + u16 queues[]; }; #define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16))) #define XPS_MIN_MAP_ALLOC ((L1_CACHE_ALIGN(offsetof(struct xps_map, queues[1])) \ @@ -778,7 +778,7 @@ struct xps_map { */ struct xps_dev_maps { struct rcu_head rcu; - struct xps_map __rcu *attr_map[0]; /* Either CPUs map or RXQs map */ + struct xps_map __rcu *attr_map[]; /* Either CPUs map or RXQs map */ }; #define XPS_CPU_DEV_MAPS_SIZE(_tcs) (sizeof(struct xps_dev_maps) + \ -- cgit v1.3-14-g43fede From aaca9408078914380fbfd8aef3c38a34b515a654 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Thu, 5 Mar 2020 09:16:40 +0200 Subject: net: sched: Make FIFO Qdisc offloadable Invoke ndo_setup_tc() as appropriate to signal init / replacement, destroying and dumping of pFIFO / bFIFO Qdisc. A lot of the FIFO logic is used for pFIFO_head_drop as well, but that's a semantically very different Qdisc that isn't really in the same boat as pFIFO / bFIFO. Split some of the functions to keep the Qdisc intact. Signed-off-by: Petr Machata Signed-off-by: Ido Schimmel Acked-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + include/net/pkt_cls.h | 15 ++++++++ net/sched/sch_fifo.c | 97 ++++++++++++++++++++++++++++++++++++++++++++--- 3 files changed, 107 insertions(+), 6 deletions(-) (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b6fedd54cd8e..654808bfad83 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -853,6 +853,7 @@ enum tc_setup_type { TC_SETUP_FT, TC_SETUP_QDISC_ETS, TC_SETUP_QDISC_TBF, + TC_SETUP_QDISC_FIFO, }; /* These structures hold the attributes of bpf state that are being passed diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 53946b509b51..341a66af8d59 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -881,4 +881,19 @@ struct tc_tbf_qopt_offload { }; }; +enum tc_fifo_command { + TC_FIFO_REPLACE, + TC_FIFO_DESTROY, + TC_FIFO_STATS, +}; + +struct tc_fifo_qopt_offload { + enum tc_fifo_command command; + u32 handle; + u32 parent; + union { + struct tc_qopt_offload_stats stats; + }; +}; + #endif diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 37c8aa75d70c..a579a4131d22 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -12,6 +12,7 @@ #include #include #include +#include /* 1 band FIFO pseudo-"scheduler" */ @@ -51,8 +52,49 @@ static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch, return NET_XMIT_CN; } -static int fifo_init(struct Qdisc *sch, struct nlattr *opt, - struct netlink_ext_ack *extack) +static void fifo_offload_init(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_fifo_qopt_offload qopt; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return; + + qopt.command = TC_FIFO_REPLACE; + qopt.handle = sch->handle; + qopt.parent = sch->parent; + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_FIFO, &qopt); +} + +static void fifo_offload_destroy(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_fifo_qopt_offload qopt; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return; + + qopt.command = TC_FIFO_DESTROY; + qopt.handle = sch->handle; + qopt.parent = sch->parent; + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_FIFO, &qopt); +} + +static int fifo_offload_dump(struct Qdisc *sch) +{ + struct tc_fifo_qopt_offload qopt; + + qopt.command = TC_FIFO_STATS; + qopt.handle = sch->handle; + qopt.parent = sch->parent; + qopt.stats.bstats = &sch->bstats; + qopt.stats.qstats = &sch->qstats; + + return qdisc_offload_dump_helper(sch, TC_SETUP_QDISC_FIFO, &qopt); +} + +static int __fifo_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) { bool bypass; bool is_bfifo = sch->ops == &bfifo_qdisc_ops; @@ -82,10 +124,35 @@ static int fifo_init(struct Qdisc *sch, struct nlattr *opt, sch->flags |= TCQ_F_CAN_BYPASS; else sch->flags &= ~TCQ_F_CAN_BYPASS; + return 0; } -static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb) +static int fifo_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + int err; + + err = __fifo_init(sch, opt, extack); + if (err) + return err; + + fifo_offload_init(sch); + return 0; +} + +static int fifo_hd_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + return __fifo_init(sch, opt, extack); +} + +static void fifo_destroy(struct Qdisc *sch) +{ + fifo_offload_destroy(sch); +} + +static int __fifo_dump(struct Qdisc *sch, struct sk_buff *skb) { struct tc_fifo_qopt opt = { .limit = sch->limit }; @@ -97,6 +164,22 @@ nla_put_failure: return -1; } +static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + int err; + + err = fifo_offload_dump(sch); + if (err) + return err; + + return __fifo_dump(sch, skb); +} + +static int fifo_hd_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + return __fifo_dump(sch, skb); +} + struct Qdisc_ops pfifo_qdisc_ops __read_mostly = { .id = "pfifo", .priv_size = 0, @@ -104,6 +187,7 @@ struct Qdisc_ops pfifo_qdisc_ops __read_mostly = { .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, .init = fifo_init, + .destroy = fifo_destroy, .reset = qdisc_reset_queue, .change = fifo_init, .dump = fifo_dump, @@ -118,6 +202,7 @@ struct Qdisc_ops bfifo_qdisc_ops __read_mostly = { .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, .init = fifo_init, + .destroy = fifo_destroy, .reset = qdisc_reset_queue, .change = fifo_init, .dump = fifo_dump, @@ -131,10 +216,10 @@ struct Qdisc_ops pfifo_head_drop_qdisc_ops __read_mostly = { .enqueue = pfifo_tail_enqueue, .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, - .init = fifo_init, + .init = fifo_hd_init, .reset = qdisc_reset_queue, - .change = fifo_init, - .dump = fifo_dump, + .change = fifo_hd_init, + .dump = fifo_hd_dump, .owner = THIS_MODULE, }; -- cgit v1.3-14-g43fede From 8537f78647c072bdb1a5dbe32e1c7e5b13ff1258 Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Wed, 11 Mar 2020 12:59:03 +0100 Subject: netfilter: Introduce egress hook Commit e687ad60af09 ("netfilter: add netfilter ingress hook after handle_ing() under unique static key") introduced the ability to classify packets on ingress. Allow the same on egress. Position the hook immediately before a packet is handed to tc and then sent out on an interface, thereby mirroring the ingress order. This order allows marking packets in the netfilter egress hook and subsequently using the mark in tc. Another benefit of this order is consistency with a lot of existing documentation which says that egress tc is performed after netfilter hooks. Egress hooks already exist for the most common protocols, such as NF_INET_LOCAL_OUT or NF_ARP_OUT, and those are to be preferred because they are executed earlier during packet processing. However for more exotic protocols, there is currently no provision to apply netfilter on egress. A common workaround is to enslave the interface to a bridge and use ebtables, or to resort to tc. But when the ingress hook was introduced, consensus was that users should be given the choice to use netfilter or tc, whichever tool suits their needs best: https://lore.kernel.org/netdev/20150430153317.GA3230@salvia/ This hook is also useful for NAT46/NAT64, tunneling and filtering of locally generated af_packet traffic such as dhclient. There have also been occasional user requests for a netfilter egress hook in the past, e.g.: https://www.spinics.net/lists/netfilter/msg50038.html Performance measurements with pktgen surprisingly show a speedup rather than a slowdown with this commit: * Without this commit: Result: OK: 34240933(c34238375+d2558) usec, 100000000 (60byte,0frags) 2920481pps 1401Mb/sec (1401830880bps) errors: 0 * With this commit: Result: OK: 33997299(c33994193+d3106) usec, 100000000 (60byte,0frags) 2941410pps 1411Mb/sec (1411876800bps) errors: 0 * Without this commit + tc egress: Result: OK: 39022386(c39019547+d2839) usec, 100000000 (60byte,0frags) 2562631pps 1230Mb/sec (1230062880bps) errors: 0 * With this commit + tc egress: Result: OK: 37604447(c37601877+d2570) usec, 100000000 (60byte,0frags) 2659259pps 1276Mb/sec (1276444320bps) errors: 0 * With this commit + nft egress: Result: OK: 41436689(c41434088+d2600) usec, 100000000 (60byte,0frags) 2413320pps 1158Mb/sec (1158393600bps) errors: 0 Tested on a bare-metal Core i7-3615QM, each measurement was performed three times to verify that the numbers are stable. Commands to perform a measurement: modprobe pktgen echo "add_device lo@3" > /proc/net/pktgen/kpktgend_3 samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh -i 'lo@3' -n 100000000 Commands for testing tc egress: tc qdisc add dev lo clsact tc filter add dev lo egress protocol ip prio 1 u32 match ip dst 4.3.2.1/32 Commands for testing nft egress: nft add table netdev t nft add chain netdev t co \{ type filter hook egress device lo priority 0 \; \} nft add rule netdev t co ip daddr 4.3.2.1/32 drop All testing was performed on the loopback interface to avoid distorting measurements by the packet handling in the low-level Ethernet driver. Signed-off-by: Lukas Wunner Cc: Daniel Borkmann Signed-off-by: Pablo Neira Ayuso --- include/linux/netdevice.h | 4 ++++ include/linux/netfilter_netdev.h | 27 +++++++++++++++++++++++++++ include/uapi/linux/netfilter.h | 1 + net/core/dev.c | 23 ++++++++++++++++++++--- net/netfilter/Kconfig | 8 ++++++++ net/netfilter/core.c | 24 ++++++++++++++++++++---- net/netfilter/nft_chain_filter.c | 4 +++- 7 files changed, 83 insertions(+), 8 deletions(-) (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 654808bfad83..15f1e32b430c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1751,6 +1751,7 @@ enum netdev_priv_flags { * @xps_maps: XXX: need comments on this one * @miniq_egress: clsact qdisc specific data for * egress processing + * @nf_hooks_egress: netfilter hooks executed for egress packets * @qdisc_hash: qdisc hash table * @watchdog_timeo: Represents the timeout that is used by * the watchdog (see dev_watchdog()) @@ -2026,6 +2027,9 @@ struct net_device { #ifdef CONFIG_NET_CLS_ACT struct mini_Qdisc __rcu *miniq_egress; #endif +#ifdef CONFIG_NETFILTER_EGRESS + struct nf_hook_entries __rcu *nf_hooks_egress; +#endif #ifdef CONFIG_NET_SCHED DECLARE_HASHTABLE (qdisc_hash, 4); diff --git a/include/linux/netfilter_netdev.h b/include/linux/netfilter_netdev.h index 49e26479642e..92d3611a782e 100644 --- a/include/linux/netfilter_netdev.h +++ b/include/linux/netfilter_netdev.h @@ -47,6 +47,9 @@ static inline void nf_hook_netdev_init(struct net_device *dev) #ifdef CONFIG_NETFILTER_INGRESS RCU_INIT_POINTER(dev->nf_hooks_ingress, NULL); #endif +#ifdef CONFIG_NETFILTER_EGRESS + RCU_INIT_POINTER(dev->nf_hooks_egress, NULL); +#endif } #ifdef CONFIG_NETFILTER_INGRESS @@ -72,4 +75,28 @@ static inline int nf_hook_ingress(struct sk_buff *skb) return 0; } #endif /* CONFIG_NETFILTER_INGRESS */ + +#ifdef CONFIG_NETFILTER_EGRESS +static inline bool nf_hook_egress_active(const struct sk_buff *skb) +{ + return nf_hook_netdev_active(NF_NETDEV_EGRESS, + skb->dev->nf_hooks_egress); +} + +static inline int nf_hook_egress(struct sk_buff *skb) +{ + return nf_hook_netdev(skb, NF_NETDEV_EGRESS, + skb->dev->nf_hooks_egress); +} +#else /* CONFIG_NETFILTER_EGRESS */ +static inline int nf_hook_egress_active(struct sk_buff *skb) +{ + return 0; +} + +static inline int nf_hook_egress(struct sk_buff *skb) +{ + return 0; +} +#endif /* CONFIG_NETFILTER_EGRESS */ #endif /* _NETFILTER_INGRESS_H_ */ diff --git a/include/uapi/linux/netfilter.h b/include/uapi/linux/netfilter.h index ca9e63d6e0e4..d1616574c54f 100644 --- a/include/uapi/linux/netfilter.h +++ b/include/uapi/linux/netfilter.h @@ -50,6 +50,7 @@ enum nf_inet_hooks { enum nf_dev_hooks { NF_NETDEV_INGRESS, + NF_NETDEV_EGRESS, NF_NETDEV_NUMHOOKS }; diff --git a/net/core/dev.c b/net/core/dev.c index 13d562f67e9c..a2da72a77c20 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3773,6 +3773,7 @@ EXPORT_SYMBOL(dev_loopback_xmit); static struct sk_buff * sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) { +#ifdef CONFIG_NET_CLS_ACT struct mini_Qdisc *miniq = rcu_dereference_bh(dev->miniq_egress); struct tcf_result cl_res; @@ -3806,11 +3807,24 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) default: break; } - +#endif /* CONFIG_NET_CLS_ACT */ return skb; } #endif /* CONFIG_NET_EGRESS */ +static inline int nf_egress(struct sk_buff *skb) +{ + if (nf_hook_egress_active(skb)) { + int ret; + + rcu_read_lock(); + ret = nf_hook_egress(skb); + rcu_read_unlock(); + return ret; + } + return 0; +} + #ifdef CONFIG_XPS static int __get_xps_queue_idx(struct net_device *dev, struct sk_buff *skb, struct xps_dev_maps *dev_maps, unsigned int tci) @@ -3997,13 +4011,16 @@ static int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) qdisc_pkt_len_init(skb); #ifdef CONFIG_NET_CLS_ACT skb->tc_at_ingress = 0; -# ifdef CONFIG_NET_EGRESS +#endif +#ifdef CONFIG_NET_EGRESS if (static_branch_unlikely(&egress_needed_key)) { + if (nf_egress(skb) < 0) + goto out; + 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. diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 468fea1aebba..f4c68f60f241 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -10,6 +10,14 @@ config NETFILTER_INGRESS This allows you to classify packets from ingress using the Netfilter infrastructure. +config NETFILTER_EGRESS + bool "Netfilter egress support" + default y + select NET_EGRESS + help + This allows you to classify packets before transmission using the + Netfilter infrastructure. + config NETFILTER_NETLINK tristate diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 78f046ec506f..85e9c959aba7 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -306,6 +306,12 @@ nf_hook_entry_head(struct net *net, int pf, unsigned int hooknum, if (dev && dev_net(dev) == net) return &dev->nf_hooks_ingress; } +#endif +#ifdef CONFIG_NETFILTER_EGRESS + if (hooknum == NF_NETDEV_EGRESS) { + if (dev && dev_net(dev) == net) + return &dev->nf_hooks_egress; + } #endif WARN_ON_ONCE(1); return NULL; @@ -318,11 +324,13 @@ static int __nf_register_net_hook(struct net *net, int pf, struct nf_hook_entries __rcu **pp; if (pf == NFPROTO_NETDEV) { -#ifndef CONFIG_NETFILTER_INGRESS - if (reg->hooknum == NF_NETDEV_INGRESS) + if ((!IS_ENABLED(CONFIG_NETFILTER_INGRESS) && + reg->hooknum == NF_NETDEV_INGRESS) || + (!IS_ENABLED(CONFIG_NETFILTER_EGRESS) && + reg->hooknum == NF_NETDEV_EGRESS)) return -EOPNOTSUPP; -#endif - if (reg->hooknum != NF_NETDEV_INGRESS || + if ((reg->hooknum != NF_NETDEV_INGRESS && + reg->hooknum != NF_NETDEV_EGRESS) || !reg->dev || dev_net(reg->dev) != net) return -EINVAL; } @@ -348,6 +356,10 @@ static int __nf_register_net_hook(struct net *net, int pf, if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) net_inc_ingress_queue(); #endif +#ifdef CONFIG_NETFILTER_EGRESS + if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_EGRESS) + net_inc_egress_queue(); +#endif #ifdef CONFIG_JUMP_LABEL static_key_slow_inc(&nf_hooks_needed[pf][reg->hooknum]); #endif @@ -406,6 +418,10 @@ static void __nf_unregister_net_hook(struct net *net, int pf, if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) net_dec_ingress_queue(); #endif +#ifdef CONFIG_NETFILTER_EGRESS + if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_EGRESS) + net_dec_egress_queue(); +#endif #ifdef CONFIG_JUMP_LABEL static_key_slow_dec(&nf_hooks_needed[pf][reg->hooknum]); #endif diff --git a/net/netfilter/nft_chain_filter.c b/net/netfilter/nft_chain_filter.c index c78d01bc02e9..67ce6dbb5496 100644 --- a/net/netfilter/nft_chain_filter.c +++ b/net/netfilter/nft_chain_filter.c @@ -277,9 +277,11 @@ static const struct nft_chain_type nft_chain_filter_netdev = { .name = "filter", .type = NFT_CHAIN_T_DEFAULT, .family = NFPROTO_NETDEV, - .hook_mask = (1 << NF_NETDEV_INGRESS), + .hook_mask = (1 << NF_NETDEV_INGRESS) | + (1 << NF_NETDEV_EGRESS), .hooks = { [NF_NETDEV_INGRESS] = nft_do_chain_netdev, + [NF_NETDEV_EGRESS] = nft_do_chain_netdev, }, }; -- cgit v1.3-14-g43fede From 357b6cc5834eabc1be7c28a9faae7da061df097d Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 18 Mar 2020 10:33:22 +0100 Subject: netfilter: revert introduction of egress hook This reverts the following commits: 8537f78647c0 ("netfilter: Introduce egress hook") 5418d3881e1f ("netfilter: Generalize ingress hook") b030f194aed2 ("netfilter: Rename ingress hook include file") >From the discussion in [0], the author's main motivation to add a hook in fast path is for an out of tree kernel module, which is a red flag to begin with. Other mentioned potential use cases like NAT{64,46} is on future extensions w/o concrete code in the tree yet. Revert as suggested [1] given the weak justification to add more hooks to critical fast-path. [0] https://lore.kernel.org/netdev/cover.1583927267.git.lukas@wunner.de/ [1] https://lore.kernel.org/netdev/20200318.011152.72770718915606186.davem@davemloft.net/ Signed-off-by: Daniel Borkmann Cc: David Miller Cc: Pablo Neira Ayuso Cc: Alexei Starovoitov Nacked-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 -- include/linux/netfilter_ingress.h | 58 ++++++++++++++++++++++ include/linux/netfilter_netdev.h | 102 -------------------------------------- include/uapi/linux/netfilter.h | 1 - net/core/dev.c | 27 ++-------- net/netfilter/Kconfig | 8 --- net/netfilter/core.c | 24 ++------- net/netfilter/nft_chain_filter.c | 4 +- 8 files changed, 68 insertions(+), 160 deletions(-) create mode 100644 include/linux/netfilter_ingress.h delete mode 100644 include/linux/netfilter_netdev.h (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 15f1e32b430c..654808bfad83 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1751,7 +1751,6 @@ enum netdev_priv_flags { * @xps_maps: XXX: need comments on this one * @miniq_egress: clsact qdisc specific data for * egress processing - * @nf_hooks_egress: netfilter hooks executed for egress packets * @qdisc_hash: qdisc hash table * @watchdog_timeo: Represents the timeout that is used by * the watchdog (see dev_watchdog()) @@ -2027,9 +2026,6 @@ struct net_device { #ifdef CONFIG_NET_CLS_ACT struct mini_Qdisc __rcu *miniq_egress; #endif -#ifdef CONFIG_NETFILTER_EGRESS - struct nf_hook_entries __rcu *nf_hooks_egress; -#endif #ifdef CONFIG_NET_SCHED DECLARE_HASHTABLE (qdisc_hash, 4); diff --git a/include/linux/netfilter_ingress.h b/include/linux/netfilter_ingress.h new file mode 100644 index 000000000000..a13774be2eb5 --- /dev/null +++ b/include/linux/netfilter_ingress.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _NETFILTER_INGRESS_H_ +#define _NETFILTER_INGRESS_H_ + +#include +#include + +#ifdef CONFIG_NETFILTER_INGRESS +static inline bool nf_hook_ingress_active(const struct sk_buff *skb) +{ +#ifdef CONFIG_JUMP_LABEL + if (!static_key_false(&nf_hooks_needed[NFPROTO_NETDEV][NF_NETDEV_INGRESS])) + return false; +#endif + return rcu_access_pointer(skb->dev->nf_hooks_ingress); +} + +/* caller must hold rcu_read_lock */ +static inline int nf_hook_ingress(struct sk_buff *skb) +{ + struct nf_hook_entries *e = rcu_dereference(skb->dev->nf_hooks_ingress); + struct nf_hook_state state; + int ret; + + /* Must recheck the ingress hook head, in the event it became NULL + * after the check in nf_hook_ingress_active evaluated to true. + */ + if (unlikely(!e)) + return 0; + + nf_hook_state_init(&state, NF_NETDEV_INGRESS, + NFPROTO_NETDEV, skb->dev, NULL, NULL, + dev_net(skb->dev), NULL); + ret = nf_hook_slow(skb, &state, e, 0); + if (ret == 0) + return -1; + + return ret; +} + +static inline void nf_hook_ingress_init(struct net_device *dev) +{ + RCU_INIT_POINTER(dev->nf_hooks_ingress, NULL); +} +#else /* CONFIG_NETFILTER_INGRESS */ +static inline int nf_hook_ingress_active(struct sk_buff *skb) +{ + return 0; +} + +static inline int nf_hook_ingress(struct sk_buff *skb) +{ + return 0; +} + +static inline void nf_hook_ingress_init(struct net_device *dev) {} +#endif /* CONFIG_NETFILTER_INGRESS */ +#endif /* _NETFILTER_INGRESS_H_ */ diff --git a/include/linux/netfilter_netdev.h b/include/linux/netfilter_netdev.h deleted file mode 100644 index 92d3611a782e..000000000000 --- a/include/linux/netfilter_netdev.h +++ /dev/null @@ -1,102 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _NETFILTER_NETDEV_H_ -#define _NETFILTER_NETDEV_H_ - -#include -#include - -#ifdef CONFIG_NETFILTER -static __always_inline bool nf_hook_netdev_active(enum nf_dev_hooks hooknum, - struct nf_hook_entries __rcu *hooks) -{ -#ifdef CONFIG_JUMP_LABEL - if (!static_key_false(&nf_hooks_needed[NFPROTO_NETDEV][hooknum])) - return false; -#endif - return rcu_access_pointer(hooks); -} - -/* caller must hold rcu_read_lock */ -static __always_inline int nf_hook_netdev(struct sk_buff *skb, - enum nf_dev_hooks hooknum, - struct nf_hook_entries __rcu *hooks) -{ - struct nf_hook_entries *e = rcu_dereference(hooks); - struct nf_hook_state state; - int ret; - - /* Must recheck the hook head, in the event it became NULL - * after the check in nf_hook_netdev_active evaluated to true. - */ - if (unlikely(!e)) - return 0; - - nf_hook_state_init(&state, hooknum, - NFPROTO_NETDEV, skb->dev, NULL, NULL, - dev_net(skb->dev), NULL); - ret = nf_hook_slow(skb, &state, e, 0); - if (ret == 0) - return -1; - - return ret; -} -#endif /* CONFIG_NETFILTER */ - -static inline void nf_hook_netdev_init(struct net_device *dev) -{ -#ifdef CONFIG_NETFILTER_INGRESS - RCU_INIT_POINTER(dev->nf_hooks_ingress, NULL); -#endif -#ifdef CONFIG_NETFILTER_EGRESS - RCU_INIT_POINTER(dev->nf_hooks_egress, NULL); -#endif -} - -#ifdef CONFIG_NETFILTER_INGRESS -static inline bool nf_hook_ingress_active(const struct sk_buff *skb) -{ - return nf_hook_netdev_active(NF_NETDEV_INGRESS, - skb->dev->nf_hooks_ingress); -} - -static inline int nf_hook_ingress(struct sk_buff *skb) -{ - return nf_hook_netdev(skb, NF_NETDEV_INGRESS, - skb->dev->nf_hooks_ingress); -} -#else /* CONFIG_NETFILTER_INGRESS */ -static inline int nf_hook_ingress_active(struct sk_buff *skb) -{ - return 0; -} - -static inline int nf_hook_ingress(struct sk_buff *skb) -{ - return 0; -} -#endif /* CONFIG_NETFILTER_INGRESS */ - -#ifdef CONFIG_NETFILTER_EGRESS -static inline bool nf_hook_egress_active(const struct sk_buff *skb) -{ - return nf_hook_netdev_active(NF_NETDEV_EGRESS, - skb->dev->nf_hooks_egress); -} - -static inline int nf_hook_egress(struct sk_buff *skb) -{ - return nf_hook_netdev(skb, NF_NETDEV_EGRESS, - skb->dev->nf_hooks_egress); -} -#else /* CONFIG_NETFILTER_EGRESS */ -static inline int nf_hook_egress_active(struct sk_buff *skb) -{ - return 0; -} - -static inline int nf_hook_egress(struct sk_buff *skb) -{ - return 0; -} -#endif /* CONFIG_NETFILTER_EGRESS */ -#endif /* _NETFILTER_INGRESS_H_ */ diff --git a/include/uapi/linux/netfilter.h b/include/uapi/linux/netfilter.h index d1616574c54f..ca9e63d6e0e4 100644 --- a/include/uapi/linux/netfilter.h +++ b/include/uapi/linux/netfilter.h @@ -50,7 +50,6 @@ enum nf_inet_hooks { enum nf_dev_hooks { NF_NETDEV_INGRESS, - NF_NETDEV_EGRESS, NF_NETDEV_NUMHOOKS }; diff --git a/net/core/dev.c b/net/core/dev.c index aeb8ccbbe93b..021e18251465 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -135,7 +135,7 @@ #include #include #include -#include +#include #include #include #include @@ -3773,7 +3773,6 @@ EXPORT_SYMBOL(dev_loopback_xmit); static struct sk_buff * sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) { -#ifdef CONFIG_NET_CLS_ACT struct mini_Qdisc *miniq = rcu_dereference_bh(dev->miniq_egress); struct tcf_result cl_res; @@ -3807,24 +3806,11 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) default: break; } -#endif /* CONFIG_NET_CLS_ACT */ + return skb; } #endif /* CONFIG_NET_EGRESS */ -static inline int nf_egress(struct sk_buff *skb) -{ - if (nf_hook_egress_active(skb)) { - int ret; - - rcu_read_lock(); - ret = nf_hook_egress(skb); - rcu_read_unlock(); - return ret; - } - return 0; -} - #ifdef CONFIG_XPS static int __get_xps_queue_idx(struct net_device *dev, struct sk_buff *skb, struct xps_dev_maps *dev_maps, unsigned int tci) @@ -4011,16 +3997,13 @@ static int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) qdisc_pkt_len_init(skb); #ifdef CONFIG_NET_CLS_ACT skb->tc_at_ingress = 0; -#endif -#ifdef CONFIG_NET_EGRESS +# ifdef CONFIG_NET_EGRESS if (static_branch_unlikely(&egress_needed_key)) { - if (nf_egress(skb) < 0) - goto out; - 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. @@ -9867,7 +9850,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, if (!dev->ethtool_ops) dev->ethtool_ops = &default_ethtool_ops; - nf_hook_netdev_init(dev); + nf_hook_ingress_init(dev); return dev; diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index f4c68f60f241..468fea1aebba 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -10,14 +10,6 @@ config NETFILTER_INGRESS This allows you to classify packets from ingress using the Netfilter infrastructure. -config NETFILTER_EGRESS - bool "Netfilter egress support" - default y - select NET_EGRESS - help - This allows you to classify packets before transmission using the - Netfilter infrastructure. - config NETFILTER_NETLINK tristate diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 85e9c959aba7..78f046ec506f 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -306,12 +306,6 @@ nf_hook_entry_head(struct net *net, int pf, unsigned int hooknum, if (dev && dev_net(dev) == net) return &dev->nf_hooks_ingress; } -#endif -#ifdef CONFIG_NETFILTER_EGRESS - if (hooknum == NF_NETDEV_EGRESS) { - if (dev && dev_net(dev) == net) - return &dev->nf_hooks_egress; - } #endif WARN_ON_ONCE(1); return NULL; @@ -324,13 +318,11 @@ static int __nf_register_net_hook(struct net *net, int pf, struct nf_hook_entries __rcu **pp; if (pf == NFPROTO_NETDEV) { - if ((!IS_ENABLED(CONFIG_NETFILTER_INGRESS) && - reg->hooknum == NF_NETDEV_INGRESS) || - (!IS_ENABLED(CONFIG_NETFILTER_EGRESS) && - reg->hooknum == NF_NETDEV_EGRESS)) +#ifndef CONFIG_NETFILTER_INGRESS + if (reg->hooknum == NF_NETDEV_INGRESS) return -EOPNOTSUPP; - if ((reg->hooknum != NF_NETDEV_INGRESS && - reg->hooknum != NF_NETDEV_EGRESS) || +#endif + if (reg->hooknum != NF_NETDEV_INGRESS || !reg->dev || dev_net(reg->dev) != net) return -EINVAL; } @@ -356,10 +348,6 @@ static int __nf_register_net_hook(struct net *net, int pf, if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) net_inc_ingress_queue(); #endif -#ifdef CONFIG_NETFILTER_EGRESS - if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_EGRESS) - net_inc_egress_queue(); -#endif #ifdef CONFIG_JUMP_LABEL static_key_slow_inc(&nf_hooks_needed[pf][reg->hooknum]); #endif @@ -418,10 +406,6 @@ static void __nf_unregister_net_hook(struct net *net, int pf, if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_INGRESS) net_dec_ingress_queue(); #endif -#ifdef CONFIG_NETFILTER_EGRESS - if (pf == NFPROTO_NETDEV && reg->hooknum == NF_NETDEV_EGRESS) - net_dec_egress_queue(); -#endif #ifdef CONFIG_JUMP_LABEL static_key_slow_dec(&nf_hooks_needed[pf][reg->hooknum]); #endif diff --git a/net/netfilter/nft_chain_filter.c b/net/netfilter/nft_chain_filter.c index 67ce6dbb5496..c78d01bc02e9 100644 --- a/net/netfilter/nft_chain_filter.c +++ b/net/netfilter/nft_chain_filter.c @@ -277,11 +277,9 @@ static const struct nft_chain_type nft_chain_filter_netdev = { .name = "filter", .type = NFT_CHAIN_T_DEFAULT, .family = NFPROTO_NETDEV, - .hook_mask = (1 << NF_NETDEV_INGRESS) | - (1 << NF_NETDEV_EGRESS), + .hook_mask = (1 << NF_NETDEV_INGRESS), .hooks = { [NF_NETDEV_INGRESS] = nft_do_chain_netdev, - [NF_NETDEV_EGRESS] = nft_do_chain_netdev, }, }; -- cgit v1.3-14-g43fede From 30e9bb8472f4454d0544020574bb03d96ffa0e52 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Wed, 25 Mar 2020 15:52:31 +0300 Subject: net: add a reference to MACsec ops in net_device This patch adds a reference to MACsec ops to the net_device structure, allowing net device drivers to implement offloading operations for MACsec. Signed-off-by: Antoine Tenart Signed-off-by: Mark Starovoytov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 654808bfad83..b521500b244d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -53,6 +53,8 @@ struct netpoll_info; struct device; struct phy_device; struct dsa_port; +struct macsec_context; +struct macsec_ops; struct sfp_bus; /* 802.11 specific */ @@ -1819,6 +1821,8 @@ enum netdev_priv_flags { * that follow this device when it is moved * to another network namespace. * + * @macsec_ops: MACsec offloading ops + * * FIXME: cleanup struct net_device such that network protocol info * moves out. */ @@ -2113,6 +2117,11 @@ struct net_device { unsigned wol_enabled:1; struct list_head net_notifier_list; + +#if IS_ENABLED(CONFIG_MACSEC) + /* MACsec management functions */ + const struct macsec_ops *macsec_ops; +#endif }; #define to_net_dev(d) container_of(d, struct net_device, dev) -- cgit v1.3-14-g43fede From 92234c8f15c8d96ad7e52afdc5994cba6be68eb9 Mon Sep 17 00:00:00 2001 From: Toke Høiland-Jørgensen Date: Wed, 25 Mar 2020 18:23:26 +0100 Subject: xdp: Support specifying expected existing program when attaching XDP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While it is currently possible for userspace to specify that an existing XDP program should not be replaced when attaching to an interface, there is no mechanism to safely replace a specific XDP program with another. This patch adds a new netlink attribute, IFLA_XDP_EXPECTED_FD, which can be set along with IFLA_XDP_FD. If set, the kernel will check that the program currently loaded on the interface matches the expected one, and fail the operation if it does not. This corresponds to a 'cmpxchg' memory operation. Setting the new attribute with a negative value means that no program is expected to be attached, which corresponds to setting the UPDATE_IF_NOEXIST flag. A new companion flag, XDP_FLAGS_REPLACE, is also added to explicitly request checking of the EXPECTED_FD attribute. This is needed for userspace to discover whether the kernel supports the new attribute. Signed-off-by: Toke Høiland-Jørgensen Signed-off-by: Alexei Starovoitov Reviewed-by: Jakub Kicinski Link: https://lore.kernel.org/bpf/158515700640.92963.3551295145441017022.stgit@toke.dk --- include/linux/netdevice.h | 2 +- include/uapi/linux/if_link.h | 4 +++- net/core/dev.c | 26 +++++++++++++++++++++----- net/core/rtnetlink.c | 14 ++++++++++++++ 4 files changed, 39 insertions(+), 7 deletions(-) (limited to 'include/linux/netdevice.h') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 654808bfad83..b503d468f0df 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3768,7 +3768,7 @@ struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf); int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack, - int fd, u32 flags); + int fd, int expected_fd, u32 flags); u32 __dev_xdp_query(struct net_device *dev, bpf_op_t xdp_op, enum bpf_netdev_command cmd); int xdp_umem_query(struct net_device *dev, u16 queue_id); diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 61e0801c82df..c2f768c8d65b 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -972,11 +972,12 @@ enum { #define XDP_FLAGS_SKB_MODE (1U << 1) #define XDP_FLAGS_DRV_MODE (1U << 2) #define XDP_FLAGS_HW_MODE (1U << 3) +#define XDP_FLAGS_REPLACE (1U << 4) #define XDP_FLAGS_MODES (XDP_FLAGS_SKB_MODE | \ XDP_FLAGS_DRV_MODE | \ XDP_FLAGS_HW_MODE) #define XDP_FLAGS_MASK (XDP_FLAGS_UPDATE_IF_NOEXIST | \ - XDP_FLAGS_MODES) + XDP_FLAGS_MODES | XDP_FLAGS_REPLACE) /* These are stored into IFLA_XDP_ATTACHED on dump. */ enum { @@ -996,6 +997,7 @@ enum { IFLA_XDP_DRV_PROG_ID, IFLA_XDP_SKB_PROG_ID, IFLA_XDP_HW_PROG_ID, + IFLA_XDP_EXPECTED_FD, __IFLA_XDP_MAX, }; diff --git a/net/core/dev.c b/net/core/dev.c index d84541c24446..651a3c28d33a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -8655,15 +8655,17 @@ static void dev_xdp_uninstall(struct net_device *dev) * @dev: device * @extack: netlink extended ack * @fd: new program fd or negative value to clear + * @expected_fd: old program fd that userspace expects to replace or clear * @flags: xdp-related flags * * Set or clear a bpf program for a device */ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack, - int fd, u32 flags) + int fd, int expected_fd, u32 flags) { const struct net_device_ops *ops = dev->netdev_ops; enum bpf_netdev_command query; + u32 prog_id, expected_id = 0; struct bpf_prog *prog = NULL; bpf_op_t bpf_op, bpf_chk; bool offload; @@ -8684,15 +8686,29 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack, if (bpf_op == bpf_chk) bpf_chk = generic_xdp_install; - if (fd >= 0) { - u32 prog_id; + prog_id = __dev_xdp_query(dev, bpf_op, query); + if (flags & XDP_FLAGS_REPLACE) { + if (expected_fd >= 0) { + prog = bpf_prog_get_type_dev(expected_fd, + BPF_PROG_TYPE_XDP, + bpf_op == ops->ndo_bpf); + if (IS_ERR(prog)) + return PTR_ERR(prog); + expected_id = prog->aux->id; + bpf_prog_put(prog); + } + if (prog_id != expected_id) { + NL_SET_ERR_MSG(extack, "Active program does not match expected"); + return -EEXIST; + } + } + if (fd >= 0) { if (!offload && __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG)) { NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time"); return -EEXIST; } - prog_id = __dev_xdp_query(dev, bpf_op, query); if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) && prog_id) { NL_SET_ERR_MSG(extack, "XDP program already attached"); return -EBUSY; @@ -8715,7 +8731,7 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack, return 0; } } else { - if (!__dev_xdp_query(dev, bpf_op, query)) + if (!prog_id) return 0; } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 14e6ea21c378..709ebbf8ab5b 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1872,7 +1872,9 @@ static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { }; static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = { + [IFLA_XDP_UNSPEC] = { .strict_start_type = IFLA_XDP_EXPECTED_FD }, [IFLA_XDP_FD] = { .type = NLA_S32 }, + [IFLA_XDP_EXPECTED_FD] = { .type = NLA_S32 }, [IFLA_XDP_ATTACHED] = { .type = NLA_U8 }, [IFLA_XDP_FLAGS] = { .type = NLA_U32 }, [IFLA_XDP_PROG_ID] = { .type = NLA_U32 }, @@ -2799,8 +2801,20 @@ static int do_setlink(const struct sk_buff *skb, } if (xdp[IFLA_XDP_FD]) { + int expected_fd = -1; + + if (xdp_flags & XDP_FLAGS_REPLACE) { + if (!xdp[IFLA_XDP_EXPECTED_FD]) { + err = -EINVAL; + goto errout; + } + expected_fd = + nla_get_s32(xdp[IFLA_XDP_EXPECTED_FD]); + } + err = dev_change_xdp_fd(dev, extack, nla_get_s32(xdp[IFLA_XDP_FD]), + expected_fd, xdp_flags); if (err) goto errout; -- cgit v1.3-14-g43fede