From b515d2637276a3810d6595e10ab02c13bfd0b63a Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 16 Apr 2021 11:27:59 +0200 Subject: xfrm: xfrm_state_mtu should return at least 1280 for ipv6 Jianwen reported that IPv6 Interoperability tests are failing in an IPsec case where one of the links between the IPsec peers has an MTU of 1280. The peer generates a packet larger than this MTU, the router replies with a "Packet too big" message indicating an MTU of 1280. When the peer tries to send another large packet, xfrm_state_mtu returns 1280 - ipsec_overhead, which causes ip6_setup_cork to fail with EINVAL. We can fix this by forcing xfrm_state_mtu to return IPV6_MIN_MTU when IPv6 is used. After going through IPsec, the packet will then be fragmented to obey the actual network's PMTU, just before leaving the host. Currently, TFC padding is capped to PMTU - overhead to avoid fragementation: after padding and encapsulation, we still fit within the PMTU. That behavior is preserved in this patch. Fixes: 91657eafb64b ("xfrm: take net hdr len into account for esp payload size calculation") Reported-by: Jianwen Ji Signed-off-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index c58a6d4eb610..6232a5f048bd 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1546,6 +1546,7 @@ void xfrm_sad_getinfo(struct net *net, struct xfrmk_sadinfo *si); void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si); u32 xfrm_replay_seqhi(struct xfrm_state *x, __be32 net_seq); int xfrm_init_replay(struct xfrm_state *x); +u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu); u32 xfrm_state_mtu(struct xfrm_state *x, int mtu); int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload); int xfrm_init_state(struct xfrm_state *x); -- cgit v1.2.3 From fe9f1d8779cb47046e76ea209b6eece7ec56d1b4 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Sun, 25 Apr 2021 21:47:12 +0200 Subject: xfrm: add state hashtable keyed by seq When creating new states with seq set in xfrm_usersa_info, we walk through all the states already installed in that netns to find a matching ACQUIRE state (__xfrm_find_acq_byseq, called from xfrm_state_add). This causes severe slowdowns on systems with a large number of states. This patch introduces a hashtable using x->km.seq as key, so that the corresponding state can be found in a reasonable time. Signed-off-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- include/net/netns/xfrm.h | 1 + include/net/xfrm.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/netns/xfrm.h b/include/net/netns/xfrm.h index e816b6a3ef2b..e946366e8ba5 100644 --- a/include/net/netns/xfrm.h +++ b/include/net/netns/xfrm.h @@ -42,6 +42,7 @@ struct netns_xfrm { struct hlist_head __rcu *state_bydst; struct hlist_head __rcu *state_bysrc; struct hlist_head __rcu *state_byspi; + struct hlist_head __rcu *state_byseq; unsigned int state_hmask; unsigned int state_num; struct work_struct state_hash_work; diff --git a/include/net/xfrm.h b/include/net/xfrm.h index c58a6d4eb610..6e11db6fa0ab 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -154,6 +154,7 @@ struct xfrm_state { }; struct hlist_node bysrc; struct hlist_node byspi; + struct hlist_node byseq; refcount_t refcnt; spinlock_t lock; -- cgit v1.2.3 From 5796254e467bf1cff002df65fbb53ecef6a0e060 Mon Sep 17 00:00:00 2001 From: Yejune Deng Date: Mon, 17 May 2021 20:22:05 +0800 Subject: net: Remove the member netns_ok Every protocol has the 'netns_ok' member and it is euqal to 1. The 'if (!prot->netns_ok)' always false in inet_add_protocol(). Signed-off-by: Yejune Deng Signed-off-by: David S. Miller --- include/net/protocol.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/protocol.h b/include/net/protocol.h index 2b778e1d2d8f..f51c06ae365f 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -43,7 +43,6 @@ struct net_protocol { int (*err_handler)(struct sk_buff *skb, u32 info); unsigned int no_policy:1, - netns_ok:1, /* does the protocol do more stringent * icmp tag validation than simple * socket lookup? -- cgit v1.2.3 From ce5c9c20d364f156c885efed8c71fca2945db00f Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 17 May 2021 21:15:18 +0300 Subject: ipv4: Add a sysctl to control multipath hash fields A subsequent patch will add a new multipath hash policy where the packet fields used for multipath hash calculation are determined by user space. This patch adds a sysctl that allows user space to set these fields. The packet fields are represented using a bitmask and are common between IPv4 and IPv6 to allow user space to use the same numbering across both protocols. For example, to hash based on standard 5-tuple: # sysctl -w net.ipv4.fib_multipath_hash_fields=0x0037 net.ipv4.fib_multipath_hash_fields = 0x0037 The kernel rejects unknown fields, for example: # sysctl -w net.ipv4.fib_multipath_hash_fields=0x1000 sysctl: setting key "net.ipv4.fib_multipath_hash_fields": Invalid argument More fields can be added in the future, if needed. Signed-off-by: Ido Schimmel Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip_fib.h | 43 +++++++++++++++++++++++++++++++++++++++++++ include/net/netns/ipv4.h | 1 + 2 files changed, 44 insertions(+) (limited to 'include/net') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index a914f33f3ed5..3ab2563b1a23 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -466,6 +466,49 @@ int fib_sync_up(struct net_device *dev, unsigned char nh_flags); void fib_sync_mtu(struct net_device *dev, u32 orig_mtu); void fib_nhc_update_mtu(struct fib_nh_common *nhc, u32 new, u32 orig); +/* Fields used for sysctl_fib_multipath_hash_fields. + * Common to IPv4 and IPv6. + * + * Add new fields at the end. This is user API. + */ +#define FIB_MULTIPATH_HASH_FIELD_SRC_IP BIT(0) +#define FIB_MULTIPATH_HASH_FIELD_DST_IP BIT(1) +#define FIB_MULTIPATH_HASH_FIELD_IP_PROTO BIT(2) +#define FIB_MULTIPATH_HASH_FIELD_FLOWLABEL BIT(3) +#define FIB_MULTIPATH_HASH_FIELD_SRC_PORT BIT(4) +#define FIB_MULTIPATH_HASH_FIELD_DST_PORT BIT(5) +#define FIB_MULTIPATH_HASH_FIELD_INNER_SRC_IP BIT(6) +#define FIB_MULTIPATH_HASH_FIELD_INNER_DST_IP BIT(7) +#define FIB_MULTIPATH_HASH_FIELD_INNER_IP_PROTO BIT(8) +#define FIB_MULTIPATH_HASH_FIELD_INNER_FLOWLABEL BIT(9) +#define FIB_MULTIPATH_HASH_FIELD_INNER_SRC_PORT BIT(10) +#define FIB_MULTIPATH_HASH_FIELD_INNER_DST_PORT BIT(11) + +#define FIB_MULTIPATH_HASH_FIELD_OUTER_MASK \ + (FIB_MULTIPATH_HASH_FIELD_SRC_IP | \ + FIB_MULTIPATH_HASH_FIELD_DST_IP | \ + FIB_MULTIPATH_HASH_FIELD_IP_PROTO | \ + FIB_MULTIPATH_HASH_FIELD_FLOWLABEL | \ + FIB_MULTIPATH_HASH_FIELD_SRC_PORT | \ + FIB_MULTIPATH_HASH_FIELD_DST_PORT) + +#define FIB_MULTIPATH_HASH_FIELD_INNER_MASK \ + (FIB_MULTIPATH_HASH_FIELD_INNER_SRC_IP | \ + FIB_MULTIPATH_HASH_FIELD_INNER_DST_IP | \ + FIB_MULTIPATH_HASH_FIELD_INNER_IP_PROTO | \ + FIB_MULTIPATH_HASH_FIELD_INNER_FLOWLABEL | \ + FIB_MULTIPATH_HASH_FIELD_INNER_SRC_PORT | \ + FIB_MULTIPATH_HASH_FIELD_INNER_DST_PORT) + +#define FIB_MULTIPATH_HASH_FIELD_ALL_MASK \ + (FIB_MULTIPATH_HASH_FIELD_OUTER_MASK | \ + FIB_MULTIPATH_HASH_FIELD_INNER_MASK) + +#define FIB_MULTIPATH_HASH_FIELD_DEFAULT_MASK \ + (FIB_MULTIPATH_HASH_FIELD_SRC_IP | \ + FIB_MULTIPATH_HASH_FIELD_DST_IP | \ + FIB_MULTIPATH_HASH_FIELD_IP_PROTO) + #ifdef CONFIG_IP_ROUTE_MULTIPATH int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4, const struct sk_buff *skb, struct flow_keys *flkeys); diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index f6af8d96d3c6..746c80cd4257 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -210,6 +210,7 @@ struct netns_ipv4 { #endif #endif #ifdef CONFIG_IP_ROUTE_MULTIPATH + u32 sysctl_fib_multipath_hash_fields; u8 sysctl_fib_multipath_use_neigh; u8 sysctl_fib_multipath_hash_policy; #endif -- cgit v1.2.3 From ed13923f980ef84dde0b9010b9e09052dc31a909 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 17 May 2021 21:15:22 +0300 Subject: ipv6: Add a sysctl to control multipath hash fields A subsequent patch will add a new multipath hash policy where the packet fields used for multipath hash calculation are determined by user space. This patch adds a sysctl that allows user space to set these fields. The packet fields are represented using a bitmask and are common between IPv4 and IPv6 to allow user space to use the same numbering across both protocols. For example, to hash based on standard 5-tuple: # sysctl -w net.ipv6.fib_multipath_hash_fields=0x0037 net.ipv6.fib_multipath_hash_fields = 0x0037 To avoid introducing holes in 'struct netns_sysctl_ipv6', move the 'bindv6only' field after the multipath hash fields. The kernel rejects unknown fields, for example: # sysctl -w net.ipv6.fib_multipath_hash_fields=0x1000 sysctl: setting key "net.ipv6.fib_multipath_hash_fields": Invalid argument Signed-off-by: Ido Schimmel Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ipv6.h | 8 ++++++++ include/net/netns/ipv6.h | 3 ++- 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 448bf2b34759..f2d0ecc257bb 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -926,11 +926,19 @@ static inline int ip6_multipath_hash_policy(const struct net *net) { return net->ipv6.sysctl.multipath_hash_policy; } +static inline u32 ip6_multipath_hash_fields(const struct net *net) +{ + return net->ipv6.sysctl.multipath_hash_fields; +} #else static inline int ip6_multipath_hash_policy(const struct net *net) { return 0; } +static inline u32 ip6_multipath_hash_fields(const struct net *net) +{ + return 0; +} #endif /* diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 6153c8067009..bde0b7adb4a3 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -28,8 +28,9 @@ struct netns_sysctl_ipv6 { int ip6_rt_gc_elasticity; int ip6_rt_mtu_expires; int ip6_rt_min_advmss; - u8 bindv6only; + u32 multipath_hash_fields; u8 multipath_hash_policy; + u8 bindv6only; u8 flowlabel_consistency; u8 auto_flowlabels; int icmpv6_time; -- cgit v1.2.3 From e624d4ed4aa8cc3c69d1359b0aaea539203ed266 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Wed, 19 May 2021 17:07:45 +0800 Subject: xdp: Extend xdp_redirect_map with broadcast support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds two flags BPF_F_BROADCAST and BPF_F_EXCLUDE_INGRESS to extend xdp_redirect_map for broadcast support. With BPF_F_BROADCAST the packet will be broadcasted to all the interfaces in the map. with BPF_F_EXCLUDE_INGRESS the ingress interface will be excluded when do broadcasting. When getting the devices in dev hash map via dev_map_hash_get_next_key(), there is a possibility that we fall back to the first key when a device was removed. This will duplicate packets on some interfaces. So just walk the whole buckets to avoid this issue. For dev array map, we also walk the whole map to find valid interfaces. Function bpf_clear_redirect_map() was removed in commit ee75aef23afe ("bpf, xdp: Restructure redirect actions"). Add it back as we need to use ri->map again. With test topology: +-------------------+ +-------------------+ | Host A (i40e 10G) | ---------- | eno1(i40e 10G) | +-------------------+ | | | Host B | +-------------------+ | | | Host C (i40e 10G) | ---------- | eno2(i40e 10G) | +-------------------+ | | | +------+ | | veth0 -- | Peer | | | veth1 -- | | | | veth2 -- | NS | | | +------+ | +-------------------+ On Host A: # pktgen/pktgen_sample03_burst_single_flow.sh -i eno1 -d $dst_ip -m $dst_mac -s 64 On Host B(Intel(R) Xeon(R) CPU E5-2690 v3 @ 2.60GHz, 128G Memory): Use xdp_redirect_map and xdp_redirect_map_multi in samples/bpf for testing. All the veth peers in the NS have a XDP_DROP program loaded. The forward_map max_entries in xdp_redirect_map_multi is modify to 4. Testing the performance impact on the regular xdp_redirect path with and without patch (to check impact of additional check for broadcast mode): 5.12 rc4 | redirect_map i40e->i40e | 2.0M | 9.7M 5.12 rc4 | redirect_map i40e->veth | 1.7M | 11.8M 5.12 rc4 + patch | redirect_map i40e->i40e | 2.0M | 9.6M 5.12 rc4 + patch | redirect_map i40e->veth | 1.7M | 11.7M Testing the performance when cloning packets with the redirect_map_multi test, using a redirect map size of 4, filled with 1-3 devices: 5.12 rc4 + patch | redirect_map multi i40e->veth (x1) | 1.7M | 11.4M 5.12 rc4 + patch | redirect_map multi i40e->veth (x2) | 1.1M | 4.3M 5.12 rc4 + patch | redirect_map multi i40e->veth (x3) | 0.8M | 2.6M Signed-off-by: Hangbin Liu Signed-off-by: Daniel Borkmann Acked-by: Toke Høiland-Jørgensen Acked-by: Martin KaFai Lau Acked-by: John Fastabend Acked-by: Jesper Dangaard Brouer Link: https://lore.kernel.org/bpf/20210519090747.1655268-3-liuhangbin@gmail.com --- include/net/xdp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/xdp.h b/include/net/xdp.h index a5bc214a49d9..5533f0ab2afc 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -170,6 +170,7 @@ struct sk_buff *__xdp_build_skb_from_frame(struct xdp_frame *xdpf, struct sk_buff *xdp_build_skb_from_frame(struct xdp_frame *xdpf, struct net_device *dev); int xdp_alloc_skb_bulk(void **skbs, int n_skb, gfp_t gfp); +struct xdp_frame *xdpf_clone(struct xdp_frame *xdpf); static inline void xdp_convert_frame_to_buff(struct xdp_frame *frame, struct xdp_buff *xdp) -- cgit v1.2.3 From 0974cff3eb66764cc86b60f1071958acc432a4e8 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 May 2021 22:29:55 +0200 Subject: netfilter: add and use nft_set_do_lookup helper Followup patch will add a CONFIG_RETPOLINE wrapper to avoid the ops->lookup() indirection cost for retpoline builds. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables_core.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h index fd10a7862fdc..5eb699454490 100644 --- a/include/net/netfilter/nf_tables_core.h +++ b/include/net/netfilter/nf_tables_core.h @@ -88,6 +88,13 @@ extern const struct nft_set_type nft_set_bitmap_type; extern const struct nft_set_type nft_set_pipapo_type; extern const struct nft_set_type nft_set_pipapo_avx2_type; +static inline bool +nft_set_do_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext) +{ + return set->ops->lookup(net, set, key, ext); +} + struct nft_expr; struct nft_regs; struct nft_pktinfo; -- cgit v1.2.3 From f227925e53c3ecc168027e0015ab0a953d1bf013 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 May 2021 22:29:56 +0200 Subject: netfilter: nf_tables: prefer direct calls for set lookups Extend nft_set_do_lookup() to use direct calls when retpoline feature is enabled. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables_core.h | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h index 5eb699454490..46c8d5bb5d8d 100644 --- a/include/net/netfilter/nf_tables_core.h +++ b/include/net/netfilter/nf_tables_core.h @@ -3,6 +3,7 @@ #define _NET_NF_TABLES_CORE_H #include +#include extern struct nft_expr_type nft_imm_type; extern struct nft_expr_type nft_cmp_type; @@ -88,12 +89,35 @@ extern const struct nft_set_type nft_set_bitmap_type; extern const struct nft_set_type nft_set_pipapo_type; extern const struct nft_set_type nft_set_pipapo_avx2_type; +#ifdef CONFIG_RETPOLINE +bool nft_rhash_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +bool nft_bitmap_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +bool nft_hash_lookup_fast(const struct net *net, + const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +bool nft_hash_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +bool nft_set_do_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +#else static inline bool nft_set_do_lookup(const struct net *net, const struct nft_set *set, const u32 *key, const struct nft_set_ext **ext) { return set->ops->lookup(net, set, key, ext); } +#endif + +/* called from nft_pipapo_avx2.c */ +bool nft_pipapo_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); +/* called from nft_set_pipapo.c */ +bool nft_pipapo_avx2_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext); struct nft_expr; struct nft_regs; -- cgit v1.2.3 From 85554eb981e5a8b0b8947611193aef1737081ef2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 28 May 2021 12:30:05 +0200 Subject: netfilter: nf_tables: add and use nft_sk helper This allows to change storage placement later on without changing readers. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 27eeb613bb4e..af1228f58e5a 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -29,6 +29,11 @@ struct nft_pktinfo { struct xt_action_param xt; }; +static inline struct sock *nft_sk(const struct nft_pktinfo *pkt) +{ + return pkt->xt.state->sk; +} + static inline struct net *nft_net(const struct nft_pktinfo *pkt) { return pkt->xt.state->net; -- cgit v1.2.3 From 2d7b4ace0754ebaaf71c6824880178d46aa0ab33 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 28 May 2021 12:30:06 +0200 Subject: netfilter: nf_tables: add and use nft_thoff helper This allows to change storage placement later on without changing readers. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index af1228f58e5a..10c1b8759990 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -34,6 +34,11 @@ static inline struct sock *nft_sk(const struct nft_pktinfo *pkt) return pkt->xt.state->sk; } +static inline unsigned int nft_thoff(const struct nft_pktinfo *pkt) +{ + return pkt->xt.thoff; +} + static inline struct net *nft_net(const struct nft_pktinfo *pkt) { return pkt->xt.state->net; -- cgit v1.2.3 From f06ad944b6a92dd9ce95f2e5f4164a8e70d32af5 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 28 May 2021 12:30:07 +0200 Subject: netfilter: nf_tables: remove unused arg in nft_set_pktinfo_unspec() The functions pass extra skb arg, but either its not used or the helpers can already access it via pkt->skb. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 3 +-- include/net/netfilter/nf_tables_ipv4.h | 28 ++++++++++++---------------- include/net/netfilter/nf_tables_ipv6.h | 30 +++++++++++++----------------- 3 files changed, 26 insertions(+), 35 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 10c1b8759990..958b8e68bb1a 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -72,8 +72,7 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt, pkt->xt.state = state; } -static inline void nft_set_pktinfo_unspec(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline void nft_set_pktinfo_unspec(struct nft_pktinfo *pkt) { pkt->tprot_set = false; pkt->tprot = 0; diff --git a/include/net/netfilter/nf_tables_ipv4.h b/include/net/netfilter/nf_tables_ipv4.h index 1f7bea39ad1b..b185a9216bf1 100644 --- a/include/net/netfilter/nf_tables_ipv4.h +++ b/include/net/netfilter/nf_tables_ipv4.h @@ -5,8 +5,7 @@ #include #include -static inline void nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline void nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt) { struct iphdr *ip; @@ -17,14 +16,13 @@ static inline void nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt, pkt->xt.fragoff = ntohs(ip->frag_off) & IP_OFFSET; } -static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt) { struct iphdr *iph, _iph; u32 len, thoff; - iph = skb_header_pointer(skb, skb_network_offset(skb), sizeof(*iph), - &_iph); + iph = skb_header_pointer(pkt->skb, skb_network_offset(pkt->skb), + sizeof(*iph), &_iph); if (!iph) return -1; @@ -33,7 +31,7 @@ static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt, len = ntohs(iph->tot_len); thoff = iph->ihl * 4; - if (skb->len < len) + if (pkt->skb->len < len) return -1; else if (len < thoff) return -1; @@ -46,29 +44,27 @@ static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt, return 0; } -static inline void nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline void nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt) { - if (__nft_set_pktinfo_ipv4_validate(pkt, skb) < 0) - nft_set_pktinfo_unspec(pkt, skb); + if (__nft_set_pktinfo_ipv4_validate(pkt) < 0) + nft_set_pktinfo_unspec(pkt); } -static inline int nft_set_pktinfo_ipv4_ingress(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline int nft_set_pktinfo_ipv4_ingress(struct nft_pktinfo *pkt) { struct iphdr *iph; u32 len, thoff; - if (!pskb_may_pull(skb, sizeof(*iph))) + if (!pskb_may_pull(pkt->skb, sizeof(*iph))) return -1; - iph = ip_hdr(skb); + iph = ip_hdr(pkt->skb); if (iph->ihl < 5 || iph->version != 4) goto inhdr_error; len = ntohs(iph->tot_len); thoff = iph->ihl * 4; - if (skb->len < len) { + if (pkt->skb->len < len) { __IP_INC_STATS(nft_net(pkt), IPSTATS_MIB_INTRUNCATEDPKTS); return -1; } else if (len < thoff) { diff --git a/include/net/netfilter/nf_tables_ipv6.h b/include/net/netfilter/nf_tables_ipv6.h index 867de29f3f7a..bf132d488b17 100644 --- a/include/net/netfilter/nf_tables_ipv6.h +++ b/include/net/netfilter/nf_tables_ipv6.h @@ -6,8 +6,7 @@ #include #include -static inline void nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline void nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt) { unsigned int flags = IP6_FH_F_AUTH; int protohdr, thoff = 0; @@ -15,7 +14,7 @@ static inline void nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt, protohdr = ipv6_find_hdr(pkt->skb, &thoff, -1, &frag_off, &flags); if (protohdr < 0) { - nft_set_pktinfo_unspec(pkt, skb); + nft_set_pktinfo_unspec(pkt); return; } @@ -25,8 +24,7 @@ static inline void nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt, pkt->xt.fragoff = frag_off; } -static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt) { #if IS_ENABLED(CONFIG_IPV6) unsigned int flags = IP6_FH_F_AUTH; @@ -36,8 +34,8 @@ static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt, int protohdr; u32 pkt_len; - ip6h = skb_header_pointer(skb, skb_network_offset(skb), sizeof(*ip6h), - &_ip6h); + ip6h = skb_header_pointer(pkt->skb, skb_network_offset(pkt->skb), + sizeof(*ip6h), &_ip6h); if (!ip6h) return -1; @@ -45,7 +43,7 @@ static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt, return -1; pkt_len = ntohs(ip6h->payload_len); - if (pkt_len + sizeof(*ip6h) > skb->len) + if (pkt_len + sizeof(*ip6h) > pkt->skb->len) return -1; protohdr = ipv6_find_hdr(pkt->skb, &thoff, -1, &frag_off, &flags); @@ -63,15 +61,13 @@ static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt, #endif } -static inline void nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline void nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt) { - if (__nft_set_pktinfo_ipv6_validate(pkt, skb) < 0) - nft_set_pktinfo_unspec(pkt, skb); + if (__nft_set_pktinfo_ipv6_validate(pkt) < 0) + nft_set_pktinfo_unspec(pkt); } -static inline int nft_set_pktinfo_ipv6_ingress(struct nft_pktinfo *pkt, - struct sk_buff *skb) +static inline int nft_set_pktinfo_ipv6_ingress(struct nft_pktinfo *pkt) { #if IS_ENABLED(CONFIG_IPV6) unsigned int flags = IP6_FH_F_AUTH; @@ -82,15 +78,15 @@ static inline int nft_set_pktinfo_ipv6_ingress(struct nft_pktinfo *pkt, int protohdr; u32 pkt_len; - if (!pskb_may_pull(skb, sizeof(*ip6h))) + if (!pskb_may_pull(pkt->skb, sizeof(*ip6h))) return -1; - ip6h = ipv6_hdr(skb); + ip6h = ipv6_hdr(pkt->skb); if (ip6h->version != 6) goto inhdr_error; pkt_len = ntohs(ip6h->payload_len); - if (pkt_len + sizeof(*ip6h) > skb->len) { + if (pkt_len + sizeof(*ip6h) > pkt->skb->len) { idev = __in6_dev_get(nft_in(pkt)); __IP6_INC_STATS(nft_net(pkt), idev, IPSTATS_MIB_INTRUNCATEDPKTS); return -1; -- cgit v1.2.3 From 897389de48283d413728dae7520973872cef8eb2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 28 May 2021 12:30:08 +0200 Subject: netfilter: nf_tables: remove xt_action_param from nft_pktinfo Init it on demand in the nft_compat expression. This reduces size of nft_pktinfo from 48 to 24 bytes on x86_64. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 25 +++++++++++++------------ include/net/netfilter/nf_tables_ipv4.h | 12 ++++++------ include/net/netfilter/nf_tables_ipv6.h | 12 ++++++------ 3 files changed, 25 insertions(+), 24 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 958b8e68bb1a..6783164428f1 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -23,45 +23,46 @@ struct module; struct nft_pktinfo { struct sk_buff *skb; + const struct nf_hook_state *state; bool tprot_set; u8 tprot; - /* for x_tables compatibility */ - struct xt_action_param xt; + u16 fragoff; + unsigned int thoff; }; static inline struct sock *nft_sk(const struct nft_pktinfo *pkt) { - return pkt->xt.state->sk; + return pkt->state->sk; } static inline unsigned int nft_thoff(const struct nft_pktinfo *pkt) { - return pkt->xt.thoff; + return pkt->thoff; } static inline struct net *nft_net(const struct nft_pktinfo *pkt) { - return pkt->xt.state->net; + return pkt->state->net; } static inline unsigned int nft_hook(const struct nft_pktinfo *pkt) { - return pkt->xt.state->hook; + return pkt->state->hook; } static inline u8 nft_pf(const struct nft_pktinfo *pkt) { - return pkt->xt.state->pf; + return pkt->state->pf; } static inline const struct net_device *nft_in(const struct nft_pktinfo *pkt) { - return pkt->xt.state->in; + return pkt->state->in; } static inline const struct net_device *nft_out(const struct nft_pktinfo *pkt) { - return pkt->xt.state->out; + return pkt->state->out; } static inline void nft_set_pktinfo(struct nft_pktinfo *pkt, @@ -69,15 +70,15 @@ static inline void nft_set_pktinfo(struct nft_pktinfo *pkt, const struct nf_hook_state *state) { pkt->skb = skb; - pkt->xt.state = state; + pkt->state = state; } static inline void nft_set_pktinfo_unspec(struct nft_pktinfo *pkt) { pkt->tprot_set = false; pkt->tprot = 0; - pkt->xt.thoff = 0; - pkt->xt.fragoff = 0; + pkt->thoff = 0; + pkt->fragoff = 0; } /** diff --git a/include/net/netfilter/nf_tables_ipv4.h b/include/net/netfilter/nf_tables_ipv4.h index b185a9216bf1..eb4c094cd54d 100644 --- a/include/net/netfilter/nf_tables_ipv4.h +++ b/include/net/netfilter/nf_tables_ipv4.h @@ -12,8 +12,8 @@ static inline void nft_set_pktinfo_ipv4(struct nft_pktinfo *pkt) ip = ip_hdr(pkt->skb); pkt->tprot_set = true; pkt->tprot = ip->protocol; - pkt->xt.thoff = ip_hdrlen(pkt->skb); - pkt->xt.fragoff = ntohs(ip->frag_off) & IP_OFFSET; + pkt->thoff = ip_hdrlen(pkt->skb); + pkt->fragoff = ntohs(ip->frag_off) & IP_OFFSET; } static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt) @@ -38,8 +38,8 @@ static inline int __nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt) pkt->tprot_set = true; pkt->tprot = iph->protocol; - pkt->xt.thoff = thoff; - pkt->xt.fragoff = ntohs(iph->frag_off) & IP_OFFSET; + pkt->thoff = thoff; + pkt->fragoff = ntohs(iph->frag_off) & IP_OFFSET; return 0; } @@ -73,8 +73,8 @@ static inline int nft_set_pktinfo_ipv4_ingress(struct nft_pktinfo *pkt) pkt->tprot_set = true; pkt->tprot = iph->protocol; - pkt->xt.thoff = thoff; - pkt->xt.fragoff = ntohs(iph->frag_off) & IP_OFFSET; + pkt->thoff = thoff; + pkt->fragoff = ntohs(iph->frag_off) & IP_OFFSET; return 0; diff --git a/include/net/netfilter/nf_tables_ipv6.h b/include/net/netfilter/nf_tables_ipv6.h index bf132d488b17..7595e02b00ba 100644 --- a/include/net/netfilter/nf_tables_ipv6.h +++ b/include/net/netfilter/nf_tables_ipv6.h @@ -20,8 +20,8 @@ static inline void nft_set_pktinfo_ipv6(struct nft_pktinfo *pkt) pkt->tprot_set = true; pkt->tprot = protohdr; - pkt->xt.thoff = thoff; - pkt->xt.fragoff = frag_off; + pkt->thoff = thoff; + pkt->fragoff = frag_off; } static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt) @@ -52,8 +52,8 @@ static inline int __nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt) pkt->tprot_set = true; pkt->tprot = protohdr; - pkt->xt.thoff = thoff; - pkt->xt.fragoff = frag_off; + pkt->thoff = thoff; + pkt->fragoff = frag_off; return 0; #else @@ -98,8 +98,8 @@ static inline int nft_set_pktinfo_ipv6_ingress(struct nft_pktinfo *pkt) pkt->tprot_set = true; pkt->tprot = protohdr; - pkt->xt.thoff = thoff; - pkt->xt.fragoff = frag_off; + pkt->thoff = thoff; + pkt->fragoff = frag_off; return 0; -- cgit v1.2.3 From 6e1e89418a5ccdfb325aed538307c2f9dba6ef51 Mon Sep 17 00:00:00 2001 From: Shaokun Zhang Date: Sat, 29 May 2021 15:52:02 +0800 Subject: xfrm: Remove the repeated declaration Function 'xfrm_parse_spi' is declared twice, so remove the repeated declaration. Cc: Steffen Klassert Cc: Herbert Xu Cc: "David S. Miller" Signed-off-by: Shaokun Zhang Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 6232a5f048bd..b30623678430 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1571,7 +1571,6 @@ int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type); int xfrm4_transport_finish(struct sk_buff *skb, int async); int xfrm4_rcv(struct sk_buff *skb); -int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, __be32 *spi, __be32 *seq); static inline int xfrm4_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi) { -- cgit v1.2.3 From 9c5eee0afca09cbde6bd00f77876754aaa552970 Mon Sep 17 00:00:00 2001 From: Boris Sukholitko Date: Tue, 1 Jun 2021 15:30:50 +0300 Subject: net/sched: act_vlan: Fix modify to allow 0 Currently vlan modification action checks existence of vlan priority by comparing it to 0. Therefore it is impossible to modify existing vlan tag to have priority 0. For example, the following tc command will change the vlan id but will not affect vlan priority: tc filter add dev eth1 ingress matchall action vlan modify id 300 \ priority 0 pipe mirred egress redirect dev eth2 The incoming packet on eth1: ethertype 802.1Q (0x8100), vlan 200, p 4, ethertype IPv4 will be changed to: ethertype 802.1Q (0x8100), vlan 300, p 4, ethertype IPv4 although the user has intended to have p == 0. The fix is to add tcfv_push_prio_exists flag to struct tcf_vlan_params and rely on it when deciding to set the priority. Fixes: 45a497f2d149a4a8061c (net/sched: act_vlan: Introduce TCA_VLAN_ACT_MODIFY vlan action) Signed-off-by: Boris Sukholitko Signed-off-by: David S. Miller --- include/net/tc_act/tc_vlan.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/tc_act/tc_vlan.h b/include/net/tc_act/tc_vlan.h index f051046ba034..f94b8bc26f9e 100644 --- a/include/net/tc_act/tc_vlan.h +++ b/include/net/tc_act/tc_vlan.h @@ -16,6 +16,7 @@ struct tcf_vlan_params { u16 tcfv_push_vid; __be16 tcfv_push_proto; u8 tcfv_push_prio; + bool tcfv_push_prio_exists; struct rcu_head rcu; }; -- cgit v1.2.3 From 4677efc486e1872f62d4632c50f7183f82296fa6 Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Wed, 2 Jun 2021 15:17:19 +0300 Subject: devlink: Introduce rate object Allow registering rate object for devlink ports with dedicated devlink_rate_leaf_{create|destroy}() API. Implement new netlink DEVLINK_CMD_RATE_GET command that is used to retrieve rate object info. Add new DEVLINK_CMD_RATE_{NEW|DEL} commands that are used for notifications when creating/deleting leaf rate object. Rate API is intended to be used for rate limiting of individual devlink ports (leafs) and their aggregates (nodes). Example: $ devlink port show pci/0000:03:00.0/0 pci/0000:03:00.0/1 $ devlink port function rate show pci/0000:03:00.0/0: type leaf pci/0000:03:00.0/1: type leaf Co-developed-by: Vlad Buslov Signed-off-by: Vlad Buslov Signed-off-by: Dmytro Linkin Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/devlink.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 7c984cadfec4..2f5954d96c3e 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -34,6 +34,7 @@ struct devlink_ops; struct devlink { struct list_head list; struct list_head port_list; + struct list_head rate_list; struct list_head sb_list; struct list_head dpipe_table_list; struct list_head resource_list; @@ -133,6 +134,15 @@ struct devlink_port_attrs { }; }; +struct devlink_rate { + struct list_head list; + enum devlink_rate_type type; + struct devlink *devlink; + void *priv; + + struct devlink_port *devlink_port; +}; + struct devlink_port { struct list_head list; struct list_head param_list; @@ -152,6 +162,8 @@ struct devlink_port { struct delayed_work type_warn_dw; struct list_head reporter_list; struct mutex reporters_lock; /* Protects reporter_list */ + + struct devlink_rate *devlink_rate; }; struct devlink_port_new_attrs { @@ -1512,6 +1524,8 @@ void devlink_port_attrs_pci_vf_set(struct devlink_port *devlink_port, u32 contro void devlink_port_attrs_pci_sf_set(struct devlink_port *devlink_port, u32 controller, u16 pf, u32 sf, bool external); +int devlink_rate_leaf_create(struct devlink_port *port, void *priv); +void devlink_rate_leaf_destroy(struct devlink_port *devlink_port); int devlink_sb_register(struct devlink *devlink, unsigned int sb_index, u32 size, u16 ingress_pools_count, u16 egress_pools_count, u16 ingress_tc_count, -- cgit v1.2.3 From 1897db2ec3109eb1dd07b357c95c5e03d54e41b9 Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Wed, 2 Jun 2021 15:17:22 +0300 Subject: devlink: Allow setting tx rate for devlink rate leaf objects Implement support for DEVLINK_CMD_RATE_SET command with new attributes DEVLINK_ATTR_RATE_TX_{SHARE|MAX} that are used to set devlink rate shared/max tx rate values. Extend devlink ops with new callbacks rate_leaf_tx_{share|max}_set() to allow supporting drivers to implement rate control through devlink. New attributes are optional. Driver implementations are allowed to support either or both of them. Shared rate example: $ devlink port function rate set netdevsim/netdevsim10/0 tx_share 10mbit $ devlink port function rate show netdevsim/netdevsim10/0 netdevsim/netdevsim10/0: type leaf tx_share 10mbit Max rate example: $ devlink port function rate set netdevsim/netdevsim10/0 tx_max 100mbit $ devlink port function rate show netdevsim/netdevsim10/0 netdevsim/netdevsim10/0: type leaf tx_max 100mbit Co-developed-by: Vlad Buslov Signed-off-by: Vlad Buslov Signed-off-by: Dmytro Linkin Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/devlink.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 2f5954d96c3e..46d553545f83 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -139,6 +139,8 @@ struct devlink_rate { enum devlink_rate_type type; struct devlink *devlink; void *priv; + u64 tx_share; + u64 tx_max; struct devlink_port *devlink_port; }; @@ -1465,6 +1467,14 @@ struct devlink_ops { struct devlink_port *port, enum devlink_port_fn_state state, struct netlink_ext_ack *extack); + + /** + * Rate control callbacks. + */ + int (*rate_leaf_tx_share_set)(struct devlink_rate *devlink_rate, void *priv, + u64 tx_share, struct netlink_ext_ack *extack); + int (*rate_leaf_tx_max_set)(struct devlink_rate *devlink_rate, void *priv, + u64 tx_max, struct netlink_ext_ack *extack); }; static inline void *devlink_priv(struct devlink *devlink) -- cgit v1.2.3 From a8ecb93ef03de4c59fb6289f99bc9616a852c917 Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Wed, 2 Jun 2021 15:17:25 +0300 Subject: devlink: Introduce rate nodes Implement support for DEVLINK_CMD_RATE_{NEW|DEL} commands that are used to create and delete devlink rate nodes. Add new attribute DEVLINK_ATTR_RATE_NODE_NAME that specify node name string. The node name is an alphanumeric identifier. No valid node name can be a devlink port index, eg. decimal number. Extend devlink ops with new callbacks rate_node_{new|del}() and rate_node_tx_{share|max}_set() to allow supporting drivers to implement ports rate grouping and setting tx rate of rate nodes through devlink. Expose devlink_rate_nodes_destroy() function to allow vendor driver do proper cleanup of internally allocated resources for the nodes if the driver goes down or due to any other reasons which requires nodes to be destroyed. Disallow moving device from switchdev to legacy mode if any node exists on that device. User must explicitly delete nodes before switching mode. Example: $ devlink port function rate add netdevsim/netdevsim10/group1 $ devlink port function rate set netdevsim/netdevsim10/group1 \ tx_share 10mbit tx_max 100mbit Add + set command can be combined: $ devlink port function rate add netdevsim/netdevsim10/group1 \ tx_share 10mbit tx_max 100mbit $ devlink port function rate show netdevsim/netdevsim10/group1 netdevsim/netdevsim10/group1: type node tx_share 10mbit tx_max 100mbit $ devlink port function rate del netdevsim/netdevsim10/group1 Co-developed-by: Vlad Buslov Signed-off-by: Vlad Buslov Signed-off-by: Dmytro Linkin Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/devlink.h | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 46d553545f83..13162b579124 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -142,7 +142,10 @@ struct devlink_rate { u64 tx_share; u64 tx_max; - struct devlink_port *devlink_port; + union { + struct devlink_port *devlink_port; + char *name; + }; }; struct devlink_port { @@ -1475,6 +1478,14 @@ struct devlink_ops { u64 tx_share, struct netlink_ext_ack *extack); int (*rate_leaf_tx_max_set)(struct devlink_rate *devlink_rate, void *priv, u64 tx_max, struct netlink_ext_ack *extack); + int (*rate_node_tx_share_set)(struct devlink_rate *devlink_rate, void *priv, + u64 tx_share, struct netlink_ext_ack *extack); + int (*rate_node_tx_max_set)(struct devlink_rate *devlink_rate, void *priv, + u64 tx_max, struct netlink_ext_ack *extack); + int (*rate_node_new)(struct devlink_rate *rate_node, void **priv, + struct netlink_ext_ack *extack); + int (*rate_node_del)(struct devlink_rate *rate_node, void *priv, + struct netlink_ext_ack *extack); }; static inline void *devlink_priv(struct devlink *devlink) @@ -1536,6 +1547,7 @@ void devlink_port_attrs_pci_sf_set(struct devlink_port *devlink_port, bool external); int devlink_rate_leaf_create(struct devlink_port *port, void *priv); void devlink_rate_leaf_destroy(struct devlink_port *devlink_port); +void devlink_rate_nodes_destroy(struct devlink *devlink); int devlink_sb_register(struct devlink *devlink, unsigned int sb_index, u32 size, u16 ingress_pools_count, u16 egress_pools_count, u16 ingress_tc_count, -- cgit v1.2.3 From d7555984507822458b32a6405881038241d140be Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Wed, 2 Jun 2021 15:17:28 +0300 Subject: devlink: Allow setting parent node of rate objects Refactor DEVLINK_CMD_RATE_{GET|SET} command handlers to support setting a node as a parent for another rate object (leaf or node) by means of new attribute DEVLINK_ATTR_RATE_PARENT_NODE_NAME. Extend devlink ops with new callbacks rate_{leaf|node}_parent_set() to set node as a parent for rate object to allow supporting drivers to implement rate grouping through devlink. Driver implementations are allowed to support leafs or node children only. Invoking callback with NULL as parent should be threated by the driver as unset parent action. Extend rate object struct with reference counter to disallow deleting a node with any child pointing to it. User should unset parent for the child explicitly. Example: $ devlink port function rate add netdevsim/netdevsim10/group1 $ devlink port function rate add netdevsim/netdevsim10/group2 $ devlink port function rate set netdevsim/netdevsim10/group1 parent group2 $ devlink port function rate show netdevsim/netdevsim10/group1 netdevsim/netdevsim10/group1: type node parent group2 $ devlink port function rate set netdevsim/netdevsim10/group1 noparent Co-developed-by: Vlad Buslov Signed-off-by: Vlad Buslov Signed-off-by: Dmytro Linkin Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/devlink.h | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 13162b579124..eb045f1b5d1d 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -142,9 +142,13 @@ struct devlink_rate { u64 tx_share; u64 tx_max; + struct devlink_rate *parent; union { struct devlink_port *devlink_port; - char *name; + struct { + char *name; + refcount_t refcnt; + }; }; }; @@ -1486,6 +1490,14 @@ struct devlink_ops { struct netlink_ext_ack *extack); int (*rate_node_del)(struct devlink_rate *rate_node, void *priv, struct netlink_ext_ack *extack); + int (*rate_leaf_parent_set)(struct devlink_rate *child, + struct devlink_rate *parent, + void *priv_child, void *priv_parent, + struct netlink_ext_ack *extack); + int (*rate_node_parent_set)(struct devlink_rate *child, + struct devlink_rate *parent, + void *priv_child, void *priv_parent, + struct netlink_ext_ack *extack); }; static inline void *devlink_priv(struct devlink *devlink) -- cgit v1.2.3 From 371087aa476ab0ac0072303ac94a3bba2d7b0a1d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 3 Jun 2021 16:24:27 -0700 Subject: sock: expose so_timestamp options for mptcp This exports SO_TIMESTAMP_* function for re-use by MPTCP. Without this there is too much copy & paste needed to support this from mptcp setsockopt path. Acked-by: Paolo Abeni Signed-off-by: Florian Westphal Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/sock.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 0e962d8bc73b..7e0116b1a73f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2743,6 +2743,7 @@ static inline bool sk_dev_equal_l3scope(struct sock *sk, int dif) void sock_def_readable(struct sock *sk); int sock_bindtoindex(struct sock *sk, int ifindex, bool lock_sk); +void sock_set_timestamp(struct sock *sk, int optname, bool valbool); void sock_enable_timestamps(struct sock *sk); void sock_no_linger(struct sock *sk); void sock_set_keepalive(struct sock *sk); -- cgit v1.2.3 From ced122d90f52eb6ff37272e32941845d46ac64c6 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 3 Jun 2021 16:24:28 -0700 Subject: sock: expose so_timestamping options for mptcp Similar to previous patch: expose SO_TIMESTAMPING helper so we do not have to copy & paste this into the mptcp core. Acked-by: Paolo Abeni Signed-off-by: Florian Westphal Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/sock.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 7e0116b1a73f..9b341c2c924f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2744,6 +2744,8 @@ void sock_def_readable(struct sock *sk); int sock_bindtoindex(struct sock *sk, int ifindex, bool lock_sk); void sock_set_timestamp(struct sock *sk, int optname, bool valbool); +int sock_set_timestamping(struct sock *sk, int optname, int val); + void sock_enable_timestamps(struct sock *sk); void sock_no_linger(struct sock *sk); void sock_set_keepalive(struct sock *sk); -- cgit v1.2.3 From 892bfd3ded0ef0f895ed6356d0f85e2009421747 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 3 Jun 2021 16:24:31 -0700 Subject: tcp: export timestamp helpers for mptcp MPTCP is builtin, so no need to add EXPORT_SYMBOL()s. It will be used to support SO_TIMESTAMP(NS) ancillary messages in the mptcp receive path. Acked-by: Paolo Abeni Signed-off-by: Florian Westphal Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/tcp.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index d05193cb0d99..e668f1bf780d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -412,6 +412,10 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int flags, int *addr_len); int tcp_set_rcvlowat(struct sock *sk, int val); int tcp_set_window_clamp(struct sock *sk, int val); +void tcp_update_recv_tstamps(struct sk_buff *skb, + struct scm_timestamping_internal *tss); +void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, + struct scm_timestamping_internal *tss); void tcp_data_ready(struct sock *sk); #ifdef CONFIG_MMU int tcp_mmap(struct file *file, struct socket *sock, -- cgit v1.2.3 From 0418b989a467885292c294923dec66951c1c1398 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 2 Jun 2021 12:39:07 +0200 Subject: netfilter: nftables: add nf_ct_pernet() helper function Consolidate call to net_generic(net, nf_conntrack_net_id) in this wrapper function. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 06dc6db70d18..cc663c68ddc4 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -346,6 +346,13 @@ nf_ct_set(struct sk_buff *skb, struct nf_conn *ct, enum ip_conntrack_info info) skb_set_nfct(skb, (unsigned long)ct | info); } +extern unsigned int nf_conntrack_net_id; + +static inline struct nf_conntrack_net *nf_ct_pernet(const struct net *net) +{ + return net_generic(net, nf_conntrack_net_id); +} + #define NF_CT_STAT_INC(net, count) __this_cpu_inc((net)->ct.stat->count) #define NF_CT_STAT_INC_ATOMIC(net, count) this_cpu_inc((net)->ct.stat->count) #define NF_CT_STAT_ADD_ATOMIC(net, count, v) this_cpu_add((net)->ct.stat->count, (v)) -- cgit v1.2.3 From ef8ed5ea091bf21648d0c4c1fa4a962d079eab2b Mon Sep 17 00:00:00 2001 From: Oz Shlomo Date: Thu, 3 Jun 2021 15:12:33 +0300 Subject: netfilter: conntrack: Introduce tcp offload timeout configuration TCP connections may be offloaded from nf conntrack to nf flow table. Offloaded connections are aged after 30 seconds of inactivity. Once aged, ownership is returned to conntrack with a hard coded pickup time of 120 seconds, after which the connection may be deleted. eted. The current aging intervals may be too aggressive for some users. Provide users with the ability to control the nf flow table offload aging and pickup time intervals via sysctl parameter as a pre-step for configuring the nf flow table GC timeout intervals. Signed-off-by: Oz Shlomo Reviewed-by: Paul Blakey Signed-off-by: Pablo Neira Ayuso --- include/net/netns/conntrack.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index ad0a95c2335e..3a391e27ec60 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -27,6 +27,10 @@ struct nf_tcp_net { u8 tcp_loose; u8 tcp_be_liberal; u8 tcp_max_retrans; +#if IS_ENABLED(CONFIG_NF_FLOW_TABLE) + unsigned int offload_timeout; + unsigned int offload_pickup; +#endif }; enum udp_conntrack { -- cgit v1.2.3 From 975c57504da1114551fdb3a91ed61dda7739613e Mon Sep 17 00:00:00 2001 From: Oz Shlomo Date: Thu, 3 Jun 2021 15:12:34 +0300 Subject: netfilter: conntrack: Introduce udp offload timeout configuration UDP connections may be offloaded from nf conntrack to nf flow table. Offloaded connections are aged after 30 seconds of inactivity. Once aged, ownership is returned to conntrack with a hard coded pickup time of 30 seconds, after which the connection may be deleted. eted. The current aging intervals may be too aggressive for some users. Provide users with the ability to control the nf flow table offload aging and pickup time intervals via sysctl parameter as a pre-step for configuring the nf flow table GC timeout intervals. Signed-off-by: Oz Shlomo Reviewed-by: Paul Blakey Signed-off-by: Pablo Neira Ayuso --- include/net/netns/conntrack.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 3a391e27ec60..c3094b83a525 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -41,6 +41,10 @@ enum udp_conntrack { struct nf_udp_net { unsigned int timeouts[UDP_CT_MAX]; +#if IS_ENABLED(CONFIG_NF_FLOW_TABLE) + unsigned int offload_timeout; + unsigned int offload_pickup; +#endif }; struct nf_icmp_net { -- cgit v1.2.3 From 1d91d2e1a7f767aa8c11d8507ecf268f787734ec Mon Sep 17 00:00:00 2001 From: Oz Shlomo Date: Thu, 3 Jun 2021 15:12:35 +0300 Subject: netfilter: flowtable: Set offload timeouts according to proto values Currently the aging period for tcp/udp connections is hard coded to 30 seconds. Aged tcp/udp connections configure a hard coded 120/30 seconds pickup timeout for conntrack. This configuration may be too aggressive or permissive for some users. Dynamically configure the nf flow table GC timeout intervals according to the user defined values. Signed-off-by: Oz Shlomo Reviewed-by: Paul Blakey Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_flow_table.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index 48ef7460ff30..a3647fadf1cc 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -177,6 +177,8 @@ struct flow_offload { #define NF_FLOW_TIMEOUT (30 * HZ) #define nf_flowtable_time_stamp (u32)jiffies +unsigned long flow_offload_get_timeout(struct flow_offload *flow); + static inline __s32 nf_flow_timeout_delta(unsigned int timeout) { return (__s32)(timeout - nf_flowtable_time_stamp); -- cgit v1.2.3 From 6a5bcd84e886a9a91982e515c539529c28acdcc2 Mon Sep 17 00:00:00 2001 From: Ilias Apalodimas Date: Mon, 7 Jun 2021 21:02:38 +0200 Subject: page_pool: Allow drivers to hint on SKB recycling Up to now several high speed NICs have custom mechanisms of recycling the allocated memory they use for their payloads. Our page_pool API already has recycling capabilities that are always used when we are running in 'XDP mode'. So let's tweak the API and the kernel network stack slightly and allow the recycling to happen even during the standard operation. The API doesn't take into account 'split page' policies used by those drivers currently, but can be extended once we have users for that. The idea is to be able to intercept the packet on skb_release_data(). If it's a buffer coming from our page_pool API recycle it back to the pool for further usage or just release the packet entirely. To achieve that we introduce a bit in struct sk_buff (pp_recycle:1) and a field in struct page (page->pp) to store the page_pool pointer. Storing the information in page->pp allows us to recycle both SKBs and their fragments. We could have skipped the skb bit entirely, since identical information can bederived from struct page. However, in an effort to affect the free path as less as possible, reading a single bit in the skb which is already in cache, is better that trying to derive identical information for the page stored data. The driver or page_pool has to take care of the sync operations on it's own during the buffer recycling since the buffer is, after opting-in to the recycling, never unmapped. Since the gain on the drivers depends on the architecture, we are not enabling recycling by default if the page_pool API is used on a driver. In order to enable recycling the driver must call skb_mark_for_recycle() to store the information we need for recycling in page->pp and enabling the recycling bit, or page_pool_store_mem_info() for a fragment. Co-developed-by: Jesper Dangaard Brouer Signed-off-by: Jesper Dangaard Brouer Co-developed-by: Matteo Croce Signed-off-by: Matteo Croce Signed-off-by: Ilias Apalodimas Signed-off-by: David S. Miller --- include/net/page_pool.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index b4b6de909c93..3dd62dd73027 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -146,6 +146,8 @@ inline enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool) return pool->p.dma_dir; } +bool page_pool_return_skb_page(struct page *page); + struct page_pool *page_pool_create(const struct page_pool_params *params); #ifdef CONFIG_PAGE_POOL @@ -251,4 +253,11 @@ static inline void page_pool_ring_unlock(struct page_pool *pool) spin_unlock_bh(&pool->ring.producer_lock); } +/* Store mem_info on struct page and use it while recycling skb frags */ +static inline +void page_pool_store_mem_info(struct page *page, struct page_pool *pp) +{ + page->pp = pp; +} + #endif /* _NET_PAGE_POOL_H */ -- cgit v1.2.3 From 152bca090243f2aebbf4c0a2aa723ab610e6f3c4 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 5 Jun 2021 12:54:43 +0200 Subject: xfrm: remove description from xfrm_type struct Its set but never read. Reduces size of xfrm_type to 64 bytes on 64bit. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 6e11db6fa0ab..1aad78c5f2d5 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -388,7 +388,6 @@ void xfrm_flush_gc(void); void xfrm_state_delete_tunnel(struct xfrm_state *x); struct xfrm_type { - char *description; struct module *owner; u8 proto; u8 flags; @@ -410,7 +409,6 @@ int xfrm_register_type(const struct xfrm_type *type, unsigned short family); void xfrm_unregister_type(const struct xfrm_type *type, unsigned short family); struct xfrm_type_offload { - char *description; struct module *owner; u8 proto; void (*encap)(struct xfrm_state *, struct sk_buff *pskb); -- cgit v1.2.3 From 594e450b3f4435a9d663df3d48d7fa34e685cbd1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 5 Jun 2021 10:19:30 -0400 Subject: csum_and_copy_to_iter(): massage into form closer to csum_and_copy_from_iter() Namely, have off counted starting from 0 rather than from csstate->off. To compensate we need to shift the initial value (csstate->sum) (rotate by 8 bits, as usual for csum) and do the same after we are finished adding the pieces up. What we get out of that is a bit more redundancy in our variables - from is always equal to addr + off, which will be useful several commits down the road. Signed-off-by: Al Viro --- include/net/checksum.h | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/checksum.h b/include/net/checksum.h index 0d05b9e8690b..5b96d5bd6e54 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -80,16 +80,18 @@ static inline __sum16 csum16_sub(__sum16 csum, __be16 addend) return csum16_add(csum, ~addend); } -static inline __wsum -csum_block_add(__wsum csum, __wsum csum2, int offset) +static inline __wsum csum_shift(__wsum sum, int offset) { - u32 sum = (__force u32)csum2; - /* rotate sum to align it with a 16b boundary */ if (offset & 1) - sum = ror32(sum, 8); + return (__force __wsum)ror32((__force u32)sum, 8); + return sum; +} - return csum_add(csum, (__force __wsum)sum); +static inline __wsum +csum_block_add(__wsum csum, __wsum csum2, int offset) +{ + return csum_add(csum, csum_shift(csum2, offset)); } static inline __wsum -- cgit v1.2.3 From d1002d2490e3ebc30dd3ba747656cfa90c87e984 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 11 Jun 2021 12:50:13 +0200 Subject: xfrm: remove hdr_offset indirection After previous patches all remaining users set the function pointer to the same function: xfrm6_find_1stfragopt. So remove this function pointer and call ip6_find_1stfragopt directly. Reduces size of xfrm_type to 64 bytes on 64bit platforms. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 1aad78c5f2d5..c8890da00b8a 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -402,7 +402,6 @@ struct xfrm_type { int (*output)(struct xfrm_state *, struct sk_buff *pskb); int (*reject)(struct xfrm_state *, struct sk_buff *, const struct flowi *); - int (*hdr_offset)(struct xfrm_state *, struct sk_buff *, u8 **); }; int xfrm_register_type(const struct xfrm_type *type, unsigned short family); @@ -1605,8 +1604,6 @@ __be32 xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr); __be32 xfrm6_tunnel_spi_lookup(struct net *net, const xfrm_address_t *saddr); int xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb); int xfrm6_output_finish(struct sock *sk, struct sk_buff *skb); -int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb, - u8 **prevhdr); #ifdef CONFIG_XFRM void xfrm6_local_rxpmtu(struct sk_buff *skb, u32 mtu); -- cgit v1.2.3 From 4e50025129efabb07714c1f27a80526897da374b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 11 Jun 2021 22:01:24 +0300 Subject: net: dsa: generalize overhead for taggers that use both headers and trailers Some really really weird switches just couldn't decide whether to use a normal or a tail tagger, so they just did both. This creates problems for DSA, because we only have the concept of an 'overhead' which can be applied to the headroom or to the tailroom of the skb (like for example during the central TX reallocation procedure), depending on the value of bool tail_tag, but not to both. We need to generalize DSA to cater for these odd switches by transforming the 'overhead / tail_tag' pair into 'needed_headroom / needed_tailroom'. The DSA master's MTU is increased to account for both. The flow dissector code is modified such that it only calls the DSA adjustment callback if the tagger has a non-zero header length. Taggers are trivially modified to declare either needed_headroom or needed_tailroom, based on the tail_tag value that they currently declare. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index e1a2610a0e06..0a10f6fffc3d 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -91,7 +91,8 @@ struct dsa_device_ops { * as regular on the master net device. */ bool (*filter)(const struct sk_buff *skb, struct net_device *dev); - unsigned int overhead; + unsigned int needed_headroom; + unsigned int needed_tailroom; const char *name; enum dsa_tag_protocol proto; /* Some tagging protocols either mangle or shift the destination MAC @@ -100,7 +101,6 @@ struct dsa_device_ops { * its RX filter. */ bool promisc_on_master; - bool tail_tag; }; /* This structure defines the control interfaces that are overlayed by the @@ -926,7 +926,7 @@ static inline void dsa_tag_generic_flow_dissect(const struct sk_buff *skb, { #if IS_ENABLED(CONFIG_NET_DSA) const struct dsa_device_ops *ops = skb->dev->dsa_ptr->tag_ops; - int tag_len = ops->overhead; + int tag_len = ops->needed_headroom; *offset = tag_len; *proto = ((__be16 *)skb->data)[(tag_len / 2) - 1]; -- cgit v1.2.3 From 4913b8ebf8a9c56ce66466b4daa07d7d4678cdd8 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 11 Jun 2021 22:01:29 +0300 Subject: net: dsa: add support for the SJA1110 native tagging protocol The SJA1110 has improved a few things compared to SJA1105: - To send a control packet from the host port with SJA1105, one needed to program a one-shot "management route" over SPI. This is no longer true with SJA1110, you can actually send "in-band control extensions" in the packets sent by DSA, these are in fact DSA tags which contain the destination port and switch ID. - When receiving a control packet from the switch with SJA1105, the source port and switch ID were written in bytes 3 and 4 of the destination MAC address of the frame (which was a very poor shot at a DSA header). If the control packet also had an RX timestamp, that timestamp was sent in an actual follow-up packet, so there were reordering concerns on multi-core/multi-queue DSA masters, where the metadata frame with the RX timestamp might get processed before the actual packet to which that timestamp belonged (there is no way to pair a packet to its timestamp other than the order in which they were received). On SJA1110, this is no longer true, control packets have the source port, switch ID and timestamp all in the DSA tags. - Timestamps from the switch were partial: to get a 64-bit timestamp as required by PTP stacks, one would need to take the partial 24-bit or 32-bit timestamp from the packet, then read the current PTP time very quickly, and then patch in the high bits of the current PTP time into the captured partial timestamp, to reconstruct what the full 64-bit timestamp must have been. That is awful because packet processing is done in NAPI context, but reading the current PTP time is done over SPI and therefore needs sleepable context. But it also aggravated a few things: - Not only is there a DSA header in SJA1110, but there is a DSA trailer in fact, too. So DSA needs to be extended to support taggers which have both a header and a trailer. Very unconventional - my understanding is that the trailer exists because the timestamps couldn't be prepared in time for putting them in the header area. - Like SJA1105, not all packets sent to the CPU have the DSA tag added to them, only control packets do: * the ones which match the destination MAC filters/traps in MAC_FLTRES1 and MAC_FLTRES0 * the ones which match FDB entries which have TRAP or TAKETS bits set So we could in theory hack something up to request the switch to take timestamps for all packets that reach the CPU, and those would be DSA-tagged and contain the source port / switch ID by virtue of the fact that there needs to be a timestamp trailer provided. BUT: - The SJA1110 does not parse its own DSA tags in a way that is useful for routing in cross-chip topologies, a la Marvell. And the sja1105 driver already supports cross-chip bridging from the SJA1105 days. It does that by automatically setting up the DSA links as VLAN trunks which contain all the necessary tag_8021q RX VLANs that must be communicated between the switches that span the same bridge. So when using tag_8021q on sja1105, it is possible to have 2 switches with ports sw0p0, sw0p1, sw1p0, sw1p1, and 2 VLAN-unaware bridges br0 and br1, and br0 can take sw0p0 and sw1p0, and br1 can take sw0p1 and sw1p1, and forwarding will happen according to the expected rules of the Linux bridge. We like that, and we don't want that to go away, so as a matter of fact, the SJA1110 tagger still needs to support tag_8021q. So the sja1110 tagger is a hybrid between tag_8021q for data packets, and the native hardware support for control packets. On RX, packets have a 13-byte trailer if they contain an RX timestamp. That trailer is padded in such a way that its byte 8 (the start of the "residence time" field - not parsed by Linux because we don't care) is aligned on a 16 byte boundary. So the padding has a variable length between 0 and 15 bytes. The DSA header contains the offset of the beginning of the padding relative to the beginning of the frame (and the end of the padding is obviously the end of the packet minus 13 bytes, the length of the trailer). So we discard it. Packets which don't have a trailer contain the source port and switch ID information in the header (they are "trap-to-host" packets). Packets which have a trailer contain the source port and switch ID in the trailer. On TX, the destination port mask and switch ID is always in the trailer, so we always need to say in the header that a trailer is present. The header needs a custom EtherType and this was chosen as 0xdadc, after 0xdada which is for Marvell and 0xdadb which is for VLANs in VLAN-unaware mode on SJA1105 (and SJA1110 in fact too). Because we use tag_8021q in concert with the native tagging protocol, control packets will have 2 DSA tags. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 0a10f6fffc3d..289d68e82da0 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -50,6 +50,7 @@ struct phylink_link_state; #define DSA_TAG_PROTO_OCELOT_8021Q_VALUE 20 #define DSA_TAG_PROTO_SEVILLE_VALUE 21 #define DSA_TAG_PROTO_BRCM_LEGACY_VALUE 22 +#define DSA_TAG_PROTO_SJA1110_VALUE 23 enum dsa_tag_protocol { DSA_TAG_PROTO_NONE = DSA_TAG_PROTO_NONE_VALUE, @@ -75,6 +76,7 @@ enum dsa_tag_protocol { DSA_TAG_PROTO_XRS700X = DSA_TAG_PROTO_XRS700X_VALUE, DSA_TAG_PROTO_OCELOT_8021Q = DSA_TAG_PROTO_OCELOT_8021Q_VALUE, DSA_TAG_PROTO_SEVILLE = DSA_TAG_PROTO_SEVILLE_VALUE, + DSA_TAG_PROTO_SJA1110 = DSA_TAG_PROTO_SJA1110_VALUE, }; struct packet_type; -- cgit v1.2.3 From 9942c192b256bc11cc903f89f4057bc97434dee9 Mon Sep 17 00:00:00 2001 From: Arseny Krasnov Date: Fri, 11 Jun 2021 14:10:34 +0300 Subject: af_vsock: implement SEQPACKET receive loop Add receive loop for SEQPACKET. It looks like receive loop for STREAM, but there are differences: 1) It doesn't call notify callbacks. 2) It doesn't care about 'SO_SNDLOWAT' and 'SO_RCVLOWAT' values, because there is no sense for these values in SEQPACKET case. 3) It waits until whole record is received. 4) It processes and sets 'MSG_TRUNC' flag. So to avoid extra conditions for two types of socket inside one loop, two independent functions were created. Signed-off-by: Arseny Krasnov Signed-off-by: David S. Miller --- include/net/af_vsock.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index b1c717286993..4d7cf6b2aca2 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -135,6 +135,10 @@ struct vsock_transport { bool (*stream_is_active)(struct vsock_sock *); bool (*stream_allow)(u32 cid, u32 port); + /* SEQ_PACKET. */ + ssize_t (*seqpacket_dequeue)(struct vsock_sock *vsk, struct msghdr *msg, + int flags); + /* Notification. */ int (*notify_poll_in)(struct vsock_sock *, size_t, bool *); int (*notify_poll_out)(struct vsock_sock *, size_t, bool *); -- cgit v1.2.3 From fbe70c480796d9052fcc786c76e6b029acb1c7bc Mon Sep 17 00:00:00 2001 From: Arseny Krasnov Date: Fri, 11 Jun 2021 14:10:49 +0300 Subject: af_vsock: implement send logic for SEQPACKET Update current stream enqueue function for SEQPACKET support: 1) Call transport's seqpacket enqueue callback. 2) Return value from enqueue function is whole record length or error for SOCK_SEQPACKET. Signed-off-by: Arseny Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- include/net/af_vsock.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 4d7cf6b2aca2..d6745d8b8f3e 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -138,6 +138,8 @@ struct vsock_transport { /* SEQ_PACKET. */ ssize_t (*seqpacket_dequeue)(struct vsock_sock *vsk, struct msghdr *msg, int flags); + int (*seqpacket_enqueue)(struct vsock_sock *vsk, struct msghdr *msg, + size_t len); /* Notification. */ int (*notify_poll_in)(struct vsock_sock *, size_t, bool *); -- cgit v1.2.3 From 0798e78b102b79ed9fe4b2beeb18cf0db117c79b Mon Sep 17 00:00:00 2001 From: Arseny Krasnov Date: Fri, 11 Jun 2021 14:11:04 +0300 Subject: af_vsock: rest of SEQPACKET support Add socket ops for SEQPACKET type and .seqpacket_allow() callback to query transports if they support SEQPACKET. Also split path for data check for STREAM and SEQPACKET branches. Signed-off-by: Arseny Krasnov Signed-off-by: David S. Miller --- include/net/af_vsock.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index d6745d8b8f3e..ab207677e0a8 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -140,6 +140,8 @@ struct vsock_transport { int flags); int (*seqpacket_enqueue)(struct vsock_sock *vsk, struct msghdr *msg, size_t len); + bool (*seqpacket_allow)(u32 remote_cid); + u32 (*seqpacket_has_data)(struct vsock_sock *vsk); /* Notification. */ int (*notify_poll_in)(struct vsock_sock *, size_t, bool *); -- cgit v1.2.3 From 8c713dc93ca9a423d6af8849c9254742a1070c37 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 12 Jun 2021 10:20:54 +0200 Subject: rtnetlink: add alloc() method to rtnl_link_ops In order to make rtnetlink ops that can create different kinds of devices, like what we want to add to the WWAN framework, the priv_size and setup parameters aren't quite sufficient. Make this easier to manage by allowing ops to allocate their own netdev via an @alloc method that gets the tb netlink data. Signed-off-by: Johannes Berg Signed-off-by: Sergey Ryazanov Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/net') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 479f60ef54c0..384e800665f2 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -37,6 +37,9 @@ static inline int rtnl_msg_family(const struct nlmsghdr *nlh) * @maxtype: Highest device specific netlink attribute number * @policy: Netlink policy for device specific attribute validation * @validate: Optional validation function for netlink/changelink parameters + * @alloc: netdev allocation function, can be %NULL and is then used + * in place of alloc_netdev_mqs(), in this case @priv_size + * and @setup are unused. Returns a netdev or ERR_PTR(). * @priv_size: sizeof net_device private space * @setup: net_device setup function * @newlink: Function for configuring and registering a new device @@ -63,6 +66,11 @@ struct rtnl_link_ops { const char *kind; size_t priv_size; + struct net_device *(*alloc)(struct nlattr *tb[], + const char *ifname, + unsigned char name_assign_type, + unsigned int num_tx_queues, + unsigned int num_rx_queues); void (*setup)(struct net_device *dev); bool netns_refund; -- cgit v1.2.3 From ddee9dbc3d7aec1cd9fdcc671db2dd0016fd0f3d Mon Sep 17 00:00:00 2001 From: Oleksandr Mazur Date: Mon, 14 Jun 2021 16:01:12 +0300 Subject: net: core: devlink: add dropped stats traps field Whenever query statistics is issued for trap, devlink subsystem would also fill-in statistics 'dropped' field. This field indicates the number of packets HW dropped and failed to report to the device driver, and thus - to the devlink subsystem itself. In case if device driver didn't register callback for hard drop statistics querying, 'dropped' field will be omitted and not filled. Signed-off-by: Oleksandr Mazur Signed-off-by: David S. Miller --- include/net/devlink.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index eb045f1b5d1d..57b738b78073 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1347,6 +1347,16 @@ struct devlink_ops { const struct devlink_trap_group *group, enum devlink_trap_action action, struct netlink_ext_ack *extack); + /** + * @trap_drop_counter_get: Trap drop counter get function. + * + * Should be used by device drivers to report number of packets + * that have been dropped, and cannot be passed to the devlink + * subsystem by the underlying device. + */ + int (*trap_drop_counter_get)(struct devlink *devlink, + const struct devlink_trap *trap, + u64 *p_drops); /** * @trap_policer_init: Trap policer initialization function. * -- cgit v1.2.3 From f9ac779f881c2ec3d1cdcd7fa9d4f9442bf60e80 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 12 Jun 2021 21:32:14 +0900 Subject: net: Introduce net.ipv4.tcp_migrate_req. This commit adds a new sysctl option: net.ipv4.tcp_migrate_req. If this option is enabled or eBPF program is attached, we will be able to migrate child sockets from a listener to another in the same reuseport group after close() or shutdown() syscalls. Signed-off-by: Kuniyuki Iwashima Signed-off-by: Daniel Borkmann Reviewed-by: Benjamin Herrenschmidt Reviewed-by: Eric Dumazet Acked-by: Martin KaFai Lau Link: https://lore.kernel.org/bpf/20210612123224.12525-2-kuniyu@amazon.co.jp --- include/net/netns/ipv4.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 746c80cd4257..b8620519eace 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -126,6 +126,7 @@ struct netns_ipv4 { u8 sysctl_tcp_syn_retries; u8 sysctl_tcp_synack_retries; u8 sysctl_tcp_syncookies; + u8 sysctl_tcp_migrate_req; int sysctl_tcp_reordering; u8 sysctl_tcp_retries1; u8 sysctl_tcp_retries2; -- cgit v1.2.3 From 5c040eaf5d1753aafe12989ca712175df0b9c436 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 12 Jun 2021 21:32:15 +0900 Subject: tcp: Add num_closed_socks to struct sock_reuseport. As noted in the following commit, a closed listener has to hold the reference to the reuseport group for socket migration. This patch adds a field (num_closed_socks) to struct sock_reuseport to manage closed sockets within the same reuseport group. Moreover, this and the following commits introduce some helper functions to split socks[] into two sections and keep TCP_LISTEN and TCP_CLOSE sockets in each section. Like a double-ended queue, we will place TCP_LISTEN sockets from the front and TCP_CLOSE sockets from the end. TCP_LISTEN----------> <-------TCP_CLOSE +---+---+ --- +---+ --- +---+ --- +---+ | 0 | 1 | ... | i | ... | j | ... | k | +---+---+ --- +---+ --- +---+ --- +---+ i = num_socks - 1 j = max_socks - num_closed_socks k = max_socks - 1 This patch also extends reuseport_add_sock() and reuseport_grow() to support num_closed_socks. Signed-off-by: Kuniyuki Iwashima Signed-off-by: Daniel Borkmann Reviewed-by: Eric Dumazet Acked-by: Martin KaFai Lau Link: https://lore.kernel.org/bpf/20210612123224.12525-3-kuniyu@amazon.co.jp --- include/net/sock_reuseport.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h index 505f1e18e9bf..0e558ca7afbf 100644 --- a/include/net/sock_reuseport.h +++ b/include/net/sock_reuseport.h @@ -13,8 +13,9 @@ extern spinlock_t reuseport_lock; struct sock_reuseport { struct rcu_head rcu; - u16 max_socks; /* length of socks */ - u16 num_socks; /* elements in socks */ + u16 max_socks; /* length of socks */ + u16 num_socks; /* elements in socks */ + u16 num_closed_socks; /* closed elements in socks */ /* The last synq overflow event timestamp of this * reuse->socks[] group. */ -- cgit v1.2.3 From 333bb73f620e1a5f2e0b8df2c0d25300fab36d89 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 12 Jun 2021 21:32:16 +0900 Subject: tcp: Keep TCP_CLOSE sockets in the reuseport group. When we close a listening socket, to migrate its connections to another listener in the same reuseport group, we have to handle two kinds of child sockets. One is that a listening socket has a reference to, and the other is not. The former is the TCP_ESTABLISHED/TCP_SYN_RECV sockets, and they are in the accept queue of their listening socket. So we can pop them out and push them into another listener's queue at close() or shutdown() syscalls. On the other hand, the latter, the TCP_NEW_SYN_RECV socket is during the three-way handshake and not in the accept queue. Thus, we cannot access such sockets at close() or shutdown() syscalls. Accordingly, we have to migrate immature sockets after their listening socket has been closed. Currently, if their listening socket has been closed, TCP_NEW_SYN_RECV sockets are freed at receiving the final ACK or retransmitting SYN+ACKs. At that time, if we could select a new listener from the same reuseport group, no connection would be aborted. However, we cannot do that because reuseport_detach_sock() sets NULL to sk_reuseport_cb and forbids access to the reuseport group from closed sockets. This patch allows TCP_CLOSE sockets to remain in the reuseport group and access it while any child socket references them. The point is that reuseport_detach_sock() was called twice from inet_unhash() and sk_destruct(). This patch replaces the first reuseport_detach_sock() with reuseport_stop_listen_sock(), which checks if the reuseport group is capable of migration. If capable, it decrements num_socks, moves the socket backwards in socks[] and increments num_closed_socks. When all connections are migrated, sk_destruct() calls reuseport_detach_sock() to remove the socket from socks[], decrement num_closed_socks, and set NULL to sk_reuseport_cb. By this change, closed or shutdowned sockets can keep sk_reuseport_cb. Consequently, calling listen() after shutdown() can cause EADDRINUSE or EBUSY in inet_csk_bind_conflict() or reuseport_add_sock() which expects such sockets not to have the reuseport group. Therefore, this patch also loosens such validation rules so that a socket can listen again if it has a reuseport group with num_closed_socks more than 0. When such sockets listen again, we handle them in reuseport_resurrect(). If there is an existing reuseport group (reuseport_add_sock() path), we move the socket from the old group to the new one and free the old one if necessary. If there is no existing group (reuseport_alloc() path), we allocate a new reuseport group, detach sk from the old one, and free it if necessary, not to break the current shutdown behaviour: - we cannot carry over the eBPF prog of shutdowned sockets - we cannot attach/detach an eBPF prog to/from listening sockets via shutdowned sockets Note that when the number of sockets gets over U16_MAX, we try to detach a closed socket randomly to make room for the new listening socket in reuseport_grow(). Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/bpf/20210612123224.12525-4-kuniyu@amazon.co.jp --- include/net/sock_reuseport.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h index 0e558ca7afbf..1333d0cddfbc 100644 --- a/include/net/sock_reuseport.h +++ b/include/net/sock_reuseport.h @@ -32,6 +32,7 @@ extern int reuseport_alloc(struct sock *sk, bool bind_inany); extern int reuseport_add_sock(struct sock *sk, struct sock *sk2, bool bind_inany); extern void reuseport_detach_sock(struct sock *sk); +void reuseport_stop_listen_sock(struct sock *sk); extern struct sock *reuseport_select_sock(struct sock *sk, u32 hash, struct sk_buff *skb, -- cgit v1.2.3 From 1cd62c21572c1df6e7090ea4cabf4cf509616dbb Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 12 Jun 2021 21:32:17 +0900 Subject: tcp: Add reuseport_migrate_sock() to select a new listener. reuseport_migrate_sock() does the same check done in reuseport_listen_stop_sock(). If the reuseport group is capable of migration, reuseport_migrate_sock() selects a new listener by the child socket hash and increments the listener's sk_refcnt beforehand. Thus, if we fail in the migration, we have to decrement it later. We will support migration by eBPF in the later commits. Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/bpf/20210612123224.12525-5-kuniyu@amazon.co.jp --- include/net/sock_reuseport.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h index 1333d0cddfbc..473b0b0fa4ab 100644 --- a/include/net/sock_reuseport.h +++ b/include/net/sock_reuseport.h @@ -37,6 +37,9 @@ extern struct sock *reuseport_select_sock(struct sock *sk, u32 hash, struct sk_buff *skb, int hdr_len); +struct sock *reuseport_migrate_sock(struct sock *sk, + struct sock *migrating_sk, + struct sk_buff *skb); extern int reuseport_attach_prog(struct sock *sk, struct bpf_prog *prog); extern int reuseport_detach_prog(struct sock *sk); -- cgit v1.2.3 From 848ca9182a7d25bb54955c3aab9a3a2742bf9678 Mon Sep 17 00:00:00 2001 From: Jussi Maki Date: Tue, 15 Jun 2021 08:54:15 +0000 Subject: net: bonding: Use per-cpu rr_tx_counter The round-robin rr_tx_counter was shared across CPUs leading to significant cache thrashing at high packet rates. This patch switches the round-robin packet counter to use a per-cpu variable to decide the destination slave. On a test with 2x100Gbit ICE nic with pktgen_sample_04_many_flows.sh (-s 64 -t 32) the tx rate was 19.6Mpps before and 22.3Mpps after this patch. "perf top -e cache_misses" before: 12.31% [bonding] [k] bond_xmit_roundrobin_slave_get 10.59% [sch_fq_codel] [k] fq_codel_dequeue 9.34% [kernel] [k] skb_release_data after: 15.42% [sch_fq_codel] [k] fq_codel_dequeue 10.06% [kernel] [k] __memset 9.12% [kernel] [k] skb_release_data Signed-off-by: Jussi Maki Signed-off-by: David S. Miller --- include/net/bonding.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bonding.h b/include/net/bonding.h index 019e998d944a..15335732e166 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -232,7 +232,7 @@ struct bonding { char proc_file_name[IFNAMSIZ]; #endif /* CONFIG_PROC_FS */ struct list_head bond_list; - u32 rr_tx_counter; + u32 __percpu *rr_tx_counter; struct ad_bond_info ad_info; struct alb_bond_info alb_info; struct bond_params params; -- cgit v1.2.3 From 1b50dd478f495c2112d1dd5655b2317d53a0723b Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Tue, 15 Jun 2021 09:20:08 +0200 Subject: xfrm: delete xfrm4_output_finish xfrm6_output_finish declarations These function declarations are not needed any more. The definitions were deleted. Fixes: 2ab6096db2f1 ("xfrm: remove output_finish indirection from xfrm_state_afinfo") Signed-off-by: Antony Antony Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index c8890da00b8a..3a01570410ab 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1579,7 +1579,6 @@ static inline int xfrm4_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi) } int xfrm4_output(struct net *net, struct sock *sk, struct sk_buff *skb); -int xfrm4_output_finish(struct sock *sk, struct sk_buff *skb); int xfrm4_protocol_register(struct xfrm4_protocol *handler, unsigned char protocol); int xfrm4_protocol_deregister(struct xfrm4_protocol *handler, unsigned char protocol); int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short family); @@ -1603,7 +1602,6 @@ int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler, unsigned short family) __be32 xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr); __be32 xfrm6_tunnel_spi_lookup(struct net *net, const xfrm_address_t *saddr); int xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb); -int xfrm6_output_finish(struct sock *sk, struct sk_buff *skb); #ifdef CONFIG_XFRM void xfrm6_local_rxpmtu(struct sk_buff *skb, u32 mtu); -- cgit v1.2.3 From 194730a9beb52d2b030ea45e12d94868d4a0e6fd Mon Sep 17 00:00:00 2001 From: Guvenc Gulce Date: Wed, 16 Jun 2021 16:52:58 +0200 Subject: net/smc: Make SMC statistics network namespace aware Make the gathered SMC statistics network namespace aware, for each namespace collect an own set of statistic information. Signed-off-by: Guvenc Gulce Signed-off-by: Karsten Graul Signed-off-by: David S. Miller --- include/net/net_namespace.h | 4 ++++ include/net/netns/smc.h | 16 ++++++++++++++++ 2 files changed, 20 insertions(+) create mode 100644 include/net/netns/smc.h (limited to 'include/net') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index fa5887143f0d..befc5b93f311 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -170,6 +171,9 @@ struct net { struct sock *crypto_nlsk; #endif struct sock *diag_nlsk; +#if IS_ENABLED(CONFIG_SMC) + struct netns_smc smc; +#endif } __randomize_layout; #include diff --git a/include/net/netns/smc.h b/include/net/netns/smc.h new file mode 100644 index 000000000000..ea8a9cf2619b --- /dev/null +++ b/include/net/netns/smc.h @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __NETNS_SMC_H__ +#define __NETNS_SMC_H__ +#include +#include + +struct smc_stats_rsn; +struct smc_stats; +struct netns_smc { + /* per cpu counters for SMC */ + struct smc_stats __percpu *smc_stats; + /* protect fback_rsn */ + struct mutex mutex_fback_rsn; + struct smc_stats_rsn *fback_rsn; +}; +#endif -- cgit v1.2.3 From 836382dc24717af203ce06703530528827086955 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 16 Jun 2021 22:25:05 +0200 Subject: netfilter: nf_tables: add last expression Add a new optional expression that tells you when last matching on a given rule / set element element has happened. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables_core.h b/include/net/netfilter/nf_tables_core.h index 46c8d5bb5d8d..0fa5a6d98a00 100644 --- a/include/net/netfilter/nf_tables_core.h +++ b/include/net/netfilter/nf_tables_core.h @@ -16,6 +16,7 @@ extern struct nft_expr_type nft_range_type; extern struct nft_expr_type nft_meta_type; extern struct nft_expr_type nft_rt_type; extern struct nft_expr_type nft_exthdr_type; +extern struct nft_expr_type nft_last_type; #ifdef CONFIG_NETWORK_SECMARK extern struct nft_object_type nft_secmark_obj_type; -- cgit v1.2.3 From 62eec0d73393a136b4523952cecbda1438f1f1b9 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 16 Jun 2021 22:06:19 +0200 Subject: netfilter: conntrack: pass hook state to log functions The packet logger backend is unable to provide the incoming (or outgoing) interface name because that information isn't available. Pass the hook state, it contains the network namespace, the protocol family, the network interfaces and other things. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_l4proto.h | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 96f9cf81f46b..1f47bef51722 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -159,22 +159,26 @@ unsigned int nf_ct_port_nlattr_tuple_size(void); extern const struct nla_policy nf_ct_port_nla_policy[]; #ifdef CONFIG_SYSCTL -__printf(3, 4) __cold +__printf(4, 5) __cold void nf_ct_l4proto_log_invalid(const struct sk_buff *skb, const struct nf_conn *ct, + const struct nf_hook_state *state, const char *fmt, ...); -__printf(5, 6) __cold +__printf(4, 5) __cold void nf_l4proto_log_invalid(const struct sk_buff *skb, - struct net *net, - u16 pf, u8 protonum, + const struct nf_hook_state *state, + u8 protonum, const char *fmt, ...); #else -static inline __printf(5, 6) __cold -void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net, - u16 pf, u8 protonum, const char *fmt, ...) {} -static inline __printf(3, 4) __cold +static inline __printf(4, 5) __cold +void nf_l4proto_log_invalid(const struct sk_buff *skb, + const struct nf_hook_state *state, + u8 protonum, + const char *fmt, ...) {} +static inline __printf(4, 5) __cold void nf_ct_l4proto_log_invalid(const struct sk_buff *skb, const struct nf_conn *ct, + const struct nf_hook_state *state, const char *fmt, ...) { } #endif /* CONFIG_SYSCTL */ -- cgit v1.2.3 From 2f99619820c2269534eb2c0cde44870313c6d353 Mon Sep 17 00:00:00 2001 From: Magnus Karlsson Date: Thu, 17 Jun 2021 11:22:55 +0200 Subject: xsk: Fix missing validation for skb and unaligned mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix a missing validation of a Tx descriptor when executing in skb mode and the umem is in unaligned mode. A descriptor could point to a buffer straddling the end of the umem, thus effectively tricking the kernel to read outside the allowed umem region. This could lead to a kernel crash if that part of memory is not mapped. In zero-copy mode, the descriptor validation code rejects such descriptors by checking a bit in the DMA address that tells us if the next page is physically contiguous or not. For the last page in the umem, this bit is not set, therefore any descriptor pointing to a packet straddling this last page boundary will be rejected. However, the skb path does not use this bit since it copies out data and can do so to two different pages. (It also does not have the array of DMA address, so it cannot even store this bit.) The code just returned that the packet is always physically contiguous. But this is unfortunately also returned for the last page in the umem, which means that packets that cross the end of the umem are being allowed, which they should not be. Fix this by introducing a check for this in the SKB path only, not penalizing the zero-copy path. Fixes: 2b43470add8c ("xsk: Introduce AF_XDP buffer allocation API") Signed-off-by: Magnus Karlsson Signed-off-by: Daniel Borkmann Acked-by: Björn Töpel Link: https://lore.kernel.org/bpf/20210617092255.3487-1-magnus.karlsson@gmail.com --- include/net/xsk_buff_pool.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h index eaa8386dbc63..7a9a23e7a604 100644 --- a/include/net/xsk_buff_pool.h +++ b/include/net/xsk_buff_pool.h @@ -147,11 +147,16 @@ static inline bool xp_desc_crosses_non_contig_pg(struct xsk_buff_pool *pool, { bool cross_pg = (addr & (PAGE_SIZE - 1)) + len > PAGE_SIZE; - if (pool->dma_pages_cnt && cross_pg) { + if (likely(!cross_pg)) + return false; + + if (pool->dma_pages_cnt) { return !(pool->dma_pages[addr >> PAGE_SHIFT] & XSK_NEXT_PG_CONTIG_MASK); } - return false; + + /* skb path */ + return addr + len > pool->addrs_cnt; } static inline u64 xp_aligned_extract_addr(struct xsk_buff_pool *pool, u64 addr) -- cgit v1.2.3 From d0cc298745f5abb3c43319cb9485daf3471d6f94 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Thu, 17 Jun 2021 16:46:08 -0700 Subject: mptcp: generate the data checksum This patch added a new member named csum in struct mptcp_ext, implemented a new function named mptcp_generate_data_checksum(). Generate the data checksum in mptcp_sendmsg_frag, save it in mpext->csum. Note that we must generate the csum for zero window probe, too. Do the csum update incrementally, to avoid multiple csum computation when the data is appended to existing skb. Note that in a later patch we will skip unneeded csum related operation. Changes not included here to keep the delta small. Co-developed-by: Paolo Abeni Signed-off-by: Paolo Abeni Signed-off-by: Geliang Tang Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 83f23774b908..23bbd439e115 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -23,6 +23,7 @@ struct mptcp_ext { u64 data_seq; u32 subflow_seq; u16 data_len; + __sum16 csum; u8 use_map:1, dsn64:1, data_fin:1, -- cgit v1.2.3 From 06fe1719aa501e3b574b1b2b3a7ad2ddac5fb9cb Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Thu, 17 Jun 2021 16:46:09 -0700 Subject: mptcp: add csum_reqd in mptcp_out_options This patch added a new member csum_reqd in struct mptcp_out_options and struct mptcp_subflow_request_sock. Initialized it with the helper function mptcp_is_checksum_enabled(). In mptcp_write_options, if this field is enabled, send out the MP_CAPABLE suboption with the MPTCP_CAP_CHECKSUM_REQD flag. Acked-by: Paolo Abeni Signed-off-by: Geliang Tang Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 23bbd439e115..33af68eea96f 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -64,8 +64,9 @@ struct mptcp_out_options { struct mptcp_rm_list rm_list; u8 join_id; u8 backup; - u8 reset_reason:4; - u8 reset_transient:1; + u8 reset_reason:4, + reset_transient:1, + csum_reqd:1; u32 nonce; u64 thmac; u32 token; -- cgit v1.2.3 From 208e8f66926c5d73e3f359385c1dd49dbc48d067 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Thu, 17 Jun 2021 16:46:14 -0700 Subject: mptcp: receive checksum for MP_CAPABLE with data This patch added a new member named csum in struct mptcp_options_received. When parsing the MP_CAPABLE with data, if the checksum is enabled, adjust the expected_opsize. If the receiving option length matches the length with the data checksum, get the checksum value and save it in mp_opt->csum. And in mptcp_incoming_options, pass it to mpext->csum. We always parse any csum/nocsum combination and delay the presence check to later code, to allow reset if missing. Additionally, in the TX path, use the newly introduce ext field to avoid MPTCP csum recomputation on TCP retransmission and unneeded csum update on when setting the data fin_flag. Co-developed-by: Paolo Abeni Signed-off-by: Paolo Abeni Signed-off-by: Geliang Tang Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 33af68eea96f..d61bbbf11979 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -32,7 +32,8 @@ struct mptcp_ext { mpc_map:1, frozen:1, reset_transient:1; - u8 reset_reason:4; + u8 reset_reason:4, + csum_reqd:1; }; #define MPTCP_RM_IDS_MAX 8 -- cgit v1.2.3 From cfc61c598e43772cc4f76b8fc40c5ec70675716b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 18 Jun 2021 15:51:56 +0200 Subject: xfrm: replay: avoid xfrm replay notify indirection replay protection is implemented using a callback structure and then called via x->repl->notify(), x->repl->recheck(), and so on. all the differect functions are always built-in, so this could be direct calls instead. This first patch prepares for removal of the x->repl structure. Add an enum with the three available replay modes to the xfrm_state structure and then replace all x->repl->notify() calls by the new xfrm_replay_notify() helper. The helper checks the enum internally to adapt behaviour as needed. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 3a01570410ab..9a79e41defa7 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -145,6 +145,12 @@ enum { XFRM_MODE_FLAG_TUNNEL = 1, }; +enum xfrm_replay_mode { + XFRM_REPLAY_MODE_LEGACY, + XFRM_REPLAY_MODE_BMP, + XFRM_REPLAY_MODE_ESN, +}; + /* Full description of state of transformer. */ struct xfrm_state { possible_net_t xs_net; @@ -218,6 +224,8 @@ struct xfrm_state { /* The functions for replay detection. */ const struct xfrm_replay *repl; + /* replay detection mode */ + enum xfrm_replay_mode repl_mode; /* internal flag that only holds state for delayed aevent at the * moment */ @@ -305,7 +313,6 @@ struct xfrm_replay { int (*recheck)(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); - void (*notify)(struct xfrm_state *x, int event); int (*overflow)(struct xfrm_state *x, struct sk_buff *skb); }; @@ -1715,6 +1722,8 @@ static inline int xfrm_policy_id2dir(u32 index) } #ifdef CONFIG_XFRM +void xfrm_replay_notify(struct xfrm_state *x, int event); + static inline int xfrm_aevent_is_on(struct net *net) { struct sock *nlsk; -- cgit v1.2.3 From c7f877833c9f361be8e88d6b140d8314e80892aa Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 18 Jun 2021 15:51:57 +0200 Subject: xfrm: replay: remove advance indirection Similar to other patches: add a new helper to avoid an indirection. v2: fix 'net/xfrm/xfrm_replay.c:519:13: warning: 'seq' may be used uninitialized in this function' warning. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 9a79e41defa7..a7f997b13198 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -306,7 +306,6 @@ struct km_event { }; struct xfrm_replay { - void (*advance)(struct xfrm_state *x, __be32 net_seq); int (*check)(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); @@ -1722,6 +1721,7 @@ static inline int xfrm_policy_id2dir(u32 index) } #ifdef CONFIG_XFRM +void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq); void xfrm_replay_notify(struct xfrm_state *x, int event); static inline int xfrm_aevent_is_on(struct net *net) -- cgit v1.2.3 From 25cfb8bc97c2b8447f86b1ad376ee672b6b173d4 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 18 Jun 2021 15:51:58 +0200 Subject: xfrm: replay: remove recheck indirection Adds new xfrm_replay_recheck() helper and calls it from xfrm input path instead of the indirection. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index a7f997b13198..3a219b34cb8c 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -309,9 +309,6 @@ struct xfrm_replay { int (*check)(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); - int (*recheck)(struct xfrm_state *x, - struct sk_buff *skb, - __be32 net_seq); int (*overflow)(struct xfrm_state *x, struct sk_buff *skb); }; @@ -1723,6 +1720,7 @@ static inline int xfrm_policy_id2dir(u32 index) #ifdef CONFIG_XFRM void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq); void xfrm_replay_notify(struct xfrm_state *x, int event); +int xfrm_replay_recheck(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); static inline int xfrm_aevent_is_on(struct net *net) { -- cgit v1.2.3 From adfc2fdbae30d42edebad01d0ea1eed43036f1fe Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 18 Jun 2021 15:51:59 +0200 Subject: xfrm: replay: avoid replay indirection Add and use xfrm_replay_check helper instead of indirection. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 3a219b34cb8c..0206d80ec291 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -306,9 +306,6 @@ struct km_event { }; struct xfrm_replay { - int (*check)(struct xfrm_state *x, - struct sk_buff *skb, - __be32 net_seq); int (*overflow)(struct xfrm_state *x, struct sk_buff *skb); }; @@ -1719,6 +1716,7 @@ static inline int xfrm_policy_id2dir(u32 index) #ifdef CONFIG_XFRM void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq); +int xfrm_replay_check(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); void xfrm_replay_notify(struct xfrm_state *x, int event); int xfrm_replay_recheck(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); -- cgit v1.2.3 From b5a1d1fe0cbb9d20ba661134a09561af1dc9ebf5 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 18 Jun 2021 15:52:00 +0200 Subject: xfrm: replay: remove last replay indirection This replaces the overflow indirection with the new xfrm_replay_overflow helper. After this, the 'repl' pointer in xfrm_state is no longer needed and can be removed as well. xfrm_replay_overflow() is added in two incarnations, one is used when the kernel is compiled with xfrm hardware offload support enabled, the other when its disabled. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 0206d80ec291..d2a0559c255f 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -221,9 +221,6 @@ struct xfrm_state { struct xfrm_replay_state preplay; struct xfrm_replay_state_esn *preplay_esn; - /* The functions for replay detection. */ - const struct xfrm_replay *repl; - /* replay detection mode */ enum xfrm_replay_mode repl_mode; /* internal flag that only holds state for delayed aevent at the @@ -305,10 +302,6 @@ struct km_event { struct net *net; }; -struct xfrm_replay { - int (*overflow)(struct xfrm_state *x, struct sk_buff *skb); -}; - struct xfrm_if_cb { struct xfrm_if *(*decode_session)(struct sk_buff *skb, unsigned short family); @@ -1718,6 +1711,7 @@ static inline int xfrm_policy_id2dir(u32 index) void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq); int xfrm_replay_check(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); void xfrm_replay_notify(struct xfrm_state *x, int event); +int xfrm_replay_overflow(struct xfrm_state *x, struct sk_buff *skb); int xfrm_replay_recheck(struct xfrm_state *x, struct sk_buff *skb, __be32 net_seq); static inline int xfrm_aevent_is_on(struct net *net) -- cgit v1.2.3 From 89837eb4b2463c556a123437f242d6c2bc62ce81 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Thu, 17 Jun 2021 09:04:14 +0800 Subject: net: sched: add barrier to ensure correct ordering for lockless qdisc The spin_trylock() was assumed to contain the implicit barrier needed to ensure the correct ordering between STATE_MISSED setting/clearing and STATE_MISSED checking in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"). But it turns out that spin_trylock() only has load-acquire semantic, for strongly-ordered system(like x86), the compiler barrier implicitly contained in spin_trylock() seems enough to ensure the correct ordering. But for weakly-orderly system (like arm64), the store-release semantic is needed to ensure the correct ordering as clear_bit() and test_bit() is store operation, see queued_spin_lock(). So add the explicit barrier to ensure the correct ordering for the above case. Fixes: a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc") Signed-off-by: Yunsheng Lin Acked-by: Jakub Kicinski Signed-off-by: David S. Miller --- include/net/sch_generic.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 1e625519ae96..57710303908c 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -163,6 +163,12 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) if (spin_trylock(&qdisc->seqlock)) goto nolock_empty; + /* Paired with smp_mb__after_atomic() to make sure + * STATE_MISSED checking is synchronized with clearing + * in pfifo_fast_dequeue(). + */ + smp_mb__before_atomic(); + /* If the MISSED flag is set, it means other thread has * set the MISSED flag before second spin_trylock(), so * we can return false here to avoid multi cpus doing @@ -180,6 +186,12 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) */ set_bit(__QDISC_STATE_MISSED, &qdisc->state); + /* spin_trylock() only has load-acquire semantic, so use + * smp_mb__after_atomic() to ensure STATE_MISSED is set + * before doing the second spin_trylock(). + */ + smp_mb__after_atomic(); + /* Retry again in case other CPU may not see the new flag * after it releases the lock at the end of qdisc_run_end(). */ -- cgit v1.2.3 From a8986681ccada614a30df7248390780e7708a763 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 21 Jun 2021 19:42:15 +0300 Subject: net: dsa: export the dsa_port_is_{user,cpu,dsa} helpers The difference between dsa_is_user_port and dsa_port_is_user is that the former needs to look up the list of ports of the DSA switch tree in order to find the struct dsa_port, while the latter directly receives it as an argument. dsa_is_user_port is already in widespread use and has its place, so there isn't any chance of converting all callers to a single form. But being able to do: dsa_port_is_user(dp) instead of dsa_is_user_port(dp->ds, dp->index) is much more efficient too, especially when the "dp" comes from an iterator over the DSA switch tree - this reduces the complexity from quadratic to linear. Move these helpers from dsa2.c to include/net/dsa.h so that others can use them too. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- include/net/dsa.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 289d68e82da0..ea47783d5695 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -409,6 +409,21 @@ static inline struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) return NULL; } +static inline bool dsa_port_is_dsa(struct dsa_port *port) +{ + return port->type == DSA_PORT_TYPE_DSA; +} + +static inline bool dsa_port_is_cpu(struct dsa_port *port) +{ + return port->type == DSA_PORT_TYPE_CPU; +} + +static inline bool dsa_port_is_user(struct dsa_port *dp) +{ + return dp->type == DSA_PORT_TYPE_USER; +} + static inline bool dsa_is_unused_port(struct dsa_switch *ds, int p) { return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_UNUSED; -- cgit v1.2.3 From 745a32117b5a0799ce1dd28d5a74dc2b7bf37692 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:47 -0400 Subject: sctp: add pad chunk and its make function and event table This chunk is defined in rfc4820#section-3, and used to pad an SCTP packet. The receiver must discard this chunk and continue processing the rest of the chunks in the packet. Add it now, as it will be bundled with a heartbeat chunk to probe pmtu in the following patches. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index fd223c94589a..09c59154634d 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -230,6 +230,7 @@ struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *asoc, const struct sctp_chunk *chunk, const void *payload, const size_t paylen); +struct sctp_chunk *sctp_make_pad(const struct sctp_association *asoc, int len); struct sctp_chunk *sctp_make_op_error(const struct sctp_association *asoc, const struct sctp_chunk *chunk, __be16 cause_code, const void *payload, -- cgit v1.2.3 From d1e462a7a5f359cbb9a0e8fbfafcfb6657034105 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:48 -0400 Subject: sctp: add probe_interval in sysctl and sock/asoc/transport PLPMTUD can be enabled by doing 'sysctl -w net.sctp.probe_interval=n'. 'n' is the interval for PLPMTUD probe timer in milliseconds, and it can't be less than 5000 if it's not 0. All asoc/transport's PLPMTUD in a new socket will be enabled by default. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/netns/sctp.h | 3 +++ include/net/sctp/constants.h | 2 ++ include/net/sctp/structs.h | 3 +++ 3 files changed, 8 insertions(+) (limited to 'include/net') diff --git a/include/net/netns/sctp.h b/include/net/netns/sctp.h index a0f315effa94..40240722cdca 100644 --- a/include/net/netns/sctp.h +++ b/include/net/netns/sctp.h @@ -84,6 +84,9 @@ struct netns_sctp { /* HB.interval - 30 seconds */ unsigned int hb_interval; + /* The interval for PLPMTUD probe timer */ + unsigned int probe_interval; + /* Association.Max.Retrans - 10 attempts * Path.Max.Retrans - 5 attempts (per destination address) * Max.Init.Retransmits - 8 attempts diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 14a0d22c9113..449cf9cb428b 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -424,4 +424,6 @@ enum { */ #define SCTP_AUTH_RANDOM_LENGTH 32 +#define SCTP_PROBE_TIMER_MIN 5000 + #endif /* __sctp_constants_h__ */ diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 1aa585216f34..bf5d22deaefb 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -177,6 +177,7 @@ struct sctp_sock { * will be inherited by all new associations. */ __u32 hbinterval; + __u32 probe_interval; __be16 udp_port; __be16 encap_port; @@ -858,6 +859,7 @@ struct sctp_transport { * the destination address every heartbeat interval. */ unsigned long hbinterval; + unsigned long probe_interval; /* SACK delay timeout */ unsigned long sackdelay; @@ -1795,6 +1797,7 @@ struct sctp_association { * will be inherited by all new transports. */ unsigned long hbinterval; + unsigned long probe_interval; __be16 encap_port; -- cgit v1.2.3 From d9e2e410ae301d4b540e965daca51de0e65e8a26 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:50 -0400 Subject: sctp: add the constants/variables and states and some APIs for transport These are 4 constants described in rfc8899#section-5.1.2: MAX_PROBES, MIN_PLPMTU, MAX_PLPMTU, BASE_PLPMTU; And 2 variables described in rfc8899#section-5.1.3: PROBED_SIZE, PROBE_COUNT; And 5 states described in rfc8899#section-5.2: DISABLED, BASE, SEARCH, SEARCH_COMPLETE, ERROR; And these 4 APIs are used to reset/update PLPMTUD, check if PLPMTUD is enabled, and calculate the additional headers length for a transport. Note the member 'probe_high' in transport will be set to the probe size when a probe fails with this probe size in the next patches. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/constants.h | 17 ++++++++++++++++ include/net/sctp/sctp.h | 48 +++++++++++++++++++++++++++++++++++++++++--- include/net/sctp/structs.h | 8 ++++++++ 3 files changed, 70 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 449cf9cb428b..85f6a105c59d 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -200,6 +200,23 @@ enum sctp_sock_state { SCTP_SS_CLOSING = TCP_CLOSE_WAIT, }; +enum sctp_plpmtud_state { + SCTP_PL_DISABLED, + SCTP_PL_BASE, + SCTP_PL_SEARCH, + SCTP_PL_COMPLETE, + SCTP_PL_ERROR, +}; + +#define SCTP_BASE_PLPMTU 1200 +#define SCTP_MAX_PLPMTU 9000 +#define SCTP_MIN_PLPMTU 512 + +#define SCTP_MAX_PROBES 3 + +#define SCTP_PL_BIG_STEP 32 +#define SCTP_PL_MIN_STEP 4 + /* These functions map various type to printable names. */ const char *sctp_cname(const union sctp_subtype id); /* chunk types */ const char *sctp_oname(const union sctp_subtype id); /* other events */ diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 86f74f2fe6de..08347d3f004f 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -573,14 +573,15 @@ static inline struct dst_entry *sctp_transport_dst_check(struct sctp_transport * /* Calculate max payload size given a MTU, or the total overhead if * given MTU is zero */ -static inline __u32 sctp_mtu_payload(const struct sctp_sock *sp, - __u32 mtu, __u32 extra) +static inline __u32 __sctp_mtu_payload(const struct sctp_sock *sp, + const struct sctp_transport *t, + __u32 mtu, __u32 extra) { __u32 overhead = sizeof(struct sctphdr) + extra; if (sp) { overhead += sp->pf->af->net_header_len; - if (sp->udp_port) + if (sp->udp_port && (!t || t->encap_port)) overhead += sizeof(struct udphdr); } else { overhead += sizeof(struct ipv6hdr); @@ -592,6 +593,12 @@ static inline __u32 sctp_mtu_payload(const struct sctp_sock *sp, return mtu ? mtu - overhead : overhead; } +static inline __u32 sctp_mtu_payload(const struct sctp_sock *sp, + __u32 mtu, __u32 extra) +{ + return __sctp_mtu_payload(sp, NULL, mtu, extra); +} + static inline __u32 sctp_dst_mtu(const struct dst_entry *dst) { return SCTP_TRUNC4(max_t(__u32, dst_mtu(dst), @@ -615,6 +622,41 @@ static inline __u32 sctp_min_frag_point(struct sctp_sock *sp, __u16 datasize) return sctp_mtu_payload(sp, SCTP_DEFAULT_MINSEGMENT, datasize); } +static inline int sctp_transport_pl_hlen(struct sctp_transport *t) +{ + return __sctp_mtu_payload(sctp_sk(t->asoc->base.sk), t, 0, 0); +} + +static inline void sctp_transport_pl_reset(struct sctp_transport *t) +{ + if (t->probe_interval && (t->param_flags & SPP_PMTUD_ENABLE) && + (t->state == SCTP_ACTIVE || t->state == SCTP_UNKNOWN)) { + if (t->pl.state == SCTP_PL_DISABLED) { + t->pl.state = SCTP_PL_BASE; + t->pl.pmtu = SCTP_BASE_PLPMTU; + t->pl.probe_size = SCTP_BASE_PLPMTU; + } + } else { + if (t->pl.state != SCTP_PL_DISABLED) + t->pl.state = SCTP_PL_DISABLED; + } +} + +static inline void sctp_transport_pl_update(struct sctp_transport *t) +{ + if (t->pl.state == SCTP_PL_DISABLED) + return; + + t->pl.state = SCTP_PL_BASE; + t->pl.pmtu = SCTP_BASE_PLPMTU; + t->pl.probe_size = SCTP_BASE_PLPMTU; +} + +static inline bool sctp_transport_pl_enabled(struct sctp_transport *t) +{ + return t->pl.state != SCTP_PL_DISABLED; +} + static inline bool sctp_newsk_ready(const struct sock *sk) { return sock_flag(sk, SOCK_DEAD) || sk->sk_socket; diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index bf5d22deaefb..85d3566c2227 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -978,6 +978,14 @@ struct sctp_transport { char cacc_saw_newack; } cacc; + struct { + __u16 pmtu; + __u16 probe_size; + __u16 probe_high; + __u8 probe_count; + __u8 state; + } pl; /* plpmtud related */ + /* 64-bit random number sent with heartbeat. */ __u64 hb_nonce; -- cgit v1.2.3 From 92548ec2f1f92d0c0b60ce59592b645571672568 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:51 -0400 Subject: sctp: add the probe timer in transport for PLPMTUD There are 3 timers described in rfc8899#section-5.1.1: PROBE_TIMER, PMTU_RAISE_TIMER, CONFIRMATION_TIMER This patches adds a 'probe_timer' in transport, and it works as either PROBE_TIMER or PMTU_RAISE_TIMER. At most time, it works as PROBE_TIMER and expires every a 'probe_interval' time to send the HB probe packet. When transport pl enters COMPLETE state, it works as PMTU_RAISE_TIMER and expires in 'probe_interval * 30' time to go back to SEARCH state and do searching again. SCTP HB is an acknowledged packet, CONFIRMATION_TIMER is not needed. The timer will start when transport pl enters BASE state and stop when it enters DISABLED state. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/command.h | 1 + include/net/sctp/constants.h | 1 + include/net/sctp/sctp.h | 9 ++++++++- include/net/sctp/sm.h | 2 ++ include/net/sctp/structs.h | 4 ++++ 5 files changed, 16 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 5e848884ff61..2058fabffbf6 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -59,6 +59,7 @@ enum sctp_verb { SCTP_CMD_HB_TIMERS_START, /* Start the heartbeat timers. */ SCTP_CMD_HB_TIMER_UPDATE, /* Update a heartbeat timers. */ SCTP_CMD_HB_TIMERS_STOP, /* Stop the heartbeat timers. */ + SCTP_CMD_PROBE_TIMER_UPDATE, /* Update a probe timer. */ SCTP_CMD_TRANSPORT_HB_SENT, /* Reset the status of a transport. */ SCTP_CMD_TRANSPORT_IDLE, /* Do manipulations on idle transport */ SCTP_CMD_TRANSPORT_ON, /* Mark the transport as active. */ diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 85f6a105c59d..265fffa33dad 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -77,6 +77,7 @@ enum sctp_event_timeout { SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD, SCTP_EVENT_TIMEOUT_HEARTBEAT, SCTP_EVENT_TIMEOUT_RECONF, + SCTP_EVENT_TIMEOUT_PROBE, SCTP_EVENT_TIMEOUT_SACK, SCTP_EVENT_TIMEOUT_AUTOCLOSE, }; diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 08347d3f004f..f7e083602c10 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -635,10 +635,14 @@ static inline void sctp_transport_pl_reset(struct sctp_transport *t) t->pl.state = SCTP_PL_BASE; t->pl.pmtu = SCTP_BASE_PLPMTU; t->pl.probe_size = SCTP_BASE_PLPMTU; + sctp_transport_reset_probe_timer(t); } } else { - if (t->pl.state != SCTP_PL_DISABLED) + if (t->pl.state != SCTP_PL_DISABLED) { + if (del_timer(&t->probe_timer)) + sctp_transport_put(t); t->pl.state = SCTP_PL_DISABLED; + } } } @@ -647,6 +651,9 @@ static inline void sctp_transport_pl_update(struct sctp_transport *t) if (t->pl.state == SCTP_PL_DISABLED) return; + if (del_timer(&t->probe_timer)) + sctp_transport_put(t); + t->pl.state = SCTP_PL_BASE; t->pl.pmtu = SCTP_BASE_PLPMTU; t->pl.probe_size = SCTP_BASE_PLPMTU; diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 09c59154634d..45542e2bac93 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -151,6 +151,7 @@ sctp_state_fn_t sctp_sf_cookie_wait_icmp_abort; /* Prototypes for timeout event state functions. */ sctp_state_fn_t sctp_sf_do_6_3_3_rtx; sctp_state_fn_t sctp_sf_send_reconf; +sctp_state_fn_t sctp_sf_send_probe; sctp_state_fn_t sctp_sf_do_6_2_sack; sctp_state_fn_t sctp_sf_autoclose_timer_expire; @@ -311,6 +312,7 @@ int sctp_do_sm(struct net *net, enum sctp_event_type event_type, void sctp_generate_t3_rtx_event(struct timer_list *t); void sctp_generate_heartbeat_event(struct timer_list *t); void sctp_generate_reconf_event(struct timer_list *t); +void sctp_generate_probe_event(struct timer_list *t); void sctp_generate_proto_unreach_event(struct timer_list *t); void sctp_ootb_pkt_free(struct sctp_packet *packet); diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 85d3566c2227..a3772f8ee7f6 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -936,6 +936,9 @@ struct sctp_transport { /* Timer to handler reconf chunk rtx */ struct timer_list reconf_timer; + /* Timer to send a probe HB packet for PLPMTUD */ + struct timer_list probe_timer; + /* Since we're using per-destination retransmission timers * (see above), we're also using per-destination "transmitted" * queues. This probably ought to be a private struct @@ -1003,6 +1006,7 @@ void sctp_transport_free(struct sctp_transport *); void sctp_transport_reset_t3_rtx(struct sctp_transport *); void sctp_transport_reset_hb_timer(struct sctp_transport *); void sctp_transport_reset_reconf_timer(struct sctp_transport *transport); +void sctp_transport_reset_probe_timer(struct sctp_transport *transport); int sctp_transport_hold(struct sctp_transport *); void sctp_transport_put(struct sctp_transport *); void sctp_transport_update_rto(struct sctp_transport *, __u32); -- cgit v1.2.3 From fe59379b9ab7ddad157f5379fa47dbf84c9b5e09 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:52 -0400 Subject: sctp: do the basic send and recv for PLPMTUD probe This patch does exactly what rfc8899#section-6.2.1.2 says: The SCTP sender needs to be able to determine the total size of a probe packet. The HEARTBEAT chunk could carry a Heartbeat Information parameter that includes, besides the information suggested in [RFC4960], the probe size to help an implementation associate a HEARTBEAT ACK with the size of probe that was sent. The sender could also use other methods, such as sending a nonce and verifying the information returned also contains the corresponding nonce. The length of the PAD chunk is computed by reducing the probing size by the size of the SCTP common header and the HEARTBEAT chunk. Note that HB ACK chunk will carry back whatever HB chunk carried, including the probe_size we put it in; We also check hbinfo->probe_size in the HB ACK against link->pl.probe_size to validate this HB ACK chunk. v1->v2: - Remove the unused 'sp' and add static for sctp_packet_bundle_pad(). Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 3 ++- include/net/sctp/structs.h | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 45542e2bac93..2eb6d7c2c931 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -226,7 +226,8 @@ struct sctp_chunk *sctp_make_new_encap_port( const struct sctp_association *asoc, const struct sctp_chunk *chunk); struct sctp_chunk *sctp_make_heartbeat(const struct sctp_association *asoc, - const struct sctp_transport *transport); + const struct sctp_transport *transport, + __u32 probe_size); struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *asoc, const struct sctp_chunk *chunk, const void *payload, diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index a3772f8ee7f6..f7b056f5af37 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -386,6 +386,7 @@ struct sctp_sender_hb_info { union sctp_addr daddr; unsigned long sent_at; __u64 hb_nonce; + __u32 probe_size; }; int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, @@ -657,6 +658,7 @@ struct sctp_chunk { data_accepted:1, /* At least 1 chunk accepted */ auth:1, /* IN: was auth'ed | OUT: needs auth */ has_asconf:1, /* IN: have seen an asconf before */ + pmtu_probe:1, /* Used by PLPMTUD, can be set in s HB chunk */ tsn_missing_report:2, /* Data chunk missing counter. */ fast_retransmit:2; /* Is this chunk fast retransmitted? */ }; -- cgit v1.2.3 From 1dc68c194571acc4027de5f8378227d0c0ff7e13 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:53 -0400 Subject: sctp: do state transition when PROBE_COUNT == MAX_PROBES on HB send path The state transition is described in rfc8899#section-5.2, PROBE_COUNT == MAX_PROBES means the probe fails for MAX times, and the state transition includes: - Base -> Error, occurs when BASE_PLPMTU Confirmation Fails, pl.pmtu is set to SCTP_MIN_PLPMTU, probe_size is still SCTP_BASE_PLPMTU; - Search -> Base, occurs when Black Hole Detected, pl.pmtu is set to SCTP_BASE_PLPMTU, probe_size is set back to SCTP_BASE_PLPMTU; - Search Complete -> Base, occurs when Black Hole Detected pl.pmtu is set to SCTP_BASE_PLPMTU, probe_size is set back to SCTP_BASE_PLPMTU; Note a black hole is encountered when a sender is unaware that packets are not being delivered to the destination endpoint. So it includes the probe failures with equal probe_size to pl.pmtu, and definitely not include that with greater probe_size than pl.pmtu. The later one is the normal probe failure where probe_size should decrease back to pl.pmtu and pl.probe_high is set. pl.probe_high would be used on HB ACK recv path in the next patch. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index f7b056f5af37..31165720b28a 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1023,6 +1023,7 @@ bool sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu); void sctp_transport_immediate_rtx(struct sctp_transport *); void sctp_transport_dst_release(struct sctp_transport *t); void sctp_transport_dst_confirm(struct sctp_transport *t); +void sctp_transport_pl_send(struct sctp_transport *t); /* This is the structure we use to queue packets as they come into -- cgit v1.2.3 From b87641aff9e772fda15d3386d159646eada2ceef Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:04:54 -0400 Subject: sctp: do state transition when a probe succeeds on HB ACK recv path As described in rfc8899#section-5.2, when a probe succeeds, there might be the following state transitions: - Base -> Search, occurs when probe succeeds with BASE_PLPMTU, pl.pmtu is not changing, pl.probe_size increases by SCTP_PL_BIG_STEP, - Error -> Search, occurs when probe succeeds with BASE_PLPMTU, pl.pmtu is changed from SCTP_MIN_PLPMTU to SCTP_BASE_PLPMTU, pl.probe_size increases by SCTP_PL_BIG_STEP. - Search -> Search Complete, occurs when probe succeeds with the probe size SCTP_MAX_PLPMTU less than pl.probe_high, pl.pmtu is not changing, but update *pathmtu* with it, pl.probe_size is set back to pl.pmtu to double check it. - Search Complete -> Search, occurs when probe succeeds with the probe size equal to pl.pmtu, pl.pmtu is not changing, pl.probe_size increases by SCTP_PL_MIN_STEP. So search process can be described as: 1. When it just enters 'Search' state, *pathmtu* is not updated with pl.pmtu, and probe_size increases by a big step (SCTP_PL_BIG_STEP) each round. 2. Until pl.probe_high is set when a probe fails, and probe_size decreases back to pl.pmtu, as described in the last patch. 3. When the probe with the new size succeeds, probe_size changes to increase by a small step (SCTP_PL_MIN_STEP) due to pl.probe_high is set. 4. Until probe_size is next to pl.probe_high, the searching finishes and it goes to 'Complete' state and updates *pathmtu* with pl.pmtu, and then probe_size is set to pl.pmtu to confirm by once more probe. 5. This probe occurs after "30 * probe_inteval", a much longer time than that in Search state. Once it is done it goes to 'Search' state again with probe_size increased by SCTP_PL_MIN_STEP. As we can see above, during the searching, pl.pmtu changes while *pathmtu* doesn't. *pathmtu* is only updated when the search finishes by which it gets an optimal value for it. A big step is used at the beginning until it gets close to the optimal value, then it changes to a small step until it has this optimal value. The small step is also used in 'Complete' until it goes to 'Search' state again and the probe with 'pmtu + the small step' succeeds, which means a higher size could be used. Then probe_size changes to increase by a big step again until it gets close to the next optimal value. Note that anytime when black hole is detected, it goes directly to 'Base' state with pl.pmtu set to SCTP_BASE_PLPMTU, as described in the last patch. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 31165720b28a..9eaa701cda23 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1024,6 +1024,7 @@ void sctp_transport_immediate_rtx(struct sctp_transport *); void sctp_transport_dst_release(struct sctp_transport *t); void sctp_transport_dst_confirm(struct sctp_transport *t); void sctp_transport_pl_send(struct sctp_transport *t); +void sctp_transport_pl_recv(struct sctp_transport *t); /* This is the structure we use to queue packets as they come into -- cgit v1.2.3 From 9e47df005cab63e545671dba8dfd6852fff1c2cf Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 22 Jun 2021 14:05:00 -0400 Subject: sctp: process sctp over udp icmp err on sctp side Previously, sctp over udp was using udp tunnel's icmp err process, which only does sk lookup on sctp side. However for sctp's icmp error process, there are more things to do, like syncing assoc pmtu/retransmit packets for toobig type err, and starting proto_unreach_timer for unreach type err etc. Now after adding PLPMTUD, which also requires to process toobig type err on sctp side. This patch is to process icmp err on sctp side by parsing the type/code/info in .encap_err_lookup and call sctp's icmp processing functions. Note as the 'redirect' err process needs to know the outer ip(v6) header's, we have to leave it to udp(v6)_err to handle it. Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index f7e083602c10..69bab88ad66b 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -145,6 +145,8 @@ struct sock *sctp_err_lookup(struct net *net, int family, struct sk_buff *, struct sctphdr *, struct sctp_association **, struct sctp_transport **); void sctp_err_finish(struct sock *, struct sctp_transport *); +int sctp_udp_v4_err(struct sock *sk, struct sk_buff *skb); +int sctp_udp_v6_err(struct sock *sk, struct sk_buff *skb); void sctp_icmp_frag_needed(struct sock *, struct sctp_association *, struct sctp_transport *t, __u32 pmtu); void sctp_icmp_redirect(struct sock *, struct sctp_transport *, -- cgit v1.2.3 From bab6b88e056038f618b2fb977d95b05ad3da8d0c Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Tue, 22 Jun 2021 12:25:19 -0700 Subject: mptcp: add allow_join_id0 in mptcp_out_options This patch defined a new flag MPTCP_CAP_DENY_JOIN_ID0 for the third bit, labeled "C" of the MP_CAPABLE option. Add a new flag allow_join_id0 in struct mptcp_out_options. If this flag is set, send out the MP_CAPABLE option with the flag MPTCP_CAP_DENY_JOIN_ID0. Acked-by: Paolo Abeni Signed-off-by: Geliang Tang Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index d61bbbf11979..cb580b06152f 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -67,7 +67,8 @@ struct mptcp_out_options { u8 backup; u8 reset_reason:4, reset_transient:1, - csum_reqd:1; + csum_reqd:1, + allow_join_id0:1; u32 nonce; u64 thmac; u32 token; -- cgit v1.2.3 From fa4535238fb5f306f95de89371a993057b32b2a4 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Mon, 14 Jun 2021 17:33:48 +0300 Subject: net/xfrm: Add inner_ipproto into sec_path The inner_ipproto saves the inner IP protocol of the plain text packet. This allows vendor's IPsec feature making offload decision at skb's features_check and configuring hardware at ndo_start_xmit. For example, ConnectX6-DX IPsec device needs the plaintext's IP protocol to support partial checksum offload on VXLAN/GENEVE packet over IPsec transport mode tunnel. Signed-off-by: Raed Salem Signed-off-by: Huy Nguyen Cc: Steffen Klassert Acked-by: Steffen Klassert Signed-off-by: Saeed Mahameed --- include/net/xfrm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index c58a6d4eb610..1d803e890c76 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1024,6 +1024,7 @@ struct xfrm_offload { #define CRYPTO_INVALID_PROTOCOL 128 __u8 proto; + __u8 inner_ipproto; }; struct sec_path { -- cgit v1.2.3 From 21b7805434f6598eaf70329f78cf3da0bd4aa3e9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 6 May 2021 22:12:00 +0200 Subject: cfg80211: remove CFG80211_MAX_NUM_DIFFERENT_CHANNELS We no longer need to put any limits here, hardware will and mac80211-hwsim can do whatever it likes. The reason we had this was some accounting code (still mentioned in the comment) but that code was deleted in commit c781944b71f8 ("cfg80211: Remove unused cfg80211_can_use_iftype_chan()"). Link: https://lore.kernel.org/r/20210506221159.d1d61db1d31c.Iac4da68d54b9f1fdc18a03586bbe06aeb9515425@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 58c2cd417e89..60325b62daae 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1245,8 +1245,6 @@ struct cfg80211_csa_settings { u8 count; }; -#define CFG80211_MAX_NUM_DIFFERENT_CHANNELS 10 - /** * struct iface_combination_params - input parameters for interface combinations * -- cgit v1.2.3 From cff7b5ca25353bef5909e357a9912f3d44b32af5 Mon Sep 17 00:00:00 2001 From: Philipp Borgers Date: Wed, 19 May 2021 14:20:17 +0200 Subject: mac80211: add ieee80211_is_tx_data helper function Add a helper function that checks if a frame is a data frame. Frames with hardware encapsulation enabled are data frames. Signed-off-by: Philipp Borgers Link: https://lore.kernel.org/r/20210519122019.92359-2-borgers@mi.fu-berlin.de Signed-off-by: Johannes Berg --- include/net/mac80211.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index e89530d0d9c6..4e876e4598e3 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -6752,4 +6752,22 @@ struct sk_buff *ieee80211_get_fils_discovery_tmpl(struct ieee80211_hw *hw, struct sk_buff * ieee80211_get_unsol_bcast_probe_resp_tmpl(struct ieee80211_hw *hw, struct ieee80211_vif *vif); + +/** + * ieee80211_is_tx_data - check if frame is a data frame + * + * The function is used to check if a frame is a data frame. Frames with + * hardware encapsulation enabled are data frames. + * + * @skb: the frame to be transmitted. + */ +static inline bool ieee80211_is_tx_data(struct sk_buff *skb) +{ + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + struct ieee80211_hdr *hdr = (void *) skb->data; + + return info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP || + ieee80211_is_data(hdr->frame_control); +} + #endif /* MAC80211_H */ -- cgit v1.2.3 From 358ae88881adc3ac1544104272eb7e9408f80b39 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Wed, 16 Jun 2021 23:28:26 +0300 Subject: cfg80211: expose the rfkill device to the low level driver This will allow the low level driver to query the rfkill state. Signed-off-by: Emmanuel Grumbach Link: https://lore.kernel.org/r/20210616202826.9833-1-emmanuel.grumbach@intel.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 60325b62daae..b3bc58ec9098 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -22,6 +22,7 @@ #include #include #include +#include #include /** @@ -4943,6 +4944,7 @@ struct wiphy_iftype_akm_suites { * configuration through the %NL80211_TID_CONFIG_ATTR_RETRY_SHORT and * %NL80211_TID_CONFIG_ATTR_RETRY_LONG attributes * @sar_capa: SAR control capabilities + * @rfkill: a pointer to the rfkill structure */ struct wiphy { struct mutex mtx; @@ -5085,6 +5087,8 @@ struct wiphy { const struct cfg80211_sar_capa *sar_capa; + struct rfkill *rfkill; + char priv[] __aligned(NETDEV_ALIGN); }; @@ -6659,7 +6663,10 @@ void wiphy_rfkill_start_polling(struct wiphy *wiphy); * wiphy_rfkill_stop_polling - stop polling rfkill * @wiphy: the wiphy */ -void wiphy_rfkill_stop_polling(struct wiphy *wiphy); +static inline void wiphy_rfkill_stop_polling(struct wiphy *wiphy) +{ + rfkill_pause_polling(wiphy->rfkill); +} /** * DOC: Vendor commands -- cgit v1.2.3 From 08a46c6420013c4ecb61262b4869fdd7e82f918a Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 17 Jun 2021 18:31:11 +0200 Subject: mac80211: move A-MPDU session check from minstrel_ht to mac80211 This avoids calling back into tx handlers from within the rate control module. Preparation for deferring rate control until tx dequeue Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20210617163113.75815-1-nbd@nbd.name Signed-off-by: Johannes Berg --- include/net/mac80211.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 4e876e4598e3..c09cd0e4a6b3 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -6184,6 +6184,11 @@ enum rate_control_capabilities { * otherwise the NSS difference doesn't bother us. */ RATE_CTRL_CAPA_VHT_EXT_NSS_BW = BIT(0), + /** + * @RATE_CTRL_CAPA_AMPDU_TRIGGER: + * mac80211 should start A-MPDU sessions on tx + */ + RATE_CTRL_CAPA_AMPDU_TRIGGER = BIT(1), }; struct rate_control_ops { -- cgit v1.2.3 From d8b261548dcf1058646cc48159c88d42d4b9a3b6 Mon Sep 17 00:00:00 2001 From: Shaul Triebitz Date: Fri, 18 Jun 2021 13:41:35 +0300 Subject: mac80211: add to bss_conf if broadcast TWT is supported Add to struct ieee80211_bss_conf a twt_broadcast field. Set it to true if both STA and AP support broadcast TWT. Signed-off-by: Shaul Triebitz Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.f7c105237541.I50b302044e2b35e5ed4d3fb8bc7bd3d8bb89b1e1@changeid Signed-off-by: Johannes Berg --- include/net/mac80211.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index c09cd0e4a6b3..5d3ce1bd5753 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -7,7 +7,7 @@ * Copyright 2007-2010 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2015 - 2017 Intel Deutschland GmbH - * Copyright (C) 2018 - 2020 Intel Corporation + * Copyright (C) 2018 - 2021 Intel Corporation */ #ifndef MAC80211_H @@ -526,6 +526,7 @@ struct ieee80211_fils_discovery { * @twt_responder: does this BSS support TWT requester (relevant for managed * mode only, set if the AP advertises TWT responder role) * @twt_protected: does this BSS support protected TWT frames + * @twt_broadcast: does this BSS support broadcast TWT * @assoc: association status * @ibss_joined: indicates whether this station is part of an IBSS * or not @@ -642,6 +643,7 @@ struct ieee80211_bss_conf { bool twt_requester; bool twt_responder; bool twt_protected; + bool twt_broadcast; /* association related data */ bool assoc, ibss_joined; bool ibss_creator; -- cgit v1.2.3 From dd3e4fc75b4ab8186a133cfe9d49666a2f8186e0 Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Fri, 18 Jun 2021 13:41:36 +0300 Subject: nl80211/cfg80211: add BSS color to NDP ranging parameters In NDP ranging, the initiator need to set the BSS color in the NDP to the BSS color of the responder. Add the BSS color as a parameter for NDP ranging. Signed-off-by: Avraham Stern Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.f097a6144b59.I27dec8b994df52e691925ea61be4dd4fa6d396c0@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index b3bc58ec9098..c6812945b4b8 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -7,7 +7,7 @@ * Copyright 2006-2010 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright 2015-2017 Intel Deutschland GmbH - * Copyright (C) 2018-2020 Intel Corporation + * Copyright (C) 2018-2021 Intel Corporation */ #include @@ -3521,7 +3521,10 @@ struct cfg80211_pmsr_result { * If neither @trigger_based nor @non_trigger_based is set, * EDCA based ranging will be used. * @lmr_feedback: negotiate for I2R LMR feedback. Only valid if either - * @trigger_based or @non_trigger_based is set. + * @trigger_based or @non_trigger_based is set. + * @bss_color: the bss color of the responder. Optional. Set to zero to + * indicate the driver should set the BSS color. Only valid if + * @non_trigger_based or @trigger_based is set. * * See also nl80211 for the respective attribute documentation. */ @@ -3539,6 +3542,7 @@ struct cfg80211_pmsr_ftm_request_peer { u8 burst_duration; u8 ftms_per_burst; u8 ftmr_retries; + u8 bss_color; }; /** -- cgit v1.2.3 From f253683e602996b250db7a3a7b77e0e908c9dbbc Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 18 Jun 2021 13:41:51 +0300 Subject: cfg80211: remove ieee80211_get_he_sta_cap() This function turned out to be too easy to misuse since it doesn't consider the interface type. Remove it now that we no longer use it in mac80211. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.8c9c72f914b0.I68e9c0626dc77a0f67f238a05ae16a0b77b09895@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index c6812945b4b8..7b4ef45d49b0 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -534,18 +534,6 @@ ieee80211_get_he_iftype_cap(const struct ieee80211_supported_band *sband, return NULL; } -/** - * ieee80211_get_he_sta_cap - return HE capabilities for an sband's STA - * @sband: the sband to search for the STA on - * - * Return: pointer to the struct ieee80211_sta_he_cap, or NULL is none found - */ -static inline const struct ieee80211_sta_he_cap * -ieee80211_get_he_sta_cap(const struct ieee80211_supported_band *sband) -{ - return ieee80211_get_he_iftype_cap(sband, NL80211_IFTYPE_STATION); -} - /** * ieee80211_get_he_6ghz_capa - return HE 6 GHz capabilities * @sband: the sband to search for the STA on -- cgit v1.2.3 From be989891e4f2ff5649bf22ab05a7cdd3a287e34b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 18 Jun 2021 13:41:39 +0300 Subject: cfg80211: add cfg80211_any_usable_channels() This helper function checks if there are any usable channels on any of the given bands with the given properties (as expressed by disallowed channel flags). Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.2b613addaa85.Idaf8b859089490537878a7de5c7453a873a3f638@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 7b4ef45d49b0..481e4e24800f 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -894,6 +894,17 @@ ieee80211_chandef_max_power(struct cfg80211_chan_def *chandef) return chandef->chan->max_power; } +/** + * cfg80211_any_usable_channels - check for usable channels + * @wiphy: the wiphy to check for + * @band_mask: which bands to check on + * @prohibited_flags: which channels to not consider usable, + * %IEEE80211_CHAN_DISABLED is always taken into account + */ +bool cfg80211_any_usable_channels(struct wiphy *wiphy, + unsigned long band_mask, + u32 prohibited_flags); + /** * enum survey_info_flags - survey information flags * -- cgit v1.2.3 From f4f8650588d35deafaa4a4e28cceb3557a71e711 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 18 Jun 2021 13:41:52 +0300 Subject: cfg80211: allow advertising vendor-specific capabilities There may be cases where vendor-specific elements need to be used over the air. Rather than have driver or firmware add them and possibly cause problems that way, add them to the iftype-data band capabilities. This way we can advertise to userspace first, and use them in mac80211 next. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.e8c4f0347276.Iee5964682b3e9ec51fc1cd57a7c62383eaf6ddd7@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 481e4e24800f..c93a2cd77920 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -371,11 +371,18 @@ struct ieee80211_sta_he_cap { * @he_cap: holds the HE capabilities * @he_6ghz_capa: HE 6 GHz capabilities, must be filled in for a * 6 GHz band channel (and 0 may be valid value). + * @vendor_elems: vendor element(s) to advertise + * @vendor_elems.data: vendor element(s) data + * @vendor_elems.len: vendor element(s) length */ struct ieee80211_sband_iftype_data { u16 types_mask; struct ieee80211_sta_he_cap he_cap; struct ieee80211_he_6ghz_capa he_6ghz_capa; + struct { + const u8 *data; + unsigned int len; + } vendor_elems; }; /** -- cgit v1.2.3 From 15fae3410f1d879b18e08fe8ef293d538549dfcb Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 18 Jun 2021 13:41:55 +0300 Subject: mac80211: notify driver on mgd TX completion We have mgd_prepare_tx(), but sometimes drivers may want/need to take action when the exchange finishes, whether successfully or not. Add a notification to the driver on completion, i.e. call the new method mgd_complete_tx(). To unify the two scenarios, and to add more information, make both of them take a struct that has the duration (prepare only), subtype (both) and success (complete only). Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/iwlwifi.20210618133832.5d94e78f6230.I6dc979606b6f28701b740d7aab725f7853a5a155@changeid Signed-off-by: Johannes Berg --- include/net/mac80211.h | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 5d3ce1bd5753..9afbcac61c2a 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -3346,6 +3346,21 @@ enum ieee80211_reconfig_type { IEEE80211_RECONFIG_TYPE_SUSPEND, }; +/** + * struct ieee80211_prep_tx_info - prepare TX information + * @duration: if non-zero, hint about the required duration, + * only used with the mgd_prepare_tx() method. + * @subtype: frame subtype (auth, (re)assoc, deauth, disassoc) + * @success: whether the frame exchange was successful, only + * used with the mgd_complete_tx() method, and then only + * valid for auth and (re)assoc. + */ +struct ieee80211_prep_tx_info { + u16 duration; + u16 subtype; + u8 success:1; +}; + /** * struct ieee80211_ops - callbacks from mac80211 to the driver * @@ -3758,9 +3773,13 @@ enum ieee80211_reconfig_type { * frame in case that no beacon was heard from the AP/P2P GO. * The callback will be called before each transmission and upon return * mac80211 will transmit the frame right away. - * If duration is greater than zero, mac80211 hints to the driver the - * duration for which the operation is requested. + * Additional information is passed in the &struct ieee80211_prep_tx_info + * data. If duration there is greater than zero, mac80211 hints to the + * driver the duration for which the operation is requested. * The callback is optional and can (should!) sleep. + * @mgd_complete_tx: Notify the driver that the response frame for a previously + * transmitted frame announced with @mgd_prepare_tx was received, the data + * is filled similarly to @mgd_prepare_tx though the duration is not used. * * @mgd_protect_tdls_discover: Protect a TDLS discovery session. After sending * a TDLS discovery-request, we expect a reply to arrive on the AP's @@ -4111,7 +4130,10 @@ struct ieee80211_ops { void (*mgd_prepare_tx)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, - u16 duration); + struct ieee80211_prep_tx_info *info); + void (*mgd_complete_tx)(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_prep_tx_info *info); void (*mgd_protect_tdls_discover)(struct ieee80211_hw *hw, struct ieee80211_vif *vif); -- cgit v1.2.3 From 2433647bc8d983a543e7d31b41ca2de1c7e2c198 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Toke=20H=C3=B8iland-J=C3=B8rgensen?= Date: Wed, 23 Jun 2021 15:47:55 +0200 Subject: mac80211: Switch to a virtual time-based airtime scheduler MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This switches the airtime scheduler in mac80211 to use a virtual time-based scheduler instead of the round-robin scheduler used before. This has a couple of advantages: - No need to sync up the round-robin scheduler in firmware/hardware with the round-robin airtime scheduler. - If several stations are eligible for transmission we can schedule both of them; no need to hard-block the scheduling rotation until the head of the queue has used up its quantum. - The check of whether a station is eligible for transmission becomes simpler (in ieee80211_txq_may_transmit()). The drawback is that scheduling becomes slightly more expensive, as we need to maintain an rbtree of TXQs sorted by virtual time. This means that ieee80211_register_airtime() becomes O(logN) in the number of currently scheduled TXQs because it can change the order of the scheduled stations. We mitigate this overhead by only resorting when a station changes position in the tree, and hopefully N rarely grows too big (it's only TXQs currently backlogged, not all associated stations), so it shouldn't be too big of an issue. To prevent divisions in the fast path, we maintain both station sums and pre-computed reciprocals of the sums. This turns the fast-path operation into a multiplication, with divisions only happening as the number of active stations change (to re-compute the current sum of all active station weights). To prevent this re-computation of the reciprocal from happening too frequently, we use a time-based notion of station activity, instead of updating the weight every time a station gets scheduled or de-scheduled. As queues can oscillate between empty and occupied quite frequently, this can significantly cut down on the number of re-computations. It also has the added benefit of making the station airtime calculation independent on whether the queue happened to have drained at the time an airtime value was accounted. Co-developed-by: Yibo Zhao Signed-off-by: Yibo Zhao Signed-off-by: Toke Høiland-Jørgensen Link: https://lore.kernel.org/r/20210623134755.235545-1-toke@redhat.com Signed-off-by: Johannes Berg --- include/net/mac80211.h | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 9afbcac61c2a..d8a1d09a2141 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -6605,9 +6605,6 @@ static inline void ieee80211_txq_schedule_end(struct ieee80211_hw *hw, u8 ac) { } -void __ieee80211_schedule_txq(struct ieee80211_hw *hw, - struct ieee80211_txq *txq, bool force); - /** * ieee80211_schedule_txq - schedule a TXQ for transmission * @@ -6620,11 +6617,7 @@ void __ieee80211_schedule_txq(struct ieee80211_hw *hw, * The driver may call this function if it has buffered packets for * this TXQ internally. */ -static inline void -ieee80211_schedule_txq(struct ieee80211_hw *hw, struct ieee80211_txq *txq) -{ - __ieee80211_schedule_txq(hw, txq, true); -} +void ieee80211_schedule_txq(struct ieee80211_hw *hw, struct ieee80211_txq *txq); /** * ieee80211_return_txq - return a TXQ previously acquired by ieee80211_next_txq() @@ -6636,12 +6629,8 @@ ieee80211_schedule_txq(struct ieee80211_hw *hw, struct ieee80211_txq *txq) * The driver may set force=true if it has buffered packets for this TXQ * internally. */ -static inline void -ieee80211_return_txq(struct ieee80211_hw *hw, struct ieee80211_txq *txq, - bool force) -{ - __ieee80211_schedule_txq(hw, txq, force); -} +void ieee80211_return_txq(struct ieee80211_hw *hw, struct ieee80211_txq *txq, + bool force); /** * ieee80211_txq_may_transmit - check whether TXQ is allowed to transmit -- cgit v1.2.3 From 761025b51c540ae1fc9516b5dafa55cd109e4871 Mon Sep 17 00:00:00 2001 From: Dmitry Osipenko Date: Wed, 12 May 2021 00:15:48 +0300 Subject: cfg80211: Add wiphy_info_once() Add wiphy_info_once() helper that prints info message only once. Signed-off-by: Dmitry Osipenko Acked-by: Johannes Berg Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20210511211549.30571-1-digetx@gmail.com --- include/net/cfg80211.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 58c2cd417e89..1e0bf249b601 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -8154,6 +8154,8 @@ bool cfg80211_iftype_allowed(struct wiphy *wiphy, enum nl80211_iftype iftype, dev_notice(&(wiphy)->dev, format, ##args) #define wiphy_info(wiphy, format, args...) \ dev_info(&(wiphy)->dev, format, ##args) +#define wiphy_info_once(wiphy, format, args...) \ + dev_info_once(&(wiphy)->dev, format, ##args) #define wiphy_err_ratelimited(wiphy, format, args...) \ dev_err_ratelimited(&(wiphy)->dev, format, ##args) -- cgit v1.2.3 From dd25296afaf60b5140ddfa9e3d8e5d9df7076754 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Tue, 22 Jun 2021 14:49:55 +0800 Subject: net: sched: avoid unnecessary seqcount operation for lockless qdisc qdisc->running seqcount operation is mainly used to do heuristic locking on q->busylock for locked qdisc, see qdisc_is_running() and __dev_xmit_skb(). So avoid doing seqcount operation for qdisc with TCQ_F_NOLOCK flag. Acked-by: Jakub Kicinski Tested-by: Vladimir Oltean # flexcan Signed-off-by: Yunsheng Lin Signed-off-by: David S. Miller --- include/net/sch_generic.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 1e625519ae96..3ed6bcc4be72 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -188,6 +188,7 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) nolock_empty: WRITE_ONCE(qdisc->empty, false); + return true; } else if (qdisc_is_running(qdisc)) { return false; } @@ -201,7 +202,6 @@ nolock_empty: static inline void qdisc_run_end(struct Qdisc *qdisc) { - write_seqcount_end(&qdisc->running); if (qdisc->flags & TCQ_F_NOLOCK) { spin_unlock(&qdisc->seqlock); @@ -210,6 +210,8 @@ static inline void qdisc_run_end(struct Qdisc *qdisc) clear_bit(__QDISC_STATE_MISSED, &qdisc->state); __netif_schedule(qdisc); } + } else { + write_seqcount_end(&qdisc->running); } } -- cgit v1.2.3 From c4fef01ba4793a85b2d38a472bddd1e3b56d9585 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Tue, 22 Jun 2021 14:49:56 +0800 Subject: net: sched: implement TCQ_F_CAN_BYPASS for lockless qdisc Currently pfifo_fast has both TCQ_F_CAN_BYPASS and TCQ_F_NOLOCK flag set, but queue discipline by-pass does not work for lockless qdisc because skb is always enqueued to qdisc even when the qdisc is empty, see __dev_xmit_skb(). This patch calls sch_direct_xmit() to transmit the skb directly to the driver for empty lockless qdisc, which aviod enqueuing and dequeuing operation. As qdisc->empty is not reliable to indicate a empty qdisc because there is a time window between enqueuing and setting qdisc->empty. So we use the MISSED state added in commit a90c57f2cedd ("net: sched: fix packet stuck problem for lockless qdisc"), which indicate there is lock contention, suggesting that it is better not to do the qdisc bypass in order to avoid packet out of order problem. In order to make MISSED state reliable to indicate a empty qdisc, we need to ensure that testing and clearing of MISSED state is within the protection of qdisc->seqlock, only setting MISSED state can be done without the protection of qdisc->seqlock. A MISSED state testing is added without the protection of qdisc->seqlock to aviod doing unnecessary spin_trylock() for contention case. As the enqueuing is not within the protection of qdisc->seqlock, there is still a potential data race as mentioned by Jakub [1]: thread1 thread2 thread3 qdisc_run_begin() # true qdisc_run_begin(q) set(MISSED) pfifo_fast_dequeue clear(MISSED) # recheck the queue qdisc_run_end() enqueue skb1 qdisc empty # true qdisc_run_begin() # true sch_direct_xmit() # skb2 qdisc_run_begin() set(MISSED) When above happens, skb1 enqueued by thread2 is transmited after skb2 is transmited by thread3 because MISSED state setting and enqueuing is not under the qdisc->seqlock. If qdisc bypass is disabled, skb1 has better chance to be transmited quicker than skb2. This patch does not take care of the above data race, because we view this as similar as below: Even at the same time CPU1 and CPU2 write the skb to two socket which both heading to the same qdisc, there is no guarantee that which skb will hit the qdisc first, because there is a lot of factor like interrupt/softirq/cache miss/scheduling afffecting that. There are below cases that need special handling: 1. When MISSED state is cleared before another round of dequeuing in pfifo_fast_dequeue(), and __qdisc_run() might not be able to dequeue all skb in one round and call __netif_schedule(), which might result in a non-empty qdisc without MISSED set. In order to avoid this, the MISSED state is set for lockless qdisc and __netif_schedule() will be called at the end of qdisc_run_end. 2. The MISSED state also need to be set for lockless qdisc instead of calling __netif_schedule() directly when requeuing a skb for a similar reason. 3. For netdev queue stopped case, the MISSED case need clearing while the netdev queue is stopped, otherwise there may be unnecessary __netif_schedule() calling. So a new DRAINING state is added to indicate this case, which also indicate a non-empty qdisc. 4. As there is already netif_xmit_frozen_or_stopped() checking in dequeue_skb() and sch_direct_xmit(), which are both within the protection of qdisc->seqlock, but the same checking in __dev_xmit_skb() is without the protection, which might cause empty indication of a lockless qdisc to be not reliable. So remove the checking in __dev_xmit_skb(), and the checking in the protection of qdisc->seqlock seems enough to avoid the cpu consumption problem for netdev queue stopped case. 1. https://lkml.org/lkml/2021/5/29/215 Acked-by: Jakub Kicinski Tested-by: Vladimir Oltean # flexcan Signed-off-by: Yunsheng Lin Signed-off-by: David S. Miller --- include/net/sch_generic.h | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 3ed6bcc4be72..177f240d59d4 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -37,8 +37,15 @@ enum qdisc_state_t { __QDISC_STATE_SCHED, __QDISC_STATE_DEACTIVATED, __QDISC_STATE_MISSED, + __QDISC_STATE_DRAINING, }; +#define QDISC_STATE_MISSED BIT(__QDISC_STATE_MISSED) +#define QDISC_STATE_DRAINING BIT(__QDISC_STATE_DRAINING) + +#define QDISC_STATE_NON_EMPTY (QDISC_STATE_MISSED | \ + QDISC_STATE_DRAINING) + struct qdisc_size_table { struct rcu_head rcu; struct list_head list; @@ -145,6 +152,11 @@ static inline bool qdisc_is_running(struct Qdisc *qdisc) return (raw_read_seqcount(&qdisc->running) & 1) ? true : false; } +static inline bool nolock_qdisc_is_empty(const struct Qdisc *qdisc) +{ + return !(READ_ONCE(qdisc->state) & QDISC_STATE_NON_EMPTY); +} + static inline bool qdisc_is_percpu_stats(const struct Qdisc *q) { return q->flags & TCQ_F_CPUSTATS; @@ -206,10 +218,8 @@ static inline void qdisc_run_end(struct Qdisc *qdisc) spin_unlock(&qdisc->seqlock); if (unlikely(test_bit(__QDISC_STATE_MISSED, - &qdisc->state))) { - clear_bit(__QDISC_STATE_MISSED, &qdisc->state); + &qdisc->state))) __netif_schedule(qdisc); - } } else { write_seqcount_end(&qdisc->running); } -- cgit v1.2.3 From d3e0f57501bde8a9585aff79afcffd99e6a5d91c Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Tue, 22 Jun 2021 14:49:57 +0800 Subject: net: sched: remove qdisc->empty for lockless qdisc As MISSED and DRAINING state are used to indicate a non-empty qdisc, qdisc->empty is not longer needed, so remove it. Acked-by: Jakub Kicinski Tested-by: Vladimir Oltean # flexcan Signed-off-by: Yunsheng Lin Signed-off-by: David S. Miller --- include/net/sch_generic.h | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 177f240d59d4..c99ffe9cc88f 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -117,8 +117,6 @@ struct Qdisc { spinlock_t busylock ____cacheline_aligned_in_smp; spinlock_t seqlock; - /* for NOLOCK qdisc, true if there are no enqueued skbs */ - bool empty; struct rcu_head rcu; /* private data */ @@ -165,7 +163,7 @@ static inline bool qdisc_is_percpu_stats(const struct Qdisc *q) static inline bool qdisc_is_empty(const struct Qdisc *qdisc) { if (qdisc_is_percpu_stats(qdisc)) - return READ_ONCE(qdisc->empty); + return nolock_qdisc_is_empty(qdisc); return !READ_ONCE(qdisc->q.qlen); } @@ -173,7 +171,7 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) { if (qdisc->flags & TCQ_F_NOLOCK) { if (spin_trylock(&qdisc->seqlock)) - goto nolock_empty; + return true; /* If the MISSED flag is set, it means other thread has * set the MISSED flag before second spin_trylock(), so @@ -195,12 +193,7 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) /* Retry again in case other CPU may not see the new flag * after it releases the lock at the end of qdisc_run_end(). */ - if (!spin_trylock(&qdisc->seqlock)) - return false; - -nolock_empty: - WRITE_ONCE(qdisc->empty, false); - return true; + return spin_trylock(&qdisc->seqlock); } else if (qdisc_is_running(qdisc)) { return false; } -- cgit v1.2.3 From 10ed7ce42b13790ba85f8e10110d89a2bce58807 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 23 Jun 2021 15:06:34 +0900 Subject: net/tls: Remove the __TLS_DEC_STATS() macro. The commit d26b698dd3cd ("net/tls: add skeleton of MIB statistics") introduced __TLS_DEC_STATS(), but it is not used and __SNMP_DEC_STATS() is not defined also. Let's remove it. Signed-off-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- include/net/tls.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index 8341a8d1e807..8d398a5de3ee 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -79,8 +79,6 @@ __SNMP_INC_STATS((net)->mib.tls_statistics, field) #define TLS_INC_STATS(net, field) \ SNMP_INC_STATS((net)->mib.tls_statistics, field) -#define __TLS_DEC_STATS(net, field) \ - __SNMP_DEC_STATS((net)->mib.tls_statistics, field) #define TLS_DEC_STATS(net, field) \ SNMP_DEC_STATS((net)->mib.tls_statistics, field) -- cgit v1.2.3 From 782347b6bcad07ddb574422e01e22c92e05928c8 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Toke=20H=C3=B8iland-J=C3=B8rgensen?= Date: Thu, 24 Jun 2021 18:05:55 +0200 Subject: xdp: Add proper __rcu annotations to redirect map entries MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit XDP_REDIRECT works by a three-step process: the bpf_redirect() and bpf_redirect_map() helpers will lookup the target of the redirect and store it (along with some other metadata) in a per-CPU struct bpf_redirect_info. Next, when the program returns the XDP_REDIRECT return code, the driver will call xdp_do_redirect() which will use the information thus stored to actually enqueue the frame into a bulk queue structure (that differs slightly by map type, but shares the same principle). Finally, before exiting its NAPI poll loop, the driver will call xdp_do_flush(), which will flush all the different bulk queues, thus completing the redirect. Pointers to the map entries will be kept around for this whole sequence of steps, protected by RCU. However, there is no top-level rcu_read_lock() in the core code; instead drivers add their own rcu_read_lock() around the XDP portions of the code, but somewhat inconsistently as Martin discovered[0]. However, things still work because everything happens inside a single NAPI poll sequence, which means it's between a pair of calls to local_bh_disable()/local_bh_enable(). So Paul suggested[1] that we could document this intention by using rcu_dereference_check() with rcu_read_lock_bh_held() as a second parameter, thus allowing sparse and lockdep to verify that everything is done correctly. This patch does just that: we add an __rcu annotation to the map entry pointers and remove the various comments explaining the NAPI poll assurance strewn through devmap.c in favour of a longer explanation in filter.c. The goal is to have one coherent documentation of the entire flow, and rely on the RCU annotations as a "standard" way of communicating the flow in the map code (which can additionally be understood by sparse and lockdep). The RCU annotation replacements result in a fairly straight-forward replacement where READ_ONCE() becomes rcu_dereference_check(), WRITE_ONCE() becomes rcu_assign_pointer() and xchg() and cmpxchg() gets wrapped in the proper constructs to cast the pointer back and forth between __rcu and __kernel address space (for the benefit of sparse). The one complication is that xskmap has a few constructions where double-pointers are passed back and forth; these simply all gain __rcu annotations, and only the final reference/dereference to the inner-most pointer gets changed. With this, everything can be run through sparse without eliciting complaints, and lockdep can verify correctness even without the use of rcu_read_lock() in the drivers. Subsequent patches will clean these up from the drivers. [0] https://lore.kernel.org/bpf/20210415173551.7ma4slcbqeyiba2r@kafai-mbp.dhcp.thefacebook.com/ [1] https://lore.kernel.org/bpf/20210419165837.GA975577@paulmck-ThinkPad-P17-Gen-1/ Signed-off-by: Toke Høiland-Jørgensen Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20210624160609.292325-6-toke@redhat.com --- include/net/xdp_sock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h index 9c0722c6d7ac..fff069d2ed1b 100644 --- a/include/net/xdp_sock.h +++ b/include/net/xdp_sock.h @@ -37,7 +37,7 @@ struct xdp_umem { struct xsk_map { struct bpf_map map; spinlock_t lock; /* Synchronize map updates */ - struct xdp_sock *xsk_map[]; + struct xdp_sock __rcu *xsk_map[]; }; struct xdp_sock { -- cgit v1.2.3 From 1f7fe5121127e037b86592ba42ce36515ea0e3f7 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Thu, 24 Jun 2021 11:38:28 +0200 Subject: net: macsec: fix the length used to copy the key for offloading The key length used when offloading macsec to Ethernet or PHY drivers was set to MACSEC_KEYID_LEN (16), which is an issue as: - This was never meant to be the key length. - The key length can be > 16. Fix this by using MACSEC_MAX_KEY_LEN to store the key (the max length accepted in uAPI) and secy->key_len to copy it. Fixes: 3cf3227a21d1 ("net: macsec: hardware offloading infrastructure") Reported-by: Lior Nahmanson Signed-off-by: Antoine Tenart Signed-off-by: David S. Miller --- include/net/macsec.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/macsec.h b/include/net/macsec.h index 52874cdfe226..d6fa6b97f6ef 100644 --- a/include/net/macsec.h +++ b/include/net/macsec.h @@ -241,7 +241,7 @@ struct macsec_context { struct macsec_rx_sc *rx_sc; struct { unsigned char assoc_num; - u8 key[MACSEC_KEYID_LEN]; + u8 key[MACSEC_MAX_KEY_LEN]; union { struct macsec_rx_sa *rx_sa; struct macsec_tx_sa *tx_sa; -- cgit v1.2.3 From 0dac127c05579854405ef14480936b32371ddaed Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 24 Jun 2021 11:48:08 -0400 Subject: sctp: do black hole detection in search complete state Currently the PLPMUTD probe will stop for a long period (interval * 30) after it enters search complete state. If there's a pmtu change on the route path, it takes a long time to be aware if the ICMP TooBig packet is lost or filtered. As it says in rfc8899#section-4.3: "A DPLPMTUD method MUST NOT rely solely on this method." (ICMP PTB message). This patch is to enable the other method for search complete state: "A PL can use the DPLPMTUD probing mechanism to periodically generate probe packets of the size of the current PLPMTU." With this patch, the probe will continue with the current pmtu every 'interval' until the PMTU_RAISE_TIMER 'timeout', which we implement by adding raise_count to raise the probe size when it counts to 30 and removing the SCTP_PL_COMPLETE check for PMTU_RAISE_TIMER. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 9eaa701cda23..c4a4c1754be8 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -987,7 +987,8 @@ struct sctp_transport { __u16 pmtu; __u16 probe_size; __u16 probe_high; - __u8 probe_count; + __u8 probe_count:3; + __u8 raise_count:5; __u8 state; } pl; /* plpmtud related */ -- cgit v1.2.3 From 1c6ed31b1696d9b5462ba5ce15b83f5ea955600c Mon Sep 17 00:00:00 2001 From: Yu Liu Date: Fri, 9 Apr 2021 15:04:06 -0700 Subject: Bluetooth: Return whether a connection is outbound When an MGMT_EV_DEVICE_CONNECTED event is reported back to the user space we will set the flags to tell if the established connection is outbound or not. This is useful for the user space to log better metrics and error messages. Reviewed-by: Miao-chen Chou Reviewed-by: Alain Michaud Signed-off-by: Yu Liu Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 +- include/net/bluetooth/mgmt.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index c73ac52af186..8f5f390363f5 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1768,7 +1768,7 @@ 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, - u32 flags, u8 *name, u8 name_len); + u8 *name, u8 name_len); void mgmt_device_disconnected(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type, u8 addr_type, u8 reason, bool mgmt_connected); diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index a7cffb069565..a03c62b1dc2f 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -939,6 +939,7 @@ struct mgmt_ev_auth_failed { #define MGMT_DEV_FOUND_CONFIRM_NAME 0x01 #define MGMT_DEV_FOUND_LEGACY_PAIRING 0x02 #define MGMT_DEV_FOUND_NOT_CONNECTABLE 0x04 +#define MGMT_DEV_FOUND_INITIATED_CONN 0x08 #define MGMT_EV_DEVICE_FOUND 0x0012 struct mgmt_ev_device_found { -- cgit v1.2.3 From de75cd0d9b2f3250d5f25846bb5632ccce6275f4 Mon Sep 17 00:00:00 2001 From: Manish Mandlik Date: Thu, 29 Apr 2021 10:24:22 -0700 Subject: Bluetooth: Add ncmd=0 recovery handling During command status or command complete event, the controller may set ncmd=0 indicating that it is not accepting any more commands. In such a case, host holds off sending any more commands to the controller. If the controller doesn't recover from such condition, host will wait forever, until the user decides that the Bluetooth is broken and may power cycles the Bluetooth. This patch triggers the hardware error to reset the controller and driver when it gets into such state as there is no other wat out. Reviewed-by: Abhishek Pandit-Subedi Signed-off-by: Manish Mandlik Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 1 + include/net/bluetooth/hci_core.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index ea4ae551c426..c4b0650fb9ae 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -339,6 +339,7 @@ enum { #define HCI_PAIRING_TIMEOUT msecs_to_jiffies(60000) /* 60 seconds */ #define HCI_INIT_TIMEOUT msecs_to_jiffies(10000) /* 10 seconds */ #define HCI_CMD_TIMEOUT msecs_to_jiffies(2000) /* 2 seconds */ +#define HCI_NCMD_TIMEOUT msecs_to_jiffies(4000) /* 4 seconds */ #define HCI_ACL_TX_TIMEOUT msecs_to_jiffies(45000) /* 45 seconds */ #define HCI_AUTO_OFF_TIMEOUT msecs_to_jiffies(2000) /* 2 seconds */ #define HCI_POWER_OFF_TIMEOUT msecs_to_jiffies(5000) /* 5 seconds */ diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 8f5f390363f5..43b08bebae74 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -470,6 +470,7 @@ struct hci_dev { struct delayed_work service_cache; struct delayed_work cmd_timer; + struct delayed_work ncmd_timer; struct work_struct rx_work; struct work_struct cmd_work; -- cgit v1.2.3 From 76c185a51505262fe19b5a2cd5dd70199d21949b Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Mon, 31 May 2021 16:37:21 +0800 Subject: Bluetooth: use inclusive language in hci_core.h This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced: master -> central slave -> peripheral These attributes are not used elsewhere in the code. Signed-off-by: Archie Pusaka Reviewed-by: Miao-chen Chou Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 44 ++++++++++++++++++++++---------------------- 1 file changed, 22 insertions(+), 22 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index c4b0650fb9ae..4eea590cd432 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -1839,23 +1839,23 @@ struct hci_rp_le_read_iso_tx_sync { #define HCI_OP_LE_SET_CIG_PARAMS 0x2062 struct hci_cis_params { __u8 cis_id; - __le16 m_sdu; - __le16 s_sdu; - __u8 m_phy; - __u8 s_phy; - __u8 m_rtn; - __u8 s_rtn; + __le16 c_sdu; + __le16 p_pdu; + __u8 c_phy; + __u8 p_phy; + __u8 c_rtn; + __u8 p_rtn; } __packed; struct hci_cp_le_set_cig_params { __u8 cig_id; - __u8 m_interval[3]; - __u8 s_interval[3]; - __u8 sca; + __u8 c_interval[3]; + __u8 p_interval[3]; + __u8 wc_sca; __u8 packing; __u8 framing; - __le16 m_latency; - __le16 s_latency; + __le16 c_latency; + __le16 p_latency; __u8 num_cis; struct hci_cis_params cis[]; } __packed; @@ -2260,7 +2260,7 @@ struct hci_ev_sync_train_complete { __u8 status; } __packed; -#define HCI_EV_SLAVE_PAGE_RESP_TIMEOUT 0x54 +#define HCI_EV_PERIPHERAL_PAGE_RESP_TIMEOUT 0x54 #define HCI_EV_LE_CONN_COMPLETE 0x01 struct hci_ev_le_conn_complete { @@ -2418,17 +2418,17 @@ struct hci_evt_le_cis_established { __le16 handle; __u8 cig_sync_delay[3]; __u8 cis_sync_delay[3]; - __u8 m_latency[3]; - __u8 s_latency[3]; - __u8 m_phy; - __u8 s_phy; + __u8 c_latency[3]; + __u8 p_latency[3]; + __u8 c_phy; + __u8 p_phy; __u8 nse; - __u8 m_bn; - __u8 s_bn; - __u8 m_ft; - __u8 s_ft; - __le16 m_mtu; - __le16 s_mtu; + __u8 c_bn; + __u8 p_bn; + __u8 c_ft; + __u8 p_ft; + __le16 c_mtu; + __le16 p_mtu; __le16 interval; } __packed; -- cgit v1.2.3 From 6397729bb74df3918187c5e96fb0f63c5f5292d9 Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Mon, 31 May 2021 16:37:22 +0800 Subject: Bluetooth: use inclusive language to describe CPB This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced when describing the connectionless peripheral broadcast feature: master -> central slave -> peripheral Signed-off-by: Archie Pusaka Reviewed-by: Miao-chen Chou Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 26 +++++++++++++------------- include/net/bluetooth/hci_core.h | 4 ++-- 2 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 4eea590cd432..ece96ccc42de 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -36,7 +36,7 @@ #define HCI_MAX_AMP_ASSOC_SIZE 672 -#define HCI_MAX_CSB_DATA_SIZE 252 +#define HCI_MAX_CPB_DATA_SIZE 252 /* HCI dev events */ #define HCI_DEV_REG 1 @@ -472,10 +472,10 @@ enum { #define LMP_EXTFEATURES 0x80 /* Extended LMP features */ -#define LMP_CSB_MASTER 0x01 -#define LMP_CSB_SLAVE 0x02 -#define LMP_SYNC_TRAIN 0x04 -#define LMP_SYNC_SCAN 0x08 +#define LMP_CPB_CENTRAL 0x01 +#define LMP_CPB_PERIPHERAL 0x02 +#define LMP_SYNC_TRAIN 0x04 +#define LMP_SYNC_SCAN 0x08 #define LMP_SC 0x01 #define LMP_PING 0x02 @@ -877,17 +877,17 @@ struct hci_rp_logical_link_cancel { __u8 flow_spec_id; } __packed; -#define HCI_OP_SET_CSB 0x0441 -struct hci_cp_set_csb { +#define HCI_OP_SET_CPB 0x0441 +struct hci_cp_set_cpb { __u8 enable; __u8 lt_addr; __u8 lpo_allowed; __le16 packet_type; __le16 interval_min; __le16 interval_max; - __le16 csb_sv_tout; + __le16 cpb_sv_tout; } __packed; -struct hci_rp_set_csb { +struct hci_rp_set_cpb { __u8 status; __u8 lt_addr; __le16 interval; @@ -1184,14 +1184,14 @@ struct hci_rp_delete_reserved_lt_addr { __u8 lt_addr; } __packed; -#define HCI_OP_SET_CSB_DATA 0x0c76 -struct hci_cp_set_csb_data { +#define HCI_OP_SET_CPB_DATA 0x0c76 +struct hci_cp_set_cpb_data { __u8 lt_addr; __u8 fragment; __u8 data_length; - __u8 data[HCI_MAX_CSB_DATA_SIZE]; + __u8 data[HCI_MAX_CPB_DATA_SIZE]; } __packed; -struct hci_rp_set_csb_data { +struct hci_rp_set_cpb_data { __u8 status; __u8 lt_addr; } __packed; diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 43b08bebae74..c9ec06997e1c 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1394,8 +1394,8 @@ void hci_conn_del_sysfs(struct hci_conn *conn); #define lmp_edr_5slot_capable(dev) ((dev)->features[0][5] & LMP_EDR_5SLOT) /* ----- Extended LMP capabilities ----- */ -#define lmp_csb_master_capable(dev) ((dev)->features[2][0] & LMP_CSB_MASTER) -#define lmp_csb_slave_capable(dev) ((dev)->features[2][0] & LMP_CSB_SLAVE) +#define lmp_cpb_central_capable(dev) ((dev)->features[2][0] & LMP_CPB_CENTRAL) +#define lmp_cpb_peripheral_capable(dev) ((dev)->features[2][0] & LMP_CPB_PERIPHERAL) #define lmp_sync_train_capable(dev) ((dev)->features[2][0] & LMP_SYNC_TRAIN) #define lmp_sync_scan_capable(dev) ((dev)->features[2][0] & LMP_SYNC_SCAN) #define lmp_sc_capable(dev) ((dev)->features[2][1] & LMP_SC) -- cgit v1.2.3 From ef365da1803de7891589c75304c8c36bb7cf4b98 Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Mon, 31 May 2021 16:37:23 +0800 Subject: Bluetooth: use inclusive language in HCI LE features This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced: master -> central slave -> peripheral Signed-off-by: Archie Pusaka Reviewed-by: Miao-chen Chou Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index ece96ccc42de..3abd6273a189 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -489,7 +489,7 @@ enum { /* LE features */ #define HCI_LE_ENCRYPTION 0x01 #define HCI_LE_CONN_PARAM_REQ_PROC 0x02 -#define HCI_LE_SLAVE_FEATURES 0x08 +#define HCI_LE_PERIPHERAL_FEATURES 0x08 #define HCI_LE_PING 0x10 #define HCI_LE_DATA_LEN_EXT 0x20 #define HCI_LE_LL_PRIVACY 0x40 @@ -498,8 +498,8 @@ enum { #define HCI_LE_PHY_CODED 0x08 #define HCI_LE_EXT_ADV 0x10 #define HCI_LE_CHAN_SEL_ALG2 0x40 -#define HCI_LE_CIS_MASTER 0x10 -#define HCI_LE_CIS_SLAVE 0x20 +#define HCI_LE_CIS_CENTRAL 0x10 +#define HCI_LE_CIS_PERIPHERAL 0x20 /* Connection modes */ #define HCI_CM_ACTIVE 0x0000 -- cgit v1.2.3 From fad646e16d3cafd67d3cfff8e66f77401190957e Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Mon, 31 May 2021 16:37:25 +0800 Subject: Bluetooth: use inclusive language in SMP This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced: master -> initiator slave -> responder Signed-off-by: Archie Pusaka Signed-off-by: Marcel Holtmann --- include/net/bluetooth/mgmt.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index a03c62b1dc2f..23a0524061b7 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -202,7 +202,7 @@ struct mgmt_cp_load_link_keys { struct mgmt_ltk_info { struct mgmt_addr_info addr; __u8 type; - __u8 master; + __u8 initiator; __u8 enc_size; __le16 ediv; __le64 rand; -- cgit v1.2.3 From 39bc74ca0119025e3cc24b97ebd964b5c605aa83 Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Fri, 4 Jun 2021 16:26:26 +0800 Subject: Bluetooth: use inclusive language when tracking connections This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced: master -> central slave -> peripheral Signed-off-by: Archie Pusaka Reviewed-by: Miao-chen Chou Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index c9ec06997e1c..fe5f3a9d9924 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -122,7 +122,7 @@ struct hci_conn_hash { unsigned int amp_num; unsigned int sco_num; unsigned int le_num; - unsigned int le_num_slave; + unsigned int le_num_peripheral; }; struct bdaddr_list { @@ -894,7 +894,7 @@ static inline void hci_conn_hash_add(struct hci_dev *hdev, struct hci_conn *c) case LE_LINK: h->le_num++; if (c->role == HCI_ROLE_SLAVE) - h->le_num_slave++; + h->le_num_peripheral++; break; case SCO_LINK: case ESCO_LINK: @@ -920,7 +920,7 @@ static inline void hci_conn_hash_del(struct hci_dev *hdev, struct hci_conn *c) case LE_LINK: h->le_num--; if (c->role == HCI_ROLE_SLAVE) - h->le_num_slave--; + h->le_num_peripheral--; break; case SCO_LINK: case ESCO_LINK: -- cgit v1.2.3 From 3d4f9c00492b4e21641e5140a5e78cb50b58d60b Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Fri, 4 Jun 2021 16:26:27 +0800 Subject: Bluetooth: use inclusive language when filtering devices This patch replaces some non-inclusive terms based on the appropriate language mapping table compiled by the Bluetooth SIG: https://specificationrefs.bluetooth.com/language-mapping/Appropriate_Language_Mapping_Table.pdf Specifically, these terms are replaced: blacklist -> reject list whitelist -> accept list Signed-off-by: Archie Pusaka Reviewed-by: Miao-chen Chou Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 16 ++++++++-------- include/net/bluetooth/hci_core.h | 8 ++++---- 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 3abd6273a189..2dc947341502 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -1505,7 +1505,7 @@ struct hci_cp_le_set_scan_enable { } __packed; #define HCI_LE_USE_PEER_ADDR 0x00 -#define HCI_LE_USE_WHITELIST 0x01 +#define HCI_LE_USE_ACCEPT_LIST 0x01 #define HCI_OP_LE_CREATE_CONN 0x200d struct hci_cp_le_create_conn { @@ -1525,22 +1525,22 @@ struct hci_cp_le_create_conn { #define HCI_OP_LE_CREATE_CONN_CANCEL 0x200e -#define HCI_OP_LE_READ_WHITE_LIST_SIZE 0x200f -struct hci_rp_le_read_white_list_size { +#define HCI_OP_LE_READ_ACCEPT_LIST_SIZE 0x200f +struct hci_rp_le_read_accept_list_size { __u8 status; __u8 size; } __packed; -#define HCI_OP_LE_CLEAR_WHITE_LIST 0x2010 +#define HCI_OP_LE_CLEAR_ACCEPT_LIST 0x2010 -#define HCI_OP_LE_ADD_TO_WHITE_LIST 0x2011 -struct hci_cp_le_add_to_white_list { +#define HCI_OP_LE_ADD_TO_ACCEPT_LIST 0x2011 +struct hci_cp_le_add_to_accept_list { __u8 bdaddr_type; bdaddr_t bdaddr; } __packed; -#define HCI_OP_LE_DEL_FROM_WHITE_LIST 0x2012 -struct hci_cp_le_del_from_white_list { +#define HCI_OP_LE_DEL_FROM_ACCEPT_LIST 0x2012 +struct hci_cp_le_del_from_accept_list { __u8 bdaddr_type; bdaddr_t bdaddr; } __packed; diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index fe5f3a9d9924..212f46806ce7 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -327,7 +327,7 @@ struct hci_dev { __u8 max_page; __u8 features[HCI_MAX_PAGES][8]; __u8 le_features[8]; - __u8 le_white_list_size; + __u8 le_accept_list_size; __u8 le_resolv_list_size; __u8 le_num_of_adv_sets; __u8 le_states[8]; @@ -522,14 +522,14 @@ struct hci_dev { struct hci_conn_hash conn_hash; struct list_head mgmt_pending; - struct list_head blacklist; - struct list_head whitelist; + struct list_head reject_list; + struct list_head accept_list; struct list_head uuids; struct list_head link_keys; struct list_head long_term_keys; struct list_head identity_resolving_keys; struct list_head remote_oob_data; - struct list_head le_white_list; + struct list_head le_accept_list; struct list_head le_resolv_list; struct list_head le_conn_params; struct list_head pend_le_conns; -- cgit v1.2.3 From c9ed0a7077306f9d41d74fb006ab5dbada8349c5 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 9 Jun 2021 11:09:27 -0700 Subject: Bluetooth: Fix Set Extended (Scan Response) Data These command do have variable length and the length can go up to 251, so this changes the struct to not use a fixed size and then when creating the PDU only the actual length of the data send to the controller. Fixes: a0fb3726ba551 ("Bluetooth: Use Set ext adv/scan rsp data if controller supports") Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 6 ++++-- include/net/bluetooth/hci_core.h | 8 ++++---- 2 files changed, 8 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 2dc947341502..b80415011dcd 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -1775,13 +1775,15 @@ struct hci_cp_ext_adv_set { __u8 max_events; } __packed; +#define HCI_MAX_EXT_AD_LENGTH 251 + #define HCI_OP_LE_SET_EXT_ADV_DATA 0x2037 struct hci_cp_le_set_ext_adv_data { __u8 handle; __u8 operation; __u8 frag_pref; __u8 length; - __u8 data[HCI_MAX_AD_LENGTH]; + __u8 data[]; } __packed; #define HCI_OP_LE_SET_EXT_SCAN_RSP_DATA 0x2038 @@ -1790,7 +1792,7 @@ struct hci_cp_le_set_ext_scan_rsp_data { __u8 operation; __u8 frag_pref; __u8 length; - __u8 data[HCI_MAX_AD_LENGTH]; + __u8 data[]; } __packed; #define LE_SET_ADV_DATA_OP_COMPLETE 0x03 diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 212f46806ce7..a53e94459ecd 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -228,9 +228,9 @@ struct adv_info { __u16 remaining_time; __u16 duration; __u16 adv_data_len; - __u8 adv_data[HCI_MAX_AD_LENGTH]; + __u8 adv_data[HCI_MAX_EXT_AD_LENGTH]; __u16 scan_rsp_len; - __u8 scan_rsp_data[HCI_MAX_AD_LENGTH]; + __u8 scan_rsp_data[HCI_MAX_EXT_AD_LENGTH]; __s8 tx_power; __u32 min_interval; __u32 max_interval; @@ -551,9 +551,9 @@ struct hci_dev { DECLARE_BITMAP(dev_flags, __HCI_NUM_FLAGS); __s8 adv_tx_power; - __u8 adv_data[HCI_MAX_AD_LENGTH]; + __u8 adv_data[HCI_MAX_EXT_AD_LENGTH]; __u8 adv_data_len; - __u8 scan_rsp_data[HCI_MAX_AD_LENGTH]; + __u8 scan_rsp_data[HCI_MAX_EXT_AD_LENGTH]; __u8 scan_rsp_data_len; struct list_head adv_instances; -- cgit v1.2.3 From fade56410c22cacafb1be9f911a0afd3701d8366 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Fri, 25 Jun 2021 19:21:39 +0300 Subject: net: lwtunnel: handle MTU calculation in forwading Commit 14972cbd34ff ("net: lwtunnel: Handle fragmentation") moved fragmentation logic away from lwtunnel by carry encap headroom and use it in output MTU calculation. But the forwarding part was not covered and created difference in MTU for output and forwarding and further to silent drops on ipv4 forwarding path. Fix it by taking into account lwtunnel encap headroom. The same commit also introduced difference in how to treat RTAX_MTU in IPv4 and IPv6 where latter explicitly removes lwtunnel encap headroom from route MTU. Make IPv4 version do the same. Fixes: 14972cbd34ff ("net: lwtunnel: Handle fragmentation") Suggested-by: David Ahern Signed-off-by: Vadim Fedorenko Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip.h | 12 ++++++++---- include/net/ip6_route.h | 16 ++++++++++++---- 2 files changed, 20 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/ip.h b/include/net/ip.h index e20874059f82..d9683bef8684 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -31,6 +31,7 @@ #include #include #include +#include #define IPV4_MAX_PMTU 65535U /* RFC 2675, Section 5.1 */ #define IPV4_MIN_MTU 68 /* RFC 791 */ @@ -445,22 +446,25 @@ static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, /* 'forwarding = true' case should always honour route mtu */ mtu = dst_metric_raw(dst, RTAX_MTU); - if (mtu) - return mtu; + if (!mtu) + mtu = min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU); - return min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU); + return mtu - lwtunnel_headroom(dst->lwtstate, mtu); } static inline unsigned int ip_skb_dst_mtu(struct sock *sk, const struct sk_buff *skb) { + unsigned int mtu; + if (!sk || !sk_fullsock(sk) || ip_sk_use_pmtu(sk)) { bool forwarding = IPCB(skb)->flags & IPSKB_FORWARDED; return ip_dst_mtu_maybe_forward(skb_dst(skb), forwarding); } - return min(READ_ONCE(skb_dst(skb)->dev->mtu), IP_MAX_MTU); + mtu = min(READ_ONCE(skb_dst(skb)->dev->mtu), IP_MAX_MTU); + return mtu - lwtunnel_headroom(skb_dst(skb)->lwtstate, mtu); } struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx, diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index f51a118bfce8..f14149df5a65 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -265,11 +265,18 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, static inline int ip6_skb_dst_mtu(struct sk_buff *skb) { + int mtu; + struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? inet6_sk(skb->sk) : NULL; - return (np && np->pmtudisc >= IPV6_PMTUDISC_PROBE) ? - skb_dst(skb)->dev->mtu : dst_mtu(skb_dst(skb)); + if (np && np->pmtudisc >= IPV6_PMTUDISC_PROBE) { + mtu = READ_ONCE(skb_dst(skb)->dev->mtu); + mtu -= lwtunnel_headroom(skb_dst(skb)->lwtstate, mtu); + } else + mtu = dst_mtu(skb_dst(skb)); + + return mtu; } static inline bool ip6_sk_accept_pmtu(const struct sock *sk) @@ -317,7 +324,7 @@ static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) if (dst_metric_locked(dst, RTAX_MTU)) { mtu = dst_metric_raw(dst, RTAX_MTU); if (mtu) - return mtu; + goto out; } mtu = IPV6_MIN_MTU; @@ -327,7 +334,8 @@ static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) mtu = idev->cnf.mtu6; rcu_read_unlock(); - return mtu; +out: + return mtu - lwtunnel_headroom(dst->lwtstate, mtu); } u32 ip6_mtu_from_fib6(const struct fib6_result *res, -- cgit v1.2.3 From 69bfac968a06aab5927160f8736485f85c3e8ee8 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 27 Jun 2021 14:54:24 +0300 Subject: net: switchdev: add a context void pointer to struct switchdev_notifier_info In the case where the driver asks for a replay of a certain type of event (port object or attribute) for a bridge port that is a LAG, it may do so because this port has just joined the LAG. But there might already be other switchdev ports in that LAG, and it is preferable that those preexisting switchdev ports do not act upon the replayed event. The solution is to add a context to switchdev events, which is NULL most of the time (when the bridge layer initiates the call) but which can be set to a value controlled by the switchdev driver when a replay is requested. The driver can then check the context to figure out if all ports within the LAG should act upon the switchdev event, or just the ones that match the context. We have to modify all switchdev_handle_* helper functions as well as the prototypes in the drivers that use these helpers too, because these helpers hide the underlying struct switchdev_notifier_info from us and there is no way to retrieve the context otherwise. The context structure will be populated and used in later patches. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/switchdev.h | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index f1a5a9a3634d..e4cac9218ce1 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -202,6 +202,7 @@ enum switchdev_notifier_type { struct switchdev_notifier_info { struct net_device *dev; struct netlink_ext_ack *extack; + const void *ctx; }; struct switchdev_notifier_fdb_info { @@ -268,19 +269,19 @@ void switchdev_port_fwd_mark_set(struct net_device *dev, int switchdev_handle_port_obj_add(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, bool (*check_cb)(const struct net_device *dev), - int (*add_cb)(struct net_device *dev, + int (*add_cb)(struct net_device *dev, const void *ctx, const struct switchdev_obj *obj, struct netlink_ext_ack *extack)); int switchdev_handle_port_obj_del(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, bool (*check_cb)(const struct net_device *dev), - int (*del_cb)(struct net_device *dev, + int (*del_cb)(struct net_device *dev, const void *ctx, const struct switchdev_obj *obj)); int switchdev_handle_port_attr_set(struct net_device *dev, struct switchdev_notifier_port_attr_info *port_attr_info, bool (*check_cb)(const struct net_device *dev), - int (*set_cb)(struct net_device *dev, + int (*set_cb)(struct net_device *dev, const void *ctx, const struct switchdev_attr *attr, struct netlink_ext_ack *extack)); #else @@ -352,7 +353,7 @@ static inline int switchdev_handle_port_obj_add(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, bool (*check_cb)(const struct net_device *dev), - int (*add_cb)(struct net_device *dev, + int (*add_cb)(struct net_device *dev, const void *ctx, const struct switchdev_obj *obj, struct netlink_ext_ack *extack)) { @@ -363,7 +364,7 @@ static inline int switchdev_handle_port_obj_del(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, bool (*check_cb)(const struct net_device *dev), - int (*del_cb)(struct net_device *dev, + int (*del_cb)(struct net_device *dev, const void *ctx, const struct switchdev_obj *obj)) { return 0; @@ -373,7 +374,7 @@ static inline int switchdev_handle_port_attr_set(struct net_device *dev, struct switchdev_notifier_port_attr_info *port_attr_info, bool (*check_cb)(const struct net_device *dev), - int (*set_cb)(struct net_device *dev, + int (*set_cb)(struct net_device *dev, const void *ctx, const struct switchdev_attr *attr, struct netlink_ext_ack *extack)) { -- cgit v1.2.3 From 9ea3e52c5bc8bb4a084938dc1e3160643438927a Mon Sep 17 00:00:00 2001 From: gushengxian Date: Sat, 26 Jun 2021 04:56:06 -0700 Subject: flow_offload: action should not be NULL when it is referenced "action" should not be NULL when it is referenced. Signed-off-by: gushengxian <13145886936@163.com> Signed-off-by: gushengxian Signed-off-by: David S. Miller --- include/net/flow_offload.h | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h index dc5c1e69cd9f..69c9eabf8325 100644 --- a/include/net/flow_offload.h +++ b/include/net/flow_offload.h @@ -319,12 +319,14 @@ flow_action_mixed_hw_stats_check(const struct flow_action *action, if (flow_offload_has_one_action(action)) return true; - flow_action_for_each(i, action_entry, action) { - if (i && action_entry->hw_stats != last_hw_stats) { - NL_SET_ERR_MSG_MOD(extack, "Mixing HW stats types for actions is not supported"); - return false; + if (action) { + flow_action_for_each(i, action_entry, action) { + if (i && action_entry->hw_stats != last_hw_stats) { + NL_SET_ERR_MSG_MOD(extack, "Mixing HW stats types for actions is not supported"); + return false; + } + last_hw_stats = action_entry->hw_stats; } - last_hw_stats = action_entry->hw_stats; } return true; } -- cgit v1.2.3 From 1fd07f33c3ea2b4aa77426f13e8cb91d4f55af8f Mon Sep 17 00:00:00 2001 From: Andreas Roeseler Date: Sat, 26 Jun 2021 09:07:46 -0500 Subject: ipv6: ICMPV6: add response to ICMPV6 RFC 8335 PROBE messages This patch builds off of commit 2b246b2569cd2ac6ff700d0dce56b8bae29b1842 and adds functionality to respond to ICMPV6 PROBE requests. Add icmp_build_probe function to construct PROBE requests for both ICMPV4 and ICMPV6. Modify icmpv6_rcv to detect ICMPV6 PROBE messages and call the icmpv6_echo_reply handler. Modify icmpv6_echo_reply to build a PROBE response message based on the queried interface. This patch has been tested using a branch of the iputils git repo which can be found here: https://github.com/Juniper-Clinic-2020/iputils/tree/probe-request Signed-off-by: Andreas Roeseler Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- include/net/icmp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/icmp.h b/include/net/icmp.h index fd84adc47963..caddf4a59ad1 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -57,5 +57,6 @@ int icmp_rcv(struct sk_buff *skb); int icmp_err(struct sk_buff *skb, u32 info); int icmp_init(void); void icmp_out_count(struct net *net, unsigned char type); +bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr); #endif /* _ICMP_H */ -- cgit v1.2.3 From 0c5dc070ff3d6246d22ddd931f23a6266249e3db Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Mon, 28 Jun 2021 16:13:41 -0300 Subject: sctp: validate from_addr_param return Ilja reported that, simply putting it, nothing was validating that from_addr_param functions were operating on initialized memory. That is, the parameter itself was being validated by sctp_walk_params, but it doesn't check for types and their specific sizes and it could be a 0-length one, causing from_addr_param to potentially work over the next parameter or even uninitialized memory. The fix here is to, in all calls to from_addr_param, check if enough space is there for the wanted IP address type. Reported-by: Ilja Van Sprundel Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 1aa585216f34..d49593c72a55 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -461,7 +461,7 @@ struct sctp_af { int saddr); void (*from_sk) (union sctp_addr *, struct sock *sk); - void (*from_addr_param) (union sctp_addr *, + bool (*from_addr_param) (union sctp_addr *, union sctp_addr_param *, __be16 port, int iif); int (*to_addr_param) (const union sctp_addr *, -- cgit v1.2.3 From 63609c8fac40810b0b14c9512d47b11965cea37f Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 29 Jun 2021 17:06:48 +0300 Subject: net: dsa: introduce dsa_is_upstream_port and dsa_switch_is_upstream_of In preparation for the new cross-chip notifiers for host addresses, let's introduce some more topology helpers which we are going to use to discern switches that are in our path towards the dedicated CPU port from switches that aren't. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index ea47783d5695..5f632cfd33c7 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -491,6 +491,32 @@ static inline unsigned int dsa_upstream_port(struct dsa_switch *ds, int port) return dsa_towards_port(ds, cpu_dp->ds->index, cpu_dp->index); } +/* Return true if this is the local port used to reach the CPU port */ +static inline bool dsa_is_upstream_port(struct dsa_switch *ds, int port) +{ + if (dsa_is_unused_port(ds, port)) + return false; + + return port == dsa_upstream_port(ds, port); +} + +/* Return true if @upstream_ds is an upstream switch of @downstream_ds, meaning + * that the routing port from @downstream_ds to @upstream_ds is also the port + * which @downstream_ds uses to reach its dedicated CPU. + */ +static inline bool dsa_switch_is_upstream_of(struct dsa_switch *upstream_ds, + struct dsa_switch *downstream_ds) +{ + int routing_port; + + if (upstream_ds == downstream_ds) + return true; + + routing_port = dsa_routing_port(downstream_ds, upstream_ds->index); + + return dsa_is_upstream_port(downstream_ds, routing_port); +} + static inline bool dsa_port_is_vlan_filtering(const struct dsa_port *dp) { const struct dsa_switch *ds = dp->ds; -- cgit v1.2.3 From 161ca59d39e909d37eeeaf14bc1165b114790d00 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 29 Jun 2021 17:06:50 +0300 Subject: net: dsa: reference count the MDB entries at the cross-chip notifier level Ever since the cross-chip notifiers were introduced, the design was meant to be simplistic and just get the job done without worrying too much about dangling resources left behind. For example, somebody installs an MDB entry on sw0p0 in this daisy chain topology. It gets installed using ds->ops->port_mdb_add() on sw0p0, sw1p4 and sw2p4. | sw0p0 sw0p1 sw0p2 sw0p3 sw0p4 [ user ] [ user ] [ user ] [ dsa ] [ cpu ] [ x ] [ ] [ ] [ ] [ ] | +---------+ | sw1p0 sw1p1 sw1p2 sw1p3 sw1p4 [ user ] [ user ] [ user ] [ dsa ] [ dsa ] [ ] [ ] [ ] [ ] [ x ] | +---------+ | sw2p0 sw2p1 sw2p2 sw2p3 sw2p4 [ user ] [ user ] [ user ] [ user ] [ dsa ] [ ] [ ] [ ] [ ] [ x ] Then the same person deletes that MDB entry. The cross-chip notifier for deletion only matches sw0p0: | sw0p0 sw0p1 sw0p2 sw0p3 sw0p4 [ user ] [ user ] [ user ] [ dsa ] [ cpu ] [ x ] [ ] [ ] [ ] [ ] | +---------+ | sw1p0 sw1p1 sw1p2 sw1p3 sw1p4 [ user ] [ user ] [ user ] [ dsa ] [ dsa ] [ ] [ ] [ ] [ ] [ ] | +---------+ | sw2p0 sw2p1 sw2p2 sw2p3 sw2p4 [ user ] [ user ] [ user ] [ user ] [ dsa ] [ ] [ ] [ ] [ ] [ ] Why? Because the DSA links are 'trunk' ports, if we just go ahead and delete the MDB from sw1p4 and sw2p4 directly, we might delete those multicast entries when they are still needed. Just consider the fact that somebody does: - add a multicast MAC address towards sw0p0 [ via the cross-chip notifiers it gets installed on the DSA links too ] - add the same multicast MAC address towards sw0p1 (another port of that same switch) - delete the same multicast MAC address from sw0p0. At this point, if we deleted the MAC address from the DSA links, it would be flooded, even though there is still an entry on switch 0 which needs it not to. So that is why deletions only match the targeted source port and nothing on DSA links. Of course, dangling resources means that the hardware tables will eventually run out given enough additions/removals, but hey, at least it's simple. But there is a bigger concern which needs to be addressed, and that is our support for SWITCHDEV_OBJ_ID_HOST_MDB. DSA simply translates such an object into a dsa_port_host_mdb_add() which ends up as ds->ops->port_mdb_add() on the upstream port, and a similar thing happens on deletion: dsa_port_host_mdb_del() will trigger ds->ops->port_mdb_del() on the upstream port. When there are 2 VLAN-unaware bridges spanning the same switch (which is a use case DSA proudly supports), each bridge will install its own SWITCHDEV_OBJ_ID_HOST_MDB entries. But upon deletion, DSA goes ahead and emits a DSA_NOTIFIER_MDB_DEL for dp->cpu_dp, which is shared between the user ports enslaved to br0 and the user ports enslaved to br1. Not good. The host-trapped multicast addresses installed by br1 will be deleted when any state changes in br0 (IGMP timers expire, or ports leave, etc). To avoid this, we could of course go the route of the zero-sum game and delete the DSA_NOTIFIER_MDB_DEL call for dp->cpu_dp. But the better design is to just admit that on shared ports like DSA links and CPU ports, we should be reference counting calls, even if this consumes some dynamic memory which DSA has traditionally avoided. On the flip side, the hardware tables of switches are limited in size, so it would be good if the OS managed them properly instead of having them eventually overflow. To address the memory usage concern, we only apply the refcounting of MDB entries on ports that are really shared (CPU ports and DSA links) and not on user ports. In a typical single-switch setup, this means only the CPU port (and the host MDB entries are not that many, really). The name of the newly introduced data structures (dsa_mac_addr) is chosen in such a way that will be reusable for host FDB entries (next patch). With this change, we can finally have the same matching logic for the MDB additions and deletions, as well as for their host-trapped variants. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 5f632cfd33c7..2c50546f9667 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -285,6 +285,11 @@ struct dsa_port { */ const struct dsa_netdevice_ops *netdev_ops; + /* List of MAC addresses that must be forwarded on this port. + * These are only valid on CPU ports and DSA links. + */ + struct list_head mdbs; + bool setup; }; @@ -299,6 +304,13 @@ struct dsa_link { struct list_head list; }; +struct dsa_mac_addr { + unsigned char addr[ETH_ALEN]; + u16 vid; + refcount_t refcount; + struct list_head list; +}; + struct dsa_switch { bool setup; -- cgit v1.2.3 From 3f6e32f92a027e91f001070ec324dd3b534d948c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 29 Jun 2021 17:06:52 +0300 Subject: net: dsa: reference count the FDB addresses at the cross-chip notifier level The same concerns expressed for host MDB entries are valid for host FDBs just as well: - in the case of multiple bridges spanning the same switch chip, deleting a host FDB entry that belongs to one bridge will result in breakage to the other bridge - not deleting FDB entries across DSA links means that the switch's hardware tables will eventually run out, given enough wear&tear So do the same thing and introduce reference counting for CPU ports and DSA links using the same data structures as we have for MDB entries. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 2c50546f9667..33f40c1ec379 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -288,6 +288,7 @@ struct dsa_port { /* List of MAC addresses that must be forwarded on this port. * These are only valid on CPU ports and DSA links. */ + struct list_head fdbs; struct list_head mdbs; bool setup; -- cgit v1.2.3 From e3ae2365efc14269170a6326477e669332271ab3 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Sun, 27 Jun 2021 18:48:21 -0400 Subject: net: sock: introduce sk_error_report This patch introduces a function wrapper to call the sk_error_report callback. That will prepare to add additional handling whenever sk_error_report is called, for example to trace socket errors. Signed-off-by: Alexander Aring Signed-off-by: David S. Miller --- include/net/sock.h | 2 ++ include/net/tls.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index ced2fc965ec7..8bdd80027ffb 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2281,6 +2281,8 @@ static inline int sock_error(struct sock *sk) return -err; } +void sk_error_report(struct sock *sk); + static inline unsigned long sock_wspace(struct sock *sk) { int amt = 0; diff --git a/include/net/tls.h b/include/net/tls.h index 8d398a5de3ee..be4b3e1cac46 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -469,7 +469,7 @@ static inline bool tls_is_sk_tx_device_offloaded(struct sock *sk) static inline void tls_err_abort(struct sock *sk, int err) { sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); } static inline bool tls_bigint_increment(unsigned char *seq, int len) -- cgit v1.2.3 From 3f8ad50a9e43b6a59070e6c9c5eec79626f81095 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 29 Jun 2021 06:53:14 -0700 Subject: tcp: change ICSK_CA_PRIV_SIZE definition Instead of a magic number (13 currently) and having to change it every other year, use sizeof_field() macro. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/inet_connection_sock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 3c8c59471bc1..b06c2d02ec84 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -135,7 +135,7 @@ struct inet_connection_sock { u32 icsk_user_timeout; u64 icsk_ca_priv[104 / sizeof(u64)]; -#define ICSK_CA_PRIV_SIZE (13 * sizeof(u64)) +#define ICSK_CA_PRIV_SIZE sizeof_field(struct inet_connection_sock, icsk_ca_priv) }; #define ICSK_TIME_RETRANS 1 /* Retransmit timer */ -- cgit v1.2.3 From 0dbffbb5335a1e3aa6855e4ee317e25e669dd302 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 29 Jun 2021 07:12:45 -0700 Subject: net: annotate data race around sk_ll_usec sk_ll_usec is read locklessly from sk_can_busy_loop() while another thread can change its value in sock_setsockopt() This is correct but needs annotations. BUG: KCSAN: data-race in __skb_try_recv_datagram / sock_setsockopt write to 0xffff88814eb5f904 of 4 bytes by task 14011 on cpu 0: sock_setsockopt+0x1287/0x2090 net/core/sock.c:1175 __sys_setsockopt+0x14f/0x200 net/socket.c:2100 __do_sys_setsockopt net/socket.c:2115 [inline] __se_sys_setsockopt net/socket.c:2112 [inline] __x64_sys_setsockopt+0x62/0x70 net/socket.c:2112 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47 entry_SYSCALL_64_after_hwframe+0x44/0xae read to 0xffff88814eb5f904 of 4 bytes by task 14001 on cpu 1: sk_can_busy_loop include/net/busy_poll.h:41 [inline] __skb_try_recv_datagram+0x14f/0x320 net/core/datagram.c:273 unix_dgram_recvmsg+0x14c/0x870 net/unix/af_unix.c:2101 unix_seqpacket_recvmsg+0x5a/0x70 net/unix/af_unix.c:2067 ____sys_recvmsg+0x15d/0x310 include/linux/uio.h:244 ___sys_recvmsg net/socket.c:2598 [inline] do_recvmmsg+0x35c/0x9f0 net/socket.c:2692 __sys_recvmmsg net/socket.c:2771 [inline] __do_sys_recvmmsg net/socket.c:2794 [inline] __se_sys_recvmmsg net/socket.c:2787 [inline] __x64_sys_recvmmsg+0xcf/0x150 net/socket.c:2787 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0x00000000 -> 0x00000101 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 14001 Comm: syz-executor.3 Not tainted 5.13.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: David S. Miller --- include/net/busy_poll.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h index 73af4a64a599..40296ed976a9 100644 --- a/include/net/busy_poll.h +++ b/include/net/busy_poll.h @@ -38,7 +38,7 @@ static inline bool net_busy_loop_on(void) static inline bool sk_can_busy_loop(const struct sock *sk) { - return sk->sk_ll_usec && !signal_pending(current); + return READ_ONCE(sk->sk_ll_usec) && !signal_pending(current); } bool sk_busy_loop_end(void *p, unsigned long start_time); -- cgit v1.2.3 From 1d11fa231cabeae09a95cb3e4cf1d9dd34e00f08 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 29 Jun 2021 23:34:08 -0400 Subject: sctp: move 198 addresses from unusable to private scope MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The doc draft-stewart-tsvwg-sctp-ipv4-00 that restricts 198 addresses was never published. These addresses as private addresses should be allowed to use in SCTP. As Michael Tuexen suggested, this patch is to move 198 addresses from unusable to private scope. Reported-by: Sérgio Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/constants.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 265fffa33dad..5859e0a16a58 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -360,8 +360,7 @@ enum { #define SCTP_SCOPE_POLICY_MAX SCTP_SCOPE_POLICY_LINK /* Based on IPv4 scoping , - * SCTP IPv4 unusable addresses: 0.0.0.0/8, 224.0.0.0/4, 198.18.0.0/24, - * 192.88.99.0/24. + * SCTP IPv4 unusable addresses: 0.0.0.0/8, 224.0.0.0/4, 192.88.99.0/24. * Also, RFC 8.4, non-unicast addresses are not considered valid SCTP * addresses. */ @@ -369,7 +368,6 @@ enum { ((htonl(INADDR_BROADCAST) == a) || \ ipv4_is_multicast(a) || \ ipv4_is_zeronet(a) || \ - ipv4_is_test_198(a) || \ ipv4_is_anycast_6to4(a)) /* Flags used for the bind address copy functions. */ -- cgit v1.2.3 From d463126e23f112629edb01594141ca437a92a108 Mon Sep 17 00:00:00 2001 From: Yangbo Lu Date: Wed, 30 Jun 2021 16:11:59 +0800 Subject: net: sock: extend SO_TIMESTAMPING for PHC binding Since PTP virtual clock support is added, there can be several PTP virtual clocks based on one PTP physical clock for timestamping. This patch is to extend SO_TIMESTAMPING API to support PHC (PTP Hardware Clock) binding by adding a new flag SOF_TIMESTAMPING_BIND_PHC. When PTP virtual clocks are in use, user space can configure to bind one for timestamping, but PTP physical clock is not supported and not needed to bind. This patch is preparation for timestamp conversion from raw timestamp to a specific PTP virtual clock time in core net. Signed-off-by: Yangbo Lu Signed-off-by: David S. Miller --- include/net/sock.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 8bdd80027ffb..f23cb259b0e2 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -316,7 +316,9 @@ struct bpf_local_storage; * @sk_timer: sock cleanup timer * @sk_stamp: time stamp of last packet received * @sk_stamp_seq: lock for accessing sk_stamp on 32 bit architectures only - * @sk_tsflags: SO_TIMESTAMPING socket options + * @sk_tsflags: SO_TIMESTAMPING flags + * @sk_bind_phc: SO_TIMESTAMPING bind PHC index of PTP virtual clock + * for timestamping * @sk_tskey: counter to disambiguate concurrent tstamp requests * @sk_zckey: counter to order MSG_ZEROCOPY notifications * @sk_socket: Identd and reporting IO signals @@ -493,6 +495,7 @@ struct sock { seqlock_t sk_stamp_seq; #endif u16 sk_tsflags; + int sk_bind_phc; u8 sk_shutdown; u32 sk_tskey; atomic_t sk_zckey; @@ -2755,7 +2758,8 @@ void sock_def_readable(struct sock *sk); int sock_bindtoindex(struct sock *sk, int ifindex, bool lock_sk); void sock_set_timestamp(struct sock *sk, int optname, bool valbool); -int sock_set_timestamping(struct sock *sk, int optname, int val); +int sock_set_timestamping(struct sock *sk, int optname, + struct so_timestamping timestamping); void sock_enable_timestamps(struct sock *sk); void sock_no_linger(struct sock *sk); -- cgit v1.2.3 From ca75bcf0a83b6cc7f53a593d98ec7121c4839b43 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 1 Jul 2021 10:15:09 +0200 Subject: net: remove the caif_hsi driver The caif_hsi driver relies on a cfhsi_get_ops symbol using symbol_get, but this symbol is not provided anywhere in the kernel tree. Remove this driver given that it is dead code. Signed-off-by: Christoph Hellwig Reviewed-by: Greg Kroah-Hartman Signed-off-by: David S. Miller --- include/net/caif/caif_hsi.h | 200 -------------------------------------------- 1 file changed, 200 deletions(-) delete mode 100644 include/net/caif/caif_hsi.h (limited to 'include/net') diff --git a/include/net/caif/caif_hsi.h b/include/net/caif/caif_hsi.h deleted file mode 100644 index 552cf68d28d2..000000000000 --- a/include/net/caif/caif_hsi.h +++ /dev/null @@ -1,200 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (C) ST-Ericsson AB 2010 - * Author: Daniel Martensson / daniel.martensson@stericsson.com - * Dmitry.Tarnyagin / dmitry.tarnyagin@stericsson.com - */ - -#ifndef CAIF_HSI_H_ -#define CAIF_HSI_H_ - -#include -#include -#include - -/* - * Maximum number of CAIF frames that can reside in the same HSI frame. - */ -#define CFHSI_MAX_PKTS 15 - -/* - * Maximum number of bytes used for the frame that can be embedded in the - * HSI descriptor. - */ -#define CFHSI_MAX_EMB_FRM_SZ 96 - -/* - * Decides if HSI buffers should be prefilled with 0xFF pattern for easier - * debugging. Both TX and RX buffers will be filled before the transfer. - */ -#define CFHSI_DBG_PREFILL 0 - -/* Structure describing a HSI packet descriptor. */ -#pragma pack(1) /* Byte alignment. */ -struct cfhsi_desc { - u8 header; - u8 offset; - u16 cffrm_len[CFHSI_MAX_PKTS]; - u8 emb_frm[CFHSI_MAX_EMB_FRM_SZ]; -}; -#pragma pack() /* Default alignment. */ - -/* Size of the complete HSI packet descriptor. */ -#define CFHSI_DESC_SZ (sizeof(struct cfhsi_desc)) - -/* - * Size of the complete HSI packet descriptor excluding the optional embedded - * CAIF frame. - */ -#define CFHSI_DESC_SHORT_SZ (CFHSI_DESC_SZ - CFHSI_MAX_EMB_FRM_SZ) - -/* - * Maximum bytes transferred in one transfer. - */ -#define CFHSI_MAX_CAIF_FRAME_SZ 4096 - -#define CFHSI_MAX_PAYLOAD_SZ (CFHSI_MAX_PKTS * CFHSI_MAX_CAIF_FRAME_SZ) - -/* Size of the complete HSI TX buffer. */ -#define CFHSI_BUF_SZ_TX (CFHSI_DESC_SZ + CFHSI_MAX_PAYLOAD_SZ) - -/* Size of the complete HSI RX buffer. */ -#define CFHSI_BUF_SZ_RX ((2 * CFHSI_DESC_SZ) + CFHSI_MAX_PAYLOAD_SZ) - -/* Bitmasks for the HSI descriptor. */ -#define CFHSI_PIGGY_DESC (0x01 << 7) - -#define CFHSI_TX_STATE_IDLE 0 -#define CFHSI_TX_STATE_XFER 1 - -#define CFHSI_RX_STATE_DESC 0 -#define CFHSI_RX_STATE_PAYLOAD 1 - -/* Bitmasks for power management. */ -#define CFHSI_WAKE_UP 0 -#define CFHSI_WAKE_UP_ACK 1 -#define CFHSI_WAKE_DOWN_ACK 2 -#define CFHSI_AWAKE 3 -#define CFHSI_WAKELOCK_HELD 4 -#define CFHSI_SHUTDOWN 5 -#define CFHSI_FLUSH_FIFO 6 - -#ifndef CFHSI_INACTIVITY_TOUT -#define CFHSI_INACTIVITY_TOUT (1 * HZ) -#endif /* CFHSI_INACTIVITY_TOUT */ - -#ifndef CFHSI_WAKE_TOUT -#define CFHSI_WAKE_TOUT (3 * HZ) -#endif /* CFHSI_WAKE_TOUT */ - -#ifndef CFHSI_MAX_RX_RETRIES -#define CFHSI_MAX_RX_RETRIES (10 * HZ) -#endif - -/* Structure implemented by the CAIF HSI driver. */ -struct cfhsi_cb_ops { - void (*tx_done_cb) (struct cfhsi_cb_ops *drv); - void (*rx_done_cb) (struct cfhsi_cb_ops *drv); - void (*wake_up_cb) (struct cfhsi_cb_ops *drv); - void (*wake_down_cb) (struct cfhsi_cb_ops *drv); -}; - -/* Structure implemented by HSI device. */ -struct cfhsi_ops { - int (*cfhsi_up) (struct cfhsi_ops *dev); - int (*cfhsi_down) (struct cfhsi_ops *dev); - int (*cfhsi_tx) (u8 *ptr, int len, struct cfhsi_ops *dev); - int (*cfhsi_rx) (u8 *ptr, int len, struct cfhsi_ops *dev); - int (*cfhsi_wake_up) (struct cfhsi_ops *dev); - int (*cfhsi_wake_down) (struct cfhsi_ops *dev); - int (*cfhsi_get_peer_wake) (struct cfhsi_ops *dev, bool *status); - int (*cfhsi_fifo_occupancy) (struct cfhsi_ops *dev, size_t *occupancy); - int (*cfhsi_rx_cancel)(struct cfhsi_ops *dev); - struct cfhsi_cb_ops *cb_ops; -}; - -/* Structure holds status of received CAIF frames processing */ -struct cfhsi_rx_state { - int state; - int nfrms; - int pld_len; - int retries; - bool piggy_desc; -}; - -/* Priority mapping */ -enum { - CFHSI_PRIO_CTL = 0, - CFHSI_PRIO_VI, - CFHSI_PRIO_VO, - CFHSI_PRIO_BEBK, - CFHSI_PRIO_LAST, -}; - -struct cfhsi_config { - u32 inactivity_timeout; - u32 aggregation_timeout; - u32 head_align; - u32 tail_align; - u32 q_high_mark; - u32 q_low_mark; -}; - -/* Structure implemented by CAIF HSI drivers. */ -struct cfhsi { - struct caif_dev_common cfdev; - struct net_device *ndev; - struct platform_device *pdev; - struct sk_buff_head qhead[CFHSI_PRIO_LAST]; - struct cfhsi_cb_ops cb_ops; - struct cfhsi_ops *ops; - int tx_state; - struct cfhsi_rx_state rx_state; - struct cfhsi_config cfg; - int rx_len; - u8 *rx_ptr; - u8 *tx_buf; - u8 *rx_buf; - u8 *rx_flip_buf; - spinlock_t lock; - int flow_off_sent; - struct list_head list; - struct work_struct wake_up_work; - struct work_struct wake_down_work; - struct work_struct out_of_sync_work; - struct workqueue_struct *wq; - wait_queue_head_t wake_up_wait; - wait_queue_head_t wake_down_wait; - wait_queue_head_t flush_fifo_wait; - struct timer_list inactivity_timer; - struct timer_list rx_slowpath_timer; - - /* TX aggregation */ - int aggregation_len; - struct timer_list aggregation_timer; - - unsigned long bits; -}; -extern struct platform_driver cfhsi_driver; - -/** - * enum ifla_caif_hsi - CAIF HSI NetlinkRT parameters. - * @IFLA_CAIF_HSI_INACTIVITY_TOUT: Inactivity timeout before - * taking the HSI wakeline down, in milliseconds. - * When using RT Netlink to create, destroy or configure a CAIF HSI interface, - * enum ifla_caif_hsi is used to specify the configuration attributes. - */ -enum ifla_caif_hsi { - __IFLA_CAIF_HSI_UNSPEC, - __IFLA_CAIF_HSI_INACTIVITY_TOUT, - __IFLA_CAIF_HSI_AGGREGATION_TOUT, - __IFLA_CAIF_HSI_HEAD_ALIGN, - __IFLA_CAIF_HSI_TAIL_ALIGN, - __IFLA_CAIF_HSI_QHIGH_WATERMARK, - __IFLA_CAIF_HSI_QLOW_WATERMARK, - __IFLA_CAIF_HSI_MAX -}; - -struct cfhsi_ops *cfhsi_get_ops(void); - -#endif /* CAIF_HSI_H_ */ -- cgit v1.2.3 From 71158bb1f2d2da61385c58fc1114e1a1c19984ba Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 30 Jun 2021 13:42:13 +0200 Subject: tcp: consistently disable header prediction for mptcp The MPTCP receive path is hooked only into the TCP slow-path. The DSS presence allows plain MPTCP traffic to hit that consistently. Since commit e1ff9e82e2ea ("net: mptcp: improve fallback to TCP"), when an MPTCP socket falls back to TCP, it can hit the TCP receive fast-path, and delay or stop triggering the event notification. Address the issue explicitly disabling the header prediction for MPTCP sockets. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/200 Fixes: e1ff9e82e2ea ("net: mptcp: improve fallback to TCP") Signed-off-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index e668f1bf780d..17df9b047ee4 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -686,6 +686,10 @@ static inline u32 __tcp_set_rto(const struct tcp_sock *tp) static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd) { + /* mptcp hooks are only on the slow path */ + if (sk_is_mptcp((struct sock *)tp)) + return; + tp->pred_flags = htonl((tp->tcp_header_len << 26) | ntohl(TCP_FLAG_ACK) | snd_wnd); -- cgit v1.2.3 From a23f89a9990684a0ca0cac4a2857c15d338ebe2d Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Thu, 1 Jul 2021 08:02:24 +0300 Subject: netfilter: conntrack: nf_ct_gre_keymap_flush() removal nf_ct_gre_keymap_flush() is useless. It is called from nf_conntrack_cleanup_net_list() only and tries to remove nf_ct_gre_keymap entries from pernet gre keymap list. Though: a) at this point the list should already be empty, all its entries were deleted during the conntracks cleanup, because nf_conntrack_cleanup_net_list() executes nf_ct_iterate_cleanup(kill_all) before nf_conntrack_proto_pernet_fini(): nf_conntrack_cleanup_net_list +- nf_ct_iterate_cleanup | nf_ct_put | nf_conntrack_put | nf_conntrack_destroy | destroy_conntrack | destroy_gre_conntrack | nf_ct_gre_keymap_destroy `- nf_conntrack_proto_pernet_fini nf_ct_gre_keymap_flush b) Let's say we find that the keymap list is not empty. This means netns still has a conntrack associated with gre, in which case we should not free its memory, because this will lead to a double free and related crashes. However I doubt it could have gone unnoticed for years, obviously this does not happen in real life. So I think we can remove both nf_ct_gre_keymap_flush() and nf_conntrack_proto_pernet_fini(). Signed-off-by: Vasily Averin Acked-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_core.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 09f2efea0b97..13807ea94cd2 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -30,7 +30,6 @@ void nf_conntrack_cleanup_net(struct net *net); void nf_conntrack_cleanup_net_list(struct list_head *net_exit_list); void nf_conntrack_proto_pernet_init(struct net *net); -void nf_conntrack_proto_pernet_fini(struct net *net); int nf_conntrack_proto_init(void); void nf_conntrack_proto_fini(void); -- cgit v1.2.3 From 2580d3f40022642452dd8422bfb8c22e54cf84bb Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 28 Jun 2021 15:34:28 +0200 Subject: xfrm: Fix RCU vs hash_resize_mutex lock inversion xfrm_bydst_resize() calls synchronize_rcu() while holding hash_resize_mutex. But then on PREEMPT_RT configurations, xfrm_policy_lookup_bytype() may acquire that mutex while running in an RCU read side critical section. This results in a deadlock. In fact the scope of hash_resize_mutex is way beyond the purpose of xfrm_policy_lookup_bytype() to just fetch a coherent and stable policy for a given destination/direction, along with other details. The lower level net->xfrm.xfrm_policy_lock, which among other things protects per destination/direction references to policy entries, is enough to serialize and benefit from priority inheritance against the write side. As a bonus, it makes it officially a per network namespace synchronization business where a policy table resize on namespace A shouldn't block a policy lookup on namespace B. Fixes: 77cc278f7b20 (xfrm: policy: Use sequence counters with associated lock) Cc: stable@vger.kernel.org Cc: Ahmed S. Darwish Cc: Peter Zijlstra (Intel) Cc: Varad Gautam Cc: Steffen Klassert Cc: Herbert Xu Cc: David S. Miller Signed-off-by: Frederic Weisbecker Signed-off-by: Steffen Klassert --- include/net/netns/xfrm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/netns/xfrm.h b/include/net/netns/xfrm.h index e816b6a3ef2b..9b376b87bd54 100644 --- a/include/net/netns/xfrm.h +++ b/include/net/netns/xfrm.h @@ -74,6 +74,7 @@ struct netns_xfrm { #endif spinlock_t xfrm_state_lock; seqcount_spinlock_t xfrm_state_hash_generation; + seqcount_spinlock_t xfrm_policy_hash_generation; spinlock_t xfrm_policy_lock; struct mutex xfrm_cfg_mutex; -- cgit v1.2.3 From 40fc3054b45820c28ea3c65e2c86d041dc244a8a Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Fri, 2 Jul 2021 02:47:00 +0300 Subject: net: ipv6: fix return value of ip6_skb_dst_mtu Commit 628a5c561890 ("[INET]: Add IP(V6)_PMTUDISC_RPOBE") introduced ip6_skb_dst_mtu with return value of signed int which is inconsistent with actually returned values. Also 2 users of this function actually assign its value to unsigned int variable and only __xfrm6_output assigns result of this function to signed variable but actually uses as unsigned in further comparisons and calls. Change this function to return unsigned int value. Fixes: 628a5c561890 ("[INET]: Add IP(V6)_PMTUDISC_RPOBE") Reviewed-by: David Ahern Signed-off-by: Vadim Fedorenko Signed-off-by: David S. Miller --- include/net/ip6_route.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index f14149df5a65..625a38ccb5d9 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -263,7 +263,7 @@ static inline bool ipv6_anycast_destination(const struct dst_entry *dst, int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, int (*output)(struct net *, struct sock *, struct sk_buff *)); -static inline int ip6_skb_dst_mtu(struct sk_buff *skb) +static inline unsigned int ip6_skb_dst_mtu(struct sk_buff *skb) { int mtu; -- cgit v1.2.3 From 1da4cd82dd180224503e745ccf3220e3490d8897 Mon Sep 17 00:00:00 2001 From: Ali Abdallah Date: Thu, 27 May 2021 09:19:06 +0200 Subject: netfilter: conntrack: add new sysctl to disable RST check This patch adds a new sysctl tcp_ignore_invalid_rst to disable marking out of segments RSTs as INVALID. Signed-off-by: Ali Abdallah Acked-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netns/conntrack.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index c3094b83a525..37e5300c7e5a 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -27,6 +27,7 @@ struct nf_tcp_net { u8 tcp_loose; u8 tcp_be_liberal; u8 tcp_max_retrans; + u8 tcp_ignore_invalid_rst; #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) unsigned int offload_timeout; unsigned int offload_pickup; -- cgit v1.2.3 From 9a5605505d9c7dbfdb89cc29a8f5fc5cf9fd2334 Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Mon, 5 Jul 2021 15:38:12 +0000 Subject: bonding: Add struct bond_ipesc to manage SA bonding has been supporting ipsec offload. When SA is added, bonding just passes SA to its own active real interface. But it doesn't manage SA. So, when events(add/del real interface, active real interface change, etc) occur, bonding can't handle that well because It doesn't manage SA. So some problems(panic, UAF, refcnt leak)occur. In order to make it stable, it should manage SA. That's the reason why struct bond_ipsec is added. When a new SA is added to bonding interface, it is stored in the bond_ipsec list. And the SA is passed to a current active real interface. If events occur, it uses bond_ipsec data to handle these events. bond->ipsec_list is protected by bond->ipsec_lock. If a current active real interface is changed, the following logic works. 1. delete all SAs from old active real interface 2. Add all SAs to the new active real interface. 3. If a new active real interface doesn't support ipsec offload or SA's option, it sets real_dev to NULL. Fixes: 18cb261afd7b ("bonding: support hardware encryption offload to slaves") Signed-off-by: Taehee Yoo Signed-off-by: David S. Miller --- include/net/bonding.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bonding.h b/include/net/bonding.h index 15335732e166..625d9c72dee3 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -201,6 +201,11 @@ struct bond_up_slave { */ #define BOND_LINK_NOCHANGE -1 +struct bond_ipsec { + struct list_head list; + struct xfrm_state *xs; +}; + /* * Here are the locking policies for the two bonding locks: * Get rcu_read_lock when reading or RTNL when writing slave list. @@ -249,7 +254,9 @@ struct bonding { #endif /* CONFIG_DEBUG_FS */ struct rtnl_link_stats64 bond_stats; #ifdef CONFIG_XFRM_OFFLOAD - struct xfrm_state *xs; + struct list_head ipsec_list; + /* protecting ipsec_list */ + spinlock_t ipsec_lock; #endif /* CONFIG_XFRM_OFFLOAD */ }; -- cgit v1.2.3 From 7445cf31d2e25e3f8ad7b1c5342e624c09ab23a2 Mon Sep 17 00:00:00 2001 From: Zvi Effron Date: Wed, 7 Jul 2021 22:16:54 +0000 Subject: bpf: Add function for XDP meta data length check This commit prepares to use the XDP meta data length check in multiple places by making it into a static inline function instead of a literal. Co-developed-by: Cody Haas Co-developed-by: Lisa Watanabe Signed-off-by: Cody Haas Signed-off-by: Lisa Watanabe Signed-off-by: Zvi Effron Signed-off-by: Alexei Starovoitov Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20210707221657.3985075-2-zeffron@riotgames.com --- include/net/xdp.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/xdp.h b/include/net/xdp.h index 5533f0ab2afc..ad5b02dcb6f4 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -276,6 +276,11 @@ xdp_data_meta_unsupported(const struct xdp_buff *xdp) return unlikely(xdp->data_meta > xdp->data); } +static inline bool xdp_metalen_invalid(unsigned long metalen) +{ + return (metalen & (sizeof(__u32) - 1)) || (metalen > 32); +} + struct xdp_attachment_info { struct bpf_prog *prog; u32 flags; -- cgit v1.2.3 From 67a9c94317402b826fc3db32afc8f39336803d97 Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Fri, 9 Jul 2021 17:35:18 +0000 Subject: net: validate lwtstate->data before returning from skb_tunnel_info() skb_tunnel_info() returns pointer of lwtstate->data as ip_tunnel_info type without validation. lwtstate->data can have various types such as mpls_iptunnel_encap, etc and these are not compatible. So skb_tunnel_info() should validate before returning that pointer. Splat looks like: BUG: KASAN: slab-out-of-bounds in vxlan_get_route+0x418/0x4b0 [vxlan] Read of size 2 at addr ffff888106ec2698 by task ping/811 CPU: 1 PID: 811 Comm: ping Not tainted 5.13.0+ #1195 Call Trace: dump_stack_lvl+0x56/0x7b print_address_description.constprop.8.cold.13+0x13/0x2ee ? vxlan_get_route+0x418/0x4b0 [vxlan] ? vxlan_get_route+0x418/0x4b0 [vxlan] kasan_report.cold.14+0x83/0xdf ? vxlan_get_route+0x418/0x4b0 [vxlan] vxlan_get_route+0x418/0x4b0 [vxlan] [ ... ] vxlan_xmit_one+0x148b/0x32b0 [vxlan] [ ... ] vxlan_xmit+0x25c5/0x4780 [vxlan] [ ... ] dev_hard_start_xmit+0x1ae/0x6e0 __dev_queue_xmit+0x1f39/0x31a0 [ ... ] neigh_xmit+0x2f9/0x940 mpls_xmit+0x911/0x1600 [mpls_iptunnel] lwtunnel_xmit+0x18f/0x450 ip_finish_output2+0x867/0x2040 [ ... ] Fixes: 61adedf3e3f1 ("route: move lwtunnel state to dst_entry") Signed-off-by: Taehee Yoo Signed-off-by: David S. Miller --- include/net/dst_metadata.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h index 56cb3c38569a..14efa0ded75d 100644 --- a/include/net/dst_metadata.h +++ b/include/net/dst_metadata.h @@ -45,7 +45,9 @@ skb_tunnel_info(const struct sk_buff *skb) return &md_dst->u.tun_info; dst = skb_dst(skb); - if (dst && dst->lwtstate) + if (dst && dst->lwtstate && + (dst->lwtstate->type == LWTUNNEL_ENCAP_IP || + dst->lwtstate->type == LWTUNNEL_ENCAP_IP6)) return lwt_tun_info(dst->lwtstate); return NULL; -- cgit v1.2.3 From 6787b7e350d3552651a3422d3d8980fbc8d65368 Mon Sep 17 00:00:00 2001 From: Jianguo Wu Date: Fri, 9 Jul 2021 17:20:49 -0700 Subject: mptcp: avoid processing packet if a subflow reset If check_fully_established() causes a subflow reset, it should not continue to process the packet in tcp_data_queue(). Add a return value to mptcp_incoming_options(), and return false if a subflow has been reset, else return true. Then drop the packet in tcp_data_queue()/tcp_rcv_state_process() if mptcp_incoming_options() return false. Fixes: d582484726c4 ("mptcp: fix fallback for MP_JOIN subflows") Signed-off-by: Jianguo Wu Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index cb580b06152f..8b5af683a818 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -105,7 +105,7 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size, bool mptcp_established_options(struct sock *sk, struct sk_buff *skb, unsigned int *size, unsigned int remaining, struct mptcp_out_options *opts); -void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb); +bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb); void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp, struct mptcp_out_options *opts); @@ -227,9 +227,10 @@ static inline bool mptcp_established_options(struct sock *sk, return false; } -static inline void mptcp_incoming_options(struct sock *sk, +static inline bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) { + return true; } static inline void mptcp_skb_ext_move(struct sk_buff *to, -- cgit v1.2.3 From c63829182c37c2d6d0608976d15fa61ebebe9e6b Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Sun, 4 Jul 2021 12:02:47 -0700 Subject: af_unix: Implement ->psock_update_sk_prot() Now we can implement unix_bpf_update_proto() to update sk_prot, especially prot->close(). Signed-off-by: Cong Wang Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20210704190252.11866-7-xiyou.wangcong@gmail.com --- include/net/af_unix.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index f42fdddecd41..cca645846af1 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -89,4 +89,14 @@ void unix_sysctl_unregister(struct net *net); static inline int unix_sysctl_register(struct net *net) { return 0; } static inline void unix_sysctl_unregister(struct net *net) {} #endif + +#ifdef CONFIG_BPF_SYSCALL +extern struct proto unix_proto; + +int unix_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); +void __init unix_bpf_build_proto(void); +#else +static inline void __init unix_bpf_build_proto(void) +{} +#endif #endif -- cgit v1.2.3 From 9825d866ce0d11009513e06824885340062c166b Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Sun, 4 Jul 2021 12:02:48 -0700 Subject: af_unix: Implement unix_dgram_bpf_recvmsg() We have to implement unix_dgram_bpf_recvmsg() to replace the original ->recvmsg() to retrieve skmsg from ingress_msg. AF_UNIX is again special here because the lack of sk_prot->recvmsg(). I simply add a special case inside unix_dgram_recvmsg() to call sk->sk_prot->recvmsg() directly. Signed-off-by: Cong Wang Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20210704190252.11866-8-xiyou.wangcong@gmail.com --- include/net/af_unix.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index cca645846af1..435a2c3d5a6f 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -82,6 +82,8 @@ static inline struct unix_sock *unix_sk(const struct sock *sk) long unix_inq_len(struct sock *sk); long unix_outq_len(struct sock *sk); +int __unix_dgram_recvmsg(struct sock *sk, struct msghdr *msg, size_t size, + int flags); #ifdef CONFIG_SYSCTL int unix_sysctl_register(struct net *net); void unix_sysctl_unregister(struct net *net); -- cgit v1.2.3 From 749468760b952e555529ca8a71256b991455101e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 19 Jul 2021 02:20:28 -0700 Subject: net/tcp_fastopen: remove obsolete extern After cited commit, sysctl_tcp_fastopen_blackhole_timeout is no longer a global variable. Fixes: 3733be14a32b ("ipv4: Namespaceify tcp_fastopen_blackhole_timeout knob") Signed-off-by: Eric Dumazet Cc: Haishuang Yan Cc: Yuchung Cheng Cc: Neal Cardwell Acked-by: Wei Wang Link: https://lore.kernel.org/r/20210719092028.3016745-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 17df9b047ee4..784d5c3ef1c5 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1709,7 +1709,6 @@ struct tcp_fastopen_context { struct rcu_head rcu; }; -extern unsigned int sysctl_tcp_fastopen_blackhole_timeout; void tcp_fastopen_active_disable(struct sock *sk); bool tcp_fastopen_active_should_disable(struct sock *sk); void tcp_fastopen_active_disable_ofo_check(struct sock *sk); -- cgit v1.2.3 From e93abb840a2c356ed2809c31fcedb058601ac2e4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 19 Jul 2021 03:11:07 -0700 Subject: net/tcp_fastopen: remove tcp_fastopen_ctx_lock Remove the (per netns) spinlock in favor of xchg() atomic operations. Signed-off-by: Eric Dumazet Cc: Yuchung Cheng Cc: Neal Cardwell Acked-by: Wei Wang Link: https://lore.kernel.org/r/20210719101107.3203943-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski --- include/net/netns/ipv4.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index b8620519eace..2f65701a43c9 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -174,7 +174,6 @@ struct netns_ipv4 { int sysctl_tcp_fastopen; const struct tcp_congestion_ops __rcu *tcp_congestion_control; struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; - spinlock_t tcp_fastopen_ctx_lock; unsigned int sysctl_tcp_fastopen_blackhole_timeout; atomic_t tfo_active_disable_times; unsigned long tfo_active_disable_stamp; -- cgit v1.2.3 From d7b1fd520d5d4271f4ab9b1671afbdcd868039d3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 19 Jul 2021 20:14:48 +0300 Subject: net: dsa: let the core manage the tag_8021q context The basic problem description is as follows: Be there 3 switches in a daisy chain topology: | sw0p0 sw0p1 sw0p2 sw0p3 sw0p4 [ user ] [ user ] [ user ] [ dsa ] [ cpu ] | +---------+ | sw1p0 sw1p1 sw1p2 sw1p3 sw1p4 [ user ] [ user ] [ user ] [ dsa ] [ dsa ] | +---------+ | sw2p0 sw2p1 sw2p2 sw2p3 sw2p4 [ user ] [ user ] [ user ] [ user ] [ dsa ] The CPU will not be able to ping through the user ports of the bottom-most switch (like for example sw2p0), simply because tag_8021q was not coded up for this scenario - it has always assumed DSA switch trees with a single switch. To add support for the topology above, we must admit that the RX VLAN of sw2p0 must be added on some ports of switches 0 and 1 as well. This is in fact a textbook example of thing that can use the cross-chip notifier framework that DSA has set up in switch.c. There is only one problem: core DSA (switch.c) is not able right now to make the connection between a struct dsa_switch *ds and a struct dsa_8021q_context *ctx. Right now, it is drivers who call into tag_8021q.c and always provide a struct dsa_8021q_context *ctx pointer, and tag_8021q.c calls them back with the .tag_8021q_vlan_{add,del} methods. But with cross-chip notifiers, it is possible for tag_8021q to call drivers without drivers having ever asked for anything. A good example is right above: when sw2p0 wants to set itself up for tag_8021q, the .tag_8021q_vlan_add method needs to be called for switches 1 and 0, so that they transport sw2p0's VLANs towards the CPU without dropping them. So instead of letting drivers manage the tag_8021q context, add a tag_8021q_ctx pointer inside of struct dsa_switch, which will be populated when dsa_tag_8021q_register() returns success. The patch is fairly long-winded because we are partly reverting commit 5899ee367ab3 ("net: dsa: tag_8021q: add a context structure") which made the driver-facing tag_8021q API use "ctx" instead of "ds". Now that we can access "ctx" directly from "ds", this is no longer needed. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 33f40c1ec379..e213572f6341 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -352,6 +352,9 @@ struct dsa_switch { unsigned int ageing_time_min; unsigned int ageing_time_max; + /* Storage for drivers using tag_8021q */ + struct dsa_8021q_context *tag_8021q_ctx; + /* devlink used to represent this switch device */ struct devlink *devlink; -- cgit v1.2.3 From 5da11eb407340233a6111c563419e19685a062a4 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 19 Jul 2021 20:14:49 +0300 Subject: net: dsa: make tag_8021q operations part of the core Make tag_8021q a more central element of DSA and move the 2 driver specific operations outside of struct dsa_8021q_context (which is supposed to hold dynamic data and not really constant function pointers). Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index e213572f6341..9e5593885357 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -872,6 +872,13 @@ struct dsa_switch_ops { const struct switchdev_obj_ring_role_mrp *mrp); int (*port_mrp_del_ring_role)(struct dsa_switch *ds, int port, const struct switchdev_obj_ring_role_mrp *mrp); + + /* + * tag_8021q operations + */ + int (*tag_8021q_vlan_add)(struct dsa_switch *ds, int port, u16 vid, + u16 flags); + int (*tag_8021q_vlan_del)(struct dsa_switch *ds, int port, u16 vid); }; #define DSA_DEVLINK_PARAM_DRIVER(_id, _name, _type, _cmodes) \ -- cgit v1.2.3 From c6451cda100d4ebbc3f6819e1161ce0e38ce7746 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 19 Jul 2021 16:51:38 +0300 Subject: net: switchdev: introduce helper for checking dynamically learned FDB entries It is a bit difficult to understand what DSA checks when it tries to avoid installing dynamically learned addresses on foreign interfaces as local host addresses, so create a generic switchdev helper that can be reused and is generally more readable. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/switchdev.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index e4cac9218ce1..745eb25fb8c4 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -238,6 +238,12 @@ switchdev_notifier_info_to_extack(const struct switchdev_notifier_info *info) return info->extack; } +static inline bool +switchdev_fdb_is_dynamically_learned(const struct switchdev_notifier_fdb_info *fdb_info) +{ + return !fdb_info->added_by_user && !fdb_info->is_local; +} + #ifdef CONFIG_NET_SWITCHDEV void switchdev_deferred_process(void); -- cgit v1.2.3 From 8ca07176ab00a6d06a9b254dcbb2514b4d607e9c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 19 Jul 2021 16:51:39 +0300 Subject: net: switchdev: introduce a fanout helper for SWITCHDEV_FDB_{ADD,DEL}_TO_DEVICE Currently DSA has an issue with FDB entries pointing towards the bridge in the presence of br_fdb_replay() being called at port join and leave time. In particular, each bridge port will ask for a replay for the FDB entries pointing towards the bridge when it joins, and for another replay when it leaves. This means that for example, a bridge with 4 switch ports will notify DSA 4 times of the bridge MAC address. But if the MAC address of the bridge changes during the normal runtime of the system, the bridge notifies switchdev [ once ] of the deletion of the old MAC address as a local FDB towards the bridge, and of the insertion [ again once ] of the new MAC address as a local FDB. This is a problem, because DSA keeps the old MAC address as a host FDB entry with refcount 4 (4 ports asked for it using br_fdb_replay). So the old MAC address will not be deleted. Additionally, the new MAC address will only be installed with refcount 1, and when the first switch port leaves the bridge (leaving 3 others as still members), it will delete with it the new MAC address of the bridge from the local FDB entries kept by DSA (because the br_fdb_replay call on deletion will bring the entry's refcount from 1 to 0). So the problem, really, is that the number of br_fdb_replay() calls is not matched with the refcount that a host FDB is offloaded to DSA during normal runtime. An elegant way to solve the problem would be to make the switchdev notification emitted by br_fdb_change_mac_address() result in a host FDB kept by DSA which has a refcount exactly equal to the number of ports under that bridge. Then, no matter how many DSA ports join or leave that bridge, the host FDB entry will always be deleted when there are exactly zero remaining DSA switch ports members of the bridge. To implement the proposed solution, we remember that the switchdev objects and port attributes have some helpers provided by switchdev, which can be optionally called by drivers: switchdev_handle_port_obj_{add,del} and switchdev_handle_port_attr_set. These helpers: - fan out a switchdev object/attribute emitted for the bridge towards all the lower interfaces that pass the check_cb(). - fan out a switchdev object/attribute emitted for a bridge port that is a LAG towards all the lower interfaces that pass the check_cb(). In other words, this is the model we need for the FDB events too: something that will keep an FDB entry emitted towards a physical port as it is, but translate an FDB entry emitted towards the bridge into N FDB entries, one per physical port. Of course, there are many differences between fanning out a switchdev object (VLAN) on 3 lower interfaces of a LAG and fanning out an FDB entry on 3 lower interfaces of a LAG. Intuitively, an FDB entry towards a LAG should be treated specially, because FDB entries are unicast, we can't just install the same address towards 3 destinations. It is imaginable that drivers might want to treat this case specifically, so create some methods for this case and do not recurse into the LAG lower ports, just the bridge ports. DSA also listens for FDB entries on "foreign" interfaces, aka interfaces bridged with us which are not part of our hardware domain: think an Ethernet switch bridged with a Wi-Fi AP. For those addresses, DSA installs host FDB entries. However, there we have the same problem (those host FDB entries are installed with a refcount of only 1) and an even bigger one which we did not have with FDB entries towards the bridge: br_fdb_replay() is currently not called for FDB entries on foreign interfaces, just for the physical port and for the bridge itself. So when DSA sniffs an address learned by the software bridge towards a foreign interface like an e1000 port, and then that e1000 leaves the bridge, DSA remains with the dangling host FDB address. That will be fixed separately by replaying all FDB entries and not just the ones towards the port and the bridge. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/switchdev.h | 56 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 745eb25fb8c4..6f57eb2e89cc 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -272,6 +272,30 @@ void switchdev_port_fwd_mark_set(struct net_device *dev, struct net_device *group_dev, bool joining); +int switchdev_handle_fdb_add_to_device(struct net_device *dev, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*add_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info), + int (*lag_add_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)); + +int switchdev_handle_fdb_del_to_device(struct net_device *dev, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*del_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info), + int (*lag_del_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)); + int switchdev_handle_port_obj_add(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, bool (*check_cb)(const struct net_device *dev), @@ -355,6 +379,38 @@ call_switchdev_blocking_notifiers(unsigned long val, return NOTIFY_DONE; } +static inline int +switchdev_handle_fdb_add_to_device(struct net_device *dev, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*add_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info), + int (*lag_add_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)) +{ + return 0; +} + +static inline int +switchdev_handle_fdb_del_to_device(struct net_device *dev, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*del_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info), + int (*lag_del_cb)(struct net_device *dev, + const struct net_device *orig_dev, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)); +{ + return 0; +} + static inline int switchdev_handle_port_obj_add(struct net_device *dev, struct switchdev_notifier_port_obj_info *port_obj_info, -- cgit v1.2.3 From 2d151d39073aff498358543801fca0f670fea981 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Sun, 18 Jul 2021 09:11:06 +0200 Subject: xfrm: Add possibility to set the default to block if we have no policy As the default we assume the traffic to pass, if we have no matching IPsec policy. With this patch, we have a possibility to change this default from allow to block. It can be configured via netlink. Each direction (input/output/forward) can be configured separately. With the default to block configuered, we need allow policies for all packet flows we accept. We do not use default policy lookup for the loopback device. v1->v2 - fix compiling when XFRM is disabled - Reported-by: kernel test robot Co-developed-by: Christian Langrock Signed-off-by: Christian Langrock Co-developed-by: Antony Antony Signed-off-by: Antony Antony Signed-off-by: Steffen Klassert --- include/net/netns/xfrm.h | 7 +++++++ include/net/xfrm.h | 36 ++++++++++++++++++++++++++++++------ 2 files changed, 37 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/xfrm.h b/include/net/netns/xfrm.h index e946366e8ba5..88c647302977 100644 --- a/include/net/netns/xfrm.h +++ b/include/net/netns/xfrm.h @@ -65,6 +65,13 @@ struct netns_xfrm { u32 sysctl_aevent_rseqth; int sysctl_larval_drop; u32 sysctl_acq_expires; + + u8 policy_default; +#define XFRM_POL_DEFAULT_IN 1 +#define XFRM_POL_DEFAULT_OUT 2 +#define XFRM_POL_DEFAULT_FWD 4 +#define XFRM_POL_DEFAULT_MASK 7 + #ifdef CONFIG_SYSCTL struct ctl_table_header *sysctl_hdr; #endif diff --git a/include/net/xfrm.h b/include/net/xfrm.h index cbff7c2a9724..2308210793a0 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1075,6 +1075,22 @@ xfrm_state_addr_cmp(const struct xfrm_tmpl *tmpl, const struct xfrm_state *x, un } #ifdef CONFIG_XFRM +static inline bool +xfrm_default_allow(struct net *net, int dir) +{ + u8 def = net->xfrm.policy_default; + + switch (dir) { + case XFRM_POLICY_IN: + return def & XFRM_POL_DEFAULT_IN ? false : true; + case XFRM_POLICY_OUT: + return def & XFRM_POL_DEFAULT_OUT ? false : true; + case XFRM_POLICY_FWD: + return def & XFRM_POL_DEFAULT_FWD ? false : true; + } + return false; +} + int __xfrm_policy_check(struct sock *, int dir, struct sk_buff *skb, unsigned short family); @@ -1088,9 +1104,13 @@ static inline int __xfrm_policy_check2(struct sock *sk, int dir, if (sk && sk->sk_policy[XFRM_POLICY_IN]) return __xfrm_policy_check(sk, ndir, skb, family); - return (!net->xfrm.policy_count[dir] && !secpath_exists(skb)) || - (skb_dst(skb) && (skb_dst(skb)->flags & DST_NOPOLICY)) || - __xfrm_policy_check(sk, ndir, skb, family); + if (xfrm_default_allow(net, dir)) + return (!net->xfrm.policy_count[dir] && !secpath_exists(skb)) || + (skb_dst(skb) && (skb_dst(skb)->flags & DST_NOPOLICY)) || + __xfrm_policy_check(sk, ndir, skb, family); + else + return (skb_dst(skb) && (skb_dst(skb)->flags & DST_NOPOLICY)) || + __xfrm_policy_check(sk, ndir, skb, family); } static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, unsigned short family) @@ -1142,9 +1162,13 @@ static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family) { struct net *net = dev_net(skb->dev); - return !net->xfrm.policy_count[XFRM_POLICY_OUT] || - (skb_dst(skb)->flags & DST_NOXFRM) || - __xfrm_route_forward(skb, family); + if (xfrm_default_allow(net, XFRM_POLICY_FWD)) + return !net->xfrm.policy_count[XFRM_POLICY_OUT] || + (skb_dst(skb)->flags & DST_NOXFRM) || + __xfrm_route_forward(skb, family); + else + return (skb_dst(skb)->flags & DST_NOXFRM) || + __xfrm_route_forward(skb, family); } static inline int xfrm4_route_forward(struct sk_buff *skb) -- cgit v1.2.3 From 94111dfc18b8b8cb3c72006e0e7b31c038709ab4 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 20 Jul 2021 20:35:56 +0300 Subject: net: switchdev: remove stray semicolon in switchdev_handle_fdb_del_to_device shim With the semicolon at the end, the compiler sees the shim function as a declaration and not as a definition, and warns: 'switchdev_handle_fdb_del_to_device' declared 'static' but never defined Reported-by: kernel test robot Fixes: 8ca07176ab00 ("net: switchdev: introduce a fanout helper for SWITCHDEV_FDB_{ADD,DEL}_TO_DEVICE") Signed-off-by: Vladimir Oltean Tested-by: Matthieu Baerts Signed-off-by: David S. Miller --- include/net/switchdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 6f57eb2e89cc..66468ff8cc0a 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -406,7 +406,7 @@ switchdev_handle_fdb_del_to_device(struct net_device *dev, const struct switchdev_notifier_fdb_info *fdb_info), int (*lag_del_cb)(struct net_device *dev, const struct net_device *orig_dev, const void *ctx, - const struct switchdev_notifier_fdb_info *fdb_info)); + const struct switchdev_notifier_fdb_info *fdb_info)) { return 0; } -- cgit v1.2.3 From 9ee11f0fff205b4b3df9750bff5e94f97c71b6a0 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Tue, 20 Jul 2021 21:42:57 +0200 Subject: ipv6: ioam: Data plane support for Pre-allocated Trace Implement support for processing the IOAM Pre-allocated Trace with IPv6, see [1] and [2]. Introduce a new IPv6 Hop-by-Hop TLV option, see IANA [3]. A new per-interface sysctl is introduced. The value is a boolean to accept (=1) or ignore (=0, by default) IPv6 IOAM options on ingress for an interface: - net.ipv6.conf.XXX.ioam6_enabled Two other sysctls are introduced to define IOAM IDs, represented by an integer. They are respectively per-namespace and per-interface: - net.ipv6.ioam6_id - net.ipv6.conf.XXX.ioam6_id The value of the first one represents the IOAM ID of the node itself (u32; max and default value = U32_MAX>>8, due to hop limit concatenation) while the other represents the IOAM ID of an interface (u16; max and default value = U16_MAX). Each "ioam6_id" sysctl has a "_wide" equivalent: - net.ipv6.ioam6_id_wide - net.ipv6.conf.XXX.ioam6_id_wide The value of the first one represents the wide IOAM ID of the node itself (u64; max and default value = U64_MAX>>8, due to hop limit concatenation) while the other represents the wide IOAM ID of an interface (u32; max and default value = U32_MAX). The use of short and wide equivalents is not exclusive, a deployment could choose to leverage both. For example, net.ipv6.conf.XXX.ioam6_id (short format) could be an identifier for a physical interface, whereas net.ipv6.conf.XXX.ioam6_id_wide (wide format) could be an identifier for a logical sub-interface. Documentation about new sysctls is provided at the end of this patchset. Two relativistic hash tables are used: one for IOAM namespaces, the other for IOAM schemas. A namespace can only have a single active schema and a schema can only be attached to a single namespace (1:1 relationship). [1] https://tools.ietf.org/html/draft-ietf-ippm-ioam-ipv6-options [2] https://tools.ietf.org/html/draft-ietf-ippm-ioam-data [3] https://www.iana.org/assignments/ipv6-parameters/ipv6-parameters.xhtml#ipv6-parameters-2 Signed-off-by: Justin Iurman Signed-off-by: David S. Miller --- include/net/ioam6.h | 64 ++++++++++++++++++++++++++++++++++++++++++++++++ include/net/netns/ipv6.h | 3 +++ 2 files changed, 67 insertions(+) create mode 100644 include/net/ioam6.h (limited to 'include/net') diff --git a/include/net/ioam6.h b/include/net/ioam6.h new file mode 100644 index 000000000000..772b91ee2e87 --- /dev/null +++ b/include/net/ioam6.h @@ -0,0 +1,64 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* + * IPv6 IOAM implementation + * + * Author: + * Justin Iurman + */ + +#ifndef _NET_IOAM6_H +#define _NET_IOAM6_H + +#include +#include +#include +#include + +struct ioam6_namespace { + struct rhash_head head; + struct rcu_head rcu; + + struct ioam6_schema __rcu *schema; + + __be16 id; + __be32 data; + __be64 data_wide; +}; + +struct ioam6_schema { + struct rhash_head head; + struct rcu_head rcu; + + struct ioam6_namespace __rcu *ns; + + u32 id; + int len; + __be32 hdr; + + u8 data[0]; +}; + +struct ioam6_pernet_data { + struct mutex lock; + struct rhashtable namespaces; + struct rhashtable schemas; +}; + +static inline struct ioam6_pernet_data *ioam6_pernet(struct net *net) +{ +#if IS_ENABLED(CONFIG_IPV6) + return net->ipv6.ioam6_data; +#else + return NULL; +#endif +} + +struct ioam6_namespace *ioam6_namespace(struct net *net, __be16 id); +void ioam6_fill_trace_data(struct sk_buff *skb, + struct ioam6_namespace *ns, + struct ioam6_trace_hdr *trace); + +int ioam6_init(void); +void ioam6_exit(void); + +#endif /* _NET_IOAM6_H */ diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index bde0b7adb4a3..a4b550380316 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -51,6 +51,8 @@ struct netns_sysctl_ipv6 { int max_dst_opts_len; int max_hbh_opts_len; int seg6_flowlabel; + u32 ioam6_id; + u64 ioam6_id_wide; bool skip_notify_on_dev_down; u8 fib_notify_on_flag_change; }; @@ -110,6 +112,7 @@ struct netns_ipv6 { spinlock_t lock; u32 seq; } ip6addrlbl_table; + struct ioam6_pernet_data *ioam6_data; }; #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) -- cgit v1.2.3 From 3edede08ff37c6a9370510508d5eeb54890baf47 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Tue, 20 Jul 2021 21:42:59 +0200 Subject: ipv6: ioam: Support for IOAM injection with lwtunnels Add support for the IOAM inline insertion (only for the host-to-host use case) which is per-route configured with lightweight tunnels. The target is iproute2 and the patch is ready. It will be posted as soon as this patchset is merged. Here is an overview: $ ip -6 ro ad fc00::1/128 encap ioam6 trace type 0x800000 ns 1 size 12 dev eth0 This example configures an IOAM Pre-allocated Trace option attached to the fc00::1/128 prefix. The IOAM namespace (ns) is 1, the size of the pre-allocated trace data block is 12 octets (size) and only the first IOAM data (bit 0: hop_limit + node id) is included in the trace (type) represented as a bitfield. The reason why the in-transit (IPv6-in-IPv6 encapsulation) use case is not implemented is explained on the patchset cover. Signed-off-by: Justin Iurman Signed-off-by: David S. Miller --- include/net/ioam6.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/ioam6.h b/include/net/ioam6.h index 772b91ee2e87..3c2993bc48c8 100644 --- a/include/net/ioam6.h +++ b/include/net/ioam6.h @@ -61,4 +61,7 @@ void ioam6_fill_trace_data(struct sk_buff *skb, int ioam6_init(void); void ioam6_exit(void); +int ioam6_iptunnel_init(void); +void ioam6_iptunnel_exit(void); + #endif /* _NET_IOAM6_H */ -- cgit v1.2.3 From 427faee167bce4ccb0b22f6815a9ac509e31d4c3 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Tue, 20 Jul 2021 23:06:27 +0300 Subject: net: ipv6: introduce ip6_dst_mtu_maybe_forward Replace ip6_dst_mtu_forward with ip6_dst_mtu_maybe_forward and reuse this code in ip6_mtu. Actually these two functions were almost duplicates, this change will simplify the maintaince of mtu calculation code. Signed-off-by: Vadim Fedorenko Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip6_route.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 625a38ccb5d9..820eae3ea95f 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -316,12 +316,13 @@ static inline bool rt6_duplicate_nexthop(struct fib6_info *a, struct fib6_info * !lwtunnel_cmp_encap(nha->fib_nh_lws, nhb->fib_nh_lws); } -static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) +static inline unsigned int ip6_dst_mtu_maybe_forward(const struct dst_entry *dst, + bool forwarding) { struct inet6_dev *idev; unsigned int mtu; - if (dst_metric_locked(dst, RTAX_MTU)) { + if (!forwarding || dst_metric_locked(dst, RTAX_MTU)) { mtu = dst_metric_raw(dst, RTAX_MTU); if (mtu) goto out; -- cgit v1.2.3 From ac6627a28dbfb5d96736544a00c3938fa7ea6dfb Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Tue, 20 Jul 2021 23:06:28 +0300 Subject: net: ipv4: Consolidate ipv4_mtu and ip_dst_mtu_maybe_forward Consolidate IPv4 MTU code the same way it is done in IPv6 to have code aligned in both address families Signed-off-by: Vadim Fedorenko Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip.h | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/ip.h b/include/net/ip.h index d9683bef8684..9192444f2964 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -436,18 +436,32 @@ static inline bool ip_sk_ignore_df(const struct sock *sk) static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, bool forwarding) { + const struct rtable *rt = container_of(dst, struct rtable, dst); struct net *net = dev_net(dst->dev); unsigned int mtu; if (net->ipv4.sysctl_ip_fwd_use_pmtu || ip_mtu_locked(dst) || - !forwarding) - return dst_mtu(dst); + !forwarding) { + mtu = rt->rt_pmtu; + if (mtu && time_before(jiffies, rt->dst.expires)) + goto out; + } /* 'forwarding = true' case should always honour route mtu */ mtu = dst_metric_raw(dst, RTAX_MTU); - if (!mtu) - mtu = min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU); + if (mtu) + goto out; + + mtu = READ_ONCE(dst->dev->mtu); + + if (unlikely(ip_mtu_locked(dst))) { + if (rt->rt_uses_gateway && mtu > 576) + mtu = 576; + } + +out: + mtu = min_t(unsigned int, mtu, IP_MAX_MTU); return mtu - lwtunnel_headroom(dst->lwtstate, mtu); } -- cgit v1.2.3 From 5b22d3669f2fa6e762c5302fc4b6051a92b81617 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 22 Jul 2021 18:55:39 +0300 Subject: net: dsa: track the number of switches in a tree In preparation of supporting data plane forwarding on behalf of a software bridge, some drivers might need to view bridges as virtual switches behind the CPU port in a cross-chip topology. Give them some help and let them know how many physical switches there are in the tree, so that they can count the virtual switches starting from that number on. Note that the first dsa_switch_ops method where this information is reliably available is .setup(). This is because of how DSA works: in a tree with 3 switches, each calling dsa_register_switch(), the first 2 will advance until dsa_tree_setup() -> dsa_tree_setup_routing_table() and exit with error code 0 because the topology is not complete. Since probing is parallel at this point, one switch does not know about the existence of the other. Then the third switch comes, and for it, dsa_tree_setup_routing_table() returns complete = true. This switch goes ahead and calls dsa_tree_setup_switches() for everybody else, calling their .setup() methods too. This acts as the synchronization point. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 9e5593885357..929bcaec4d41 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -159,6 +159,9 @@ struct dsa_switch_tree { */ struct net_device **lags; unsigned int lags_len; + + /* Track the largest switch index within a tree */ + unsigned int last_switch; }; #define dsa_lags_foreach_id(_id, _dst) \ -- cgit v1.2.3 From 123abc06e74f49d9b173a93cb2b797fb85f50ba3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 22 Jul 2021 18:55:40 +0300 Subject: net: dsa: add support for bridge TX forwarding offload For a DSA switch, to offload the forwarding process of a bridge device means to send the packets coming from the software bridge as data plane packets. This is contrary to everything that DSA has done so far, because the current taggers only know to send control packets (ones that target a specific destination port), whereas data plane packets are supposed to be forwarded according to the FDB lookup, much like packets ingressing on any regular ingress port. If the FDB lookup process returns multiple destination ports (flooding, multicast), then replication is also handled by the switch hardware - the bridge only sends a single packet and avoids the skb_clone(). DSA keeps for each bridge port a zero-based index (the number of the bridge). Multiple ports performing TX forwarding offload to the same bridge have the same dp->bridge_num value, and ports not offloading the TX data plane of a bridge have dp->bridge_num = -1. The tagger can check if the packet that is being transmitted on has skb->offload_fwd_mark = true or not. If it does, it can be sure that the packet belongs to the data plane of a bridge, further information about which can be obtained based on dp->bridge_dev and dp->bridge_num. It can then compose a DSA tag for injecting a data plane packet into that bridge number. For the switch driver side, we offer two new dsa_switch_ops methods, called .port_bridge_fwd_offload_{add,del}, which are modeled after .port_bridge_{join,leave}. These methods are provided in case the driver needs to configure the hardware to treat packets coming from that bridge software interface as data plane packets. The switchdev <-> bridge interaction happens during the netdev_master_upper_dev_link() call, so to switch drivers, the effect is that the .port_bridge_fwd_offload_add() method is called immediately after .port_bridge_join(). If the bridge number exceeds the number of bridges for which the switch driver can offload the TX data plane (and this includes the case where the driver can offload none), DSA falls back to simply returning tx_fwd_offload = false in the switchdev_bridge_port_offload() call. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 929bcaec4d41..f8eb2dc3fbef 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -162,6 +162,9 @@ struct dsa_switch_tree { /* Track the largest switch index within a tree */ unsigned int last_switch; + + /* Track the bridges with forwarding offload enabled */ + unsigned long fwd_offloading_bridges; }; #define dsa_lags_foreach_id(_id, _dst) \ @@ -262,6 +265,7 @@ struct dsa_port { bool vlan_filtering; u8 stp_state; struct net_device *bridge_dev; + int bridge_num; struct devlink_port devlink_port; bool devlink_port_setup; struct phylink *pl; @@ -413,6 +417,12 @@ struct dsa_switch { */ unsigned int num_lag_ids; + /* Drivers that support bridge forwarding offload should set this to + * the maximum number of bridges spanning the same switch tree that can + * be offloaded. + */ + unsigned int num_fwd_offloading_bridges; + size_t num_ports; }; @@ -696,6 +706,14 @@ struct dsa_switch_ops { struct net_device *bridge); void (*port_bridge_leave)(struct dsa_switch *ds, int port, struct net_device *bridge); + /* Called right after .port_bridge_join() */ + int (*port_bridge_tx_fwd_offload)(struct dsa_switch *ds, int port, + struct net_device *bridge, + int bridge_num); + /* Called right before .port_bridge_leave() */ + void (*port_bridge_tx_fwd_unoffload)(struct dsa_switch *ds, int port, + struct net_device *bridge, + int bridge_num); void (*port_stp_state_set)(struct dsa_switch *ds, int port, u8 state); void (*port_fast_age)(struct dsa_switch *ds, int port); -- cgit v1.2.3 From 62001372c2b6cdf2346afb2cf94ed3d950eee64c Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 1 Jul 2021 13:05:54 -0700 Subject: bpf: tcp: seq_file: Remove bpf_seq_afinfo from tcp_iter_state A following patch will create a separate struct to store extra bpf_iter state and it will embed the existing tcp_iter_state like this: struct bpf_tcp_iter_state { struct tcp_iter_state state; /* More bpf_iter specific states here ... */ } As a prep work, this patch removes the "struct tcp_seq_afinfo *bpf_seq_afinfo" where its purpose is to tell if it is iterating from bpf_iter instead of proc fs. Currently, if "*bpf_seq_afinfo" is not NULL, it is iterating from bpf_iter. The kernel should not filter by the addr family and leave this filtering decision to the bpf prog. Instead of adding a "*bpf_seq_afinfo" pointer, this patch uses the "seq->op == &bpf_iter_tcp_seq_ops" test to tell if it is iterating from the bpf iter. The bpf_iter_(init|fini)_tcp() is left here to prepare for the change of a following patch. Signed-off-by: Martin KaFai Lau Signed-off-by: Andrii Nakryiko Reviewed-by: Eric Dumazet Acked-by: Kuniyuki Iwashima Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20210701200554.1034982-1-kafai@fb.com --- include/net/tcp.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 17df9b047ee4..ba3034123e1d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1959,7 +1959,6 @@ struct tcp_iter_state { struct seq_net_private p; enum tcp_seq_states state; struct sock *syn_wait_sk; - struct tcp_seq_afinfo *bpf_seq_afinfo; int bucket, offset, sbucket, num; loff_t last_pos; }; -- cgit v1.2.3 From 05c0b35709c58b83d4dc515d2ac52e9c0f197d03 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 1 Jul 2021 13:06:06 -0700 Subject: tcp: seq_file: Replace listening_hash with lhash2 This patch moves the tcp seq_file iteration on listeners from the port only listening_hash to the port+addr lhash2. When iterating from the bpf iter, the next patch will need to lock the socket such that the bpf iter can call setsockopt (e.g. to change the TCP_CONGESTION). To avoid locking the bucket and then locking the sock, the bpf iter will first batch some sockets from the same bucket and then unlock the bucket. If the bucket size is small (which usually is), it is easier to batch the whole bucket such that it is less likely to miss a setsockopt on a socket due to changes in the bucket. However, the port only listening_hash could have many listeners hashed to a bucket (e.g. many individual VIP(s):443 and also multiple by the number of SO_REUSEPORT). We have seen bucket size in tens of thousands range. Also, the chance of having changes in some popular port buckets (e.g. 443) is also high. The port+addr lhash2 was introduced to solve this large listener bucket issue. Also, the listening_hash usage has already been replaced with lhash2 in the fast path inet[6]_lookup_listener(). This patch follows the same direction on moving to lhash2 and iterates the lhash2 instead of listening_hash. Signed-off-by: Martin KaFai Lau Signed-off-by: Andrii Nakryiko Reviewed-by: Eric Dumazet Acked-by: Kuniyuki Iwashima Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20210701200606.1035783-1-kafai@fb.com --- include/net/inet_hashtables.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/net') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index ca6a3ea9057e..f72ec113ae56 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -160,6 +160,12 @@ struct inet_hashinfo { ____cacheline_aligned_in_smp; }; +#define inet_lhash2_for_each_icsk_continue(__icsk) \ + hlist_for_each_entry_continue(__icsk, icsk_listen_portaddr_node) + +#define inet_lhash2_for_each_icsk(__icsk, list) \ + hlist_for_each_entry(__icsk, list, icsk_listen_portaddr_node) + #define inet_lhash2_for_each_icsk_rcu(__icsk, list) \ hlist_for_each_entry_rcu(__icsk, list, icsk_listen_portaddr_node) -- cgit v1.2.3 From 48d5440393d328cfe18b26e3609c84c0f41d6515 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:47:32 +0200 Subject: nfc: constify payload argument in nci_send_cmd() The nci_send_cmd() payload argument is passed directly to skb_put_data() which already accepts a pointer to const, so make it const as well for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nci_core.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index 1df0f8074c9d..bf573eca07ca 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -360,7 +360,7 @@ int nci_core_rsp_packet(struct nci_dev *ndev, __u16 opcode, int nci_core_ntf_packet(struct nci_dev *ndev, __u16 opcode, struct sk_buff *skb); void nci_rx_data_packet(struct nci_dev *ndev, struct sk_buff *skb); -int nci_send_cmd(struct nci_dev *ndev, __u16 opcode, __u8 plen, void *payload); +int nci_send_cmd(struct nci_dev *ndev, __u16 opcode, __u8 plen, const void *payload); int nci_send_data(struct nci_dev *ndev, __u8 conn_id, struct sk_buff *skb); int nci_conn_max_data_pkt_payload_size(struct nci_dev *ndev, __u8 conn_id); void nci_data_exchange_complete(struct nci_dev *ndev, struct sk_buff *skb, -- cgit v1.2.3 From b9c28286d8f1822a10f702fcefe0b8a1db4e6917 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:47:33 +0200 Subject: nfc: constify nci_ops The struct nci_ops is modified by NFC core in only one case: nci_allocate_device() receives too many proprietary commands (prop_ops) to configure. This is a build time known constrain, so a graceful handling of such case is not necessary. Instead, fail the nci_allocate_device() and add BUILD_BUG_ON() to places which set these. This allows to constify the struct nci_ops (consisting of function pointers) for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nci_core.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index bf573eca07ca..5dae7e2cbc49 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -194,7 +194,7 @@ struct nci_hci_dev { /* NCI Core structures */ struct nci_dev { struct nfc_dev *nfc_dev; - struct nci_ops *ops; + const struct nci_ops *ops; struct nci_hci_dev *hci_dev; int tx_headroom; @@ -267,7 +267,7 @@ struct nci_dev { }; /* ----- NCI Devices ----- */ -struct nci_dev *nci_allocate_device(struct nci_ops *ops, +struct nci_dev *nci_allocate_device(const struct nci_ops *ops, __u32 supported_protocols, int tx_headroom, int tx_tailroom); -- cgit v1.2.3 From cb8caa3c6c04c18027258ab34e09bcf87e726f22 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:47:35 +0200 Subject: nfc: constify nci_driver_ops (prop_ops and core_ops) Neither the core nor the drivers modify the passed pointer to struct nci_driver_ops (consisting of function pointers), so make it a pointer to const for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nci_core.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index 5dae7e2cbc49..e7118e0cc3b1 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -82,10 +82,10 @@ struct nci_ops { void (*hci_cmd_received)(struct nci_dev *ndev, u8 pipe, u8 cmd, struct sk_buff *skb); - struct nci_driver_ops *prop_ops; + const struct nci_driver_ops *prop_ops; size_t n_prop_ops; - struct nci_driver_ops *core_ops; + const struct nci_driver_ops *core_ops; size_t n_core_ops; }; -- cgit v1.2.3 From 15944ad2e5a1cc2ef2c6fa60b04e464dbb4ddb61 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:49:23 +0200 Subject: nfc: constify pointer to nfc_vendor_cmd Neither the core nor the drivers modify the passed pointer to struct nfc_vendor_cmd, so make it a pointer to const for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/hci.h | 2 +- include/net/nfc/nci_core.h | 2 +- include/net/nfc/nfc.h | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/hci.h b/include/net/nfc/hci.h index b35f37a57686..2daec8036be9 100644 --- a/include/net/nfc/hci.h +++ b/include/net/nfc/hci.h @@ -168,7 +168,7 @@ void nfc_hci_set_clientdata(struct nfc_hci_dev *hdev, void *clientdata); void *nfc_hci_get_clientdata(struct nfc_hci_dev *hdev); static inline int nfc_hci_set_vendor_cmds(struct nfc_hci_dev *hdev, - struct nfc_vendor_cmd *cmds, + const struct nfc_vendor_cmd *cmds, int n_cmds) { return nfc_set_vendor_cmds(hdev->ndev, cmds, n_cmds); diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index e7118e0cc3b1..00f2c60971d7 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -343,7 +343,7 @@ static inline void *nci_get_drvdata(struct nci_dev *ndev) } static inline int nci_set_vendor_cmds(struct nci_dev *ndev, - struct nfc_vendor_cmd *cmds, + const struct nfc_vendor_cmd *cmds, int n_cmds) { return nfc_set_vendor_cmds(ndev->nfc_dev, cmds, n_cmds); diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index 2cd3a261bcbc..31672021d071 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -188,7 +188,7 @@ struct nfc_dev { struct rfkill *rfkill; - struct nfc_vendor_cmd *vendor_cmds; + const struct nfc_vendor_cmd *vendor_cmds; int n_vendor_cmds; struct nfc_ops *ops; @@ -297,7 +297,7 @@ void nfc_send_to_raw_sock(struct nfc_dev *dev, struct sk_buff *skb, u8 payload_type, u8 direction); static inline int nfc_set_vendor_cmds(struct nfc_dev *dev, - struct nfc_vendor_cmd *cmds, + const struct nfc_vendor_cmd *cmds, int n_cmds) { if (dev->vendor_cmds || dev->n_vendor_cmds) -- cgit v1.2.3 From f6c802a726aea43a2937763f13560c1fd170fae6 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:49:25 +0200 Subject: nfc: constify nfc_ops Neither the core nor the drivers modify the passed pointer to struct nfc_ops, so make it a pointer to const for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nfc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index 31672021d071..85b698794b14 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -191,14 +191,14 @@ struct nfc_dev { const struct nfc_vendor_cmd *vendor_cmds; int n_vendor_cmds; - struct nfc_ops *ops; + const struct nfc_ops *ops; struct genl_info *cur_cmd_info; }; #define to_nfc_dev(_dev) container_of(_dev, struct nfc_dev, dev) extern struct class nfc_class; -struct nfc_dev *nfc_allocate_device(struct nfc_ops *ops, +struct nfc_dev *nfc_allocate_device(const struct nfc_ops *ops, u32 supported_protocols, int tx_headroom, int tx_tailroom); -- cgit v1.2.3 From 094c45c84d799fb0eb3617632097370843ba3678 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:49:26 +0200 Subject: nfc: constify nfc_hci_ops Neither the core nor the drivers modify the passed pointer to struct nfc_hci_ops, so make it a pointer to const for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/hci.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/hci.h b/include/net/nfc/hci.h index 2daec8036be9..756c11084f65 100644 --- a/include/net/nfc/hci.h +++ b/include/net/nfc/hci.h @@ -118,7 +118,7 @@ struct nfc_hci_dev { struct sk_buff_head msg_rx_queue; - struct nfc_hci_ops *ops; + const struct nfc_hci_ops *ops; struct nfc_llc *llc; @@ -151,7 +151,7 @@ struct nfc_hci_dev { }; /* hci device allocation */ -struct nfc_hci_dev *nfc_hci_allocate_device(struct nfc_hci_ops *ops, +struct nfc_hci_dev *nfc_hci_allocate_device(const struct nfc_hci_ops *ops, struct nfc_hci_init_data *init_data, unsigned long quirks, u32 protocols, -- cgit v1.2.3 From 7186aac9c22de855220ebc54df7af399647bd1fc Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 24 Jul 2021 23:49:28 +0200 Subject: nfc: constify nfc_digital_ops Neither the core nor the drivers modify the passed pointer to struct nfc_digital_ops, so make it a pointer to const for correctness and safety. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/digital.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/digital.h b/include/net/nfc/digital.h index 963db96bcbbb..bb3e8fdc0692 100644 --- a/include/net/nfc/digital.h +++ b/include/net/nfc/digital.h @@ -191,7 +191,7 @@ struct digital_poll_tech { struct nfc_digital_dev { struct nfc_dev *nfc_dev; - struct nfc_digital_ops *ops; + const struct nfc_digital_ops *ops; u32 protocols; @@ -236,7 +236,7 @@ struct nfc_digital_dev { void (*skb_add_crc)(struct sk_buff *skb); }; -struct nfc_digital_dev *nfc_digital_allocate_device(struct nfc_digital_ops *ops, +struct nfc_digital_dev *nfc_digital_allocate_device(const struct nfc_digital_ops *ops, __u32 supported_protocols, __u32 driver_capabilities, int tx_headroom, -- cgit v1.2.3 From 058e6e0ed0eace43401c945082dec1d669b5b231 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 25 Jul 2021 13:42:50 -0400 Subject: sctp: improve the code for pmtu probe send and recv update This patch does 3 things: - make sctp_transport_pl_send() and sctp_transport_pl_recv() return bool type to decide if more probe is needed to send. - pr_debug() only when probe is really needed to send. - count pl.raise_count in sctp_transport_pl_send() instead of sctp_transport_pl_recv(), and it's only incremented for the 1st probe for the same size. These are preparations for the next patch to make probes happen only when there's packet loss in Search Complete state. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 32fc4a309df5..f3d414ed208e 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1024,8 +1024,8 @@ bool sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu); void sctp_transport_immediate_rtx(struct sctp_transport *); void sctp_transport_dst_release(struct sctp_transport *t); void sctp_transport_dst_confirm(struct sctp_transport *t); -void sctp_transport_pl_send(struct sctp_transport *t); -void sctp_transport_pl_recv(struct sctp_transport *t); +bool sctp_transport_pl_send(struct sctp_transport *t); +bool sctp_transport_pl_recv(struct sctp_transport *t); /* This is the structure we use to queue packets as they come into -- cgit v1.2.3 From eacf078cf4c7aa23e9591738511f142cc39b5186 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 25 Jul 2021 13:42:51 -0400 Subject: sctp: send pmtu probe only if packet loss in Search Complete state This patch is to introduce last_rtx_chunks into sctp_transport to detect if there's any packet retransmission/loss happened by checking against asoc's rtx_data_chunks in sctp_transport_pl_send(). If there is, namely, transport->last_rtx_chunks != asoc->rtx_data_chunks, the pmtu probe will be sent out. Otherwise, increment the pl.raise_count and return when it's in Search Complete state. With this patch, if in Search Complete state, which is a long period, it doesn't need to keep probing the current pmtu unless there's data packet loss. This will save quite some traffic. v1->v2: - add the missing Fixes tag. Fixes: 0dac127c0557 ("sctp: do black hole detection in search complete state") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index f3d414ed208e..651bba654d77 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -984,6 +984,7 @@ struct sctp_transport { } cacc; struct { + __u32 last_rtx_chunks; __u16 pmtu; __u16 probe_size; __u16 probe_high; -- cgit v1.2.3 From edac6f6332d96aab59af5f27a195f55cd080f034 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 26 Jul 2021 19:55:36 +0300 Subject: Revert "net: dsa: Allow drivers to filter packets they can decode source port from" This reverts commit cc1939e4b3aaf534fb2f3706820012036825731c. Currently 2 classes of DSA drivers are able to send/receive packets directly through the DSA master: - drivers with DSA_TAG_PROTO_NONE - sja1105 Now that sja1105 has gained the ability to perform traffic termination even under the tricky case (VLAN-aware bridge), and that is much more functional (we can perform VLAN-aware bridging with foreign interfaces), there is no reason to keep this code in the receive path of the network core. So delete it. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 15 --------------- 1 file changed, 15 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index f8eb2dc3fbef..55fcac854058 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -88,11 +88,6 @@ struct dsa_device_ops { struct packet_type *pt); void (*flow_dissect)(const struct sk_buff *skb, __be16 *proto, int *offset); - /* Used to determine which traffic should match the DSA filter in - * eth_type_trans, and which, if any, should bypass it and be processed - * as regular on the master net device. - */ - bool (*filter)(const struct sk_buff *skb, struct net_device *dev); unsigned int needed_headroom; unsigned int needed_tailroom; const char *name; @@ -246,7 +241,6 @@ struct dsa_port { struct dsa_switch_tree *dst; struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt); - bool (*filter)(const struct sk_buff *skb, struct net_device *dev); enum { DSA_PORT_TYPE_UNUSED = 0, @@ -985,15 +979,6 @@ static inline bool netdev_uses_dsa(const struct net_device *dev) return false; } -static inline bool dsa_can_decode(const struct sk_buff *skb, - struct net_device *dev) -{ -#if IS_ENABLED(CONFIG_NET_DSA) - return !dev->dsa_ptr->filter || dev->dsa_ptr->filter(skb, dev); -#endif - return false; -} - /* All DSA tags that push the EtherType to the right (basically all except tail * tags, which don't break dissection) can be treated the same from the * perspective of the flow dissector. -- cgit v1.2.3 From f9b282b36dfa9b6c6d6b3e8816cdf0e4defff482 Mon Sep 17 00:00:00 2001 From: Yajun Deng Date: Tue, 27 Jul 2021 11:41:41 +0800 Subject: net: netlink: add the case when nlh is NULL Add the case when nlh is NULL in nlmsg_report(), so that the caller doesn't need to deal with this case. Signed-off-by: Yajun Deng Signed-off-by: David S. Miller --- include/net/netlink.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netlink.h b/include/net/netlink.h index 1ceec518ab49..7a2a9d3144ba 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -885,7 +885,7 @@ static inline int nlmsg_validate_deprecated(const struct nlmsghdr *nlh, */ static inline int nlmsg_report(const struct nlmsghdr *nlh) { - return !!(nlh->nlmsg_flags & NLM_F_ECHO); + return nlh ? !!(nlh->nlmsg_flags & NLM_F_ECHO) : 0; } /** -- cgit v1.2.3 From c7c9d2102c9c098916ab9e0ab248006107d00d6c Mon Sep 17 00:00:00 2001 From: Pavel Skripkin Date: Sun, 25 Jul 2021 00:11:59 +0300 Subject: net: llc: fix skb_over_panic Syzbot reported skb_over_panic() in llc_pdu_init_as_xid_cmd(). The problem was in wrong LCC header manipulations. Syzbot's reproducer tries to send XID packet. llc_ui_sendmsg() is doing following steps: 1. skb allocation with size = len + header size len is passed from userpace and header size is 3 since addr->sllc_xid is set. 2. skb_reserve() for header_len = 3 3. filling all other space with memcpy_from_msg() Ok, at this moment we have fully loaded skb, only headers needs to be filled. Then code comes to llc_sap_action_send_xid_c(). This function pushes 3 bytes for LLC PDU header and initializes it. Then comes llc_pdu_init_as_xid_cmd(). It initalizes next 3 bytes *AFTER* LLC PDU header and call skb_push(skb, 3). This looks wrong for 2 reasons: 1. Bytes rigth after LLC header are user data, so this function was overwriting payload. 2. skb_push(skb, 3) call can cause skb_over_panic() since all free space was filled in llc_ui_sendmsg(). (This can happen is user passed 686 len: 686 + 14 (eth header) + 3 (LLC header) = 703. SKB_DATA_ALIGN(703) = 704) So, in this patch I added 2 new private constansts: LLC_PDU_TYPE_U_XID and LLC_PDU_LEN_U_XID. LLC_PDU_LEN_U_XID is used to correctly reserve header size to handle LLC + XID case. LLC_PDU_TYPE_U_XID is used by llc_pdu_header_init() function to push 6 bytes instead of 3. And finally I removed skb_push() call from llc_pdu_init_as_xid_cmd(). This changes should not affect other parts of LLC, since after all steps we just transmit buffer. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-and-tested-by: syzbot+5e5a981ad7cc54c4b2b4@syzkaller.appspotmail.com Signed-off-by: Pavel Skripkin Signed-off-by: David S. Miller --- include/net/llc_pdu.h | 31 +++++++++++++++++++++++-------- 1 file changed, 23 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/llc_pdu.h b/include/net/llc_pdu.h index c0f0a13ed818..49aa79c7b278 100644 --- a/include/net/llc_pdu.h +++ b/include/net/llc_pdu.h @@ -15,9 +15,11 @@ #include /* Lengths of frame formats */ -#define LLC_PDU_LEN_I 4 /* header and 2 control bytes */ -#define LLC_PDU_LEN_S 4 -#define LLC_PDU_LEN_U 3 /* header and 1 control byte */ +#define LLC_PDU_LEN_I 4 /* header and 2 control bytes */ +#define LLC_PDU_LEN_S 4 +#define LLC_PDU_LEN_U 3 /* header and 1 control byte */ +/* header and 1 control byte and XID info */ +#define LLC_PDU_LEN_U_XID (LLC_PDU_LEN_U + sizeof(struct llc_xid_info)) /* Known SAP addresses */ #define LLC_GLOBAL_SAP 0xFF #define LLC_NULL_SAP 0x00 /* not network-layer visible */ @@ -50,9 +52,10 @@ #define LLC_PDU_TYPE_U_MASK 0x03 /* 8-bit control field */ #define LLC_PDU_TYPE_MASK 0x03 -#define LLC_PDU_TYPE_I 0 /* first bit */ -#define LLC_PDU_TYPE_S 1 /* first two bits */ -#define LLC_PDU_TYPE_U 3 /* first two bits */ +#define LLC_PDU_TYPE_I 0 /* first bit */ +#define LLC_PDU_TYPE_S 1 /* first two bits */ +#define LLC_PDU_TYPE_U 3 /* first two bits */ +#define LLC_PDU_TYPE_U_XID 4 /* private type for detecting XID commands */ #define LLC_PDU_TYPE_IS_I(pdu) \ ((!(pdu->ctrl_1 & LLC_PDU_TYPE_I_MASK)) ? 1 : 0) @@ -230,9 +233,18 @@ static inline struct llc_pdu_un *llc_pdu_un_hdr(struct sk_buff *skb) static inline void llc_pdu_header_init(struct sk_buff *skb, u8 type, u8 ssap, u8 dsap, u8 cr) { - const int hlen = type == LLC_PDU_TYPE_U ? 3 : 4; + int hlen = 4; /* default value for I and S types */ struct llc_pdu_un *pdu; + switch (type) { + case LLC_PDU_TYPE_U: + hlen = 3; + break; + case LLC_PDU_TYPE_U_XID: + hlen = 6; + break; + } + skb_push(skb, hlen); skb_reset_network_header(skb); pdu = llc_pdu_un_hdr(skb); @@ -374,7 +386,10 @@ static inline void llc_pdu_init_as_xid_cmd(struct sk_buff *skb, xid_info->fmt_id = LLC_XID_FMT_ID; /* 0x81 */ xid_info->type = svcs_supported; xid_info->rw = rx_window << 1; /* size of receive window */ - skb_put(skb, sizeof(struct llc_xid_info)); + + /* no need to push/put since llc_pdu_header_init() has already + * pushed 3 + 3 bytes + */ } /** -- cgit v1.2.3 From 3e7a1c7c561ed8508fbdb98ed5708175bbcf7938 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 27 Jul 2021 15:45:06 +0200 Subject: ip_tunnel: use ndo_siocdevprivate The various ipv4 and ipv6 tunnel drivers each implement a set of 12 SIOCDEVPRIVATE commands for managing tunnels. These all work correctly in compat mode. Move them over to the new .ndo_siocdevprivate operation. Cc: Hideaki YOSHIFUJI Cc: David Ahern Cc: Steffen Klassert Cc: Herbert Xu Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/net/ip_tunnels.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 548b65bd3973..bc3b13ec93c9 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -270,7 +270,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const u8 proto, int tunnel_hlen); int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd); -int ip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd); +int ip_tunnel_siocdevprivate(struct net_device *dev, struct ifreq *ifr, + void __user *data, int cmd); int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict); int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu); -- cgit v1.2.3 From a76053707dbf0dc020a73b4d90cd952409ef3691 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 27 Jul 2021 15:45:13 +0200 Subject: dev_ioctl: split out ndo_eth_ioctl Most users of ndo_do_ioctl are ethernet drivers that implement the MII commands SIOCGMIIPHY/SIOCGMIIREG/SIOCSMIIREG, or hardware timestamping with SIOCSHWTSTAMP/SIOCGHWTSTAMP. Separate these from the few drivers that use ndo_do_ioctl to implement SIOCBOND, SIOCBR and SIOCWANDEV commands. This is a purely cosmetic change intended to help readers find their way through the implementation. Cc: Doug Ledford Cc: Jason Gunthorpe Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Cc: Andrew Lunn Cc: Vivien Didelot Cc: Florian Fainelli Cc: Vladimir Oltean Cc: Leon Romanovsky Cc: linux-rdma@vger.kernel.org Signed-off-by: Arnd Bergmann Acked-by: Jason Gunthorpe Signed-off-by: David S. Miller --- include/net/dsa.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 55fcac854058..2af6ee2f2bfb 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -106,8 +106,8 @@ struct dsa_device_ops { * function pointers. */ struct dsa_netdevice_ops { - int (*ndo_do_ioctl)(struct net_device *dev, struct ifreq *ifr, - int cmd); + int (*ndo_eth_ioctl)(struct net_device *dev, struct ifreq *ifr, + int cmd); }; #define DSA_TAG_DRIVER_ALIAS "dsa_tag-" @@ -1019,8 +1019,8 @@ static inline int __dsa_netdevice_ops_check(struct net_device *dev) return 0; } -static inline int dsa_ndo_do_ioctl(struct net_device *dev, struct ifreq *ifr, - int cmd) +static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr, + int cmd) { const struct dsa_netdevice_ops *ops; int err; @@ -1031,11 +1031,11 @@ static inline int dsa_ndo_do_ioctl(struct net_device *dev, struct ifreq *ifr, ops = dev->dsa_ptr->netdev_ops; - return ops->ndo_do_ioctl(dev, ifr, cmd); + return ops->ndo_eth_ioctl(dev, ifr, cmd); } #else -static inline int dsa_ndo_do_ioctl(struct net_device *dev, struct ifreq *ifr, - int cmd) +static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr, + int cmd) { return -EOPNOTSUPP; } -- cgit v1.2.3 From d7907a2b1a3b89bea136025f885035a083525e41 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Wed, 28 Jul 2021 10:33:47 +0300 Subject: devlink: Remove duplicated registration check Both registered flag and devlink pointer are set at the same time and indicate the same thing - devlink/devlink_port are ready. Instead of checking ->registered use devlink pointer as an indication. Signed-off-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 57b738b78073..e48a62320407 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -55,8 +55,7 @@ struct devlink { * port, sb, dpipe, resource, params, region, traps and more. */ u8 reload_failed:1, - reload_enabled:1, - registered:1; + reload_enabled:1; char priv[0] __aligned(NETDEV_ALIGN); }; @@ -158,7 +157,6 @@ struct devlink_port { struct list_head region_list; struct devlink *devlink; unsigned int index; - bool registered; spinlock_t type_lock; /* Protects type and type_dev * pointer consistency. */ -- cgit v1.2.3 From 58ce6d5b271ab25fb2056f84a8e5546945eb5fc9 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Tue, 27 Jul 2021 06:12:04 +0900 Subject: Bluetooth: defer cleanup of resources in hci_unregister_dev() syzbot is hitting might_sleep() warning at hci_sock_dev_event() due to calling lock_sock() with rw spinlock held [1]. It seems that history of this locking problem is a trial and error. Commit b40df5743ee8aed8 ("[PATCH] bluetooth: fix socket locking in hci_sock_dev_event()") in 2.6.21-rc4 changed bh_lock_sock() to lock_sock() as an attempt to fix lockdep warning. Then, commit 4ce61d1c7a8ef4c1 ("[BLUETOOTH]: Fix locking in hci_sock_dev_event().") in 2.6.22-rc2 changed lock_sock() to local_bh_disable() + bh_lock_sock_nested() as an attempt to fix sleep in atomic context warning. Then, commit 4b5dd696f81b210c ("Bluetooth: Remove local_bh_disable() from hci_sock.c") in 3.3-rc1 removed local_bh_disable(). Then, commit e305509e678b3a4a ("Bluetooth: use correct lock to prevent UAF of hdev object") in 5.13-rc5 again changed bh_lock_sock_nested() to lock_sock() as an attempt to fix CVE-2021-3573. This difficulty comes from current implementation that hci_sock_dev_event(HCI_DEV_UNREG) is responsible for dropping all references from sockets because hci_unregister_dev() immediately reclaims resources as soon as returning from hci_sock_dev_event(HCI_DEV_UNREG). But the history suggests that hci_sock_dev_event(HCI_DEV_UNREG) was not doing what it should do. Therefore, instead of trying to detach sockets from device, let's accept not detaching sockets from device at hci_sock_dev_event(HCI_DEV_UNREG), by moving actual cleanup of resources from hci_unregister_dev() to hci_release_dev() which is called by bt_host_release when all references to this unregistered device (which is a kobject) are gone. Link: https://syzkaller.appspot.com/bug?extid=a5df189917e79d5e59c9 [1] Reported-by: syzbot Signed-off-by: Tetsuo Handa Tested-by: syzbot Fixes: e305509e678b3a4a ("Bluetooth: use correct lock to prevent UAF of hdev object") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a53e94459ecd..4abe3c494002 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1230,6 +1230,7 @@ struct hci_dev *hci_alloc_dev(void); void hci_free_dev(struct hci_dev *hdev); int hci_register_dev(struct hci_dev *hdev); void hci_unregister_dev(struct hci_dev *hdev); +void hci_release_dev(struct hci_dev *hdev); int hci_suspend_dev(struct hci_dev *hdev); int hci_resume_dev(struct hci_dev *hdev); int hci_reset_dev(struct hci_dev *hdev); -- cgit v1.2.3 From 8a886b142bd03d36612747e9aefdf0282c8b02dd Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 28 Jul 2021 18:24:00 +0200 Subject: sk_buff: track dst status in slow_gro Similar to the previous patch, but covering the dst field: the slow_gro flag is additionally set when a dst is attached to the skb RFC -> v1: - use the existing flag instead of adding a new one Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/net/dst.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/dst.h b/include/net/dst.h index 75b1e734e9c2..a057319aabef 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -277,6 +277,7 @@ static inline void skb_dst_drop(struct sk_buff *skb) static inline void __skb_dst_copy(struct sk_buff *nskb, unsigned long refdst) { + nskb->slow_gro |= !!refdst; nskb->_skb_refdst = refdst; if (!(nskb->_skb_refdst & SKB_DST_NOREF)) dst_clone(skb_dst(nskb)); @@ -316,6 +317,7 @@ static inline bool skb_dst_force(struct sk_buff *skb) dst = NULL; skb->_skb_refdst = (unsigned long)dst; + skb->slow_gro |= !!dst; } return skb->_skb_refdst != 0UL; -- cgit v1.2.3 From 5e10da5385d20c4bae587bc2921e5fdd9655d5fc Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 28 Jul 2021 18:24:03 +0200 Subject: skbuff: allow 'slow_gro' for skb carring sock reference This change leverages the infrastructure introduced by the previous patches to allow soft devices passing to the GRO engine owned skbs without impacting the fast-path. It's up to the GRO caller ensuring the slow_gro bit validity before invoking the GRO engine. The new helper skb_prepare_for_gro() is introduced for that goal. On slow_gro, skbs are aggregated only with equal sk. Additionally, skb truesize on GRO recycle and free is correctly updated so that sk wmem is not changed by the GRO processing. rfc-> v1: - fixed bad truesize on dev_gro_receive NAPI_FREE - use the existing state bit Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/net/sock.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index f23cb259b0e2..ff1be7e7e90b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2249,6 +2249,15 @@ static inline __must_check bool skb_set_owner_sk_safe(struct sk_buff *skb, struc return false; } +static inline void skb_prepare_for_gro(struct sk_buff *skb) +{ + if (skb->destructor != sock_wfree) { + skb_orphan(skb); + return; + } + skb->slow_gro = 1; +} + void sk_reset_timer(struct sock *sk, struct timer_list *timer, unsigned long expires); -- cgit v1.2.3 From dd8987a394c0730380167e0b0aebd766cf3511e1 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 29 Jul 2021 12:40:11 +0200 Subject: nfc: constify passed nfc_dev The struct nfc_dev is not modified by nfc_get_drvdata() and nfc_device_name() so it can be made a const. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nfc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index 85b698794b14..901779138e2b 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -245,7 +245,7 @@ static inline void nfc_set_drvdata(struct nfc_dev *dev, void *data) * * @dev: The nfc device */ -static inline void *nfc_get_drvdata(struct nfc_dev *dev) +static inline void *nfc_get_drvdata(const struct nfc_dev *dev) { return dev_get_drvdata(&dev->dev); } @@ -255,7 +255,7 @@ static inline void *nfc_get_drvdata(struct nfc_dev *dev) * * @dev: The nfc device whose name to return */ -static inline const char *nfc_device_name(struct nfc_dev *dev) +static inline const char *nfc_device_name(const struct nfc_dev *dev) { return dev_name(&dev->dev); } -- cgit v1.2.3 From 2c8e2e9aec7927b387540a88351b8405ee82b34a Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Thu, 29 Jul 2021 10:20:41 +0800 Subject: mctp: Add base packet definitions Simple packet header format as defined by DMTF DSP0236. Signed-off-by: Jeremy Kerr Signed-off-by: David S. Miller --- include/net/mctp.h | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) create mode 100644 include/net/mctp.h (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h new file mode 100644 index 000000000000..4c01e083be45 --- /dev/null +++ b/include/net/mctp.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Management Component Transport Protocol (MCTP) + * + * Copyright (c) 2021 Code Construct + * Copyright (c) 2021 Google + */ + +#ifndef __NET_MCTP_H +#define __NET_MCTP_H + +#include + +/* MCTP packet definitions */ +struct mctp_hdr { + u8 ver; + u8 dest; + u8 src; + u8 flags_seq_tag; +}; + +#define MCTP_VER_MIN 1 +#define MCTP_VER_MAX 1 + +/* Definitions for flags_seq_tag field */ +#define MCTP_HDR_FLAG_SOM BIT(7) +#define MCTP_HDR_FLAG_EOM BIT(6) +#define MCTP_HDR_FLAG_TO BIT(3) +#define MCTP_HDR_FLAGS GENMASK(5, 3) +#define MCTP_HDR_SEQ_SHIFT 4 +#define MCTP_HDR_SEQ_MASK GENMASK(1, 0) +#define MCTP_HDR_TAG_SHIFT 0 +#define MCTP_HDR_TAG_MASK GENMASK(2, 0) + +#endif /* __NET_MCTP_H */ -- cgit v1.2.3 From 583be982d93479ea3d85091b0fd0b01201ede87d Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Thu, 29 Jul 2021 10:20:44 +0800 Subject: mctp: Add device handling and netlink interface This change adds the infrastructure for managing MCTP netdevices; we add a pointer to the AF_MCTP-specific data to struct netdevice, and hook up the rtnetlink operations for adding and removing addresses. Includes changes from Matt Johnston . Signed-off-by: Jeremy Kerr Signed-off-by: David S. Miller --- include/net/mctp.h | 14 ++++++++++++++ include/net/mctpdevice.h | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 49 insertions(+) create mode 100644 include/net/mctpdevice.h (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 4c01e083be45..61452e03aa85 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -10,6 +10,7 @@ #define __NET_MCTP_H #include +#include /* MCTP packet definitions */ struct mctp_hdr { @@ -32,4 +33,17 @@ struct mctp_hdr { #define MCTP_HDR_TAG_SHIFT 0 #define MCTP_HDR_TAG_MASK GENMASK(2, 0) +static inline bool mctp_address_ok(mctp_eid_t eid) +{ + return eid >= 8 && eid < 255; +} + +static inline struct mctp_hdr *mctp_hdr(struct sk_buff *skb) +{ + return (struct mctp_hdr *)skb_network_header(skb); +} + +void mctp_device_init(void); +void mctp_device_exit(void); + #endif /* __NET_MCTP_H */ diff --git a/include/net/mctpdevice.h b/include/net/mctpdevice.h new file mode 100644 index 000000000000..71a11012fac7 --- /dev/null +++ b/include/net/mctpdevice.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Management Component Transport Protocol (MCTP) - device + * definitions. + * + * Copyright (c) 2021 Code Construct + * Copyright (c) 2021 Google + */ + +#ifndef __NET_MCTPDEVICE_H +#define __NET_MCTPDEVICE_H + +#include +#include +#include + +struct mctp_dev { + struct net_device *dev; + + unsigned int net; + + /* Only modified under RTNL. Reads have addrs_lock held */ + u8 *addrs; + size_t num_addrs; + spinlock_t addrs_lock; + + struct rcu_head rcu; +}; + +#define MCTP_INITIAL_DEFAULT_NET 1 + +struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev); +struct mctp_dev *__mctp_dev_get(const struct net_device *dev); + +#endif /* __NET_MCTPDEVICE_H */ -- cgit v1.2.3 From 889b7da23abf92faf34491df95733bda63639e32 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Thu, 29 Jul 2021 10:20:45 +0800 Subject: mctp: Add initial routing framework Add a simple routing table, and a couple of route output handlers, and the mctp packet_type & handler. Includes changes from Matt Johnston . Signed-off-by: Jeremy Kerr Signed-off-by: David S. Miller --- include/net/mctp.h | 75 +++++++++++++++++++++++++++++++++++++++++++++ include/net/net_namespace.h | 4 +++ include/net/netns/mctp.h | 16 ++++++++++ 3 files changed, 95 insertions(+) create mode 100644 include/net/netns/mctp.h (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 61452e03aa85..4094bec5e5db 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -11,6 +11,7 @@ #include #include +#include /* MCTP packet definitions */ struct mctp_hdr { @@ -33,6 +34,8 @@ struct mctp_hdr { #define MCTP_HDR_TAG_SHIFT 0 #define MCTP_HDR_TAG_MASK GENMASK(2, 0) +#define MCTP_HEADER_MAXLEN 4 + static inline bool mctp_address_ok(mctp_eid_t eid) { return eid >= 8 && eid < 255; @@ -43,6 +46,78 @@ static inline struct mctp_hdr *mctp_hdr(struct sk_buff *skb) return (struct mctp_hdr *)skb_network_header(skb); } +struct mctp_skb_cb { + unsigned int magic; + unsigned int net; + mctp_eid_t src; +}; + +/* skb control-block accessors with a little extra debugging for initial + * development. + * + * TODO: remove checks & mctp_skb_cb->magic; replace callers of __mctp_cb + * with mctp_cb(). + * + * __mctp_cb() is only for the initial ingress code; we should see ->magic set + * at all times after this. + */ +static inline struct mctp_skb_cb *__mctp_cb(struct sk_buff *skb) +{ + struct mctp_skb_cb *cb = (void *)skb->cb; + + cb->magic = 0x4d435450; + return cb; +} + +static inline struct mctp_skb_cb *mctp_cb(struct sk_buff *skb) +{ + struct mctp_skb_cb *cb = (void *)skb->cb; + + WARN_ON(cb->magic != 0x4d435450); + return (void *)(skb->cb); +} + +/* Route definition. + * + * These are held in the pernet->mctp.routes list, with RCU protection for + * removed routes. We hold a reference to the netdev; routes need to be + * dropped on NETDEV_UNREGISTER events. + * + * Updates to the route table are performed under rtnl; all reads under RCU, + * so routes cannot be referenced over a RCU grace period. Specifically: A + * caller cannot block between mctp_route_lookup and passing the route to + * mctp_do_route. + */ +struct mctp_route { + mctp_eid_t min, max; + + struct mctp_dev *dev; + unsigned int mtu; + int (*output)(struct mctp_route *route, + struct sk_buff *skb); + + struct list_head list; + refcount_t refs; + struct rcu_head rcu; +}; + +/* route interfaces */ +struct mctp_route *mctp_route_lookup(struct net *net, unsigned int dnet, + mctp_eid_t daddr); + +int mctp_do_route(struct mctp_route *rt, struct sk_buff *skb); + +int mctp_local_output(struct sock *sk, struct mctp_route *rt, + struct sk_buff *skb, mctp_eid_t daddr, u8 req_tag); + +/* routing <--> device interface */ +int mctp_route_add_local(struct mctp_dev *mdev, mctp_eid_t addr); +int mctp_route_remove_local(struct mctp_dev *mdev, mctp_eid_t addr); +void mctp_route_remove_dev(struct mctp_dev *mdev); + +int mctp_routes_init(void); +void mctp_routes_exit(void); + void mctp_device_init(void); void mctp_device_exit(void); diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 12cf6d7ea62c..cc54750dd3db 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -34,6 +34,7 @@ #include #include #include +#include #include #include #include @@ -167,6 +168,9 @@ struct net { #ifdef CONFIG_XDP_SOCKETS struct netns_xdp xdp; #endif +#if IS_ENABLED(CONFIG_MCTP) + struct netns_mctp mctp; +#endif #if IS_ENABLED(CONFIG_CRYPTO_USER) struct sock *crypto_nlsk; #endif diff --git a/include/net/netns/mctp.h b/include/net/netns/mctp.h new file mode 100644 index 000000000000..508459b08a59 --- /dev/null +++ b/include/net/netns/mctp.h @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * MCTP per-net structures + */ + +#ifndef __NETNS_MCTP_H__ +#define __NETNS_MCTP_H__ + +#include + +struct netns_mctp { + /* Only updated under RTNL, entries freed via RCU */ + struct list_head routes; +}; + +#endif /* __NETNS_MCTP_H__ */ -- cgit v1.2.3 From 06d2f4c583a7d892300920fc85d654d48a15e914 Mon Sep 17 00:00:00 2001 From: Matt Johnston Date: Thu, 29 Jul 2021 10:20:46 +0800 Subject: mctp: Add netlink route management This change adds RTM_GETROUTE, RTM_NEWROUTE & RTM_DELROUTE handlers, allowing management of the MCTP route table. Includes changes from Jeremy Kerr . Signed-off-by: Matt Johnston Signed-off-by: David S. Miller --- include/net/mctp.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 4094bec5e5db..bc36e37e8198 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -111,6 +111,8 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt, struct sk_buff *skb, mctp_eid_t daddr, u8 req_tag); /* routing <--> device interface */ +unsigned int mctp_default_net(struct net *net); +int mctp_default_net_set(struct net *net, unsigned int index); int mctp_route_add_local(struct mctp_dev *mdev, mctp_eid_t addr); int mctp_route_remove_local(struct mctp_dev *mdev, mctp_eid_t addr); void mctp_route_remove_dev(struct mctp_dev *mdev); -- cgit v1.2.3 From 4d8b9319282ae84f5a17b28d8b5b5d1e7e537312 Mon Sep 17 00:00:00 2001 From: Matt Johnston Date: Thu, 29 Jul 2021 10:20:47 +0800 Subject: mctp: Add neighbour implementation Add an initial neighbour table implementation, to be used in the route output path. Signed-off-by: Matt Johnston Signed-off-by: David S. Miller --- include/net/mctp.h | 25 +++++++++++++++++++++++++ include/net/mctpdevice.h | 1 + include/net/netns/mctp.h | 4 ++++ 3 files changed, 30 insertions(+) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index bc36e37e8198..53f035c8b59c 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -117,6 +117,31 @@ int mctp_route_add_local(struct mctp_dev *mdev, mctp_eid_t addr); int mctp_route_remove_local(struct mctp_dev *mdev, mctp_eid_t addr); void mctp_route_remove_dev(struct mctp_dev *mdev); +/* neighbour definitions */ +enum mctp_neigh_source { + MCTP_NEIGH_STATIC, + MCTP_NEIGH_DISCOVER, +}; + +struct mctp_neigh { + struct mctp_dev *dev; + mctp_eid_t eid; + enum mctp_neigh_source source; + + unsigned char ha[MAX_ADDR_LEN]; + + struct list_head list; + struct rcu_head rcu; +}; + +int mctp_neigh_init(void); +void mctp_neigh_exit(void); + +// ret_hwaddr may be NULL, otherwise must have space for MAX_ADDR_LEN +int mctp_neigh_lookup(struct mctp_dev *dev, mctp_eid_t eid, + void *ret_hwaddr); +void mctp_neigh_remove_dev(struct mctp_dev *mdev); + int mctp_routes_init(void); void mctp_routes_exit(void); diff --git a/include/net/mctpdevice.h b/include/net/mctpdevice.h index 71a11012fac7..57e773ff08bb 100644 --- a/include/net/mctpdevice.h +++ b/include/net/mctpdevice.h @@ -31,5 +31,6 @@ struct mctp_dev { struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev); struct mctp_dev *__mctp_dev_get(const struct net_device *dev); +struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev); #endif /* __NET_MCTPDEVICE_H */ diff --git a/include/net/netns/mctp.h b/include/net/netns/mctp.h index 508459b08a59..2f5ebeeb320e 100644 --- a/include/net/netns/mctp.h +++ b/include/net/netns/mctp.h @@ -11,6 +11,10 @@ struct netns_mctp { /* Only updated under RTNL, entries freed via RCU */ struct list_head routes; + + /* neighbour table */ + struct mutex neigh_lock; + struct list_head neighbours; }; #endif /* __NETNS_MCTP_H__ */ -- cgit v1.2.3 From 833ef3b91de692ef33b800bca6b1569c39dece74 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Thu, 29 Jul 2021 10:20:49 +0800 Subject: mctp: Populate socket implementation Start filling-out the socket syscalls: bind, sendmsg & recvmsg. This requires an input route implementation, so we add to mctp_route_input, allowing lookups on binds & message tags. This just handles single-packet messages at present, we will add fragmentation in a future change. Signed-off-by: Jeremy Kerr Signed-off-by: David S. Miller --- include/net/mctp.h | 59 ++++++++++++++++++++++++++++++++++++++++++++++++ include/net/netns/mctp.h | 13 +++++++++++ 2 files changed, 72 insertions(+) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 53f035c8b59c..f2d98f6993c0 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -12,6 +12,7 @@ #include #include #include +#include /* MCTP packet definitions */ struct mctp_hdr { @@ -46,6 +47,64 @@ static inline struct mctp_hdr *mctp_hdr(struct sk_buff *skb) return (struct mctp_hdr *)skb_network_header(skb); } +/* socket implementation */ +struct mctp_sock { + struct sock sk; + + /* bind() params */ + int bind_net; + mctp_eid_t bind_addr; + __u8 bind_type; + + /* list of mctp_sk_key, for incoming tag lookup. updates protected + * by sk->net->keys_lock + */ + struct hlist_head keys; +}; + +/* Key for matching incoming packets to sockets or reassembly contexts. + * Packets are matched on (src,dest,tag). + * + * Lifetime requirements: + * + * - keys are free()ed via RCU + * + * - a mctp_sk_key contains a reference to a struct sock; this is valid + * for the life of the key. On sock destruction (through unhash), the key is + * removed from lists (see below), and will not be observable after a RCU + * grace period. + * + * any RX occurring within that grace period may still queue to the socket, + * but will hit the SOCK_DEAD case before the socket is freed. + * + * - these mctp_sk_keys appear on two lists: + * 1) the struct mctp_sock->keys list + * 2) the struct netns_mctp->keys list + * + * updates to either list are performed under the netns_mctp->keys + * lock. + * + * - there is a single destruction path for a mctp_sk_key - through socket + * unhash (see mctp_sk_unhash). This performs the list removal under + * keys_lock. + */ +struct mctp_sk_key { + mctp_eid_t peer_addr; + mctp_eid_t local_addr; + __u8 tag; /* incoming tag match; invert TO for local */ + + /* we hold a ref to sk when set */ + struct sock *sk; + + /* routing lookup list */ + struct hlist_node hlist; + + /* per-socket list */ + struct hlist_node sklist; + + struct rcu_head rcu; +}; + struct mctp_skb_cb { unsigned int magic; unsigned int net; diff --git a/include/net/netns/mctp.h b/include/net/netns/mctp.h index 2f5ebeeb320e..14ae6d37e52a 100644 --- a/include/net/netns/mctp.h +++ b/include/net/netns/mctp.h @@ -12,6 +12,19 @@ struct netns_mctp { /* Only updated under RTNL, entries freed via RCU */ struct list_head routes; + /* Bound sockets: list of sockets bound by type. + * This list is updated from non-atomic contexts (under bind_lock), + * and read (under rcu) in packet rx + */ + struct mutex bind_lock; + struct hlist_head binds; + + /* tag allocations. This list is read and updated from atomic contexts, + * but elements are free()ed after a RCU grace-period + */ + spinlock_t keys_lock; + struct hlist_head keys; + /* neighbour table */ struct mutex neigh_lock; struct list_head neighbours; -- cgit v1.2.3 From 4a992bbd365094730a31bae1e12a6ca695336d57 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Thu, 29 Jul 2021 10:20:50 +0800 Subject: mctp: Implement message fragmentation & reassembly This change implements MCTP fragmentation (based on route & device MTU), and corresponding reassembly. The MCTP specification only allows for fragmentation on the originating message endpoint, and reassembly on the destination endpoint - intermediate nodes do not need to reassemble/refragment. Consequently, we only fragment in the local transmit path, and reassemble locally-bound packets. Messages are required to be in-order, so we simply cancel reassembly on out-of-order or missing packets. In the fragmentation path, we just break up the message into MTU-sized fragments; the skb structure is a simple copy for now, which we can later improve with a shared data implementation. For reassembly, we keep track of incoming message fragments using the existing tag infrastructure, allocating a key on the (src,dest,tag) tuple, and reassembles matching fragments into a skb->frag_list. Signed-off-by: Jeremy Kerr Signed-off-by: David S. Miller --- include/net/mctp.h | 25 ++++++++++++++++++++++--- 1 file changed, 22 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index f2d98f6993c0..0a460ba185b8 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -84,9 +84,21 @@ struct mctp_sock { * updates to either list are performed under the netns_mctp->keys * lock. * - * - there is a single destruction path for a mctp_sk_key - through socket - * unhash (see mctp_sk_unhash). This performs the list removal under - * keys_lock. + * - a key may have a sk_buff attached as part of an in-progress message + * reassembly (->reasm_head). The reassembly context is protected by + * reasm_lock, which may be acquired with the keys lock (above) held, if + * necessary. Consequently, keys lock *cannot* be acquired with the + * reasm_lock held. + * + * - there are two destruction paths for a mctp_sk_key: + * + * - through socket unhash (see mctp_sk_unhash). This performs the list + * removal under keys_lock. + * + * - where a key is established to receive a reply message: after receiving + * the (complete) reply, or during reassembly errors. Here, we clean up + * the reassembly context (marking reasm_dead, to prevent another from + * starting), and remove the socket from the netns & socket lists. */ struct mctp_sk_key { mctp_eid_t peer_addr; @@ -102,6 +114,13 @@ struct mctp_sk_key { /* per-socket list */ struct hlist_node sklist; + /* incoming fragment reassembly context */ + spinlock_t reasm_lock; + struct sk_buff *reasm_head; + struct sk_buff **reasm_tailp; + bool reasm_dead; + u8 last_seq; + struct rcu_head rcu; }; -- cgit v1.2.3 From 03f2bbc4ee57ca53b2fa1d9caabc5006e0b8f375 Mon Sep 17 00:00:00 2001 From: Matt Johnston Date: Thu, 29 Jul 2021 10:20:52 +0800 Subject: mctp: Allow per-netns default networks Currently we have a compile-time default network (MCTP_INITIAL_DEFAULT_NET). This change introduces a default_net field on the net namespace, allowing future configuration for new interfaces. Signed-off-by: Matt Johnston Signed-off-by: David S. Miller --- include/net/mctp.h | 2 ++ include/net/netns/mctp.h | 3 +++ 2 files changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 0a460ba185b8..54bbe042c973 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -37,6 +37,8 @@ struct mctp_hdr { #define MCTP_HEADER_MAXLEN 4 +#define MCTP_INITIAL_DEFAULT_NET 1 + static inline bool mctp_address_ok(mctp_eid_t eid) { return eid >= 8 && eid < 255; diff --git a/include/net/netns/mctp.h b/include/net/netns/mctp.h index 14ae6d37e52a..acedef12a35e 100644 --- a/include/net/netns/mctp.h +++ b/include/net/netns/mctp.h @@ -25,6 +25,9 @@ struct netns_mctp { spinlock_t keys_lock; struct hlist_head keys; + /* MCTP network */ + unsigned int default_net; + /* neighbour table */ struct mutex neigh_lock; struct list_head neighbours; -- cgit v1.2.3 From 3aa2605594556c676fb88744bd9845acae60683d Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Wed, 28 Jul 2021 20:08:00 +0200 Subject: net/sched: store the last executed chain also for clsact egress currently, only 'ingress' and 'clsact ingress' qdiscs store the tc 'chain id' in the skb extension. However, userspace programs (like ovs) are able to setup egress rules, and datapath gets confused in case it doesn't find the 'chain id' for a packet that's "recirculated" by tc. Change tcf_classify() to have the same semantic as tcf_classify_ingress() so that a single function can be called in ingress / egress, using the tc ingress / egress block respectively. Suggested-by: Alaa Hleilel Signed-off-by: Davide Caratti Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 22 +++++++--------------- 1 file changed, 7 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index ec7823921bd2..dc28fcb6f0a2 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -76,12 +76,10 @@ static inline struct Qdisc *tcf_block_q(struct tcf_block *block) return block->q; } -int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, - struct tcf_result *res, bool compat_mode); -int tcf_classify_ingress(struct sk_buff *skb, - const struct tcf_block *ingress_block, - const struct tcf_proto *tp, struct tcf_result *res, - bool compat_mode); +int tcf_classify(struct sk_buff *skb, + const struct tcf_block *block, + const struct tcf_proto *tp, struct tcf_result *res, + bool compat_mode); #else static inline bool tcf_block_shared(struct tcf_block *block) @@ -138,20 +136,14 @@ void tc_setup_cb_block_unregister(struct tcf_block *block, flow_setup_cb_t *cb, { } -static inline int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, +static inline int tcf_classify(struct sk_buff *skb, + const struct tcf_block *block, + const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode) { return TC_ACT_UNSPEC; } -static inline int tcf_classify_ingress(struct sk_buff *skb, - const struct tcf_block *ingress_block, - const struct tcf_proto *tp, - struct tcf_result *res, bool compat_mode) -{ - return TC_ACT_UNSPEC; -} - #endif static inline unsigned long -- cgit v1.2.3 From 79976892f7ea37f44f8bcfb6d266954e8ae0124d Mon Sep 17 00:00:00 2001 From: Yajun Deng Date: Thu, 29 Jul 2021 15:13:50 +0800 Subject: net: convert fib_treeref from int to refcount_t refcount_t type should be used instead of int when fib_treeref is used as a reference counter,and avoid use-after-free risks. Signed-off-by: Yajun Deng Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20210729071350.28919-1-yajun.deng@linux.dev Signed-off-by: Jakub Kicinski --- include/net/dn_fib.h | 2 +- include/net/ip_fib.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/dn_fib.h b/include/net/dn_fib.h index ccc6e9df178b..ddd6565957b3 100644 --- a/include/net/dn_fib.h +++ b/include/net/dn_fib.h @@ -29,7 +29,7 @@ struct dn_fib_nh { struct dn_fib_info { struct dn_fib_info *fib_next; struct dn_fib_info *fib_prev; - int fib_treeref; + refcount_t fib_treeref; refcount_t fib_clntref; int fib_dead; unsigned int fib_flags; diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 3ab2563b1a23..21c5386d4a6d 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -133,7 +133,7 @@ struct fib_info { struct hlist_node fib_lhash; struct list_head nh_list; struct net *fib_net; - int fib_treeref; + refcount_t fib_treeref; refcount_t fib_clntref; unsigned int fib_flags; unsigned char fib_dead; -- cgit v1.2.3 From 3df40eb3a2ea58bf404a38f15a7a2768e4762cb0 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 30 Jul 2021 16:41:59 +0200 Subject: nfc: constify several pointers to u8, char and sk_buff Several functions receive pointers to u8, char or sk_buff but do not modify the contents so make them const. This allows doing the same for local variables and in total makes the code a little bit safer. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Jakub Kicinski --- include/net/nfc/nfc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index 901779138e2b..5dee575fbe86 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -266,7 +266,7 @@ struct sk_buff *nfc_alloc_send_skb(struct nfc_dev *dev, struct sock *sk, struct sk_buff *nfc_alloc_recv_skb(unsigned int size, gfp_t gfp); int nfc_set_remote_general_bytes(struct nfc_dev *dev, - u8 *gt, u8 gt_len); + const u8 *gt, u8 gt_len); u8 *nfc_get_local_general_bytes(struct nfc_dev *dev, size_t *gb_len); int nfc_fw_download_done(struct nfc_dev *dev, const char *firmware_name, @@ -280,7 +280,7 @@ int nfc_dep_link_is_up(struct nfc_dev *dev, u32 target_idx, u8 comm_mode, u8 rf_mode); int nfc_tm_activated(struct nfc_dev *dev, u32 protocol, u8 comm_mode, - u8 *gb, size_t gb_len); + const u8 *gb, size_t gb_len); int nfc_tm_deactivated(struct nfc_dev *dev); int nfc_tm_data_received(struct nfc_dev *dev, struct sk_buff *skb); -- cgit v1.2.3 From ddecf5556f7fdf871fa8ce26b376e3e8ae6213b6 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 30 Jul 2021 16:42:01 +0200 Subject: nfc: nci: constify several pointers to u8, sk_buff and other structs Several functions receive pointers to u8, sk_buff or other structs but do not modify the contents so make them const. This allows doing the same for local variables and in total makes the code a little bit safer. This makes const also data passed as "unsigned long opt" argument to nci_request() function. Usual flow for such functions is: 1. Receive "u8 *" and store it (the pointer) in a structure allocated on stack (e.g. struct nci_set_config_param), 2. Call nci_request() or __nci_request() passing a callback function an the pointer to the structure via an "unsigned long opt", 3. nci_request() calls the callback which dereferences "unsigned long opt" in a read-only way. This converts all above paths to use proper pointer to const data, so entire flow is safer. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Jakub Kicinski --- include/net/nfc/nci_core.h | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index 00f2c60971d7..4770a81f4aa7 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -278,23 +278,25 @@ int nci_request(struct nci_dev *ndev, void (*req)(struct nci_dev *ndev, unsigned long opt), unsigned long opt, __u32 timeout); -int nci_prop_cmd(struct nci_dev *ndev, __u8 oid, size_t len, __u8 *payload); -int nci_core_cmd(struct nci_dev *ndev, __u16 opcode, size_t len, __u8 *payload); +int nci_prop_cmd(struct nci_dev *ndev, __u8 oid, size_t len, + const __u8 *payload); +int nci_core_cmd(struct nci_dev *ndev, __u16 opcode, size_t len, + const __u8 *payload); int nci_core_reset(struct nci_dev *ndev); int nci_core_init(struct nci_dev *ndev); int nci_recv_frame(struct nci_dev *ndev, struct sk_buff *skb); int nci_send_frame(struct nci_dev *ndev, struct sk_buff *skb); -int nci_set_config(struct nci_dev *ndev, __u8 id, size_t len, __u8 *val); +int nci_set_config(struct nci_dev *ndev, __u8 id, size_t len, const __u8 *val); int nci_nfcee_discover(struct nci_dev *ndev, u8 action); int nci_nfcee_mode_set(struct nci_dev *ndev, u8 nfcee_id, u8 nfcee_mode); int nci_core_conn_create(struct nci_dev *ndev, u8 destination_type, u8 number_destination_params, size_t params_len, - struct core_conn_create_dest_spec_params *params); + const struct core_conn_create_dest_spec_params *params); int nci_core_conn_close(struct nci_dev *ndev, u8 conn_id); -int nci_nfcc_loopback(struct nci_dev *ndev, void *data, size_t data_len, +int nci_nfcc_loopback(struct nci_dev *ndev, const void *data, size_t data_len, struct sk_buff **resp); struct nci_hci_dev *nci_hci_allocate(struct nci_dev *ndev); @@ -378,7 +380,7 @@ void nci_req_complete(struct nci_dev *ndev, int result); struct nci_conn_info *nci_get_conn_info_by_conn_id(struct nci_dev *ndev, int conn_id); int nci_get_conn_info_by_dest_type_params(struct nci_dev *ndev, u8 dest_type, - struct dest_spec_params *params); + const struct dest_spec_params *params); /* ----- NCI status code ----- */ int nci_to_errno(__u8 code); -- cgit v1.2.3 From 26713455048eb19122b1561b471d30710177ef97 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Thu, 29 Jul 2021 20:19:25 +0300 Subject: devlink: Allocate devlink directly in requested net namespace There is no need in extra call indirection and check from impossible flow where someone tries to set namespace without prior call to devlink_alloc(). Instead of this extra logic and additional EXPORT_SYMBOL, use specialized devlink allocation function that receives net namespace as an argument. Such specialized API allows clear view when devlink initialized in wrong net namespace and/or kernel users don't try to change devlink namespace under the hood. Reviewed-by: Jiri Pirko Signed-off-by: Leon Romanovsky Signed-off-by: Jakub Kicinski --- include/net/devlink.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index e48a62320407..08f4c6191e72 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1540,8 +1540,18 @@ static inline struct devlink *netdev_to_devlink(struct net_device *dev) struct ib_device; struct net *devlink_net(const struct devlink *devlink); -void devlink_net_set(struct devlink *devlink, struct net *net); -struct devlink *devlink_alloc(const struct devlink_ops *ops, size_t priv_size); +/* This call is intended for software devices that can create + * devlink instances in other namespaces than init_net. + * + * Drivers that operate on real HW must use devlink_alloc() instead. + */ +struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, + size_t priv_size, struct net *net); +static inline struct devlink *devlink_alloc(const struct devlink_ops *ops, + size_t priv_size) +{ + return devlink_alloc_ns(ops, priv_size, &init_net); +} int devlink_register(struct devlink *devlink, struct device *dev); void devlink_unregister(struct devlink *devlink); void devlink_reload_enable(struct devlink *devlink); -- cgit v1.2.3 From f2e3778db7e13fd9a83c309dd8c9fbe69ba8efbe Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 22 Jul 2021 10:48:34 +0200 Subject: netfilter: remove xt pernet data clusterip is now handled via net_generic. NOTRACK is tiny compared to rest of xt_CT feature set, even the existing deprecation warning is bigger than the actual functionality. Just remove the warning, its not worth keeping/adding a net_generic one. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/net_namespace.h | 2 -- include/net/netns/x_tables.h | 12 ------------ 2 files changed, 14 deletions(-) delete mode 100644 include/net/netns/x_tables.h (limited to 'include/net') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index cc54750dd3db..bb5fa5914032 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -23,7 +23,6 @@ #include #include #include -#include #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) #include #endif @@ -133,7 +132,6 @@ struct net { #endif #ifdef CONFIG_NETFILTER struct netns_nf nf; - struct netns_xt xt; #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) struct netns_ct ct; #endif diff --git a/include/net/netns/x_tables.h b/include/net/netns/x_tables.h deleted file mode 100644 index d02316ec2906..000000000000 --- a/include/net/netns/x_tables.h +++ /dev/null @@ -1,12 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __NETNS_X_TABLES_H -#define __NETNS_X_TABLES_H - -#include -#include - -struct netns_xt { - bool notrack_deprecated_warning; - bool clusterip_deprecated_warning; -}; -#endif -- cgit v1.2.3 From 695176bfe5dec2051f950bdac0ae0b21e29e6de3 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 29 Jul 2021 16:12:14 -0700 Subject: net_sched: refactor TC action init API TC action ->init() API has 10 parameters, it becomes harder to read. Some of them are just boolean and can be replaced by flags. Similarly for the internal API tcf_action_init() and tcf_exts_validate(). This patch converts them to flags and fold them into the upper 16 bits of "flags", whose lower 16 bits are still reserved for user-space. More specifically, the following kernel flags are introduced: TCA_ACT_FLAGS_POLICE replace 'name' in a few contexts, to distinguish whether it is compatible with policer. TCA_ACT_FLAGS_BIND replaces 'bind', to indicate whether this action is bound to a filter. TCA_ACT_FLAGS_REPLACE replaces 'ovr' in most contexts, means we are replacing an existing action. TCA_ACT_FLAGS_NO_RTNL replaces 'rtnl_held' but has the opposite meaning, because we still hold RTNL in most cases. The only user-space flag TCA_ACT_FLAGS_NO_PERCPU_STATS is untouched and still stored as before. I have tested this patch with tdc and I do not see any failure related to this patch. Tested-by: Vlad Buslov Acked-by: Jamal Hadi Salim Cc: Jiri Pirko Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/net/act_api.h | 22 ++++++++++++++-------- include/net/pkt_cls.h | 2 +- include/net/sch_generic.h | 2 +- 3 files changed, 16 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/act_api.h b/include/net/act_api.h index 086b291e9530..f19f7f4a463c 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -58,6 +58,14 @@ struct tc_action { #define TCA_ACT_HW_STATS_ANY (TCA_ACT_HW_STATS_IMMEDIATE | \ TCA_ACT_HW_STATS_DELAYED) +/* Reserve 16 bits for user-space. See TCA_ACT_FLAGS_NO_PERCPU_STATS. */ +#define TCA_ACT_FLAGS_USER_BITS 16 +#define TCA_ACT_FLAGS_USER_MASK 0xffff +#define TCA_ACT_FLAGS_POLICE (1U << TCA_ACT_FLAGS_USER_BITS) +#define TCA_ACT_FLAGS_BIND (1U << (TCA_ACT_FLAGS_USER_BITS + 1)) +#define TCA_ACT_FLAGS_REPLACE (1U << (TCA_ACT_FLAGS_USER_BITS + 2)) +#define TCA_ACT_FLAGS_NO_RTNL (1U << (TCA_ACT_FLAGS_USER_BITS + 3)) + /* Update lastuse only if needed, to avoid dirtying a cache line. * We use a temp variable to avoid fetching jiffies twice. */ @@ -99,8 +107,8 @@ struct tc_action_ops { void (*cleanup)(struct tc_action *); int (*lookup)(struct net *net, struct tc_action **a, u32 index); int (*init)(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action **act, int ovr, - int bind, bool rtnl_held, struct tcf_proto *tp, + struct nlattr *est, struct tc_action **act, + struct tcf_proto *tp, u32 flags, struct netlink_ext_ack *extack); int (*walk)(struct net *, struct sk_buff *, struct netlink_callback *, int, @@ -179,18 +187,16 @@ int tcf_action_destroy(struct tc_action *actions[], int bind); int tcf_action_exec(struct sk_buff *skb, struct tc_action **actions, int nr_actions, struct tcf_result *res); int tcf_action_init(struct net *net, struct tcf_proto *tp, struct nlattr *nla, - struct nlattr *est, char *name, int ovr, int bind, + struct nlattr *est, struct tc_action *actions[], int init_res[], size_t *attr_size, - bool rtnl_held, struct netlink_ext_ack *extack); -struct tc_action_ops *tc_action_load_ops(char *name, struct nlattr *nla, + u32 flags, struct netlink_ext_ack *extack); +struct tc_action_ops *tc_action_load_ops(struct nlattr *nla, bool police, bool rtnl_held, struct netlink_ext_ack *extack); struct tc_action *tcf_action_init_1(struct net *net, struct tcf_proto *tp, struct nlattr *nla, struct nlattr *est, - char *name, int ovr, int bind, struct tc_action_ops *a_o, int *init_res, - bool rtnl_held, - struct netlink_ext_ack *extack); + u32 flags, struct netlink_ext_ack *extack); int tcf_action_dump(struct sk_buff *skb, struct tc_action *actions[], int bind, int ref, bool terse); int tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int, int); diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index dc28fcb6f0a2..64de26b7ad39 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -319,7 +319,7 @@ tcf_exts_exec(struct sk_buff *skb, struct tcf_exts *exts, int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, struct nlattr *rate_tlv, - struct tcf_exts *exts, bool ovr, bool rtnl_held, + struct tcf_exts *exts, u32 flags, struct netlink_ext_ack *extack); void tcf_exts_destroy(struct tcf_exts *exts); void tcf_exts_change(struct tcf_exts *dst, struct tcf_exts *src); diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9ed33e6840bd..c0069ac00e62 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -357,7 +357,7 @@ struct tcf_proto_ops { int (*change)(struct net *net, struct sk_buff *, struct tcf_proto*, unsigned long, u32 handle, struct nlattr **, - void **, bool, bool, + void **, u32, struct netlink_ext_ack *); int (*delete)(struct tcf_proto *tp, void *arg, bool *last, bool rtnl_held, -- cgit v1.2.3 From 35d7a6f1fb53479965e9f99e8c87edc642336eba Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sat, 31 Jul 2021 12:21:44 +0200 Subject: nfc: hci: pass callback data param as pointer in nci_request() The nci_request() receives a callback function and unsigned long data argument "opt" which is passed to the callback. Almost all of the nci_request() callers pass pointer to a stack variable as data argument. Only few pass scalar value (e.g. u8). All such callbacks do not modify passed data argument and in previous commit they were made as const. However passing pointers via unsigned long removes the const annotation. The callback could simply cast unsigned long to a pointer to writeable memory. Use "const void *" as type of this "opt" argument to solve this and prevent modifying the pointed contents. This is also consistent with generic pattern of passing data arguments - via "void *". In few places which pass scalar values, use casts via "unsigned long" to suppress any warnings. Signed-off-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- include/net/nfc/nci_core.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h index 4770a81f4aa7..a964daedc17b 100644 --- a/include/net/nfc/nci_core.h +++ b/include/net/nfc/nci_core.h @@ -276,8 +276,8 @@ int nci_register_device(struct nci_dev *ndev); void nci_unregister_device(struct nci_dev *ndev); int nci_request(struct nci_dev *ndev, void (*req)(struct nci_dev *ndev, - unsigned long opt), - unsigned long opt, __u32 timeout); + const void *opt), + const void *opt, __u32 timeout); int nci_prop_cmd(struct nci_dev *ndev, __u8 oid, size_t len, const __u8 *payload); int nci_core_cmd(struct nci_dev *ndev, __u16 opcode, size_t len, -- cgit v1.2.3 From 29a097b7747725da003245412dab61093d4e5976 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 31 Jul 2021 17:14:32 +0300 Subject: net: dsa: remove the struct packet_type argument from dsa_device_ops::rcv() No tagging driver uses this. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 2af6ee2f2bfb..7cc9507282d3 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -79,13 +79,11 @@ enum dsa_tag_protocol { DSA_TAG_PROTO_SJA1110 = DSA_TAG_PROTO_SJA1110_VALUE, }; -struct packet_type; struct dsa_switch; struct dsa_device_ops { struct sk_buff *(*xmit)(struct sk_buff *skb, struct net_device *dev); - struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt); + struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev); void (*flow_dissect)(const struct sk_buff *skb, __be16 *proto, int *offset); unsigned int needed_headroom; @@ -239,8 +237,7 @@ struct dsa_port { /* Copies for faster access in master receive hot path */ struct dsa_switch_tree *dst; - struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, - struct packet_type *pt); + struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev); enum { DSA_PORT_TYPE_UNUSED = 0, -- cgit v1.2.3 From 3a755cd8b7c601f756cbbf908b84f7cc8c04a02b Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Mon, 2 Aug 2021 11:02:19 +0800 Subject: bonding: add new option lacp_active Add an option lacp_active, which is similar with team's runner.active. This option specifies whether to send LACPDU frames periodically. If set on, the LACPDU frames are sent along with the configured lacp_rate setting. If set off, the LACPDU frames acts as "speak when spoken to". Note, the LACPDU state frames still will be sent when init or unbind port. v2: remove module parameter Signed-off-by: Hangbin Liu Signed-off-by: David S. Miller --- include/net/bond_3ad.h | 1 + include/net/bond_options.h | 1 + include/net/bonding.h | 1 + 3 files changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/bond_3ad.h b/include/net/bond_3ad.h index c8696a230b7d..38785d48baff 100644 --- a/include/net/bond_3ad.h +++ b/include/net/bond_3ad.h @@ -303,6 +303,7 @@ int __bond_3ad_get_active_agg_info(struct bonding *bond, int bond_3ad_lacpdu_recv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave); int bond_3ad_set_carrier(struct bonding *bond); +void bond_3ad_update_lacp_active(struct bonding *bond); void bond_3ad_update_lacp_rate(struct bonding *bond); void bond_3ad_update_ad_actor_settings(struct bonding *bond); int bond_3ad_stats_fill(struct sk_buff *skb, struct bond_3ad_stats *stats); diff --git a/include/net/bond_options.h b/include/net/bond_options.h index 9d382f2f0bc5..e64833a674eb 100644 --- a/include/net/bond_options.h +++ b/include/net/bond_options.h @@ -64,6 +64,7 @@ enum { BOND_OPT_AD_USER_PORT_KEY, BOND_OPT_NUM_PEER_NOTIF_ALIAS, BOND_OPT_PEER_NOTIF_DELAY, + BOND_OPT_LACP_ACTIVE, BOND_OPT_LAST }; diff --git a/include/net/bonding.h b/include/net/bonding.h index 625d9c72dee3..46df47004803 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -129,6 +129,7 @@ struct bond_params { int updelay; int downdelay; int peer_notif_delay; + int lacp_active; int lacp_fast; unsigned int min_links; int ad_select; -- cgit v1.2.3 From c87a4c542b5a796f795fec2b7a909c7d3067b11c Mon Sep 17 00:00:00 2001 From: Bijie Xu Date: Tue, 3 Aug 2021 11:40:18 +0200 Subject: net: flow_offload: correct comments mismatch with code Correct mismatch between the name of flow_offload_has_one_action() and its kdoc entry. Found using ./scripts/kernel-doc -Werror -none include/net/flow_offload.h Signed-off-by: Bijie Xu Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- include/net/flow_offload.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h index 69c9eabf8325..f3c2841566a0 100644 --- a/include/net/flow_offload.h +++ b/include/net/flow_offload.h @@ -293,7 +293,7 @@ static inline bool flow_action_has_entries(const struct flow_action *action) } /** - * flow_action_has_one_action() - check if exactly one action is present + * flow_offload_has_one_action() - check if exactly one action is present * @action: tc filter flow offload action * * Returns true if exactly one action is present. -- cgit v1.2.3 From 0161d151f3e36306219f5aa6f5f6b3877038afd3 Mon Sep 17 00:00:00 2001 From: Bijie Xu Date: Tue, 3 Aug 2021 11:40:19 +0200 Subject: net: sched: provide missing kdoc for tcf_pkt_info and tcf_ematch_ops Provide missing kdoc of fields of struct tcf_pkt_info and tcf_ematch_ops. Found using ./scripts/kernel-doc -none -Werror include/net/pkt_cls.h Signed-off-by: Bijie Xu Signed-off-by: Simon Horman Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index ec7823921bd2..298a8d10168b 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -337,6 +337,9 @@ int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts); /** * struct tcf_pkt_info - packet information + * + * @ptr: start of the pkt data + * @nexthdr: offset of the next header */ struct tcf_pkt_info { unsigned char * ptr; @@ -355,6 +358,7 @@ struct tcf_ematch_ops; * @ops: the operations lookup table of the corresponding ematch module * @datalen: length of the ematch specific configuration data * @data: ematch specific data + * @net: the network namespace */ struct tcf_ematch { struct tcf_ematch_ops * ops; -- cgit v1.2.3 From 4039146777a91e1576da2bf38e0d8a1061a1ae47 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Tue, 3 Aug 2021 12:00:16 +0200 Subject: net: ipv6: fix returned variable type in ip6_skb_dst_mtu The patch fixing the returned value of ip6_skb_dst_mtu (int -> unsigned int) was rebased between its initial review and the version applied. In the meantime fade56410c22 was applied, which added a new variable (int) used as the returned value. This lead to a mismatch between the function prototype and the variable used as the return value. Fixes: 40fc3054b458 ("net: ipv6: fix return value of ip6_skb_dst_mtu") Cc: Vadim Fedorenko Signed-off-by: Antoine Tenart Signed-off-by: David S. Miller --- include/net/ip6_route.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 625a38ccb5d9..0bf09a9bca4e 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -265,7 +265,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, static inline unsigned int ip6_skb_dst_mtu(struct sk_buff *skb) { - int mtu; + unsigned int mtu; struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? inet6_sk(skb->sk) : NULL; -- cgit v1.2.3 From 81dd3ee5962d767b913d4c4efec3f50e888463c1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 3 Aug 2021 13:40:40 +0200 Subject: appletalk: ltpc: remove static probing This driver never relies on the netdev_boot_setup_check() to get its configuration, so it can just as well do its own probing all the time. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/net/Space.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/Space.h b/include/net/Space.h index 9cce0d80d37a..e30e7a70ea99 100644 --- a/include/net/Space.h +++ b/include/net/Space.h @@ -21,7 +21,6 @@ struct net_device *mvme147lance_probe(int unit); struct net_device *tc515_probe(int unit); struct net_device *lance_probe(int unit); struct net_device *cops_probe(int unit); -struct net_device *ltpc_probe(void); /* Fibre Channel adapters */ int iph5526_probe(struct net_device *dev); -- cgit v1.2.3 From e179d78ee11a70e2675bc572f9f4e33d97233b23 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 3 Aug 2021 13:40:43 +0200 Subject: m68k: remove legacy probing There are six m68k specific drivers that use the legacy probe method in drivers/net/Space.c. However, all of these only support a single device, and they completely ignore the command line settings from netdev_boot_setup_check, so there is really no point at all. Aside from sun3_82586, these already have a module_init function that can be used for built-in mode as well, simply by removing the #ifdef. Note that the 82596 driver was previously used on ISA as well, but that got dropped long ago. Reviewed-by: Geert Uytterhoeven Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/net/Space.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/net') diff --git a/include/net/Space.h b/include/net/Space.h index e30e7a70ea99..93fd6caa4bad 100644 --- a/include/net/Space.h +++ b/include/net/Space.h @@ -8,16 +8,10 @@ struct net_device *ultra_probe(int unit); struct net_device *wd_probe(int unit); struct net_device *ne_probe(int unit); struct net_device *fmv18x_probe(int unit); -struct net_device *i82596_probe(int unit); struct net_device *ni65_probe(int unit); struct net_device *sonic_probe(int unit); struct net_device *smc_init(int unit); -struct net_device *atarilance_probe(int unit); -struct net_device *sun3lance_probe(int unit); -struct net_device *sun3_82586_probe(int unit); -struct net_device *apne_probe(int unit); struct net_device *cs89x0_probe(int unit); -struct net_device *mvme147lance_probe(int unit); struct net_device *tc515_probe(int unit); struct net_device *lance_probe(int unit); struct net_device *cops_probe(int unit); -- cgit v1.2.3 From 375df5f8c1812c59930cfed14ff4cc15929c8f2f Mon Sep 17 00:00:00 2001 From: Michael Schmitz Date: Tue, 3 Aug 2021 13:40:44 +0200 Subject: ax88796: export ax_NS8390_init() hook The block I/O code for the new X-Surf 100 ax88796 driver needs ax_NS8390_init() for error fixup in its block_output function. Export this static function through the ax_NS8390_reinit() wrapper so we can lose the lib8380.c include in the X-Surf 100 driver. [arnd: add the declaration in the header to avoid a -Wmissing-prototypes warning] Fixes: 861928f4e60e826c ("net-next: New ax88796 platform driver for Amiga X-Surf 100 Zorro board (m68k)") Signed-off-by: Michael Schmitz Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/net/ax88796.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/ax88796.h b/include/net/ax88796.h index aa52b2e8ff7b..2ed23a368602 100644 --- a/include/net/ax88796.h +++ b/include/net/ax88796.h @@ -38,4 +38,7 @@ struct ax_plat_data { int (*check_irq)(struct platform_device *pdev); }; +/* exported from ax88796.c for xsurf100.c */ +extern void ax_NS8390_reinit(struct net_device *dev); + #endif /* __NET_AX88796_PLAT_H */ -- cgit v1.2.3 From 72bcad5393a7079706fcfe02d84ed1599716d6a2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 3 Aug 2021 13:40:49 +0200 Subject: wan: remove sbni/granch driver The driver was merged in 1999 and has only ever seen treewide cleanups since then, with no indication whatsoever that anyone has actually had access to hardware for testing the patches. >From the information in the link below, it appears that the hardware is for some leased line system in Russia that has since been discontinued, and useless without any remote end to connect to. As the driver still feels like a Linux-2.2 era artifact today, it appears that the best way forward is to just delete it. Link: https://www.tms.ru/%D0%90%D0%B4%D0%B0%D0%BF%D1%82%D0%B5%D1%80_%D0%B4%D0%BB%D1%8F_%D0%B2%D1%8B%D0%B4%D0%B5%D0%BB%D0%B5%D0%BD%D0%BD%D1%8B%D1%85_%D0%BB%D0%B8%D0%BD%D0%B8%D0%B9_Granch_SBNI12-10 Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/net/Space.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/net') diff --git a/include/net/Space.h b/include/net/Space.h index 93fd6caa4bad..08ca9cef0213 100644 --- a/include/net/Space.h +++ b/include/net/Space.h @@ -18,6 +18,3 @@ struct net_device *cops_probe(int unit); /* Fibre Channel adapters */ int iph5526_probe(struct net_device *dev); - -/* SBNI adapters */ -int sbni_probe(int unit); -- cgit v1.2.3 From 102793136ce9dacae77c2d1f8dd6878b480f4b95 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 2 Aug 2021 16:56:18 -0700 Subject: Bluetooth: HCI: Add proper tracking for enable status of adv instances This adds a field to track if advertising instances are enabled or not and only clear HCI_LE_ADV flag if there is no instance left advertising. Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 4abe3c494002..b79b31359bf8 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -221,6 +221,7 @@ struct oob_data { struct adv_info { struct list_head list; + bool enabled; bool pending; __u8 instance; __u32 flags; -- cgit v1.2.3 From c45074d68a9b1e893d86520af71fab37693c3d7e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 2 Aug 2021 16:56:19 -0700 Subject: Bluetooth: Fix not generating RPA when required Code was checking if random_addr and hdev->rpa match without first checking if the RPA has not been set (BDADDR_ANY), furthermore it was clearing HCI_RPA_EXPIRED before the command completes and the RPA is actually programmed which in case of failure would leave the expired RPA still set. Since advertising instance have a similar problem the clearing of HCI_RPA_EXPIRED has been moved to hci_event.c after checking the random address is in fact the hdev->rap and then proceed to set the expire timeout. Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index b79b31359bf8..b011eeea28c3 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1413,6 +1413,10 @@ void hci_conn_del_sysfs(struct hci_conn *conn); !hci_dev_test_flag(dev, HCI_AUTO_OFF)) #define bredr_sc_enabled(dev) (lmp_sc_capable(dev) && \ hci_dev_test_flag(dev, HCI_SC_ENABLED)) +#define rpa_valid(dev) (bacmp(&dev->rpa, BDADDR_ANY) && \ + !hci_dev_test_flag(dev, HCI_RPA_EXPIRED)) +#define adv_rpa_valid(adv) (bacmp(&adv->random_addr, BDADDR_ANY) && \ + !adv->rpa_expired) #define scan_1m(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_1M) || \ ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_1M)) -- cgit v1.2.3 From 314001f0bf927015e459c9d387d62a231fe93af3 Mon Sep 17 00:00:00 2001 From: Rao Shoaib Date: Sun, 1 Aug 2021 00:57:07 -0700 Subject: af_unix: Add OOB support This patch adds OOB support for AF_UNIX sockets. The semantics is same as TCP. The last byte of a message with the OOB flag is treated as the OOB byte. The byte is separated into a skb and a pointer to the skb is stored in unix_sock. The pointer is used to enforce OOB semantics. Signed-off-by: Rao Shoaib Signed-off-by: David S. Miller --- include/net/af_unix.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 435a2c3d5a6f..4757d7f53f13 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -70,6 +70,9 @@ struct unix_sock { struct socket_wq peer_wq; wait_queue_entry_t peer_wake; struct scm_stat scm_stat; +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + struct sk_buff *oob_skb; +#endif }; static inline struct unix_sock *unix_sk(const struct sock *sk) -- cgit v1.2.3 From 8679c31e0284aa3aaba038035e443180b5bacb99 Mon Sep 17 00:00:00 2001 From: Rocco Yue Date: Tue, 3 Aug 2021 20:02:50 +0800 Subject: net: add extack arg for link ops Pass extack arg to validate_linkmsg and validate_link_af callbacks. If a netlink attribute has a reject_message, use the extended ack mechanism to carry the message back to user space. Signed-off-by: Rocco Yue Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 384e800665f2..9f48733bfd21 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -153,7 +153,8 @@ struct rtnl_af_ops { u32 ext_filter_mask); int (*validate_link_af)(const struct net_device *dev, - const struct nlattr *attr); + const struct nlattr *attr, + struct netlink_ext_ack *extack); int (*set_link_af)(struct net_device *dev, const struct nlattr *attr, struct netlink_ext_ack *extack); -- cgit v1.2.3 From 957e2235e5264c97cd6be8e2e17f2e11b41f2239 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 3 Aug 2021 23:34:08 +0300 Subject: net: make switchdev_bridge_port_{,unoffload} loosely coupled with the bridge With the introduction of explicit offloading API in switchdev in commit 2f5dc00f7a3e ("net: bridge: switchdev: let drivers inform which bridge ports are offloaded"), we started having Ethernet switch drivers calling directly into a function exported by net/bridge/br_switchdev.c, which is a function exported by the bridge driver. This means that drivers that did not have an explicit dependency on the bridge before, like cpsw and am65-cpsw, now do - otherwise it is not possible to call a symbol exported by a driver that can be built as module unless you are a module too. There was an attempt to solve the dependency issue in the form of commit b0e81817629a ("net: build all switchdev drivers as modules when the bridge is a module"). Grygorii Strashko, however, says about it: | In my opinion, the problem is a bit bigger here than just fixing the | build :( | | In case, of ^cpsw the switchdev mode is kinda optional and in many | cases (especially for testing purposes, NFS) the multi-mac mode is | still preferable mode. | | There were no such tight dependency between switchdev drivers and | bridge core before and switchdev serviced as independent, notification | based layer between them, so ^cpsw still can be "Y" and bridge can be | "M". Now for mostly every kernel build configuration the CONFIG_BRIDGE | will need to be set as "Y", or we will have to update drivers to | support build with BRIDGE=n and maintain separate builds for | networking vs non-networking testing. But is this enough? Wouldn't | it cause 'chain reaction' required to add more and more "Y" options | (like CONFIG_VLAN_8021Q)? | | PS. Just to be sure we on the same page - ARM builds will be forced | (with this patch) to have CONFIG_TI_CPSW_SWITCHDEV=m and so all our | automation testing will just fail with omap2plus_defconfig. In the light of this, it would be desirable for some configurations to avoid dependencies between switchdev drivers and the bridge, and have the switchdev mode as completely optional within the driver. Arnd Bergmann also tried to write a patch which better expressed the build time dependency for Ethernet switch drivers where the switchdev support is optional, like cpsw/am65-cpsw, and this made the drivers follow the bridge (compile as module if the bridge is a module) only if the optional switchdev support in the driver was enabled in the first place: https://patchwork.kernel.org/project/netdevbpf/patch/20210802144813.1152762-1-arnd@kernel.org/ but this still did not solve the fact that cpsw and am65-cpsw now must be built as modules when the bridge is a module - it just expressed correctly that optional dependency. But the new behavior is an apparent regression from Grygorii's perspective. So to support the use case where the Ethernet driver is built-in, NET_SWITCHDEV (a bool option) is enabled, and the bridge is a module, we need a framework that can handle the possible absence of the bridge from the running system, i.e. runtime bloatware as opposed to build-time bloatware. Luckily we already have this framework, since switchdev has been using it extensively. Events from the bridge side are transmitted to the driver side using notifier chains - this was originally done so that unrelated drivers could snoop for events emitted by the bridge towards ports that are implemented by other drivers (think of a switch driver with LAG offload that listens for switchdev events on a bonding/team interface that it offloads). There are also events which are transmitted from the driver side to the bridge side, which again are modeled using notifiers. SWITCHDEV_FDB_ADD_TO_BRIDGE is an example of this, and deals with notifying the bridge that a MAC address has been dynamically learned. So there is a precedent we can use for modeling the new framework. The difference compared to SWITCHDEV_FDB_ADD_TO_BRIDGE is that the work that the bridge needs to do when a port becomes offloaded is blocking in its nature: replay VLANs, MDBs etc. The calling context is indeed blocking (we are under rtnl_mutex), but the existing switchdev notification chain that the bridge is subscribed to is only the atomic one. So we need to subscribe the bridge to the blocking switchdev notification chain too. This patch: - keeps the driver-side perception of the switchdev_bridge_port_{,un}offload unchanged - moves the implementation of switchdev_bridge_port_{,un}offload from the bridge module into the switchdev module. - makes everybody that is subscribed to the switchdev blocking notifier chain "hear" offload & unoffload events - makes the bridge driver subscribe and handle those events - moves the bridge driver's handling of those events into 2 new functions called br_switchdev_port_{,un}offload. These functions contain in fact the core of the logic that was previously in switchdev_bridge_port_{,un}offload, just that now we go through an extra indirection layer to reach them. Unlike all the other switchdev notification structures, the structure used to carry the bridge port information, struct switchdev_notifier_brport_info, does not contain a "bool handled". This is because in the current usage pattern, we always know that a switchdev bridge port offloading event will be handled by the bridge, because the switchdev_bridge_port_offload() call was initiated by a NETDEV_CHANGEUPPER event in the first place, where info->upper_dev is a bridge. So if the bridge wasn't loaded, then the CHANGEUPPER event couldn't have happened. Signed-off-by: Vladimir Oltean Tested-by: Grygorii Strashko Signed-off-by: David S. Miller --- include/net/switchdev.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 66468ff8cc0a..60d806b6a5ae 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -180,6 +180,14 @@ struct switchdev_obj_in_state_mrp { typedef int switchdev_obj_dump_cb_t(struct switchdev_obj *obj); +struct switchdev_brport { + struct net_device *dev; + const void *ctx; + struct notifier_block *atomic_nb; + struct notifier_block *blocking_nb; + bool tx_fwd_offload; +}; + enum switchdev_notifier_type { SWITCHDEV_FDB_ADD_TO_BRIDGE = 1, SWITCHDEV_FDB_DEL_TO_BRIDGE, @@ -197,6 +205,9 @@ enum switchdev_notifier_type { SWITCHDEV_VXLAN_FDB_ADD_TO_DEVICE, SWITCHDEV_VXLAN_FDB_DEL_TO_DEVICE, SWITCHDEV_VXLAN_FDB_OFFLOADED, + + SWITCHDEV_BRPORT_OFFLOADED, + SWITCHDEV_BRPORT_UNOFFLOADED, }; struct switchdev_notifier_info { @@ -226,6 +237,11 @@ struct switchdev_notifier_port_attr_info { bool handled; }; +struct switchdev_notifier_brport_info { + struct switchdev_notifier_info info; /* must be first */ + const struct switchdev_brport brport; +}; + static inline struct net_device * switchdev_notifier_info_to_dev(const struct switchdev_notifier_info *info) { @@ -246,6 +262,17 @@ switchdev_fdb_is_dynamically_learned(const struct switchdev_notifier_fdb_info *f #ifdef CONFIG_NET_SWITCHDEV +int switchdev_bridge_port_offload(struct net_device *brport_dev, + struct net_device *dev, const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb, + bool tx_fwd_offload, + struct netlink_ext_ack *extack); +void switchdev_bridge_port_unoffload(struct net_device *brport_dev, + const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb); + void switchdev_deferred_process(void); int switchdev_port_attr_set(struct net_device *dev, const struct switchdev_attr *attr, @@ -316,6 +343,25 @@ int switchdev_handle_port_attr_set(struct net_device *dev, struct netlink_ext_ack *extack)); #else +static inline int +switchdev_bridge_port_offload(struct net_device *brport_dev, + struct net_device *dev, const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb, + bool tx_fwd_offload, + struct netlink_ext_ack *extack) +{ + return -EOPNOTSUPP; +} + +static inline void +switchdev_bridge_port_unoffload(struct net_device *brport_dev, + const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb) +{ +} + static inline void switchdev_deferred_process(void) { } -- cgit v1.2.3 From 04190bf8944deb7e3ac165a1a494db23aa0160a9 Mon Sep 17 00:00:00 2001 From: Pavel Tikhomirov Date: Wed, 4 Aug 2021 10:55:56 +0300 Subject: sock: allow reading and changing sk_userlocks with setsockopt SOCK_SNDBUF_LOCK and SOCK_RCVBUF_LOCK flags disable automatic socket buffers adjustment done by kernel (see tcp_fixup_rcvbuf() and tcp_sndbuf_expand()). If we've just created a new socket this adjustment is enabled on it, but if one changes the socket buffer size by setsockopt(SO_{SND,RCV}BUF*) it becomes disabled. CRIU needs to call setsockopt(SO_{SND,RCV}BUF*) on each socket on restore as it first needs to increase buffer sizes for packet queues restore and second it needs to restore back original buffer sizes. So after CRIU restore all sockets become non-auto-adjustable, which can decrease network performance of restored applications significantly. CRIU need to be able to restore sockets with enabled/disabled adjustment to the same state it was before dump, so let's add special setsockopt for it. Let's also export SOCK_SNDBUF_LOCK and SOCK_RCVBUF_LOCK flags to uAPI so that using these interface one can reenable automatic socket buffer adjustment on their sockets. Signed-off-by: Pavel Tikhomirov Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index ff1be7e7e90b..6e761451c927 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -68,6 +68,7 @@ #include #include #include +#include /* * This structure really needs to be cleaned up. @@ -1438,8 +1439,6 @@ static inline int __sk_prot_rehash(struct sock *sk) #define RCV_SHUTDOWN 1 #define SEND_SHUTDOWN 2 -#define SOCK_SNDBUF_LOCK 1 -#define SOCK_RCVBUF_LOCK 2 #define SOCK_BINDADDR_LOCK 4 #define SOCK_BINDPORT_LOCK 8 -- cgit v1.2.3 From db243b796439c0caba47865564d8acd18a301d18 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 4 Aug 2021 15:45:36 -0500 Subject: net/ipv4/ipv6: Replace one-element arraya with flexible-array members MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. Use an anonymous union with a couple of anonymous structs in order to keep userspace unchanged and refactor the related code accordingly: $ pahole -C group_filter net/ipv4/ip_sockglue.o struct group_filter { union { struct { __u32 gf_interface_aux; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ struct __kernel_sockaddr_storage gf_group_aux; /* 8 128 */ /* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */ __u32 gf_fmode_aux; /* 136 4 */ __u32 gf_numsrc_aux; /* 140 4 */ struct __kernel_sockaddr_storage gf_slist[1]; /* 144 128 */ }; /* 0 272 */ struct { __u32 gf_interface; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ struct __kernel_sockaddr_storage gf_group; /* 8 128 */ /* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */ __u32 gf_fmode; /* 136 4 */ __u32 gf_numsrc; /* 140 4 */ struct __kernel_sockaddr_storage gf_slist_flex[0]; /* 144 0 */ }; /* 0 144 */ }; /* 0 272 */ /* size: 272, cachelines: 5, members: 1 */ /* last cacheline: 16 bytes */ }; $ pahole -C compat_group_filter net/ipv4/ip_sockglue.o struct compat_group_filter { union { struct { __u32 gf_interface_aux; /* 0 4 */ struct __kernel_sockaddr_storage gf_group_aux __attribute__((__aligned__(4))); /* 4 128 */ /* --- cacheline 2 boundary (128 bytes) was 4 bytes ago --- */ __u32 gf_fmode_aux; /* 132 4 */ __u32 gf_numsrc_aux; /* 136 4 */ struct __kernel_sockaddr_storage gf_slist[1] __attribute__((__aligned__(4))); /* 140 128 */ } __attribute__((__packed__)) __attribute__((__aligned__(4))); /* 0 268 */ struct { __u32 gf_interface; /* 0 4 */ struct __kernel_sockaddr_storage gf_group __attribute__((__aligned__(4))); /* 4 128 */ /* --- cacheline 2 boundary (128 bytes) was 4 bytes ago --- */ __u32 gf_fmode; /* 132 4 */ __u32 gf_numsrc; /* 136 4 */ struct __kernel_sockaddr_storage gf_slist_flex[0] __attribute__((__aligned__(4))); /* 140 0 */ } __attribute__((__packed__)) __attribute__((__aligned__(4))); /* 0 140 */ } __attribute__((__aligned__(1))); /* 0 268 */ /* size: 268, cachelines: 5, members: 1 */ /* forced alignments: 1 */ /* last cacheline: 12 bytes */ } __attribute__((__packed__)); This helps with the ongoing efforts to globally enable -Warray-bounds and get us closer to being able to tighten the FORTIFY_SOURCE routines on memcpy(). [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.10/process/deprecated.html#zero-length-and-one-element-arrays Link: https://github.com/KSPP/linux/issues/79 Link: https://github.com/KSPP/linux/issues/109 Signed-off-by: Gustavo A. R. Silva Signed-off-by: David S. Miller --- include/net/compat.h | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/compat.h b/include/net/compat.h index 84805bdc4435..595fee069b82 100644 --- a/include/net/compat.h +++ b/include/net/compat.h @@ -71,13 +71,26 @@ struct compat_group_source_req { } __packed; struct compat_group_filter { - __u32 gf_interface; - struct __kernel_sockaddr_storage gf_group - __aligned(4); - __u32 gf_fmode; - __u32 gf_numsrc; - struct __kernel_sockaddr_storage gf_slist[1] - __aligned(4); + union { + struct { + __u32 gf_interface_aux; + struct __kernel_sockaddr_storage gf_group_aux + __aligned(4); + __u32 gf_fmode_aux; + __u32 gf_numsrc_aux; + struct __kernel_sockaddr_storage gf_slist[1] + __aligned(4); + } __packed; + struct { + __u32 gf_interface; + struct __kernel_sockaddr_storage gf_group + __aligned(4); + __u32 gf_fmode; + __u32 gf_numsrc; + struct __kernel_sockaddr_storage gf_slist_flex[] + __aligned(4); + } __packed; + }; } __packed; #endif /* NET_COMPAT_H */ -- cgit v1.2.3 From e11c0e258c1a87e478a7a44e7c3d9e8ea4b8438c Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 4 Aug 2021 16:43:52 -0500 Subject: net/ipv6/mcast: Use struct_size() helper Replace IP6_SFLSIZE() with struct_size() helper in order to avoid any potential type mistakes or integer overflows that, in the worst scenario, could lead to heap overflows. Signed-off-by: Gustavo A. R. Silva Signed-off-by: David S. Miller --- include/net/if_inet6.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/net') diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index 71bb4cc4d05d..42235c178b06 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -82,9 +82,6 @@ struct ip6_sf_socklist { struct in6_addr sl_addr[]; }; -#define IP6_SFLSIZE(count) (sizeof(struct ip6_sf_socklist) + \ - (count) * sizeof(struct in6_addr)) - #define IP6_SFBLOCK 10 /* allocate this many at once */ struct ipv6_mc_socklist { -- cgit v1.2.3 From 6ec566131de0e330fffd33753b325fb4d6b2d00e Mon Sep 17 00:00:00 2001 From: Tedd Ho-Jeong An Date: Wed, 4 Aug 2021 17:32:08 -0700 Subject: Bluetooth: Add support hdev to allocate private data This patch adds support hdev to allocate extra size for private data. The size of private data is specified in the hdev_alloc_size(priv_size) and the allocated buffer can be accessed with hci_get_priv(hdev). Signed-off-by: Tedd Ho-Jeong An Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index b011eeea28c3..a7d06d7da602 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1224,10 +1224,21 @@ static inline void hci_set_drvdata(struct hci_dev *hdev, void *data) dev_set_drvdata(&hdev->dev, data); } +static inline void *hci_get_priv(struct hci_dev *hdev) +{ + return (char *)hdev + sizeof(*hdev); +} + struct hci_dev *hci_dev_get(int index); struct hci_dev *hci_get_route(bdaddr_t *dst, bdaddr_t *src, u8 src_type); -struct hci_dev *hci_alloc_dev(void); +struct hci_dev *hci_alloc_dev_priv(int sizeof_priv); + +static inline struct hci_dev *hci_alloc_dev(void) +{ + return hci_alloc_dev_priv(0); +} + void hci_free_dev(struct hci_dev *hdev); int hci_register_dev(struct hci_dev *hdev); void hci_unregister_dev(struct hci_dev *hdev); -- cgit v1.2.3 From e04480920d1eec9c061841399aa6f35b6f987d8b Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Wed, 4 Aug 2021 19:26:56 +0900 Subject: Bluetooth: defer cleanup of resources in hci_unregister_dev() syzbot is hitting might_sleep() warning at hci_sock_dev_event() due to calling lock_sock() with rw spinlock held [1]. It seems that history of this locking problem is a trial and error. Commit b40df5743ee8 ("[PATCH] bluetooth: fix socket locking in hci_sock_dev_event()") in 2.6.21-rc4 changed bh_lock_sock() to lock_sock() as an attempt to fix lockdep warning. Then, commit 4ce61d1c7a8e ("[BLUETOOTH]: Fix locking in hci_sock_dev_event().") in 2.6.22-rc2 changed lock_sock() to local_bh_disable() + bh_lock_sock_nested() as an attempt to fix the sleep in atomic context warning. Then, commit 4b5dd696f81b ("Bluetooth: Remove local_bh_disable() from hci_sock.c") in 3.3-rc1 removed local_bh_disable(). Then, commit e305509e678b ("Bluetooth: use correct lock to prevent UAF of hdev object") in 5.13-rc5 again changed bh_lock_sock_nested() to lock_sock() as an attempt to fix CVE-2021-3573. This difficulty comes from current implementation that hci_sock_dev_event(HCI_DEV_UNREG) is responsible for dropping all references from sockets because hci_unregister_dev() immediately reclaims resources as soon as returning from hci_sock_dev_event(HCI_DEV_UNREG). But the history suggests that hci_sock_dev_event(HCI_DEV_UNREG) was not doing what it should do. Therefore, instead of trying to detach sockets from device, let's accept not detaching sockets from device at hci_sock_dev_event(HCI_DEV_UNREG), by moving actual cleanup of resources from hci_unregister_dev() to hci_cleanup_dev() which is called by bt_host_release() when all references to this unregistered device (which is a kobject) are gone. Since hci_sock_dev_event(HCI_DEV_UNREG) no longer resets hci_pi(sk)->hdev, we need to check whether this device was unregistered and return an error based on HCI_UNREGISTER flag. There might be subtle behavioral difference in "monitor the hdev" functionality; please report if you found something went wrong due to this patch. Link: https://syzkaller.appspot.com/bug?extid=a5df189917e79d5e59c9 [1] Reported-by: syzbot Suggested-by: Linus Torvalds Signed-off-by: Tetsuo Handa Fixes: e305509e678b ("Bluetooth: use correct lock to prevent UAF of hdev object") Acked-by: Luiz Augusto von Dentz Signed-off-by: Linus Torvalds --- include/net/bluetooth/hci_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a53e94459ecd..db4312e44d47 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1230,6 +1230,7 @@ struct hci_dev *hci_alloc_dev(void); void hci_free_dev(struct hci_dev *hdev); int hci_register_dev(struct hci_dev *hdev); void hci_unregister_dev(struct hci_dev *hdev); +void hci_cleanup_dev(struct hci_dev *hdev); int hci_suspend_dev(struct hci_dev *hdev); int hci_resume_dev(struct hci_dev *hdev); int hci_reset_dev(struct hci_dev *hdev); -- cgit v1.2.3 From c73c57081b3d59aa99093fbedced32ea02620cd3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 6 Aug 2021 03:20:08 +0300 Subject: net: dsa: don't disable multicast flooding to the CPU even without an IGMP querier Commit 08cc83cc7fd8 ("net: dsa: add support for BRIDGE_MROUTER attribute") added an option for users to turn off multicast flooding towards the CPU if they turn off the IGMP querier on a bridge which already has enslaved ports (echo 0 > /sys/class/net/br0/bridge/multicast_router). And commit a8b659e7ff75 ("net: dsa: act as passthrough for bridge port flags") simply papered over that issue, because it moved the decision to flood the CPU with multicast (or not) from the DSA core down to individual drivers, instead of taking a more radical position then. The truth is that disabling multicast flooding to the CPU is simply something we are not prepared to do now, if at all. Some reasons: - ICMP6 neighbor solicitation messages are unregistered multicast packets as far as the bridge is concerned. So if we stop flooding multicast, the outside world cannot ping the bridge device's IPv6 link-local address. - There might be foreign interfaces bridged with our DSA switch ports (sending a packet towards the host does not necessarily equal termination, but maybe software forwarding). So if there is no one interested in that multicast traffic in the local network stack, that doesn't mean nobody is. - PTP over L4 (IPv4, IPv6) is multicast, but is unregistered as far as the bridge is concerned. This should reach the CPU port. - The switch driver might not do FDB partitioning. And since we don't even bother to do more fine-grained flood disabling (such as "disable flooding _from_port_N_ towards the CPU port" as opposed to "disable flooding _from_any_port_ towards the CPU port"), this breaks standalone ports, or even multiple bridges where one has an IGMP querier and one doesn't. Reverting the logic makes all of the above work. Fixes: a8b659e7ff75 ("net: dsa: act as passthrough for bridge port flags") Fixes: 08cc83cc7fd8 ("net: dsa: add support for BRIDGE_MROUTER attribute") Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 7cc9507282d3..d7dc26d316ea 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -714,8 +714,6 @@ struct dsa_switch_ops { int (*port_bridge_flags)(struct dsa_switch *ds, int port, struct switchdev_brport_flags flags, struct netlink_ext_ack *extack); - int (*port_set_mrouter)(struct dsa_switch *ds, int port, bool mrouter, - struct netlink_ext_ack *extack); /* * VLAN support -- cgit v1.2.3 From 4592ee7f525c4683ec9e290381601fdee50ae110 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 4 Aug 2021 15:02:15 +0200 Subject: netfilter: conntrack: remove offload_pickup sysctl again These two sysctls were added because the hardcoded defaults (2 minutes, tcp, 30 seconds, udp) turned out to be too low for some setups. They appeared in 5.14-rc1 so it should be fine to remove it again. Marcelo convinced me that there should be no difference between a flow that was offloaded vs. a flow that was not wrt. timeout handling. Thus the default is changed to those for TCP established and UDP stream, 5 days and 120 seconds, respectively. Marcelo also suggested to account for the timeout value used for the offloading, this avoids increase beyond the value in the conntrack-sysctl and will also instantly expire the conntrack entry with altered sysctls. Example: nf_conntrack_udp_timeout_stream=60 nf_flowtable_udp_timeout=60 This will remove offloaded udp flows after one minute, rather than two. An earlier version of this patch also cleared the ASSURED bit to allow nf_conntrack to evict the entry via early_drop (i.e., table full). However, it looks like we can safely assume that connection timed out via HW is still in established state, so this isn't needed. Quoting Oz: [..] the hardware sends all packets with a set FIN flags to sw. [..] Connections that are aged in hardware are expected to be in the established state. In case it turns out that back-to-sw-path transition can occur for 'dodgy' connections too (e.g., one side disappeared while software-path would have been in RETRANS timeout), we can adjust this later. Cc: Oz Shlomo Cc: Paul Blakey Suggested-by: Marcelo Ricardo Leitner Signed-off-by: Florian Westphal Reviewed-by: Marcelo Ricardo Leitner Reviewed-by: Oz Shlomo Signed-off-by: Pablo Neira Ayuso --- include/net/netns/conntrack.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 37e5300c7e5a..fefd38db95b3 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -30,7 +30,6 @@ struct nf_tcp_net { u8 tcp_ignore_invalid_rst; #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) unsigned int offload_timeout; - unsigned int offload_pickup; #endif }; @@ -44,7 +43,6 @@ struct nf_udp_net { unsigned int timeouts[UDP_CT_MAX]; #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) unsigned int offload_timeout; - unsigned int offload_pickup; #endif }; -- cgit v1.2.3 From 82564f6c706a37e5f7dec962375581cc9f8fca5d Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sun, 8 Aug 2021 14:41:21 +0300 Subject: devlink: Simplify devlink port API calls Devlink port already has pointer to the devlink instance and all API calls that forward these devlink ports to the drivers perform same "devlink_port->devlink" assignment before actual call. This patch removes useless parameter and allows us in the future to create specific devlink_port_ops to manage user space access with reliable ops assignment. Signed-off-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 08f4c6191e72..ccbfb3a844aa 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1396,8 +1396,8 @@ struct devlink_ops { * * Note: @extack can be NULL when port notifier queries the port function. */ - int (*port_function_hw_addr_get)(struct devlink *devlink, struct devlink_port *port, - u8 *hw_addr, int *hw_addr_len, + int (*port_function_hw_addr_get)(struct devlink_port *port, u8 *hw_addr, + int *hw_addr_len, struct netlink_ext_ack *extack); /** * @port_function_hw_addr_set: Port function's hardware address set function. @@ -1406,7 +1406,7 @@ struct devlink_ops { * by the devlink port. Driver should return -EOPNOTSUPP if it doesn't support port * function handling for a particular port. */ - int (*port_function_hw_addr_set)(struct devlink *devlink, struct devlink_port *port, + int (*port_function_hw_addr_set)(struct devlink_port *port, const u8 *hw_addr, int hw_addr_len, struct netlink_ext_ack *extack); /** @@ -1462,8 +1462,7 @@ struct devlink_ops { * * Return: 0 on success, negative value otherwise. */ - int (*port_fn_state_get)(struct devlink *devlink, - struct devlink_port *port, + int (*port_fn_state_get)(struct devlink_port *port, enum devlink_port_fn_state *state, enum devlink_port_fn_opstate *opstate, struct netlink_ext_ack *extack); @@ -1478,8 +1477,7 @@ struct devlink_ops { * * Return: 0 on success, negative value otherwise. */ - int (*port_fn_state_set)(struct devlink *devlink, - struct devlink_port *port, + int (*port_fn_state_set)(struct devlink_port *port, enum devlink_port_fn_state state, struct netlink_ext_ack *extack); -- cgit v1.2.3 From 045c45d1f598c65806f885b59f6fbc4cebb62b15 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 8 Aug 2021 17:35:23 +0300 Subject: net: dsa: centralize fast ageing when address learning is turned off Currently DSA leaves it down to device drivers to fast age the FDB on a port when address learning is disabled on it. There are 2 reasons for doing that in the first place: - when address learning is disabled by user space, through IFLA_BRPORT_LEARNING or the brport_attr_learning sysfs, what user space typically wants to achieve is to operate in a mode with no dynamic FDB entry on that port. But if the port is already up, some addresses might have been already learned on it, and it seems silly to wait for 5 minutes for them to expire until something useful can be done. - when a port leaves a bridge and becomes standalone, DSA turns off address learning on it. This also has the nice side effect of flushing the dynamically learned bridge FDB entries on it, which is a good idea because standalone ports should not have bridge FDB entries on them. We let drivers manage fast ageing under this condition because if DSA were to do it, it would need to track each port's learning state, and act upon the transition, which it currently doesn't. But there are 2 reasons why doing it is better after all: - drivers might get it wrong and not do it (see b53_port_set_learning) - we would like to flush the dynamic entries from the software bridge too, and letting drivers do that would be another pain point So track the port learning state and trigger a fast age process automatically within DSA. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index d7dc26d316ea..995e9d3f9cfc 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -254,6 +254,7 @@ struct dsa_port { struct device_node *dn; unsigned int ageing_time; bool vlan_filtering; + bool learning; u8 stp_state; struct net_device *bridge_dev; int bridge_num; -- cgit v1.2.3 From 5313a37b881e57767bc37185bef2873862be8d47 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 8 Aug 2021 17:35:26 +0300 Subject: net: dsa: sja1105: rely on DSA core tracking of port learning state Now that DSA keeps track of the port learning state, it becomes superfluous to keep an additional variable with this information in the sja1105 driver. Remove it. The DSA core's learning state is present in struct dsa_port *dp. To avoid the antipattern where we iterate through a DSA switch's ports and then call dsa_to_port to obtain the "dp" reference (which is bad because dsa_to_port iterates through the DSA switch tree once again), just iterate through the dst->ports and operate on those directly. The sja1105 had an extra use of priv->learn_ena on non-user ports. DSA does not touch the learning state of those ports - drivers are free to do what they wish on them. Mark that information with a comment in struct dsa_port and let sja1105 set dp->learning for cascade ports. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 995e9d3f9cfc..0c2cba45fa79 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -254,6 +254,7 @@ struct dsa_port { struct device_node *dn; unsigned int ageing_time; bool vlan_filtering; + /* Managed by DSA on user ports and by drivers on CPU and DSA ports */ bool learning; u8 stp_state; struct net_device *bridge_dev; -- cgit v1.2.3 From 919d13a7e455c2e7676042d7a5f94c164e859d8a Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sun, 8 Aug 2021 21:57:43 +0300 Subject: devlink: Set device as early as possible All kernel devlink implementations call to devlink_alloc() during initialization routine for specific device which is used later as a parent device for devlink_register(). Such late device assignment causes to the situation which requires us to call to device_register() before setting other parameters, but that call opens devlink to the world and makes accessible for the netlink users. Any attempt to move devlink_register() to be the last call generates the following error due to access to the devlink->dev pointer. [ 8.758862] devlink_nl_param_fill+0x2e8/0xe50 [ 8.760305] devlink_param_notify+0x6d/0x180 [ 8.760435] __devlink_params_register+0x2f1/0x670 [ 8.760558] devlink_params_register+0x1e/0x20 The simple change of API to set devlink device in the devlink_alloc() instead of devlink_register() fixes all this above and ensures that prior to call to devlink_register() everything already set. Signed-off-by: Leon Romanovsky Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/devlink.h | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index ccbfb3a844aa..0236c77f2fd0 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1544,13 +1544,15 @@ struct net *devlink_net(const struct devlink *devlink); * Drivers that operate on real HW must use devlink_alloc() instead. */ struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, - size_t priv_size, struct net *net); + size_t priv_size, struct net *net, + struct device *dev); static inline struct devlink *devlink_alloc(const struct devlink_ops *ops, - size_t priv_size) + size_t priv_size, + struct device *dev) { - return devlink_alloc_ns(ops, priv_size, &init_net); + return devlink_alloc_ns(ops, priv_size, &init_net, dev); } -int devlink_register(struct devlink *devlink, struct device *dev); +int devlink_register(struct devlink *devlink); void devlink_unregister(struct devlink *devlink); void devlink_reload_enable(struct devlink *devlink); void devlink_reload_disable(struct devlink *devlink); -- cgit v1.2.3 From 9e2ee5c7e7c35d195e2aa0692a7241d47a433d1e Mon Sep 17 00:00:00 2001 From: Jussi Maki Date: Sat, 31 Jul 2021 05:57:34 +0000 Subject: net, bonding: Add XDP support to the bonding driver XDP is implemented in the bonding driver by transparently delegating the XDP program loading, removal and xmit operations to the bonding slave devices. The overall goal of this work is that XDP programs can be attached to a bond device *without* any further changes (or awareness) necessary to the program itself, meaning the same XDP program can be attached to a native device but also a bonding device. Semantics of XDP_TX when attached to a bond are equivalent in such setting to the case when a tc/BPF program would be attached to the bond, meaning transmitting the packet out of the bond itself using one of the bond's configured xmit methods to select a slave device (rather than XDP_TX on the slave itself). Handling of XDP_TX to transmit using the configured bonding mechanism is therefore implemented by rewriting the BPF program return value in bpf_prog_run_xdp. To avoid performance impact this check is guarded by a static key, which is incremented when a XDP program is loaded onto a bond device. This approach was chosen to avoid changes to drivers implementing XDP. If the slave device does not match the receive device, then XDP_REDIRECT is transparently used to perform the redirection in order to have the network driver release the packet from its RX ring. The bonding driver hashing functions have been refactored to allow reuse with xdp_buff's to avoid code duplication. The motivation for this change is to enable use of bonding (and 802.3ad) in hairpinning L4 load-balancers such as [1] implemented with XDP and also to transparently support bond devices for projects that use XDP given most modern NICs have dual port adapters. An alternative to this approach would be to implement 802.3ad in user-space and implement the bonding load-balancing in the XDP program itself, but is rather a cumbersome endeavor in terms of slave device management (e.g. by watching netlink) and requires separate programs for native vs bond cases for the orchestrator. A native in-kernel implementation overcomes these issues and provides more flexibility. Below are benchmark results done on two machines with 100Gbit Intel E810 (ice) NIC and with 32-core 3970X on sending machine, and 16-core 3950X on receiving machine. 64 byte packets were sent with pktgen-dpdk at full rate. Two issues [2, 3] were identified with the ice driver, so the tests were performed with iommu=off and patch [2] applied. Additionally the bonding round robin algorithm was modified to use per-cpu tx counters as high CPU load (50% vs 10%) and high rate of cache misses were caused by the shared rr_tx_counter (see patch 2/3). The statistics were collected using "sar -n dev -u 1 10". On top of that, for ice, further work is in progress on improving the XDP_TX numbers [4]. -----------------------| CPU |--| rxpck/s |--| txpck/s |---- without patch (1 dev): XDP_DROP: 3.15% 48.6Mpps XDP_TX: 3.12% 18.3Mpps 18.3Mpps XDP_DROP (RSS): 9.47% 116.5Mpps XDP_TX (RSS): 9.67% 25.3Mpps 24.2Mpps ----------------------- with patch, bond (1 dev): XDP_DROP: 3.14% 46.7Mpps XDP_TX: 3.15% 13.9Mpps 13.9Mpps XDP_DROP (RSS): 10.33% 117.2Mpps XDP_TX (RSS): 10.64% 25.1Mpps 24.0Mpps ----------------------- with patch, bond (2 devs): XDP_DROP: 6.27% 92.7Mpps XDP_TX: 6.26% 17.6Mpps 17.5Mpps XDP_DROP (RSS): 11.38% 117.2Mpps XDP_TX (RSS): 14.30% 28.7Mpps 27.4Mpps -------------------------------------------------------------- RSS: Receive Side Scaling, e.g. the packets were sent to a range of destination IPs. [1]: https://cilium.io/blog/2021/05/20/cilium-110#standalonelb [2]: https://lore.kernel.org/bpf/20210601113236.42651-1-maciej.fijalkowski@intel.com/T/#t [3]: https://lore.kernel.org/bpf/CAHn8xckNXci+X_Eb2WMv4uVYjO2331UWB2JLtXr_58z0Av8+8A@mail.gmail.com/ [4]: https://lore.kernel.org/bpf/20210805230046.28715-1-maciej.fijalkowski@intel.com/T/#t Signed-off-by: Jussi Maki Signed-off-by: Daniel Borkmann Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Cc: Maciej Fijalkowski Cc: Magnus Karlsson Link: https://lore.kernel.org/bpf/20210731055738.16820-4-joamaki@gmail.com --- include/net/bonding.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/bonding.h b/include/net/bonding.h index 625d9c72dee3..b91c365e4e95 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -258,6 +258,7 @@ struct bonding { /* protecting ipsec_list */ spinlock_t ipsec_lock; #endif /* CONFIG_XFRM_OFFLOAD */ + struct bpf_prog *xdp_prog; }; #define bond_slave_get_rcu(dev) \ -- cgit v1.2.3 From beb7f2de5728b0bd2140a652fa51f6ad85d159f7 Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Sun, 8 Aug 2021 09:52:42 +0300 Subject: psample: Add a fwd declaration for skbuff Without this there is a warning if source files include psample.h before skbuff.h or doesn't include it at all. Fixes: 6ae0a6286171 ("net: Introduce psample, a new genetlink channel for packet sampling") Signed-off-by: Roi Dayan Link: https://lore.kernel.org/r/20210808065242.1522535-1-roid@nvidia.com Signed-off-by: Jakub Kicinski --- include/net/psample.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/psample.h b/include/net/psample.h index e328c5127757..0509d2d6be67 100644 --- a/include/net/psample.h +++ b/include/net/psample.h @@ -31,6 +31,8 @@ struct psample_group *psample_group_get(struct net *net, u32 group_num); void psample_group_take(struct psample_group *group); void psample_group_put(struct psample_group *group); +struct sk_buff; + #if IS_ENABLED(CONFIG_PSAMPLE) void psample_sample_packet(struct psample_group *group, struct sk_buff *skb, -- cgit v1.2.3 From 57f05bc2ab2443b89c2e2562c05053bcc7d30e8b Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Fri, 6 Aug 2021 10:46:19 +0800 Subject: page_pool: keep pp info as long as page pool owns the page Currently, page->pp is cleared and set everytime the page is recycled, which is unnecessary. So only set the page->pp when the page is added to the page pool and only clear it when the page is released from the page pool. This is also a preparation to support allocating frag page in page pool. Reviewed-by: Ilias Apalodimas Signed-off-by: Yunsheng Lin Signed-off-by: Jakub Kicinski --- include/net/page_pool.h | 7 ------- 1 file changed, 7 deletions(-) (limited to 'include/net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index 3dd62dd73027..8d7744d1c7c1 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -253,11 +253,4 @@ static inline void page_pool_ring_unlock(struct page_pool *pool) spin_unlock_bh(&pool->ring.producer_lock); } -/* Store mem_info on struct page and use it while recycling skb frags */ -static inline -void page_pool_store_mem_info(struct page *page, struct page_pool *pp) -{ - page->pp = pp; -} - #endif /* _NET_PAGE_POOL_H */ -- cgit v1.2.3 From 0e9d2a0a3a836c37528899010e73b5be8111753e Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Fri, 6 Aug 2021 10:46:20 +0800 Subject: page_pool: add interface to manipulate frag count in page pool For 32 bit systems with 64 bit dma, dma_addr[1] is used to store the upper 32 bit dma addr, those system should be rare those days. For normal system, the dma_addr[1] in 'struct page' is not used, so we can reuse dma_addr[1] for storing frag count, which means how many frags this page might be splited to. In order to simplify the page frag support in the page pool, the PAGE_POOL_DMA_USE_PP_FRAG_COUNT macro is added to indicate the 32 bit systems with 64 bit dma, and the page frag support in page pool is disabled for such system. The newly added page_pool_set_frag_count() is called to reserve the maximum frag count before any page frag is passed to the user. The page_pool_atomic_sub_frag_count_return() is called when user is done with the page frag. Signed-off-by: Yunsheng Lin Signed-off-by: Jakub Kicinski --- include/net/page_pool.h | 46 +++++++++++++++++++++++++++++++++++++++------- 1 file changed, 39 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index 8d7744d1c7c1..42e6997e637d 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -45,7 +45,10 @@ * Please note DMA-sync-for-CPU is still * device driver responsibility */ -#define PP_FLAG_ALL (PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV) +#define PP_FLAG_PAGE_FRAG BIT(2) /* for page frag feature */ +#define PP_FLAG_ALL (PP_FLAG_DMA_MAP |\ + PP_FLAG_DMA_SYNC_DEV |\ + PP_FLAG_PAGE_FRAG) /* * Fast allocation side cache array/stack @@ -198,19 +201,48 @@ static inline void page_pool_recycle_direct(struct page_pool *pool, page_pool_put_full_page(pool, page, true); } +#define PAGE_POOL_DMA_USE_PP_FRAG_COUNT \ + (sizeof(dma_addr_t) > sizeof(unsigned long)) + static inline dma_addr_t page_pool_get_dma_addr(struct page *page) { - dma_addr_t ret = page->dma_addr[0]; - if (sizeof(dma_addr_t) > sizeof(unsigned long)) - ret |= (dma_addr_t)page->dma_addr[1] << 16 << 16; + dma_addr_t ret = page->dma_addr; + + if (PAGE_POOL_DMA_USE_PP_FRAG_COUNT) + ret |= (dma_addr_t)page->dma_addr_upper << 16 << 16; + return ret; } static inline void page_pool_set_dma_addr(struct page *page, dma_addr_t addr) { - page->dma_addr[0] = addr; - if (sizeof(dma_addr_t) > sizeof(unsigned long)) - page->dma_addr[1] = upper_32_bits(addr); + page->dma_addr = addr; + if (PAGE_POOL_DMA_USE_PP_FRAG_COUNT) + page->dma_addr_upper = upper_32_bits(addr); +} + +static inline void page_pool_set_frag_count(struct page *page, long nr) +{ + atomic_long_set(&page->pp_frag_count, nr); +} + +static inline long page_pool_atomic_sub_frag_count_return(struct page *page, + long nr) +{ + long ret; + + /* As suggested by Alexander, atomic_long_read() may cover up the + * reference count errors, so avoid calling atomic_long_read() in + * the cases of freeing or draining the page_frags, where we would + * not expect it to match or that are slowpath anyway. + */ + if (__builtin_constant_p(nr) && + atomic_long_read(&page->pp_frag_count) == nr) + return 0; + + ret = atomic_long_sub_return(nr, &page->pp_frag_count); + WARN_ON(ret < 0); + return ret; } static inline bool is_page_pool_compiled_in(void) -- cgit v1.2.3 From 53e0961da1c7bbdabd1abebb20de403ec237ec09 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Fri, 6 Aug 2021 10:46:21 +0800 Subject: page_pool: add frag page recycling support in page pool Currently page pool only support page recycling when there is only one user of the page, and the split page reusing implemented in the most driver can not use the page pool as bing-pong way of reusing requires the multi user support in page pool. Those reusing or recycling has below limitations: 1. page from page pool can only be used be one user in order for the page recycling to happen. 2. Bing-pong way of reusing in most driver does not support multi desc using different part of the same page in order to save memory. So add multi-users support and frag page recycling in page pool to overcome the above limitation. Signed-off-by: Yunsheng Lin Signed-off-by: Jakub Kicinski --- include/net/page_pool.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include/net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index 42e6997e637d..a4082406a003 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -91,6 +91,9 @@ struct page_pool { unsigned long defer_warn; u32 pages_state_hold_cnt; + unsigned int frag_offset; + struct page *frag_page; + long frag_users; /* * Data structure for allocation side @@ -140,6 +143,18 @@ static inline struct page *page_pool_dev_alloc_pages(struct page_pool *pool) return page_pool_alloc_pages(pool, gfp); } +struct page *page_pool_alloc_frag(struct page_pool *pool, unsigned int *offset, + unsigned int size, gfp_t gfp); + +static inline struct page *page_pool_dev_alloc_frag(struct page_pool *pool, + unsigned int *offset, + unsigned int size) +{ + gfp_t gfp = (GFP_ATOMIC | __GFP_NOWARN); + + return page_pool_alloc_frag(pool, offset, size, gfp); +} + /* get the stored dma direction. A driver might decide to treat this locally and * avoid the extra cache line from page_pool to determine the direction */ -- cgit v1.2.3 From 8702997074363c294a1f83928cd0c33ca57bf813 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 5 Aug 2021 12:02:43 +0200 Subject: netfilter: nf_queue: move hookfn registration out of struct net This was done to detect when the pernet->init() function was not called yet, by checking if net->nf.queue_handler is NULL. Once the nfnetlink_queue module is active, all struct net pointers contain the same address. So place this back in nf_queue.c. Handle the 'netns error unwind' test by checking nfnl_queue_net for a NULL pointer and add a comment for this. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_queue.h | 4 ++-- include/net/netns/netfilter.h | 1 - 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_queue.h b/include/net/netfilter/nf_queue.h index e770bba00066..9eed51e920e8 100644 --- a/include/net/netfilter/nf_queue.h +++ b/include/net/netfilter/nf_queue.h @@ -33,8 +33,8 @@ struct nf_queue_handler { void (*nf_hook_drop)(struct net *net); }; -void nf_register_queue_handler(struct net *net, const struct nf_queue_handler *qh); -void nf_unregister_queue_handler(struct net *net); +void nf_register_queue_handler(const struct nf_queue_handler *qh); +void nf_unregister_queue_handler(void); void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict); void nf_queue_entry_get_refs(struct nf_queue_entry *entry); diff --git a/include/net/netns/netfilter.h b/include/net/netns/netfilter.h index 15e2b13fb0c0..986a2a9cfdfa 100644 --- a/include/net/netns/netfilter.h +++ b/include/net/netns/netfilter.h @@ -12,7 +12,6 @@ struct netns_nf { #if defined CONFIG_PROC_FS struct proc_dir_entry *proc_netfilter; #endif - const struct nf_queue_handler __rcu *queue_handler; const struct nf_logger __rcu *nf_loggers[NFPROTO_NUMPROTO]; #ifdef CONFIG_SYSCTL struct ctl_table_header *nf_log_dir_header; -- cgit v1.2.3 From f13a5ad88186c142b4f6060fb06f0f8fb1674915 Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Tue, 10 Aug 2021 16:24:15 +0300 Subject: devlink: Add new "enable_eth" generic device param Add new device generic parameter to enable/disable creation of Ethernet auxiliary device and associated device functionality in the devlink instance. User who prefers to disable such functionality can disable it using below example. $ devlink dev param set pci/0000:06:00.0 \ name enable_eth value false cmode driverinit $ devlink dev reload pci/0000:06:00.0 At this point devlink instance do not create auxiliary device for the Ethernet functionality. Signed-off-by: Parav Pandit Reviewed-by: Jiri Pirko Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 0236c77f2fd0..1e3e183bb2c2 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -519,6 +519,7 @@ enum devlink_param_generic_id { DEVLINK_PARAM_GENERIC_ID_RESET_DEV_ON_DRV_PROBE, DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE, DEVLINK_PARAM_GENERIC_ID_ENABLE_REMOTE_DEV_RESET, + DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH, /* add new param generic ids above here*/ __DEVLINK_PARAM_GENERIC_ID_MAX, @@ -559,6 +560,9 @@ enum devlink_param_generic_id { #define DEVLINK_PARAM_GENERIC_ENABLE_REMOTE_DEV_RESET_NAME "enable_remote_dev_reset" #define DEVLINK_PARAM_GENERIC_ENABLE_REMOTE_DEV_RESET_TYPE DEVLINK_PARAM_TYPE_BOOL +#define DEVLINK_PARAM_GENERIC_ENABLE_ETH_NAME "enable_eth" +#define DEVLINK_PARAM_GENERIC_ENABLE_ETH_TYPE DEVLINK_PARAM_TYPE_BOOL + #define DEVLINK_PARAM_GENERIC(_id, _cmodes, _get, _set, _validate) \ { \ .id = DEVLINK_PARAM_GENERIC_ID_##_id, \ -- cgit v1.2.3 From 8ddaabee3c7994854841a9b097fd94538126c12c Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Tue, 10 Aug 2021 16:24:16 +0300 Subject: devlink: Add new "enable_rdma" generic device param Add new device generic parameter to enable/disable creation of RDMA auxiliary device and associated device functionality in the devlink instance. User who prefers to disable such functionality can disable it using below example. $ devlink dev param set pci/0000:06:00.0 \ name enable_rdma value false cmode driverinit $ devlink dev reload pci/0000:06:00.0 At this point devlink instance do not create auxiliary device for the RDMA functionality. Signed-off-by: Parav Pandit Reviewed-by: Jiri Pirko Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 1e3e183bb2c2..6f4f0416e598 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -520,6 +520,7 @@ enum devlink_param_generic_id { DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE, DEVLINK_PARAM_GENERIC_ID_ENABLE_REMOTE_DEV_RESET, DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH, + DEVLINK_PARAM_GENERIC_ID_ENABLE_RDMA, /* add new param generic ids above here*/ __DEVLINK_PARAM_GENERIC_ID_MAX, @@ -563,6 +564,9 @@ enum devlink_param_generic_id { #define DEVLINK_PARAM_GENERIC_ENABLE_ETH_NAME "enable_eth" #define DEVLINK_PARAM_GENERIC_ENABLE_ETH_TYPE DEVLINK_PARAM_TYPE_BOOL +#define DEVLINK_PARAM_GENERIC_ENABLE_RDMA_NAME "enable_rdma" +#define DEVLINK_PARAM_GENERIC_ENABLE_RDMA_TYPE DEVLINK_PARAM_TYPE_BOOL + #define DEVLINK_PARAM_GENERIC(_id, _cmodes, _get, _set, _validate) \ { \ .id = DEVLINK_PARAM_GENERIC_ID_##_id, \ -- cgit v1.2.3 From 076b2a9dbb28e8b3d9a264a8bca664794255d448 Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Tue, 10 Aug 2021 16:24:17 +0300 Subject: devlink: Add new "enable_vnet" generic device param Add new device generic parameter to enable/disable creation of VDPA net auxiliary device and associated device functionality in the devlink instance. User who prefers to disable such functionality can disable it using below example. $ devlink dev param set pci/0000:06:00.0 \ name enable_vnet value false cmode driverinit $ devlink dev reload pci/0000:06:00.0 At this point devlink instance do not create auxiliary device for the VDPA net functionality. Signed-off-by: Parav Pandit Reviewed-by: Jiri Pirko Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 6f4f0416e598..0a0becbcdc49 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -521,6 +521,7 @@ enum devlink_param_generic_id { DEVLINK_PARAM_GENERIC_ID_ENABLE_REMOTE_DEV_RESET, DEVLINK_PARAM_GENERIC_ID_ENABLE_ETH, DEVLINK_PARAM_GENERIC_ID_ENABLE_RDMA, + DEVLINK_PARAM_GENERIC_ID_ENABLE_VNET, /* add new param generic ids above here*/ __DEVLINK_PARAM_GENERIC_ID_MAX, @@ -567,6 +568,9 @@ enum devlink_param_generic_id { #define DEVLINK_PARAM_GENERIC_ENABLE_RDMA_NAME "enable_rdma" #define DEVLINK_PARAM_GENERIC_ENABLE_RDMA_TYPE DEVLINK_PARAM_TYPE_BOOL +#define DEVLINK_PARAM_GENERIC_ENABLE_VNET_NAME "enable_vnet" +#define DEVLINK_PARAM_GENERIC_ENABLE_VNET_TYPE DEVLINK_PARAM_TYPE_BOOL + #define DEVLINK_PARAM_GENERIC(_id, _cmodes, _get, _set, _validate) \ { \ .id = DEVLINK_PARAM_GENERIC_ID_##_id, \ -- cgit v1.2.3 From b40c51efefbc4a3ddec682f118adefea1ccf70dc Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Tue, 10 Aug 2021 16:24:19 +0300 Subject: devlink: Add API to register and unregister single parameter Currently device configuration parameters can be registered as an array. Due to this a constant array must be registered. A single driver supporting multiple devices each with different device capabilities end up registering all parameters even if it doesn't support it. One possible workaround a driver can do is, it registers multiple single entry arrays to overcome such limitation. Better is to provide a API that enables driver to register/unregister a single parameter. This also further helps in two ways. (1) to reduce the memory of devlink_param_entry by avoiding in registering parameters which are not supported by the device. (2) avoid generating multiple parameter add, delete, publish, unpublish, init value notifications for such unsupported parameters Signed-off-by: Parav Pandit Reviewed-by: Jiri Pirko Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 0a0becbcdc49..f6459ee77114 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1645,6 +1645,10 @@ int devlink_params_register(struct devlink *devlink, void devlink_params_unregister(struct devlink *devlink, const struct devlink_param *params, size_t params_count); +int devlink_param_register(struct devlink *devlink, + const struct devlink_param *param); +void devlink_param_unregister(struct devlink *devlink, + const struct devlink_param *param); void devlink_params_publish(struct devlink *devlink); void devlink_params_unpublish(struct devlink *devlink); int devlink_port_params_register(struct devlink_port *devlink_port, -- cgit v1.2.3 From 9c4a7665b4237621879caf115a78f69bad67b9c7 Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Tue, 10 Aug 2021 16:24:20 +0300 Subject: devlink: Add APIs to publish, unpublish individual parameter Enable drivers to publish/unpublish individual parameter. Signed-off-by: Parav Pandit Reviewed-by: Jiri Pirko Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index f6459ee77114..1151497c0ec5 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1651,6 +1651,10 @@ void devlink_param_unregister(struct devlink *devlink, const struct devlink_param *param); void devlink_params_publish(struct devlink *devlink); void devlink_params_unpublish(struct devlink *devlink); +void devlink_param_publish(struct devlink *devlink, + const struct devlink_param *param); +void devlink_param_unpublish(struct devlink *devlink, + const struct devlink_param *param); int devlink_port_params_register(struct devlink_port *devlink_port, const struct devlink_param *params, size_t params_count); -- cgit v1.2.3 From 891a88f4f5768b1e6ff52a2386d48558aa5a3f63 Mon Sep 17 00:00:00 2001 From: Jonathan Toppins Date: Tue, 10 Aug 2021 22:53:30 -0400 Subject: bonding: remove extraneous definitions from bonding.h All of the symbols either only exist in bond_options.c or nowhere at all. These symbols were verified to not exist in the code base by using `git grep` and their removal was verified by compiling bonding.ko. Signed-off-by: Jonathan Toppins Signed-off-by: Jakub Kicinski --- include/net/bonding.h | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'include/net') diff --git a/include/net/bonding.h b/include/net/bonding.h index 9f3fdc180c6c..15e083e18f75 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -150,11 +150,6 @@ struct bond_params { u8 ad_actor_system[ETH_ALEN + 2]; }; -struct bond_parm_tbl { - char *modename; - int mode; -}; - struct slave { struct net_device *dev; /* first - useful for panic debug */ struct bonding *bond; /* our master */ @@ -755,13 +750,6 @@ static inline int bond_get_targets_ip(__be32 *targets, __be32 ip) /* exported from bond_main.c */ extern unsigned int bond_net_id; -extern const struct bond_parm_tbl bond_lacp_tbl[]; -extern const struct bond_parm_tbl xmit_hashtype_tbl[]; -extern const struct bond_parm_tbl arp_validate_tbl[]; -extern const struct bond_parm_tbl arp_all_targets_tbl[]; -extern const struct bond_parm_tbl fail_over_mac_tbl[]; -extern const struct bond_parm_tbl pri_reselect_tbl[]; -extern struct bond_parm_tbl ad_select_tbl[]; /* exported from bond_netlink.c */ extern struct rtnl_link_ops bond_link_ops; -- cgit v1.2.3 From 83f0a0b7285b299e006b0698a0ddc1ffacff3e43 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Tue, 10 Aug 2021 10:38:34 +0800 Subject: mctp: Specify route types, require rtm_type in RTM_*ROUTE messages This change adds a 'type' attribute to routes, which can be parsed from a RTM_NEWROUTE message. This will help to distinguish local vs. peer routes in a future change. This means userspace will need to set a correct rtm_type in RTM_NEWROUTE and RTM_DELROUTE messages; we currently only accept RTN_UNICAST. Signed-off-by: Jeremy Kerr Link: https://lore.kernel.org/r/20210810023834.2231088-1-jk@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- include/net/mctp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index 54bbe042c973..a824d47c3c6d 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -173,6 +173,7 @@ struct mctp_route { struct mctp_dev *dev; unsigned int mtu; + unsigned char type; int (*output)(struct mctp_route *route, struct sk_buff *skb); -- cgit v1.2.3 From 8c89f7b3d3f2880c57b0bc96c72ccd98fe354399 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 6 Aug 2021 14:53:05 -0700 Subject: mac80211: Use flex-array for radiotap header bitmap In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields. The it_present member of struct ieee80211_radiotap_header is treated as a flexible array (multiple u32s can be conditionally present). In order for memcpy() to reason (or really, not reason) about the size of operations against this struct, use of bytes beyond it_present need to be treated as part of the flexible array. Add a trailing flexible array and initialize its initial index via pointer arithmetic. Cc: Johannes Berg Cc: "David S. Miller" Cc: Jakub Kicinski Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook Link: https://lore.kernel.org/r/20210806215305.2875621-1-keescook@chromium.org Signed-off-by: Johannes Berg --- include/net/ieee80211_radiotap.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h index c0854933e24f..11630351c978 100644 --- a/include/net/ieee80211_radiotap.h +++ b/include/net/ieee80211_radiotap.h @@ -43,6 +43,11 @@ struct ieee80211_radiotap_header { * @it_present: (first) present word */ __le32 it_present; + + /** + * @it_optional: all remaining presence bitmaps + */ + __le32 it_optional[]; } __packed; /* version is always 0 */ -- cgit v1.2.3 From 437ebfd90a2567aab19dce47bafc81ebd8a63324 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sat, 14 Aug 2021 12:57:28 +0300 Subject: devlink: Count struct devlink consumers The struct devlink itself is protected by internal lock and doesn't need global lock during operation. That global lock is used to protect addition/removal new devlink instances from the global list in use by all devlink consumers in the system. The future conversion of linked list to be xarray will allow us to actually delete that lock, but first we need to count all struct devlink users. The reference counting provides us a way to ensure that no new user space commands success to grab devlink instance which is going to be destroyed makes it is safe to access it without lock. Signed-off-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 1151497c0ec5..4c60d61d92da 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -56,6 +56,8 @@ struct devlink { */ u8 reload_failed:1, reload_enabled:1; + refcount_t refcount; + struct completion comp; char priv[0] __aligned(NETDEV_ALIGN); }; -- cgit v1.2.3 From 11a861d767cdd87a34397821b0fd2095893b84b3 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Sat, 14 Aug 2021 12:57:29 +0300 Subject: devlink: Use xarray to store devlink instances We can use xarray instead of linearly organized linked lists for the devlink instances. This will let us revise the locking scheme in favour of internal xarray locking that protects database. Signed-off-by: Leon Romanovsky Signed-off-by: David S. Miller --- include/net/devlink.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/devlink.h b/include/net/devlink.h index 4c60d61d92da..154cf0dbca37 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -32,7 +32,7 @@ struct devlink_dev_stats { struct devlink_ops; struct devlink { - struct list_head list; + u32 index; struct list_head port_list; struct list_head rate_list; struct list_head sb_list; -- cgit v1.2.3 From 6c9b40844751ea30c72f7a2f92f4d704bc6b2927 Mon Sep 17 00:00:00 2001 From: Cai Huoqing Date: Fri, 13 Aug 2021 20:08:02 +0800 Subject: net: Remove net/ipx.h and uapi/linux/ipx.h header files commit <47595e32869f> ("") indicated the ipx network layer as obsolete in Jan 2018, updated in the MAINTAINERS file now, after being exposed for 3 years to refactoring, so to delete uapi/linux/ipx.h and net/ipx.h header files for good. additionally, there is no module that depends on ipx.h except a broken staging driver(r8188eu) Signed-off-by: Cai Huoqing Signed-off-by: David S. Miller --- include/net/ipx.h | 171 ------------------------------------------------------ 1 file changed, 171 deletions(-) delete mode 100644 include/net/ipx.h (limited to 'include/net') diff --git a/include/net/ipx.h b/include/net/ipx.h deleted file mode 100644 index 9d1342807b59..000000000000 --- a/include/net/ipx.h +++ /dev/null @@ -1,171 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _NET_INET_IPX_H_ -#define _NET_INET_IPX_H_ -/* - * The following information is in its entirety obtained from: - * - * Novell 'IPX Router Specification' Version 1.10 - * Part No. 107-000029-001 - * - * Which is available from ftp.novell.com - */ - -#include -#include -#include -#include -#include -#include - -struct ipx_address { - __be32 net; - __u8 node[IPX_NODE_LEN]; - __be16 sock; -}; - -#define ipx_broadcast_node "\377\377\377\377\377\377" -#define ipx_this_node "\0\0\0\0\0\0" - -#define IPX_MAX_PPROP_HOPS 8 - -struct ipxhdr { - __be16 ipx_checksum __packed; -#define IPX_NO_CHECKSUM cpu_to_be16(0xFFFF) - __be16 ipx_pktsize __packed; - __u8 ipx_tctrl; - __u8 ipx_type; -#define IPX_TYPE_UNKNOWN 0x00 -#define IPX_TYPE_RIP 0x01 /* may also be 0 */ -#define IPX_TYPE_SAP 0x04 /* may also be 0 */ -#define IPX_TYPE_SPX 0x05 /* SPX protocol */ -#define IPX_TYPE_NCP 0x11 /* $lots for docs on this (SPIT) */ -#define IPX_TYPE_PPROP 0x14 /* complicated flood fill brdcast */ - struct ipx_address ipx_dest __packed; - struct ipx_address ipx_source __packed; -}; - -/* From af_ipx.c */ -extern int sysctl_ipx_pprop_broadcasting; - -struct ipx_interface { - /* IPX address */ - __be32 if_netnum; - unsigned char if_node[IPX_NODE_LEN]; - refcount_t refcnt; - - /* physical device info */ - struct net_device *if_dev; - struct datalink_proto *if_dlink; - __be16 if_dlink_type; - - /* socket support */ - unsigned short if_sknum; - struct hlist_head if_sklist; - spinlock_t if_sklist_lock; - - /* administrative overhead */ - int if_ipx_offset; - unsigned char if_internal; - unsigned char if_primary; - - struct list_head node; /* node in ipx_interfaces list */ -}; - -struct ipx_route { - __be32 ir_net; - struct ipx_interface *ir_intrfc; - unsigned char ir_routed; - unsigned char ir_router_node[IPX_NODE_LEN]; - struct list_head node; /* node in ipx_routes list */ - refcount_t refcnt; -}; - -struct ipx_cb { - u8 ipx_tctrl; - __be32 ipx_dest_net; - __be32 ipx_source_net; - struct { - __be32 netnum; - int index; - } last_hop; -}; - -#include - -struct ipx_sock { - /* struct sock has to be the first member of ipx_sock */ - struct sock sk; - struct ipx_address dest_addr; - struct ipx_interface *intrfc; - __be16 port; -#ifdef CONFIG_IPX_INTERN - unsigned char node[IPX_NODE_LEN]; -#endif - unsigned short type; - /* - * To handle special ncp connection-handling sockets for mars_nwe, - * the connection number must be stored in the socket. - */ - unsigned short ipx_ncp_conn; -}; - -static inline struct ipx_sock *ipx_sk(struct sock *sk) -{ - return (struct ipx_sock *)sk; -} - -#define IPX_SKB_CB(__skb) ((struct ipx_cb *)&((__skb)->cb[0])) - -#define IPX_MIN_EPHEMERAL_SOCKET 0x4000 -#define IPX_MAX_EPHEMERAL_SOCKET 0x7fff - -extern struct list_head ipx_routes; -extern rwlock_t ipx_routes_lock; - -extern struct list_head ipx_interfaces; -struct ipx_interface *ipx_interfaces_head(void); -extern spinlock_t ipx_interfaces_lock; - -extern struct ipx_interface *ipx_primary_net; - -int ipx_proc_init(void); -void ipx_proc_exit(void); - -const char *ipx_frame_name(__be16); -const char *ipx_device_name(struct ipx_interface *intrfc); - -static __inline__ void ipxitf_hold(struct ipx_interface *intrfc) -{ - refcount_inc(&intrfc->refcnt); -} - -void ipxitf_down(struct ipx_interface *intrfc); -struct ipx_interface *ipxitf_find_using_net(__be32 net); -int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node); -__be16 ipx_cksum(struct ipxhdr *packet, int length); -int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, - unsigned char *node); -void ipxrtr_del_routes(struct ipx_interface *intrfc); -int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, - struct msghdr *msg, size_t len, int noblock); -int ipxrtr_route_skb(struct sk_buff *skb); -struct ipx_route *ipxrtr_lookup(__be32 net); -int ipxrtr_ioctl(unsigned int cmd, void __user *arg); - -static __inline__ void ipxitf_put(struct ipx_interface *intrfc) -{ - if (refcount_dec_and_test(&intrfc->refcnt)) - ipxitf_down(intrfc); -} - -static __inline__ void ipxrtr_hold(struct ipx_route *rt) -{ - refcount_inc(&rt->refcnt); -} - -static __inline__ void ipxrtr_put(struct ipx_route *rt) -{ - if (refcount_dec_and_test(&rt->refcnt)) - kfree(rt); -} -#endif /* _NET_INET_IPX_H_ */ -- cgit v1.2.3 From 7087c4f69487f017722df7d299ef9b7709996b79 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 11 Aug 2021 16:20:16 -0700 Subject: Bluetooth: Store advertising handle so it can be re-enabled This stores the advertising handle/instance into hci_conn so it is accessible when re-enabling the advertising once disconnected. Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a7d06d7da602..a7360c8c72f8 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -629,6 +629,7 @@ struct hci_conn { __u8 init_addr_type; bdaddr_t resp_addr; __u8 resp_addr_type; + __u8 adv_instance; __u16 handle; __u16 state; __u8 mode; -- cgit v1.2.3 From 94531cfcbe79c3598acf96806627b2137ca32eb9 Mon Sep 17 00:00:00 2001 From: Jiang Wang Date: Mon, 16 Aug 2021 19:03:21 +0000 Subject: af_unix: Add unix_stream_proto for sockmap Previously, sockmap for AF_UNIX protocol only supports dgram type. This patch add unix stream type support, which is similar to unix_dgram_proto. To support sockmap, dgram and stream cannot share the same unix_proto anymore, because they have different implementations, such as unhash for stream type (which will remove closed or disconnected sockets from the map), so rename unix_proto to unix_dgram_proto and add a new unix_stream_proto. Also implement stream related sockmap functions. And add dgram key words to those dgram specific functions. Signed-off-by: Jiang Wang Signed-off-by: Andrii Nakryiko Reviewed-by: Cong Wang Acked-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20210816190327.2739291-3-jiang.wang@bytedance.com --- include/net/af_unix.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 4757d7f53f13..7d142e8a0550 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -87,6 +87,8 @@ long unix_outq_len(struct sock *sk); int __unix_dgram_recvmsg(struct sock *sk, struct msghdr *msg, size_t size, int flags); +int __unix_stream_recvmsg(struct sock *sk, struct msghdr *msg, size_t size, + int flags); #ifdef CONFIG_SYSCTL int unix_sysctl_register(struct net *net); void unix_sysctl_unregister(struct net *net); @@ -96,9 +98,11 @@ static inline void unix_sysctl_unregister(struct net *net) {} #endif #ifdef CONFIG_BPF_SYSCALL -extern struct proto unix_proto; +extern struct proto unix_dgram_proto; +extern struct proto unix_stream_proto; -int unix_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); +int unix_dgram_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); +int unix_stream_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); void __init unix_bpf_build_proto(void); #else static inline void __init unix_bpf_build_proto(void) -- cgit v1.2.3 From 0d2ab3aea50bb02ff0c9c3d53c7b2b4b21cdd59d Mon Sep 17 00:00:00 2001 From: John Crispin Date: Fri, 2 Jul 2021 19:44:07 +0200 Subject: nl80211: add support for BSS coloring This patch adds support for BSS color collisions to the wireless subsystem. Add the required functionality to nl80211 that will notify about color collisions, triggering the color change and notifying when it is completed. Co-developed-by: Lorenzo Bianconi Signed-off-by: Lorenzo Bianconi Signed-off-by: John Crispin Link: https://lore.kernel.org/r/500b3582aec8fe2c42ef46f3117b148cb7cbceb5.1625247619.git.lorenzo@kernel.org [remove unnecessary NULL initialisation] Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 92 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 92 insertions(+) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 161cdf7df1a0..62dd8422e0dc 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1252,6 +1252,27 @@ struct cfg80211_csa_settings { u8 count; }; +/** + * struct cfg80211_color_change_settings - color change settings + * + * Used for bss color change + * + * @beacon_color_change: beacon data while performing the color countdown + * @counter_offsets_beacon: offsets of the counters within the beacon (tail) + * @counter_offsets_presp: offsets of the counters within the probe response + * @beacon_next: beacon data to be used after the color change + * @count: number of beacons until the color change + * @color: the color used after the change + */ +struct cfg80211_color_change_settings { + struct cfg80211_beacon_data beacon_color_change; + u16 counter_offset_beacon; + u16 counter_offset_presp; + struct cfg80211_beacon_data beacon_next; + u8 count; + u8 color; +}; + /** * struct iface_combination_params - input parameters for interface combinations * @@ -3995,6 +4016,8 @@ struct mgmt_frame_regs { * given TIDs. This callback may sleep. * * @set_sar_specs: Update the SAR (TX power) settings. + * + * @color_change: Initiate a color change. */ struct cfg80211_ops { int (*suspend)(struct wiphy *wiphy, struct cfg80211_wowlan *wow); @@ -4322,6 +4345,9 @@ struct cfg80211_ops { const u8 *peer, u8 tids); int (*set_sar_specs)(struct wiphy *wiphy, struct cfg80211_sar_specs *sar); + int (*color_change)(struct wiphy *wiphy, + struct net_device *dev, + struct cfg80211_color_change_settings *params); }; /* @@ -8218,4 +8244,70 @@ void cfg80211_update_owe_info_event(struct net_device *netdev, */ void cfg80211_bss_flush(struct wiphy *wiphy); +/** + * cfg80211_bss_color_notify - notify about bss color event + * @dev: network device + * @gfp: allocation flags + * @cmd: the actual event we want to notify + * @count: the number of TBTTs until the color change happens + * @color_bitmap: representations of the colors that the local BSS is aware of + */ +int cfg80211_bss_color_notify(struct net_device *dev, gfp_t gfp, + enum nl80211_commands cmd, u8 count, + u64 color_bitmap); + +/** + * cfg80211_obss_color_collision_notify - notify about bss color collision + * @dev: network device + * @color_bitmap: representations of the colors that the local BSS is aware of + */ +static inline int cfg80211_obss_color_collision_notify(struct net_device *dev, + u64 color_bitmap) +{ + return cfg80211_bss_color_notify(dev, GFP_KERNEL, + NL80211_CMD_OBSS_COLOR_COLLISION, + 0, color_bitmap); +} + +/** + * cfg80211_color_change_started_notify - notify color change start + * @dev: the device on which the color is switched + * @count: the number of TBTTs until the color change happens + * + * Inform the userspace about the color change that has started. + */ +static inline int cfg80211_color_change_started_notify(struct net_device *dev, + u8 count) +{ + return cfg80211_bss_color_notify(dev, GFP_KERNEL, + NL80211_CMD_COLOR_CHANGE_STARTED, + count, 0); +} + +/** + * cfg80211_color_change_aborted_notify - notify color change abort + * @dev: the device on which the color is switched + * + * Inform the userspace about the color change that has aborted. + */ +static inline int cfg80211_color_change_aborted_notify(struct net_device *dev) +{ + return cfg80211_bss_color_notify(dev, GFP_KERNEL, + NL80211_CMD_COLOR_CHANGE_ABORTED, + 0, 0); +} + +/** + * cfg80211_color_change_notify - notify color change completion + * @dev: the device on which the color was switched + * + * Inform the userspace about the color change that has completed. + */ +static inline int cfg80211_color_change_notify(struct net_device *dev) +{ + return cfg80211_bss_color_notify(dev, GFP_KERNEL, + NL80211_CMD_COLOR_CHANGE_COMPLETED, + 0, 0); +} + #endif /* __NET_CFG80211_H */ -- cgit v1.2.3 From 5f9404abdf2ac31c8f4768c39714bfcaca389e3a Mon Sep 17 00:00:00 2001 From: John Crispin Date: Fri, 2 Jul 2021 19:44:08 +0200 Subject: mac80211: add support for BSS color change The color change announcement is very similar to how CSA works where we have an IE that includes a counter. When the counter hits 0, the new color is applied via an updated beacon. This patch makes the CSA counter functionality reusable, rather than implementing it again. This also allows for future reuse incase support for other counter IEs gets added. Co-developed-by: Lorenzo Bianconi Signed-off-by: Lorenzo Bianconi Signed-off-by: John Crispin Link: https://lore.kernel.org/r/057c1e67b82bee561ea44ce6a45a8462d3da6995.1625247619.git.lorenzo@kernel.org Signed-off-by: Johannes Berg --- include/net/mac80211.h | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index d8a1d09a2141..a23e6734d26b 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1711,6 +1711,10 @@ enum ieee80211_offload_flags { * protected by fq->lock. * @offload_flags: 802.3 -> 802.11 enapsulation offload flags, see * &enum ieee80211_offload_flags. + * @color_change_active: marks whether a color change is ongoing. Internally it is + * write-protected by sdata_lock and local->mtx so holding either is fine + * for read access. + * @color_change_color: the bss color that will be used after the change. */ struct ieee80211_vif { enum nl80211_iftype type; @@ -1739,6 +1743,9 @@ struct ieee80211_vif { bool txqs_stopped[IEEE80211_NUM_ACS]; + bool color_change_active; + u8 color_change_color; + /* must be last */ u8 drv_priv[] __aligned(sizeof(void *)); }; @@ -5007,6 +5014,16 @@ void ieee80211_csa_finish(struct ieee80211_vif *vif); */ bool ieee80211_beacon_cntdwn_is_complete(struct ieee80211_vif *vif); +/** + * ieee80211_color_change_finish - notify mac80211 about color change + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * + * After a color change announcement was scheduled and the counter in this + * announcement hits 1, this function must be called by the driver to + * notify mac80211 that the color can be changed + */ +void ieee80211_color_change_finish(struct ieee80211_vif *vif); + /** * ieee80211_proberesp_get - retrieve a Probe Response template * @hw: pointer obtained from ieee80211_alloc_hw(). @@ -6771,6 +6788,18 @@ struct sk_buff * ieee80211_get_unsol_bcast_probe_resp_tmpl(struct ieee80211_hw *hw, struct ieee80211_vif *vif); +/** + * ieeee80211_obss_color_collision_notify - notify userland about a BSS color + * collision. + * + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * @color_bitmap: a 64 bit bitmap representing the colors that the local BSS is + * aware of. + */ +void +ieeee80211_obss_color_collision_notify(struct ieee80211_vif *vif, + u64 color_bitmap); + /** * ieee80211_is_tx_data - check if frame is a data frame * -- cgit v1.2.3 From 4b1327be9fe57443295ae86fe0fcf24a18469e9f Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Tue, 17 Aug 2021 12:40:03 -0700 Subject: net-memcg: pass in gfp_t mask to mem_cgroup_charge_skmem() Add gfp_t mask as an input parameter to mem_cgroup_charge_skmem(), to give more control to the networking stack and enable it to change memcg charging behavior. In the future, the networking stack may decide to avoid oom-kills when fallbacks are more appropriate. One behavior change in mem_cgroup_charge_skmem() by this patch is to avoid force charging by default and let the caller decide when and if force charging is needed through the presence or absence of __GFP_NOFAIL. Signed-off-by: Wei Wang Reviewed-by: Shakeel Butt Signed-off-by: David S. Miller --- include/net/sock.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 6e761451c927..95b25777b53e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2400,6 +2400,11 @@ static inline gfp_t gfp_any(void) return in_softirq() ? GFP_ATOMIC : GFP_KERNEL; } +static inline gfp_t gfp_memcg_charge(void) +{ + return in_softirq() ? GFP_NOWAIT : GFP_KERNEL; +} + static inline long sock_rcvtimeo(const struct sock *sk, bool noblock) { return noblock ? 0 : sk->sk_rcvtimeo; -- cgit v1.2.3 From 74fc4f828769cca1c3be89ea92cb88feaa27ef52 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 17 Aug 2021 20:05:18 +0300 Subject: net: Fix offloading indirect devices dependency on qdisc order creation Currently, when creating an ingress qdisc on an indirect device before the driver registered for callbacks, the driver will not have a chance to register its filter configuration callbacks. To fix that, modify the code such that it keeps track of all the ingress qdiscs that call flow_indr_dev_setup_offload(). When a driver calls flow_indr_dev_register(), go through the list of tracked ingress qdiscs and call the driver callback entry point so as to give it a chance to register its callback. Reviewed-by: Jiri Pirko Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- include/net/flow_offload.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h index f3c2841566a0..5aa27acdb0b3 100644 --- a/include/net/flow_offload.h +++ b/include/net/flow_offload.h @@ -453,6 +453,7 @@ struct flow_block_offload { struct list_head *driver_block_list; struct netlink_ext_ack *extack; struct Qdisc *sch; + struct list_head *cb_list_head; }; enum tc_setup_type; -- cgit v1.2.3 From fa05bdb89b01b098aad19ec0ebc4d1cc7b11177e Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 19 Aug 2021 13:58:42 +0300 Subject: Revert "flow_offload: action should not be NULL when it is referenced" This reverts commit 9ea3e52c5bc8bb4a084938dc1e3160643438927a. Cited commit added a check to make sure 'action' is not NULL, but 'action' is already dereferenced before the check, when calling flow_offload_has_one_action(). Therefore, the check does not make any sense and results in a smatch warning: include/net/flow_offload.h:322 flow_action_mixed_hw_stats_check() warn: variable dereferenced before check 'action' (see line 319) Fix by reverting this commit. Cc: gushengxian Fixes: 9ea3e52c5bc8 ("flow_offload: action should not be NULL when it is referenced") Signed-off-by: Ido Schimmel Acked-by: Jamal Hadi Salim Link: https://lore.kernel.org/r/20210819105842.1315705-1-idosch@idosch.org Signed-off-by: Jakub Kicinski --- include/net/flow_offload.h | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h index f3c2841566a0..1b9d75aedb22 100644 --- a/include/net/flow_offload.h +++ b/include/net/flow_offload.h @@ -319,14 +319,12 @@ flow_action_mixed_hw_stats_check(const struct flow_action *action, if (flow_offload_has_one_action(action)) return true; - if (action) { - flow_action_for_each(i, action_entry, action) { - if (i && action_entry->hw_stats != last_hw_stats) { - NL_SET_ERR_MSG_MOD(extack, "Mixing HW stats types for actions is not supported"); - return false; - } - last_hw_stats = action_entry->hw_stats; + flow_action_for_each(i, action_entry, action) { + if (i && action_entry->hw_stats != last_hw_stats) { + NL_SET_ERR_MSG_MOD(extack, "Mixing HW stats types for actions is not supported"); + return false; } + last_hw_stats = action_entry->hw_stats; } return true; } -- cgit v1.2.3 From f5e165e72b29d908214e554ef57f67790ba95934 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 19 Aug 2021 20:55:00 +0300 Subject: net: dsa: track unique bridge numbers across all DSA switch trees Right now, cross-tree bridging setups work somewhat by mistake. In the case of cross-tree bridging with sja1105, all switch instances need to agree upon a common VLAN ID for forwarding a packet that belongs to a certain bridging domain. With TX forwarding offload, the VLAN ID is the bridge VLAN for VLAN-aware bridging, and the tag_8021q TX forwarding offload VID (a VLAN which has non-zero VBID bits) for VLAN-unaware bridging. The VBID for VLAN-unaware bridging is derived from the dp->bridge_num value calculated by DSA independently for each switch tree. If ports from one tree join one bridge, and ports from another tree join another bridge, DSA will assign them the same bridge_num, even though the bridges are different. If cross-tree bridging is supported, this is an issue. Modify DSA to calculate the bridge_num globally across all switch trees. This has the implication for a driver that the dp->bridge_num value that DSA will assign to its ports might not be contiguous, if there are boards with multiple DSA drivers instantiated. Additionally, all bridge_num values eat up towards each switch's ds->num_fwd_offloading_bridges maximum, which is potentially unfortunate, and can be seen as a limitation introduced by this patch. However, that is the lesser evil for now. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 0c2cba45fa79..c7ea0f61056f 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -155,9 +155,6 @@ struct dsa_switch_tree { /* Track the largest switch index within a tree */ unsigned int last_switch; - - /* Track the bridges with forwarding offload enabled */ - unsigned long fwd_offloading_bridges; }; #define dsa_lags_foreach_id(_id, _dst) \ @@ -411,8 +408,9 @@ struct dsa_switch { unsigned int num_lag_ids; /* Drivers that support bridge forwarding offload should set this to - * the maximum number of bridges spanning the same switch tree that can - * be offloaded. + * the maximum number of bridges spanning the same switch tree (or all + * trees, in the case of cross-tree bridging support) that can be + * offloaded. */ unsigned int num_fwd_offloading_bridges; -- cgit v1.2.3 From f5a4c24e689f54e66201f04d343bdd2e8a1d7923 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 23 Aug 2021 20:02:39 +0200 Subject: mac80211: introduce individual TWT support in AP mode Introduce TWT action frames parsing support to mac80211. Currently just individual TWT agreement are support in AP mode. Whenever the AP receives a TWT action frame from an associated client, after performing sanity checks, it will notify the underlay driver with requested parameters in order to check if they are supported and if there is enough room for a new agreement. The driver is expected to set the agreement result and report it to mac80211. Drivers supporting this have two new callbacks: - add_twt_setup (mandatory) - twt_teardown_request (optional) mac80211 will send an action frame reply according to the result reported by the driver. Tested-by: Peter Chiu Signed-off-by: Lorenzo Bianconi Link: https://lore.kernel.org/r/257512f2e22ba42b9f2624942a128dd8f141de4b.1629741512.git.lorenzo@kernel.org [use le16p_replace_bits(), minor cleanups, use (void *) casts, fix to use ieee80211_get_he_iftype_cap() correctly] Signed-off-by: Johannes Berg --- include/net/mac80211.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a23e6734d26b..af0fc13cea34 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -3926,6 +3926,13 @@ struct ieee80211_prep_tx_info { * @set_sar_specs: Update the SAR (TX power) settings. * @sta_set_decap_offload: Called to notify the driver when a station is allowed * to use rx decapsulation offload + * @add_twt_setup: Update hw with TWT agreement parameters received from the peer. + * This callback allows the hw to check if requested parameters + * are supported and if there is enough room for a new agreement. + * The hw is expected to set agreement result in the req_type field of + * twt structure. + * @twt_teardown_request: Update the hw with TWT teardown request received + * from the peer. */ struct ieee80211_ops { void (*tx)(struct ieee80211_hw *hw, @@ -4249,6 +4256,11 @@ struct ieee80211_ops { void (*sta_set_decap_offload)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, bool enabled); + void (*add_twt_setup)(struct ieee80211_hw *hw, + struct ieee80211_sta *sta, + struct ieee80211_twt_setup *twt); + void (*twt_teardown_request)(struct ieee80211_hw *hw, + struct ieee80211_sta *sta, u8 flowid); }; /** -- cgit v1.2.3 From 58adf9dcb15b99f047e80e10c85fb51ed3c88215 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 24 Aug 2021 00:22:58 +0300 Subject: net: dsa: let drivers state that they need VLAN filtering while standalone As explained in commit e358bef7c392 ("net: dsa: Give drivers the chance to veto certain upper devices"), the hellcreek driver uses some tricks to comply with the network stack expectations: it enforces port separation in standalone mode using VLANs. For untagged traffic, bridging between ports is prevented by using different PVIDs, and for VLAN-tagged traffic, it never accepts 8021q uppers with the same VID on two ports, so packets with one VLAN cannot leak from one port to another. That is almost fine*, and has worked because hellcreek relied on an implicit behavior of the DSA core that was changed by the previous patch: the standalone ports declare the 'rx-vlan-filter' feature as 'on [fixed]'. Since most of the DSA drivers are actually VLAN-unaware in standalone mode, that feature was actually incorrectly reflecting the hardware/driver state, so there was a desire to fix it. This leaves the hellcreek driver in a situation where it has to explicitly request this behavior from the DSA framework. We configure the ports as follows: - Standalone: 'rx-vlan-filter' is on. An 8021q upper on top of a standalone hellcreek port will go through dsa_slave_vlan_rx_add_vid and will add a VLAN to the hardware tables, giving the driver the opportunity to refuse it through .port_prechangeupper. - Bridged with vlan_filtering=0: 'rx-vlan-filter' is off. An 8021q upper on top of a bridged hellcreek port will not go through dsa_slave_vlan_rx_add_vid, because there will not be any attempt to offload this VLAN. The driver already disables VLAN awareness, so that upper should receive the traffic it needs. - Bridged with vlan_filtering=1: 'rx-vlan-filter' is on. An 8021q upper on top of a bridged hellcreek port will call dsa_slave_vlan_rx_add_vid, and can again be vetoed through .port_prechangeupper. *It is not actually completely fine, because if I follow through correctly, we can have the following situation: ip link add br0 type bridge vlan_filtering 0 ip link set lan0 master br0 # lan0 now becomes VLAN-unaware ip link set lan0 nomaster # lan0 fails to become VLAN-aware again, therefore breaking isolation This patch fixes that corner case by extending the DSA core logic, based on this requested attribute, to change the VLAN awareness state of the switch (port) when it leaves the bridge. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Acked-by: Kurt Kanzenbach Signed-off-by: David S. Miller --- include/net/dsa.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index c7ea0f61056f..f9a17145255a 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -363,6 +363,9 @@ struct dsa_switch { */ bool vlan_filtering_is_global; + /* Keep VLAN filtering enabled on ports not offloading any upper. */ + bool needs_standalone_vlan_filtering; + /* Pass .port_vlan_add and .port_vlan_del to drivers even for bridges * that have vlan_filtering=0. All drivers should ideally set this (and * then the option would get removed), but it is unknown whether this -- cgit v1.2.3 From 446e7f218b7662c912c610aae99069543aa88a40 Mon Sep 17 00:00:00 2001 From: zhang kai Date: Mon, 23 Aug 2021 11:49:00 +0800 Subject: ipv6: correct comments about fib6_node sernum correct comments in set and get fn_sernum Signed-off-by: zhang kai Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 15b7fbe6b15c..c412dde4d67d 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -267,7 +267,7 @@ static inline bool fib6_check_expired(const struct fib6_info *f6i) return false; } -/* Function to safely get fn->sernum for passed in rt +/* Function to safely get fn->fn_sernum for passed in rt * and store result in passed in cookie. * Return true if we can get cookie safely * Return false if not @@ -282,7 +282,7 @@ static inline bool fib6_get_cookie_safe(const struct fib6_info *f6i, if (fn) { *cookie = fn->fn_sernum; - /* pairs with smp_wmb() in fib6_update_sernum_upto_root() */ + /* pairs with smp_wmb() in __fib6_update_sernum_upto_root() */ smp_rmb(); status = true; } -- cgit v1.2.3 From d7b269083786dca3b1b0141bde6cea834062b691 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 24 Aug 2021 16:26:14 -0700 Subject: mptcp: shrink mptcp_out_options struct After the previous patch we can alias with a union several fields in mptcp_out_options. Such struct is stack allocated and memset() for each plain TCP out packet. Every saved byted counts. Before: pahole -EC mptcp_out_options # ... /* size: 136, cachelines: 3, members: 17 */ After: pahole -EC mptcp_out_options # ... /* size: 56, cachelines: 1, members: 9 */ Signed-off-by: Paolo Abeni Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 26 +++++++++++++++++--------- 1 file changed, 17 insertions(+), 9 deletions(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 8b5af683a818..3236010afa29 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -58,10 +58,6 @@ struct mptcp_addr_info { struct mptcp_out_options { #if IS_ENABLED(CONFIG_MPTCP) u16 suboptions; - u64 sndr_key; - u64 rcvr_key; - u64 ahmac; - struct mptcp_addr_info addr; struct mptcp_rm_list rm_list; u8 join_id; u8 backup; @@ -69,11 +65,23 @@ struct mptcp_out_options { reset_transient:1, csum_reqd:1, allow_join_id0:1; - u32 nonce; - u64 thmac; - u32 token; - u8 hmac[20]; - struct mptcp_ext ext_copy; + union { + struct { + u64 sndr_key; + u64 rcvr_key; + }; + struct { + struct mptcp_addr_info addr; + u64 ahmac; + }; + struct mptcp_ext ext_copy; + struct { + u32 nonce; + u32 token; + u64 thmac; + u8 hmac[20]; + }; + }; #endif }; -- cgit v1.2.3 From c25aeb4e095355eec3beb6a2b2b30322bd6d0dd4 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Tue, 24 Aug 2021 16:26:15 -0700 Subject: mptcp: MP_FAIL suboption sending This patch added the MP_FAIL suboption sending support. Add a new flag named send_mp_fail in struct mptcp_subflow_context. If this flag is set, send out MP_FAIL suboption. Add a new member fail_seq in struct mptcp_out_options to save the data sequence number to put into the MP_FAIL suboption. An MP_FAIL option could be included in a RST or on the subflow-level ACK. Suggested-by: Paolo Abeni Signed-off-by: Geliang Tang Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/net/mptcp.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 3236010afa29..6026bbefbffd 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -74,7 +74,10 @@ struct mptcp_out_options { struct mptcp_addr_info addr; u64 ahmac; }; - struct mptcp_ext ext_copy; + struct { + struct mptcp_ext ext_copy; + u64 fail_seq; + }; struct { u32 nonce; u32 token; -- cgit v1.2.3 From 87e5ef4b19cec86c861e3ebab3a5d840ecc2f4a4 Mon Sep 17 00:00:00 2001 From: Shaokun Zhang Date: Wed, 25 Aug 2021 14:34:31 +0800 Subject: mctp: Remove the repeated declaration Function 'mctp_dev_get_rtnl' is declared twice, so remove the repeated declaration. Cc: Jeremy Kerr Cc: Matt Johnston Cc: "David S. Miller" Cc: Jakub Kicinski Signed-off-by: Shaokun Zhang Signed-off-by: David S. Miller --- include/net/mctpdevice.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mctpdevice.h b/include/net/mctpdevice.h index 57e773ff08bb..71a11012fac7 100644 --- a/include/net/mctpdevice.h +++ b/include/net/mctpdevice.h @@ -31,6 +31,5 @@ struct mctp_dev { struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev); struct mctp_dev *__mctp_dev_get(const struct net_device *dev); -struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev); #endif /* __NET_MCTPDEVICE_H */ -- cgit v1.2.3 From 478374a3c15f369e57fdd79d64d7a1d2eb307e16 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 16 Aug 2021 17:16:22 +0200 Subject: netfilter: ecache: remove one indent level nf_conntrack_eventmask_report and nf_ct_deliver_cached_events shared most of their code. This unifies the layout by changing if (nf_ct_is_confirmed(ct)) { foo } to if (!nf_ct_is_confirmed(ct))) return foo This removes one level of indentation. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_ecache.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index d00ba6048e44..3734bacf9763 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -73,7 +73,7 @@ struct nf_ct_event { }; struct nf_ct_event_notifier { - int (*fcn)(unsigned int events, struct nf_ct_event *item); + int (*fcn)(unsigned int events, const struct nf_ct_event *item); }; int nf_conntrack_register_notifier(struct net *net, -- cgit v1.2.3 From b86c0e6429dac2458694495aeebf15f4fe6b269d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 16 Aug 2021 17:16:25 +0200 Subject: netfilter: ecache: prepare for event notifier merge This prepares for merge for ct and exp notifier structs. The 'fcn' member is renamed to something unique. Second, the register/unregister api is simplified. There is only one implementation so there is no need to do any error checking. Replace the EBUSY logic with WARN_ON_ONCE. This allows to remove error unwinding. The exp notifier register/unregister function is removed in a followup patch. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_ecache.h | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index 3734bacf9763..061a93a03b82 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -73,13 +73,12 @@ struct nf_ct_event { }; struct nf_ct_event_notifier { - int (*fcn)(unsigned int events, const struct nf_ct_event *item); + int (*ct_event)(unsigned int events, const struct nf_ct_event *item); }; -int nf_conntrack_register_notifier(struct net *net, - struct nf_ct_event_notifier *nb); -void nf_conntrack_unregister_notifier(struct net *net, - struct nf_ct_event_notifier *nb); +void nf_conntrack_register_notifier(struct net *net, + const struct nf_ct_event_notifier *nb); +void nf_conntrack_unregister_notifier(struct net *net); void nf_ct_deliver_cached_events(struct nf_conn *ct); int nf_conntrack_eventmask_report(unsigned int eventmask, struct nf_conn *ct, @@ -159,7 +158,7 @@ struct nf_exp_event { }; struct nf_exp_event_notifier { - int (*fcn)(unsigned int events, struct nf_exp_event *item); + int (*exp_event)(unsigned int events, struct nf_exp_event *item); }; int nf_ct_expect_register_notifier(struct net *net, -- cgit v1.2.3 From bd1431db0b8131098a285c8cc6a357629b4362e5 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 16 Aug 2021 17:16:26 +0200 Subject: netfilter: ecache: remove nf_exp_event_notifier structure Reuse the conntrack event notofier struct, this allows to remove the extra register/unregister functions and avoids a pointer in struct net. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_ecache.h | 23 +++++++---------------- include/net/netns/conntrack.h | 1 - 2 files changed, 7 insertions(+), 17 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index 061a93a03b82..d932e22edcb4 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -72,8 +72,15 @@ struct nf_ct_event { int report; }; +struct nf_exp_event { + struct nf_conntrack_expect *exp; + u32 portid; + int report; +}; + struct nf_ct_event_notifier { int (*ct_event)(unsigned int events, const struct nf_ct_event *item); + int (*exp_event)(unsigned int events, const struct nf_exp_event *item); }; void nf_conntrack_register_notifier(struct net *net, @@ -150,22 +157,6 @@ nf_conntrack_event(enum ip_conntrack_events event, struct nf_conn *ct) } #ifdef CONFIG_NF_CONNTRACK_EVENTS - -struct nf_exp_event { - struct nf_conntrack_expect *exp; - u32 portid; - int report; -}; - -struct nf_exp_event_notifier { - int (*exp_event)(unsigned int events, struct nf_exp_event *item); -}; - -int nf_ct_expect_register_notifier(struct net *net, - struct nf_exp_event_notifier *nb); -void nf_ct_expect_unregister_notifier(struct net *net, - struct nf_exp_event_notifier *nb); - void nf_ct_expect_event_report(enum ip_conntrack_expect_events event, struct nf_conntrack_expect *exp, u32 portid, int report); diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index fefd38db95b3..0294f3d473af 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -113,7 +113,6 @@ struct netns_ct { struct ct_pcpu __percpu *pcpu_lists; struct ip_conntrack_stat __percpu *stat; struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb; - struct nf_exp_event_notifier __rcu *nf_expect_event_cb; struct nf_ip_net nf_ct_proto; #if defined(CONFIG_NF_CONNTRACK_LABELS) unsigned int labels_used; -- cgit v1.2.3 From 723783d077e39c256a1fafebbd97cbb14207c28f Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Thu, 26 Aug 2021 10:49:47 +0800 Subject: sock: remove one redundant SKB_FRAG_PAGE_ORDER macro Both SKB_FRAG_PAGE_ORDER are defined to the same value in net/core/sock.c and drivers/vhost/net.c. Move the SKB_FRAG_PAGE_ORDER definition to net/core/sock.h, as both net/core/sock.c and drivers/vhost/net.c include it, and it seems a reasonable file to put the macro. Signed-off-by: Yunsheng Lin Acked-by: Jason Wang Signed-off-by: David S. Miller --- include/net/sock.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 95b25777b53e..66a9a90f9558 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2717,6 +2717,7 @@ extern int sysctl_optmem_max; extern __u32 sysctl_wmem_default; extern __u32 sysctl_rmem_default; +#define SKB_FRAG_PAGE_ORDER get_order(32768) DECLARE_STATIC_KEY_FALSE(net_high_order_alloc_disable_key); static inline int sk_get_wmem0(const struct sock *sk, const struct proto *proto) -- cgit v1.2.3 From 49b99da2c9ce13ffcd93fe3a0f5670791c1d76f7 Mon Sep 17 00:00:00 2001 From: Rocco Yue Date: Fri, 27 Aug 2021 23:04:12 +0800 Subject: ipv6: add IFLA_INET6_RA_MTU to expose mtu value The kernel provides a "/proc/sys/net/ipv6/conf//mtu" file, which can temporarily record the mtu value of the last received RA message when the RA mtu value is lower than the interface mtu, but this proc has following limitations: (1) when the interface mtu (/sys/class/net//mtu) is updeated, mtu6 (/proc/sys/net/ipv6/conf//mtu) will be updated to the value of interface mtu; (2) mtu6 (/proc/sys/net/ipv6/conf//mtu) only affect ipv6 connection, and not affect ipv4. Therefore, when the mtu option is carried in the RA message, there will be a problem that the user sometimes cannot obtain RA mtu value correctly by reading mtu6. After this patch set, if a RA message carries the mtu option, you can send a netlink msg which nlmsg_type is RTM_GETLINK, and then by parsing the attribute of IFLA_INET6_RA_MTU to get the mtu value carried in the RA message received on the inet6 device. In addition, you can also get a link notification when ra_mtu is updated so it doesn't have to poll. In this way, if the MTU values that the device receives from the network in the PCO IPv4 and the RA IPv6 procedures are different, the user can obtain the correct ipv6 ra_mtu value and compare the value of ra_mtu and ipv4 mtu, then the device can use the lower MTU value for both IPv4 and IPv6. Signed-off-by: Rocco Yue Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20210827150412.9267-1-rocco.yue@mediatek.com Signed-off-by: Jakub Kicinski --- include/net/if_inet6.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index 42235c178b06..653e7d0f65cb 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -210,6 +210,8 @@ struct inet6_dev { unsigned long tstamp; /* ipv6InterfaceTable update timestamp */ struct rcu_head rcu; + + unsigned int ra_mtu; }; static inline void ipv6_eth_mc_map(const struct in6_addr *addr, char *buf) -- cgit v1.2.3 From 7a3f5b0de3647c854e34269c3332d7a1e902901a Mon Sep 17 00:00:00 2001 From: Ryoga Saito Date: Tue, 17 Aug 2021 08:39:37 +0000 Subject: netfilter: add netfilter hooks to SRv6 data plane This patch introduces netfilter hooks for solving the problem that conntrack couldn't record both inner flows and outer flows. This patch also introduces a new sysctl toggle for enabling lightweight tunnel netfilter hooks. Signed-off-by: Ryoga Saito Signed-off-by: Pablo Neira Ayuso --- include/net/lwtunnel.h | 3 +++ include/net/netfilter/nf_hooks_lwtunnel.h | 7 +++++++ 2 files changed, 10 insertions(+) create mode 100644 include/net/netfilter/nf_hooks_lwtunnel.h (limited to 'include/net') diff --git a/include/net/lwtunnel.h b/include/net/lwtunnel.h index 05cfd6ff6528..6f15e6fa154e 100644 --- a/include/net/lwtunnel.h +++ b/include/net/lwtunnel.h @@ -51,6 +51,9 @@ struct lwtunnel_encap_ops { }; #ifdef CONFIG_LWTUNNEL + +DECLARE_STATIC_KEY_FALSE(nf_hooks_lwtunnel_enabled); + void lwtstate_free(struct lwtunnel_state *lws); static inline struct lwtunnel_state * diff --git a/include/net/netfilter/nf_hooks_lwtunnel.h b/include/net/netfilter/nf_hooks_lwtunnel.h new file mode 100644 index 000000000000..52e27920f829 --- /dev/null +++ b/include/net/netfilter/nf_hooks_lwtunnel.h @@ -0,0 +1,7 @@ +#include +#include + +#ifdef CONFIG_SYSCTL +int nf_hooks_lwtunnel_sysctl_handler(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos); +#endif -- cgit v1.2.3 From ca49bfd90a9dde175d2929dc1544b54841e33804 Mon Sep 17 00:00:00 2001 From: Maxim Mikityanskiy Date: Thu, 26 Aug 2021 14:54:25 +0300 Subject: sch_htb: Fix inconsistency when leaf qdisc creation fails In HTB offload mode, qdiscs of leaf classes are grafted to netdev queues. sch_htb expects the dev_queue field of these qdiscs to point to the corresponding queues. However, qdisc creation may fail, and in that case noop_qdisc is used instead. Its dev_queue doesn't point to the right queue, so sch_htb can lose track of used netdev queues, which will cause internal inconsistencies. This commit fixes this bug by keeping track of the netdev queue inside struct htb_class. All reads of cl->leaf.q->dev_queue are replaced by the new field, the two values are synced on writes, and WARNs are added to assert equality of the two values. The driver API has changed: when TC_HTB_LEAF_DEL needs to move a queue, the driver used to pass the old and new queue IDs to sch_htb. Now that there is a new field (offload_queue) in struct htb_class that needs to be updated on this operation, the driver will pass the old class ID to sch_htb instead (it already knows the new class ID). Fixes: d03b195b5aa0 ("sch_htb: Hierarchical QoS hardware offload") Signed-off-by: Maxim Mikityanskiy Reviewed-by: Tariq Toukan Link: https://lore.kernel.org/r/20210826115425.1744053-1-maximmi@nvidia.com Signed-off-by: Jakub Kicinski --- include/net/pkt_cls.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 8fb47fc61097..83a6d0792180 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -816,10 +816,9 @@ enum tc_htb_command { struct tc_htb_qopt_offload { struct netlink_ext_ack *extack; enum tc_htb_command command; - u16 classid; u32 parent_classid; + u16 classid; u16 qid; - u16 moved_qid; u64 rate; u64 ceil; }; -- cgit v1.2.3 From b9edbfe1adecfc48fd11061dce68afb03d6adbdc Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Thu, 2 Sep 2021 14:36:17 +0900 Subject: flow: fix object-size-mismatch warning in flowi{4,6}_to_flowi_common() Commit 3df98d79215ace13 ("lsm,selinux: pass flowi_common instead of flowi to the LSM hooks") introduced flowi{4,6}_to_flowi_common() functions which cause UBSAN warning when building with LLVM 11.0.1 on Ubuntu 21.04. ================================================================================ UBSAN: object-size-mismatch in ./include/net/flow.h:197:33 member access within address ffffc9000109fbd8 with insufficient space for an object of type 'struct flowi' CPU: 2 PID: 7410 Comm: systemd-resolve Not tainted 5.14.0 #51 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 02/27/2020 Call Trace: dump_stack_lvl+0x103/0x171 ubsan_type_mismatch_common+0x1de/0x390 __ubsan_handle_type_mismatch_v1+0x41/0x50 udp_sendmsg+0xda2/0x1300 ? ip_skb_dst_mtu+0x1f0/0x1f0 ? sock_rps_record_flow+0xe/0x200 ? inet_send_prepare+0x2d/0x90 sock_sendmsg+0x49/0x80 ____sys_sendmsg+0x269/0x370 __sys_sendmsg+0x15e/0x1d0 ? syscall_enter_from_user_mode+0xf0/0x1b0 do_syscall_64+0x3d/0xb0 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f7081a50497 Code: 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 RSP: 002b:00007ffc153870f8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 000000000000000c RCX: 00007f7081a50497 RDX: 0000000000000000 RSI: 00007ffc15387140 RDI: 000000000000000c RBP: 00007ffc15387140 R08: 0000563f29a5e4fc R09: 000000000000cd28 R10: 0000563f29a68a30 R11: 0000000000000246 R12: 000000000000000c R13: 0000000000000001 R14: 0000563f29a68a30 R15: 0000563f29a5e50c ================================================================================ I don't think we need to call flowi{4,6}_to_flowi() from these functions because the first member of "struct flowi4" and "struct flowi6" is struct flowi_common __fl_common; while the first member of "struct flowi" is union { struct flowi_common __fl_common; struct flowi4 ip4; struct flowi6 ip6; struct flowidn dn; } u; which should point to the same address without access to "struct flowi". Signed-off-by: Tetsuo Handa Signed-off-by: David S. Miller --- include/net/flow.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/flow.h b/include/net/flow.h index 6f5e70240071..58beb16a49b8 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -194,7 +194,7 @@ static inline struct flowi *flowi4_to_flowi(struct flowi4 *fl4) static inline struct flowi_common *flowi4_to_flowi_common(struct flowi4 *fl4) { - return &(flowi4_to_flowi(fl4)->u.__fl_common); + return &(fl4->__fl_common); } static inline struct flowi *flowi6_to_flowi(struct flowi6 *fl6) @@ -204,7 +204,7 @@ static inline struct flowi *flowi6_to_flowi(struct flowi6 *fl6) static inline struct flowi_common *flowi6_to_flowi_common(struct flowi6 *fl6) { - return &(flowi6_to_flowi(fl6)->u.__fl_common); + return &(fl6->__fl_common); } static inline struct flowi *flowidn_to_flowi(struct flowidn *fldn) -- cgit v1.2.3 From a57d8c217aadac75530b8e7ffb3a3e1b7bfd0330 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 14 Sep 2021 16:47:26 +0300 Subject: net: dsa: flush switchdev workqueue before tearing down CPU/DSA ports Sometimes when unbinding the mv88e6xxx driver on Turris MOX, these error messages appear: mv88e6085 d0032004.mdio-mii:12: port 1 failed to delete be:79:b4:9e:9e:96 vid 1 from fdb: -2 mv88e6085 d0032004.mdio-mii:12: port 1 failed to delete be:79:b4:9e:9e:96 vid 0 from fdb: -2 mv88e6085 d0032004.mdio-mii:12: port 1 failed to delete d8:58:d7:00:ca:6d vid 100 from fdb: -2 mv88e6085 d0032004.mdio-mii:12: port 1 failed to delete d8:58:d7:00:ca:6d vid 1 from fdb: -2 mv88e6085 d0032004.mdio-mii:12: port 1 failed to delete d8:58:d7:00:ca:6d vid 0 from fdb: -2 (and similarly for other ports) What happens is that DSA has a policy "even if there are bugs, let's at least not leak memory" and dsa_port_teardown() clears the dp->fdbs and dp->mdbs lists, which are supposed to be empty. But deleting that cleanup code, the warnings go away. => the FDB and MDB lists (used for refcounting on shared ports, aka CPU and DSA ports) will eventually be empty, but are not empty by the time we tear down those ports. Aka we are deleting them too soon. The addresses that DSA complains about are host-trapped addresses: the local addresses of the ports, and the MAC address of the bridge device. The problem is that offloading those entries happens from a deferred work item scheduled by the SWITCHDEV_FDB_DEL_TO_DEVICE handler, and this races with the teardown of the CPU and DSA ports where the refcounting is kept. In fact, not only it races, but fundamentally speaking, if we iterate through the port list linearly, we might end up tearing down the shared ports even before we delete a DSA user port which has a bridge upper. So as it turns out, we need to first tear down the user ports (and the unused ones, for no better place of doing that), then the shared ports (the CPU and DSA ports). In between, we need to ensure that all work items scheduled by our switchdev handlers (which only run for user ports, hence the reason why we tear them down first) have finished. Fixes: 161ca59d39e9 ("net: dsa: reference count the MDB entries at the cross-chip notifier level") Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/20210914134726.2305133-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index f9a17145255a..258867eff230 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -447,6 +447,11 @@ static inline bool dsa_port_is_user(struct dsa_port *dp) return dp->type == DSA_PORT_TYPE_USER; } +static inline bool dsa_port_is_unused(struct dsa_port *dp) +{ + return dp->type == DSA_PORT_TYPE_UNUSED; +} + static inline bool dsa_is_unused_port(struct dsa_switch *ds, int p) { return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_UNUSED; -- cgit v1.2.3 From 0650bf52b31ff35dc6430fc2e37969c36baba724 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 17 Sep 2021 16:34:33 +0300 Subject: net: dsa: be compatible with masters which unregister on shutdown Lino reports that on his system with bcmgenet as DSA master and KSZ9897 as a switch, rebooting or shutting down never works properly. What does the bcmgenet driver have special to trigger this, that other DSA masters do not? It has an implementation of ->shutdown which simply calls its ->remove implementation. Otherwise said, it unregisters its network interface on shutdown. This message can be seen in a loop, and it hangs the reboot process there: unregister_netdevice: waiting for eth0 to become free. Usage count = 3 So why 3? A usage count of 1 is normal for a registered network interface, and any virtual interface which links itself as an upper of that will increment it via dev_hold. In the case of DSA, this is the call path: dsa_slave_create -> netdev_upper_dev_link -> __netdev_upper_dev_link -> __netdev_adjacent_dev_insert -> dev_hold So a DSA switch with 3 interfaces will result in a usage count elevated by two, and netdev_wait_allrefs will wait until they have gone away. Other stacked interfaces, like VLAN, watch NETDEV_UNREGISTER events and delete themselves, but DSA cannot just vanish and go poof, at most it can unbind itself from the switch devices, but that must happen strictly earlier compared to when the DSA master unregisters its net_device, so reacting on the NETDEV_UNREGISTER event is way too late. It seems that it is a pretty established pattern to have a driver's ->shutdown hook redirect to its ->remove hook, so the same code is executed regardless of whether the driver is unbound from the device, or the system is just shutting down. As Florian puts it, it is quite a big hammer for bcmgenet to unregister its net_device during shutdown, but having a common code path with the driver unbind helps ensure it is well tested. So DSA, for better or for worse, has to live with that and engage in an arms race of implementing the ->shutdown hook too, from all individual drivers, and do something sane when paired with masters that unregister their net_device there. The only sane thing to do, of course, is to unlink from the master. However, complications arise really quickly. The pattern of redirecting ->shutdown to ->remove is not unique to bcmgenet or even to net_device drivers. In fact, SPI controllers do it too (see dspi_shutdown -> dspi_remove), and presumably, I2C controllers and MDIO controllers do it too (this is something I have not researched too deeply, but even if this is not the case today, it is certainly plausible to happen in the future, and must be taken into consideration). Since DSA switches might be SPI devices, I2C devices, MDIO devices, the insane implication is that for the exact same DSA switch device, we might have both ->shutdown and ->remove getting called. So we need to do something with that insane environment. The pattern I've come up with is "if this, then not that", so if either ->shutdown or ->remove gets called, we set the device's drvdata to NULL, and in the other hook, we check whether the drvdata is NULL and just do nothing. This is probably not necessary for platform devices, just for devices on buses, but I would really insist for consistency among drivers, because when code is copy-pasted, it is not always copy-pasted from the best sources. So depending on whether the DSA switch's ->remove or ->shutdown will get called first, we cannot really guarantee even for the same driver if rebooting will result in the same code path on all platforms. But nonetheless, we need to do something minimally reasonable on ->shutdown too to fix the bug. Of course, the ->remove will do more (a full teardown of the tree, with all data structures freed, and this is why the bug was not caught for so long). The new ->shutdown method is kept separate from dsa_unregister_switch not because we couldn't have unregistered the switch, but simply in the interest of doing something quick and to the point. The big question is: does the DSA switch's ->shutdown get called earlier than the DSA master's ->shutdown? If not, there is still a risk that we might still trigger the WARN_ON in unregister_netdevice that says we are attempting to unregister a net_device which has uppers. That's no good. Although the reference to the master net_device won't physically go away even if DSA's ->shutdown comes afterwards, remember we have a dev_hold on it. The answer to that question lies in this comment above device_link_add: * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending * on it to the ends of these lists (that does not happen to devices that have * not been registered when this function is called). so the fact that DSA uses device_link_add towards its master is not exactly for nothing. device_shutdown() walks devices_kset from the back, so this is our guarantee that DSA's shutdown happens before the master's shutdown. Fixes: 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings") Link: https://lore.kernel.org/netdev/20210909095324.12978-1-LinoSanfilippo@gmx.de/ Reported-by: Lino Sanfilippo Signed-off-by: Vladimir Oltean Tested-by: Andrew Lunn Signed-off-by: David S. Miller --- include/net/dsa.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 258867eff230..6e29c0e080f6 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -1046,6 +1046,7 @@ static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr, void dsa_unregister_switch(struct dsa_switch *ds); int dsa_register_switch(struct dsa_switch *ds); +void dsa_switch_shutdown(struct dsa_switch *ds); struct dsa_switch *dsa_switch_find(int tree_index, int sw_index); #ifdef CONFIG_PM_SLEEP int dsa_switch_suspend(struct dsa_switch *ds); -- cgit v1.2.3 From 2dcb96bacce36021c2f3eaae0cef607b5bb71ede Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 18 Sep 2021 14:42:35 +0200 Subject: net: core: Correct the sock::sk_lock.owned lockdep annotations lock_sock_fast() and lock_sock_nested() contain lockdep annotations for the sock::sk_lock.owned 'mutex'. sock::sk_lock.owned is not a regular mutex. It is just lockdep wise equivalent. In fact it's an open coded trivial mutex implementation with some interesting features. sock::sk_lock.slock is a regular spinlock protecting the 'mutex' representation sock::sk_lock.owned which is a plain boolean. If 'owned' is true, then some other task holds the 'mutex', otherwise it is uncontended. As this locking construct is obviously endangered by lock ordering issues as any other locking primitive it got lockdep annotated via a dedicated dependency map sock::sk_lock.dep_map which has to be updated at the lock and unlock sites. lock_sock_nested() is a straight forward 'mutex' lock operation: might_sleep(); spin_lock_bh(sock::sk_lock.slock) while (!try_lock(sock::sk_lock.owned)) { spin_unlock_bh(sock::sk_lock.slock); wait_for_release(); spin_lock_bh(sock::sk_lock.slock); } The lockdep annotation for sock::sk_lock.owned is for unknown reasons _after_ the lock has been acquired, i.e. after the code block above and after releasing sock::sk_lock.slock, but inside the bottom halves disabled region: spin_unlock(sock::sk_lock.slock); mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_); local_bh_enable(); The placement after the unlock is obvious because otherwise the mutex_acquire() would nest into the spin lock held region. But that's from the lockdep perspective still the wrong place: 1) The mutex_acquire() is issued _after_ the successful acquisition which is pointless because in a dead lock scenario this point is never reached which means that if the deadlock is the first instance of exposing the wrong lock order lockdep does not have a chance to detect it. 2) It only works because lockdep is rather lax on the context from which the mutex_acquire() is issued. Acquiring a mutex inside a bottom halves and therefore non-preemptible region is obviously invalid, except for a trylock which is clearly not the case here. This 'works' stops working on RT enabled kernels where the bottom halves serialization is done via a local lock, which exposes this misplacement because the 'mutex' and the local lock nest the wrong way around and lockdep complains rightfully about a lock inversion. The placement is wrong since the initial commit a5b5bb9a053a ("[PATCH] lockdep: annotate sk_locks") which introduced this. Fix it by moving the mutex_acquire() in front of the actual lock acquisition, which is what the regular mutex_lock() operation does as well. lock_sock_fast() is not that straight forward. It looks at the first glance like a convoluted trylock operation: spin_lock_bh(sock::sk_lock.slock) if (!sock::sk_lock.owned) return false; while (!try_lock(sock::sk_lock.owned)) { spin_unlock_bh(sock::sk_lock.slock); wait_for_release(); spin_lock_bh(sock::sk_lock.slock); } spin_unlock(sock::sk_lock.slock); mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_); local_bh_enable(); return true; But that's not the case: lock_sock_fast() is an interesting optimization for short critical sections which can run with bottom halves disabled and sock::sk_lock.slock held. This allows to shortcut the 'mutex' operation in the non contended case by preventing other lockers to acquire sock::sk_lock.owned because they are blocked on sock::sk_lock.slock, which in turn avoids the overhead of doing the heavy processing in release_sock() including waking up wait queue waiters. In the contended case, i.e. when sock::sk_lock.owned == true the behavior is the same as lock_sock_nested(). Semantically this shortcut means, that the task acquired the 'mutex' even if it does not touch the sock::sk_lock.owned field in the non-contended case. Not telling lockdep about this shortcut acquisition is hiding potential lock ordering violations in the fast path. As a consequence the same reasoning as for the above lock_sock_nested() case vs. the placement of the lockdep annotation applies. The current placement of the lockdep annotation was just copied from the original lock_sock(), now renamed to lock_sock_nested(), implementation. Fix this by moving the mutex_acquire() in front of the actual lock acquisition and adding the corresponding mutex_release() into unlock_sock_fast(). Also document the fast path return case with a comment. Reported-by: Sebastian Siewior Signed-off-by: Thomas Gleixner Cc: netdev@vger.kernel.org Cc: "David S. Miller" Cc: Jakub Kicinski Cc: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 66a9a90f9558..c005c3c750e8 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1640,6 +1640,7 @@ static inline void unlock_sock_fast(struct sock *sk, bool slow) release_sock(sk); __release(&sk->sk_lock.slock); } else { + mutex_release(&sk->sk_lock.dep_map, _RET_IP_); spin_unlock_bh(&sk->sk_lock.slock); } } -- cgit v1.2.3 From fd292c189a979838622d5e03e15fa688c81dd50b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 17 Sep 2021 17:29:16 +0300 Subject: net: dsa: tear down devlink port regions when tearing down the devlink port on error Commit 86f8b1c01a0a ("net: dsa: Do not make user port errors fatal") decided it was fine to ignore errors on certain ports that fail to probe, and go on with the ports that do probe fine. Commit fb6ec87f7229 ("net: dsa: Fix type was not set for devlink port") noticed that devlink_port_type_eth_set(dlp, dp->slave); does not get called, and devlink notices after a timeout of 3600 seconds and prints a WARN_ON. So it went ahead to unregister the devlink port. And because there exists an UNUSED port flavour, we actually re-register the devlink port as UNUSED. Commit 08156ba430b4 ("net: dsa: Add devlink port regions support to DSA") added devlink port regions, which are set up by the driver and not by DSA. When we trigger the devlink port deregistration and reregistration as unused, devlink now prints another WARN_ON, from here: devlink_port_unregister: WARN_ON(!list_empty(&devlink_port->region_list)); So the port still has regions, which makes sense, because they were set up by the driver, and the driver doesn't know we're unregistering the devlink port. Somebody needs to tear them down, and optionally (actually it would be nice, to be consistent) set them up again for the new devlink port. But DSA's layering stays in our way quite badly here. The options I've considered are: 1. Introduce a function in devlink to just change a port's type and flavour. No dice, devlink keeps a lot of state, it really wants the port to not be registered when you set its parameters, so changing anything can only be done by destroying what we currently have and recreating it. 2. Make DSA cache the parameters passed to dsa_devlink_port_region_create, and the region returned, keep those in a list, then when the devlink port unregister needs to take place, the existing devlink regions are destroyed by DSA, and we replay the creation of new regions using the cached parameters. Problem: mv88e6xxx keeps the region pointers in chip->ports[port].region, and these will remain stale after DSA frees them. There are many things DSA can do, but updating mv88e6xxx's private pointers is not one of them. 3. Just let the driver do it (i.e. introduce a very specific method called ds->ops->port_reinit_as_unused, which unregisters its devlink port devlink regions, then the old devlink port, then registers the new one, then the devlink port regions for it). While it does work, as opposed to the others, it's pretty horrible from an API perspective and we can do better. 4. Introduce a new pair of methods, ->port_setup and ->port_teardown, which in the case of mv88e6xxx must register and unregister the devlink port regions. Call these 2 methods when the port must be reinitialized as unused. Naturally, I went for the 4th approach. Fixes: 08156ba430b4 ("net: dsa: Add devlink port regions support to DSA") Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- include/net/dsa.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 6e29c0e080f6..d784e76113b8 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -585,8 +585,16 @@ struct dsa_switch_ops { int (*change_tag_protocol)(struct dsa_switch *ds, int port, enum dsa_tag_protocol proto); + /* Optional switch-wide initialization and destruction methods */ int (*setup)(struct dsa_switch *ds); void (*teardown)(struct dsa_switch *ds); + + /* Per-port initialization and destruction methods. Mandatory if the + * driver registers devlink port regions, optional otherwise. + */ + int (*port_setup)(struct dsa_switch *ds, int port); + void (*port_teardown)(struct dsa_switch *ds, int port); + u32 (*get_phy_flags)(struct dsa_switch *ds, int port); /* -- cgit v1.2.3 From 597aa16c782496bf74c5dc3b45ff472ade6cee64 Mon Sep 17 00:00:00 2001 From: Xiao Liang Date: Thu, 23 Sep 2021 23:03:19 +0800 Subject: net: ipv4: Fix rtnexthop len when RTA_FLOW is present Multipath RTA_FLOW is embedded in nexthop. Dump it in fib_add_nexthop() to get the length of rtnexthop correct. Fixes: b0f60193632e ("ipv4: Refactor nexthop attributes in fib_dump_info") Signed-off-by: Xiao Liang Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip_fib.h | 2 +- include/net/nexthop.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 21c5386d4a6d..ab5348e57db1 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -597,5 +597,5 @@ int ip_valid_fib_dump_req(struct net *net, const struct nlmsghdr *nlh, int fib_nexthop_info(struct sk_buff *skb, const struct fib_nh_common *nh, u8 rt_family, unsigned char *flags, bool skip_oif); int fib_add_nexthop(struct sk_buff *skb, const struct fib_nh_common *nh, - int nh_weight, u8 rt_family); + int nh_weight, u8 rt_family, u32 nh_tclassid); #endif /* _NET_FIB_H */ diff --git a/include/net/nexthop.h b/include/net/nexthop.h index 10e1777877e6..28085b995ddc 100644 --- a/include/net/nexthop.h +++ b/include/net/nexthop.h @@ -325,7 +325,7 @@ int nexthop_mpath_fill_node(struct sk_buff *skb, struct nexthop *nh, struct fib_nh_common *nhc = &nhi->fib_nhc; int weight = nhg->nh_entries[i].weight; - if (fib_add_nexthop(skb, nhc, weight, rt_family) < 0) + if (fib_add_nexthop(skb, nhc, weight, rt_family, 0) < 0) return -EMSGSIZE; } -- cgit v1.2.3 From b193e15ac69d56f35e1d8e2b5d16cbd47764d053 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=8E=8B=E8=B4=87?= Date: Fri, 24 Sep 2021 10:35:58 +0800 Subject: net: prevent user from passing illegal stab size We observed below report when playing with netlink sock: UBSAN: shift-out-of-bounds in net/sched/sch_api.c:580:10 shift exponent 249 is too large for 32-bit type CPU: 0 PID: 685 Comm: a.out Not tainted Call Trace: dump_stack_lvl+0x8d/0xcf ubsan_epilogue+0xa/0x4e __ubsan_handle_shift_out_of_bounds+0x161/0x182 __qdisc_calculate_pkt_len+0xf0/0x190 __dev_queue_xmit+0x2ed/0x15b0 it seems like kernel won't check the stab log value passing from user, and will use the insane value later to calculate pkt_len. This patch just add a check on the size/cell_log to avoid insane calculation. Reported-by: Abaci Signed-off-by: Michael Wang Signed-off-by: David S. Miller --- include/net/pkt_sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 6d7b12cba015..bf79f3a890af 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -11,6 +11,7 @@ #include #define DEFAULT_TX_QUEUE_LEN 1000 +#define STAB_SIZE_LOG_MAX 30 struct qdisc_walker { int stop; -- cgit v1.2.3 From 33092aca857bf35a8e9cac0e8340c685a4796e90 Mon Sep 17 00:00:00 2001 From: Alexander Wetzel Date: Fri, 24 Sep 2021 22:05:14 +0200 Subject: mac80211: Fix Ptk0 rekey documentation @IEEE80211_KEY_FLAG_GENERATE_IV setting is irrelevant for RX. Move the requirement to the correct section in the PTK0 rekey documentation. Signed-off-by: Alexander Wetzel Link: https://lore.kernel.org/r/20210924200514.7936-1-alexander@wetzel-home.de Signed-off-by: Johannes Berg --- include/net/mac80211.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index af0fc13cea34..618d1f427cb2 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2818,13 +2818,13 @@ void ieee80211_free_txskb(struct ieee80211_hw *hw, struct sk_buff *skb); * Mac80211 drivers should set the @NL80211_EXT_FEATURE_CAN_REPLACE_PTK0 flag * when they are able to replace in-use PTK keys according to the following * requirements: - * 1) They do not hand over frames decrypted with the old key to - mac80211 once the call to set_key() with command %DISABLE_KEY has been - completed when also setting @IEEE80211_KEY_FLAG_GENERATE_IV for any key, + * 1) They do not hand over frames decrypted with the old key to mac80211 + once the call to set_key() with command %DISABLE_KEY has been completed, 2) either drop or continue to use the old key for any outgoing frames queued at the time of the key deletion (including re-transmits), 3) never send out a frame queued prior to the set_key() %SET_KEY command - encrypted with the new key and + encrypted with the new key when also needing + @IEEE80211_KEY_FLAG_GENERATE_IV and 4) never send out a frame unencrypted when it should be encrypted. Mac80211 will not queue any new frames for a deleted key to the driver. */ -- cgit v1.2.3 From 339031bafe6b281cf2dcb8364217288b9fdab555 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 23 Sep 2021 16:44:34 +0200 Subject: netfilter: conntrack: fix boot failure with nf_conntrack.enable_hooks=1 This is a revert of 7b1957b049 ("netfilter: nf_defrag_ipv4: use net_generic infra") and a partial revert of 8b0adbe3e3 ("netfilter: nf_defrag_ipv6: use net_generic infra"). If conntrack is builtin and kernel is booted with: nf_conntrack.enable_hooks=1 .... kernel will fail to boot due to a NULL deref in nf_defrag_ipv4_enable(): Its called before the ipv4 defrag initcall is made, so net_generic() returns NULL. To resolve this, move the user refcount back to struct net so calls to those functions are possible even before their initcalls have run. Fixes: 7b1957b04956 ("netfilter: nf_defrag_ipv4: use net_generic infra") Fixes: 8b0adbe3e38d ("netfilter: nf_defrag_ipv6: use net_generic infra"). Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/ipv6/nf_defrag_ipv6.h | 1 - include/net/netns/netfilter.h | 6 ++++++ 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netfilter/ipv6/nf_defrag_ipv6.h b/include/net/netfilter/ipv6/nf_defrag_ipv6.h index 0fd8a4159662..ceadf8ba25a4 100644 --- a/include/net/netfilter/ipv6/nf_defrag_ipv6.h +++ b/include/net/netfilter/ipv6/nf_defrag_ipv6.h @@ -17,7 +17,6 @@ struct inet_frags_ctl; struct nft_ct_frag6_pernet { struct ctl_table_header *nf_frag_frags_hdr; struct fqdir *fqdir; - unsigned int users; }; #endif /* _NF_DEFRAG_IPV6_H */ diff --git a/include/net/netns/netfilter.h b/include/net/netns/netfilter.h index 986a2a9cfdfa..b593f95e9991 100644 --- a/include/net/netns/netfilter.h +++ b/include/net/netns/netfilter.h @@ -27,5 +27,11 @@ struct netns_nf { #if IS_ENABLED(CONFIG_DECNET) struct nf_hook_entries __rcu *hooks_decnet[NF_DN_NUMHOOKS]; #endif +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) + unsigned int defrag_ipv4_users; +#endif +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) + unsigned int defrag_ipv6_users; +#endif }; #endif -- cgit v1.2.3 From 49054556289e8787501630b7c7a9d407da02e296 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 29 Sep 2021 11:59:17 +0200 Subject: net: introduce and use lock_sock_fast_nested() Syzkaller reported a false positive deadlock involving the nl socket lock and the subflow socket lock: MPTCP: kernel_bind error, err=-98 ============================================ WARNING: possible recursive locking detected 5.15.0-rc1-syzkaller #0 Not tainted -------------------------------------------- syz-executor998/6520 is trying to acquire lock: ffff8880795718a0 (k-sk_lock-AF_INET){+.+.}-{0:0}, at: mptcp_close+0x267/0x7b0 net/mptcp/protocol.c:2738 but task is already holding lock: ffff8880787c8c60 (k-sk_lock-AF_INET){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1612 [inline] ffff8880787c8c60 (k-sk_lock-AF_INET){+.+.}-{0:0}, at: mptcp_close+0x23/0x7b0 net/mptcp/protocol.c:2720 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(k-sk_lock-AF_INET); lock(k-sk_lock-AF_INET); *** DEADLOCK *** May be due to missing lock nesting notation 3 locks held by syz-executor998/6520: #0: ffffffff8d176c50 (cb_lock){++++}-{3:3}, at: genl_rcv+0x15/0x40 net/netlink/genetlink.c:802 #1: ffffffff8d176d08 (genl_mutex){+.+.}-{3:3}, at: genl_lock net/netlink/genetlink.c:33 [inline] #1: ffffffff8d176d08 (genl_mutex){+.+.}-{3:3}, at: genl_rcv_msg+0x3e0/0x580 net/netlink/genetlink.c:790 #2: ffff8880787c8c60 (k-sk_lock-AF_INET){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1612 [inline] #2: ffff8880787c8c60 (k-sk_lock-AF_INET){+.+.}-{0:0}, at: mptcp_close+0x23/0x7b0 net/mptcp/protocol.c:2720 stack backtrace: CPU: 1 PID: 6520 Comm: syz-executor998 Not tainted 5.15.0-rc1-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_deadlock_bug kernel/locking/lockdep.c:2944 [inline] check_deadlock kernel/locking/lockdep.c:2987 [inline] validate_chain kernel/locking/lockdep.c:3776 [inline] __lock_acquire.cold+0x149/0x3ab kernel/locking/lockdep.c:5015 lock_acquire kernel/locking/lockdep.c:5625 [inline] lock_acquire+0x1ab/0x510 kernel/locking/lockdep.c:5590 lock_sock_fast+0x36/0x100 net/core/sock.c:3229 mptcp_close+0x267/0x7b0 net/mptcp/protocol.c:2738 inet_release+0x12e/0x280 net/ipv4/af_inet.c:431 __sock_release net/socket.c:649 [inline] sock_release+0x87/0x1b0 net/socket.c:677 mptcp_pm_nl_create_listen_socket+0x238/0x2c0 net/mptcp/pm_netlink.c:900 mptcp_nl_cmd_add_addr+0x359/0x930 net/mptcp/pm_netlink.c:1170 genl_family_rcv_msg_doit+0x228/0x320 net/netlink/genetlink.c:731 genl_family_rcv_msg net/netlink/genetlink.c:775 [inline] genl_rcv_msg+0x328/0x580 net/netlink/genetlink.c:792 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2504 genl_rcv+0x24/0x40 net/netlink/genetlink.c:803 netlink_unicast_kernel net/netlink/af_netlink.c:1314 [inline] netlink_unicast+0x533/0x7d0 net/netlink/af_netlink.c:1340 netlink_sendmsg+0x86d/0xdb0 net/netlink/af_netlink.c:1929 sock_sendmsg_nosec net/socket.c:704 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:724 sock_no_sendpage+0x101/0x150 net/core/sock.c:2980 kernel_sendpage.part.0+0x1a0/0x340 net/socket.c:3504 kernel_sendpage net/socket.c:3501 [inline] sock_sendpage+0xe5/0x140 net/socket.c:1003 pipe_to_sendpage+0x2ad/0x380 fs/splice.c:364 splice_from_pipe_feed fs/splice.c:418 [inline] __splice_from_pipe+0x43e/0x8a0 fs/splice.c:562 splice_from_pipe fs/splice.c:597 [inline] generic_splice_sendpage+0xd4/0x140 fs/splice.c:746 do_splice_from fs/splice.c:767 [inline] direct_splice_actor+0x110/0x180 fs/splice.c:936 splice_direct_to_actor+0x34b/0x8c0 fs/splice.c:891 do_splice_direct+0x1b3/0x280 fs/splice.c:979 do_sendfile+0xae9/0x1240 fs/read_write.c:1249 __do_sys_sendfile64 fs/read_write.c:1314 [inline] __se_sys_sendfile64 fs/read_write.c:1300 [inline] __x64_sys_sendfile64+0x1cc/0x210 fs/read_write.c:1300 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f215cb69969 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 e1 14 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffc96bb3868 EFLAGS: 00000246 ORIG_RAX: 0000000000000028 RAX: ffffffffffffffda RBX: 00007f215cbad072 RCX: 00007f215cb69969 RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000005 RBP: 0000000000000000 R08: 00007ffc96bb3a08 R09: 00007ffc96bb3a08 R10: 0000000100000002 R11: 0000000000000246 R12: 00007ffc96bb387c R13: 431bde82d7b634db R14: 0000000000000000 R15: 0000000000000000 the problem originates from uncorrect lock annotation in the mptcp code and is only visible since commit 2dcb96bacce3 ("net: core: Correct the sock::sk_lock.owned lockdep annotations"), but is present since the port-based endpoint support initial implementation. This patch addresses the issue introducing a nested variant of lock_sock_fast() and using it in the relevant code path. Fixes: 1729cf186d8a ("mptcp: create the listening socket for new port") Fixes: 2dcb96bacce3 ("net: core: Correct the sock::sk_lock.owned lockdep annotations") Suggested-by: Thomas Gleixner Reported-and-tested-by: syzbot+1dd53f7a89b299d59eaf@syzkaller.appspotmail.com Signed-off-by: Paolo Abeni Reviewed-by: Thomas Gleixner Signed-off-by: David S. Miller --- include/net/sock.h | 31 ++++++++++++++++++++++++++++++- 1 file changed, 30 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index c005c3c750e8..dc3f8169312e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1623,7 +1623,36 @@ void release_sock(struct sock *sk); SINGLE_DEPTH_NESTING) #define bh_unlock_sock(__sk) spin_unlock(&((__sk)->sk_lock.slock)) -bool lock_sock_fast(struct sock *sk) __acquires(&sk->sk_lock.slock); +bool __lock_sock_fast(struct sock *sk) __acquires(&sk->sk_lock.slock); + +/** + * lock_sock_fast - fast version of lock_sock + * @sk: socket + * + * This version should be used for very small section, where process wont block + * return false if fast path is taken: + * + * sk_lock.slock locked, owned = 0, BH disabled + * + * return true if slow path is taken: + * + * sk_lock.slock unlocked, owned = 1, BH enabled + */ +static inline bool lock_sock_fast(struct sock *sk) +{ + /* The sk_lock has mutex_lock() semantics here. */ + mutex_acquire(&sk->sk_lock.dep_map, 0, 0, _RET_IP_); + + return __lock_sock_fast(sk); +} + +/* fast socket lock variant for caller already holding a [different] socket lock */ +static inline bool lock_sock_fast_nested(struct sock *sk) +{ + mutex_acquire(&sk->sk_lock.dep_map, SINGLE_DEPTH_NESTING, 0, _RET_IP_); + + return __lock_sock_fast(sk); +} /** * unlock_sock_fast - complement of lock_sock_fast -- cgit v1.2.3 From 35306eb23814444bd4021f8a1c3047d3cb0c8b2b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 29 Sep 2021 15:57:50 -0700 Subject: af_unix: fix races in sk_peer_pid and sk_peer_cred accesses Jann Horn reported that SO_PEERCRED and SO_PEERGROUPS implementations are racy, as af_unix can concurrently change sk_peer_pid and sk_peer_cred. In order to fix this issue, this patch adds a new spinlock that needs to be used whenever these fields are read or written. Jann also pointed out that l2cap_sock_get_peer_pid_cb() is currently reading sk->sk_peer_pid which makes no sense, as this field is only possibly set by AF_UNIX sockets. We will have to clean this in a separate patch. This could be done by reverting b48596d1dc25 "Bluetooth: L2CAP: Add get_peer_pid callback" or implementing what was truly expected. Fixes: 109f6e39fa07 ("af_unix: Allow SO_PEERCRED to work across namespaces.") Signed-off-by: Eric Dumazet Reported-by: Jann Horn Cc: Eric W. Biederman Cc: Luiz Augusto von Dentz Cc: Marcel Holtmann Signed-off-by: David S. Miller --- include/net/sock.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index dc3f8169312e..ae929e21a376 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -488,8 +488,10 @@ struct sock { u8 sk_prefer_busy_poll; u16 sk_busy_poll_budget; #endif + spinlock_t sk_peer_lock; struct pid *sk_peer_pid; const struct cred *sk_peer_cred; + long sk_rcvtimeo; ktime_t sk_stamp; #if BITS_PER_LONG==32 -- cgit v1.2.3 From 5fb14d20f8241461d351bef73e49871e4b2330ab Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 1 Oct 2021 09:46:22 -0700 Subject: net: add kerneldoc comment for sk_peer_lock Fixes following warning: include/net/sock.h:533: warning: Function parameter or member 'sk_peer_lock' not described in 'sock' Fixes: 35306eb23814 ("af_unix: fix races in sk_peer_pid and sk_peer_cred accesses") Signed-off-by: Eric Dumazet Reported-by: Stephen Rothwell Link: https://lore.kernel.org/r/20211001164622.58520-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index ae929e21a376..ea6fbc88c8f9 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -307,6 +307,7 @@ struct bpf_local_storage; * @sk_priority: %SO_PRIORITY setting * @sk_type: socket type (%SOCK_STREAM, etc) * @sk_protocol: which protocol this socket belongs in this network family + * @sk_peer_lock: lock protecting @sk_peer_pid and @sk_peer_cred * @sk_peer_pid: &struct pid for this socket's peer * @sk_peer_cred: %SO_PEERCRED setting * @sk_rcvlowat: %SO_RCVLOWAT setting -- cgit v1.2.3 From 6fb721cf781808ee2ca5e737fb0592cc68de3381 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 26 Sep 2021 09:59:35 +0200 Subject: netfilter: nf_tables: honor NLM_F_CREATE and NLM_F_EXCL in event notification Include the NLM_F_CREATE and NLM_F_EXCL flags in netlink event notifications, otherwise userspace cannot distiguish between create and add commands. Fixes: 96518518cc41 ("netfilter: add nftables") Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 148f5d8ee5ab..a16171c5fd9e 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1202,7 +1202,7 @@ struct nft_object *nft_obj_lookup(const struct net *net, void nft_obj_notify(struct net *net, const struct nft_table *table, struct nft_object *obj, u32 portid, u32 seq, - int event, int family, int report, gfp_t gfp); + int event, u16 flags, int family, int report, gfp_t gfp); /** * struct nft_object_type - stateful object type -- cgit v1.2.3 From 075718fdaf0efe20223571236c1bf14ca35a7aa1 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 14 Oct 2021 00:50:55 -0400 Subject: sctp: fix transport encap_port update in sctp_vtag_verify transport encap_port update should be updated when sctp_vtag_verify() succeeds, namely, returns 1, not returns 0. Correct it in this patch. While at it, also fix the indentation. Fixes: a1dd2cf2f1ae ("sctp: allow changing transport encap_port by peer packets") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 2eb6d7c2c931..f37c7a558d6d 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -384,11 +384,11 @@ sctp_vtag_verify(const struct sctp_chunk *chunk, * Verification Tag value does not match the receiver's own * tag value, the receiver shall silently discard the packet... */ - if (ntohl(chunk->sctp_hdr->vtag) == asoc->c.my_vtag) - return 1; + if (ntohl(chunk->sctp_hdr->vtag) != asoc->c.my_vtag) + return 0; chunk->transport->encap_port = SCTP_INPUT_CB(chunk->skb)->encap_port; - return 0; + return 1; } /* Check VTAG of the packet matches the sender's own tag and the T bit is -- cgit v1.2.3 From a76c2315bec7afeb9bafe776fe532106fa0a8b05 Mon Sep 17 00:00:00 2001 From: Leonard Crestez Date: Fri, 15 Oct 2021 10:26:05 +0300 Subject: tcp: md5: Allow MD5SIG_FLAG_IFINDEX with ifindex=0 Multiple VRFs are generally meant to be "separate" but right now md5 keys for the default VRF also affect connections inside VRFs if the IP addresses happen to overlap. So far the combination of TCP_MD5SIG_FLAG_IFINDEX with tcpm_ifindex == 0 was an error, accept this to mean "key only applies to default VRF". This is what applications using VRFs for traffic separation want. Signed-off-by: Leonard Crestez Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/tcp.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 3166dc15d7d6..60c384569e9c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1576,6 +1576,7 @@ struct tcp_md5sig_key { u8 keylen; u8 family; /* AF_INET or AF_INET6 */ u8 prefixlen; + u8 flags; union tcp_md5_addr addr; int l3index; /* set if key added with L3 scope */ u8 key[TCP_MD5SIG_MAXKEYLEN]; @@ -1621,10 +1622,10 @@ struct tcp_md5sig_pool { int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, const struct sock *sk, const struct sk_buff *skb); int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, - int family, u8 prefixlen, int l3index, + int family, u8 prefixlen, int l3index, u8 flags, const u8 *newkey, u8 newkeylen, gfp_t gfp); int tcp_md5_do_del(struct sock *sk, const union tcp_md5_addr *addr, - int family, u8 prefixlen, int l3index); + int family, u8 prefixlen, int l3index, u8 flags); struct tcp_md5sig_key *tcp_v4_md5_lookup(const struct sock *sk, const struct sock *addr_sk); -- cgit v1.2.3 From b416beb25d9317499c823aea1522eefd8d72ea36 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Mon, 18 Oct 2021 11:29:34 +0800 Subject: mctp: unify sockaddr_mctp types Use the more precise __kernel_sa_family_t for smctp_family, to match struct sockaddr. Also, use an unsigned int for the network member; negative networks don't make much sense. We're already using unsigned for mctp_dev and mctp_skb_cb, but need to change mctp_sock to suit. Fixes: 60fc63981693 ("mctp: Add sockaddr_mctp to uapi") Signed-off-by: Jeremy Kerr Acked-by: Eugene Syromiatnikov Signed-off-by: David S. Miller --- include/net/mctp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/mctp.h b/include/net/mctp.h index a824d47c3c6d..ffd2c23bd76d 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -54,7 +54,7 @@ struct mctp_sock { struct sock sk; /* bind() params */ - int bind_net; + unsigned int bind_net; mctp_eid_t bind_addr; __u8 bind_type; -- cgit v1.2.3 From 09b1d5dc6ce1c9151777f6c4e128a59457704c97 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 25 Oct 2021 13:31:12 +0200 Subject: cfg80211: fix management registrations locking The management registrations locking was broken, the list was locked for each wdev, but cfg80211_mgmt_registrations_update() iterated it without holding all the correct spinlocks, causing list corruption. Rather than trying to fix it with fine-grained locking, just move the lock to the wiphy/rdev (still need the list on each wdev), we already need to hold the wdev lock to change it, so there's no contention on the lock in any case. This trivially fixes the bug since we hold one wdev's lock already, and now will hold the lock that protects all lists. Cc: stable@vger.kernel.org Reported-by: Jouni Malinen Fixes: 6cd536fe62ef ("cfg80211: change internal management frame registration API") Link: https://lore.kernel.org/r/20211025133111.5cf733eab0f4.I7b0abb0494ab712f74e2efcd24bb31ac33f7eee9@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 62dd8422e0dc..27336fc70467 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5376,7 +5376,6 @@ static inline void wiphy_unlock(struct wiphy *wiphy) * netdev and may otherwise be used by driver read-only, will be update * by cfg80211 on change_interface * @mgmt_registrations: list of registrations for management frames - * @mgmt_registrations_lock: lock for the list * @mgmt_registrations_need_update: mgmt registrations were updated, * need to propagate the update to the driver * @mtx: mutex used to lock data in this struct, may be used by drivers @@ -5423,7 +5422,6 @@ struct wireless_dev { u32 identifier; struct list_head mgmt_registrations; - spinlock_t mgmt_registrations_lock; u8 mgmt_registrations_need_update:1; struct mutex mtx; -- cgit v1.2.3 From 9122a70a6333705c0c35614ddc51c274ed1d3637 Mon Sep 17 00:00:00 2001 From: Cyril Strejc Date: Sun, 24 Oct 2021 22:14:25 +0200 Subject: net: multicast: calculate csum of looped-back and forwarded packets During a testing of an user-space application which transmits UDP multicast datagrams and utilizes multicast routing to send the UDP datagrams out of defined network interfaces, I've found a multicast router does not fill-in UDP checksum into locally produced, looped-back and forwarded UDP datagrams, if an original output NIC the datagrams are sent to has UDP TX checksum offload enabled. The datagrams are sent malformed out of the NIC the datagrams have been forwarded to. It is because: 1. If TX checksum offload is enabled on the output NIC, UDP checksum is not calculated by kernel and is not filled into skb data. 2. dev_loopback_xmit(), which is called solely by ip_mc_finish_output(), sets skb->ip_summed = CHECKSUM_UNNECESSARY unconditionally. 3. Since 35fc92a9 ("[NET]: Allow forwarding of ip_summed except CHECKSUM_COMPLETE"), the ip_summed value is preserved during forwarding. 4. If ip_summed != CHECKSUM_PARTIAL, checksum is not calculated during a packet egress. The minimum fix in dev_loopback_xmit(): 1. Preserves skb->ip_summed CHECKSUM_PARTIAL. This is the case when the original output NIC has TX checksum offload enabled. The effects are: a) If the forwarding destination interface supports TX checksum offloading, the NIC driver is responsible to fill-in the checksum. b) If the forwarding destination interface does NOT support TX checksum offloading, checksums are filled-in by kernel before skb is submitted to the NIC driver. c) For local delivery, checksum validation is skipped as in the case of CHECKSUM_UNNECESSARY, thanks to skb_csum_unnecessary(). 2. Translates ip_summed CHECKSUM_NONE to CHECKSUM_UNNECESSARY. It means, for CHECKSUM_NONE, the behavior is unmodified and is there to skip a looped-back packet local delivery checksum validation. Signed-off-by: Cyril Strejc Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- include/net/udp.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/udp.h b/include/net/udp.h index 360df454356c..909ecf447e0f 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -494,8 +494,9 @@ static inline struct sk_buff *udp_rcv_segment(struct sock *sk, * CHECKSUM_NONE in __udp_gso_segment. UDP GRO indeed builds partial * packets in udp_gro_complete_segment. As does UDP GSO, verified by * udp_send_skb. But when those packets are looped in dev_loopback_xmit - * their ip_summed is set to CHECKSUM_UNNECESSARY. Reset in this - * specific case, where PARTIAL is both correct and required. + * their ip_summed CHECKSUM_NONE is changed to CHECKSUM_UNNECESSARY. + * Reset in this specific case, where PARTIAL is both correct and + * required. */ if (skb->pkt_type == PACKET_LOOPBACK) skb->ip_summed = CHECKSUM_PARTIAL; -- cgit v1.2.3 From 7b50ecfcc6cdfe87488576bc3ed443dc8d083b90 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Fri, 8 Oct 2021 13:33:03 -0700 Subject: net: Rename ->stream_memory_read to ->sock_is_readable The proto ops ->stream_memory_read() is currently only used by TCP to check whether psock queue is empty or not. We need to rename it before reusing it for non-TCP protocols, and adjust the exsiting users accordingly. Signed-off-by: Cong Wang Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20211008203306.37525-2-xiyou.wangcong@gmail.com --- include/net/sock.h | 8 +++++++- include/net/tls.h | 2 +- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index ea6fbc88c8f9..463f390d90b3 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1208,7 +1208,7 @@ struct proto { #endif bool (*stream_memory_free)(const struct sock *sk, int wake); - bool (*stream_memory_read)(const struct sock *sk); + bool (*sock_is_readable)(struct sock *sk); /* Memory pressure */ void (*enter_memory_pressure)(struct sock *sk); void (*leave_memory_pressure)(struct sock *sk); @@ -2820,4 +2820,10 @@ void sock_set_sndtimeo(struct sock *sk, s64 secs); int sock_bind_add(struct sock *sk, struct sockaddr *addr, int addr_len); +static inline bool sk_is_readable(struct sock *sk) +{ + if (sk->sk_prot->sock_is_readable) + return sk->sk_prot->sock_is_readable(sk); + return false; +} #endif /* _SOCK_H */ diff --git a/include/net/tls.h b/include/net/tls.h index be4b3e1cac46..01d2e3744393 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -375,7 +375,7 @@ void tls_sw_release_resources_rx(struct sock *sk); void tls_sw_free_ctx_rx(struct tls_context *tls_ctx); int tls_sw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int flags, int *addr_len); -bool tls_sw_stream_read(const struct sock *sk); +bool tls_sw_sock_is_readable(struct sock *sk); ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); -- cgit v1.2.3 From da353fac65fede6b8b4cfe207f0d9408e3121105 Mon Sep 17 00:00:00 2001 From: Daniel Jordan Date: Wed, 27 Oct 2021 17:59:20 -0400 Subject: net/tls: Fix flipped sign in tls_err_abort() calls sk->sk_err appears to expect a positive value, a convention that ktls doesn't always follow and that leads to memory corruption in other code. For instance, [kworker] tls_encrypt_done(..., err=) tls_err_abort(.., err) sk->sk_err = err; [task] splice_from_pipe_feed ... tls_sw_do_sendpage if (sk->sk_err) { ret = -sk->sk_err; // ret is positive splice_from_pipe_feed (continued) ret = actor(...) // ret is still positive and interpreted as bytes // written, resulting in underflow of buf->len and // sd->len, leading to huge buf->offset and bogus // addresses computed in later calls to actor() Fix all tls_err_abort() callers to pass a negative error code consistently and centralize the error-prone sign flip there, throwing in a warning to catch future misuse and uninlining the function so it really does only warn once. Cc: stable@vger.kernel.org Fixes: c46234ebb4d1e ("tls: RX path for ktls") Reported-by: syzbot+b187b77c8474f9648fae@syzkaller.appspotmail.com Signed-off-by: Daniel Jordan Signed-off-by: David S. Miller --- include/net/tls.h | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index 01d2e3744393..1fffb206f09f 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -358,6 +358,7 @@ int tls_sk_query(struct sock *sk, int optname, char __user *optval, int __user *optlen); int tls_sk_attach(struct sock *sk, int optname, char __user *optval, unsigned int optlen); +void tls_err_abort(struct sock *sk, int err); int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx); void tls_sw_strparser_arm(struct sock *sk, struct tls_context *ctx); @@ -466,12 +467,6 @@ static inline bool tls_is_sk_tx_device_offloaded(struct sock *sk) #endif } -static inline void tls_err_abort(struct sock *sk, int err) -{ - sk->sk_err = err; - sk_error_report(sk); -} - static inline bool tls_bigint_increment(unsigned char *seq, int len) { int i; @@ -512,7 +507,7 @@ static inline void tls_advance_record_sn(struct sock *sk, struct cipher_context *ctx) { if (tls_bigint_increment(ctx->rec_seq, prot->rec_seq_size)) - tls_err_abort(sk, EBADMSG); + tls_err_abort(sk, -EBADMSG); if (prot->version != TLS_1_3_VERSION && prot->cipher_type != TLS_CIPHER_CHACHA20_POLY1305) -- cgit v1.2.3 From f7cc8890f30d3ddc785e2b2ddc647da5b4b3c3ec Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Wed, 27 Oct 2021 13:38:55 -0700 Subject: mptcp: fix corrupt receiver key in MPC + data + checksum MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit using packetdrill it's possible to observe that the receiver key contains random values when clients transmit MP_CAPABLE with data and checksum (as specified in RFC8684 §3.1). Fix the layout of mptcp_out_options, to avoid using the skb extension copy when writing the MP_CAPABLE sub-option. Fixes: d7b269083786 ("mptcp: shrink mptcp_out_options struct") Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/233 Reported-by: Poorva Sonparote Signed-off-by: Davide Caratti Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20211027203855.264600-1-mathew.j.martineau@linux.intel.com Signed-off-by: Jakub Kicinski --- include/net/mptcp.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 6026bbefbffd..3214848402ec 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -69,6 +69,10 @@ struct mptcp_out_options { struct { u64 sndr_key; u64 rcvr_key; + u64 data_seq; + u32 subflow_seq; + u16 data_len; + __sum16 csum; }; struct { struct mptcp_addr_info addr; -- cgit v1.2.3