From 1ba5bf993c6a3142e18e68ea6452b347f9cb5635 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Tue, 5 Jul 2016 10:18:08 +0200 Subject: xfrm: fix crash in XFRM_MSG_GETSA netlink handler If we hit any of the error conditions inside xfrm_dump_sa(), then xfrm_state_walk_init() never gets called. However, we still call xfrm_state_walk_done() from xfrm_dump_sa_done(), which will crash because the state walk was never initialized properly. We can fix this by setting cb->args[0] only after we've processed the first element and checking this before calling xfrm_state_walk_done(). Fixes: d3623099d3 ("ipsec: add support of limited SA dump") Cc: Nicolas Dichtel Cc: Steffen Klassert Signed-off-by: Vegard Nossum Acked-by: Nicolas Dichtel Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index d516845e16e3..4fb04ced5867 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -896,7 +896,8 @@ static int xfrm_dump_sa_done(struct netlink_callback *cb) struct sock *sk = cb->skb->sk; struct net *net = sock_net(sk); - xfrm_state_walk_done(walk, net); + if (cb->args[0]) + xfrm_state_walk_done(walk, net); return 0; } @@ -921,8 +922,6 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) u8 proto = 0; int err; - cb->args[0] = 1; - err = nlmsg_parse(cb->nlh, 0, attrs, XFRMA_MAX, xfrma_policy); if (err < 0) @@ -939,6 +938,7 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) proto = nla_get_u8(attrs[XFRMA_PROTO]); xfrm_state_walk_init(walk, proto, filter); + cb->args[0] = 1; } (void) xfrm_state_walk(net, walk, dump_one_state, &info); -- cgit v1.2.3-59-g8ed1b From 73efc3245fd3edb3632d82a3a9c5d5d975a02efc Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 27 Jul 2016 08:03:18 +0200 Subject: xfrm: get rid of incorrect WARN AFAICT this message is just printed whenever input validation fails. This is a normal failure and we shouldn't be dumping the stack over it. Looks like it was originally a printk that was maybe incorrectly upgraded to a WARN: commit 62db5cfd70b1ef53aa21f144a806fe3b78c84fab Author: stephen hemminger Date: Wed May 12 06:37:06 2010 +0000 xfrm: add severity to printk Cc: Stephen Hemminger Cc: Steffen Klassert Signed-off-by: Vegard Nossum Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 4fb04ced5867..1a4f142dd50a 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2117,7 +2117,7 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh, err = verify_newpolicy_info(&ua->policy); if (err) - goto bad_policy; + goto free_state; /* build an XP */ xp = xfrm_policy_construct(net, &ua->policy, attrs, &err); @@ -2149,8 +2149,6 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh, return 0; -bad_policy: - WARN(1, "BAD policy passed\n"); free_state: kfree(x); nomem: -- cgit v1.2.3-59-g8ed1b From 7677c7560c3e80fde08a7e710d378dedabf950c3 Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 27 Jul 2016 08:44:15 +0200 Subject: xfrm: get rid of another incorrect WARN During fuzzing I regularly run into this WARN(). According to Herbert Xu, this "certainly shouldn't be a WARN, it probably shouldn't print anything either". Cc: Stephen Hemminger Cc: Steffen Klassert Cc: Herbert Xu Signed-off-by: Vegard Nossum Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 1a4f142dd50a..cb65d916a345 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2051,9 +2051,6 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, if (up->hard) { xfrm_policy_delete(xp, p->dir); xfrm_audit_policy_delete(xp, 1, true); - } else { - // reset the timers here? - WARN(1, "Don't know what to do with soft policy expire\n"); } km_policy_expired(xp, p->dir, up->hard, nlh->nlmsg_pid); -- cgit v1.2.3-59-g8ed1b From 6916fb3b10b3cbe3b1f9f5b680675f53e4e299eb Mon Sep 17 00:00:00 2001 From: Tobias Brunner Date: Fri, 29 Jul 2016 09:57:32 +0200 Subject: xfrm: Ignore socket policies when rebuilding hash tables Whenever thresholds are changed the hash tables are rebuilt. This is done by enumerating all policies and hashing and inserting them into the right table according to the thresholds and direction. Because socket policies are also contained in net->xfrm.policy_all but no hash tables are defined for their direction (dir + XFRM_POLICY_MAX) this causes a NULL or invalid pointer dereference after returning from policy_hash_bysel() if the rebuild is done while any socket policies are installed. Since the rebuild after changing thresholds is scheduled this crash could even occur if the userland sets thresholds seemingly before installing any socket policies. Fixes: 53c2e285f970 ("xfrm: Do not hash socket policies") Signed-off-by: Tobias Brunner Acked-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index b5e665b3cfb0..45f9cf97ea25 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -626,6 +626,10 @@ static void xfrm_hash_rebuild(struct work_struct *work) /* re-insert all policies by order of creation */ list_for_each_entry_reverse(policy, &net->xfrm.policy_all, walk.all) { + if (xfrm_policy_id2dir(policy->index) >= XFRM_POLICY_MAX) { + /* skip socket policies */ + continue; + } newpos = NULL; chain = policy_hash_bysel(net, &policy->selector, policy->family, -- cgit v1.2.3-59-g8ed1b From 4d0bd46a4d55383f7b925e6cf7865a77e0f0e020 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 8 Aug 2016 08:45:33 +0200 Subject: Revert "wext: Fix 32 bit iwpriv compatibility issue with 64 bit Kernel" This reverts commit 3d5fdff46c4b2b9534fa2f9fc78e90a48e0ff724. Ben Hutchings pointed out that the commit isn't safe since it assumes that the structure used by the driver is iw_point, when in fact there's no way to know about that. Fortunately, the only driver in the tree that ever runs this code path is the wilc1000 staging driver, so it doesn't really matter. Clearly I should have investigated this better before applying, sorry. Reported-by: Ben Hutchings Cc: stable@vger.kernel.org [though I guess it doesn't matter much] Fixes: 3d5fdff46c4b ("wext: Fix 32 bit iwpriv compatibility issue with 64 bit Kernel") Signed-off-by: Johannes Berg --- net/wireless/wext-core.c | 25 ++----------------------- 1 file changed, 2 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c index dbb2738e356a..6250b1cfcde5 100644 --- a/net/wireless/wext-core.c +++ b/net/wireless/wext-core.c @@ -958,29 +958,8 @@ static int wireless_process_ioctl(struct net *net, struct ifreq *ifr, return private(dev, iwr, cmd, info, handler); } /* Old driver API : call driver ioctl handler */ - if (dev->netdev_ops->ndo_do_ioctl) { -#ifdef CONFIG_COMPAT - if (info->flags & IW_REQUEST_FLAG_COMPAT) { - int ret = 0; - struct iwreq iwr_lcl; - struct compat_iw_point *iwp_compat = (void *) &iwr->u.data; - - memcpy(&iwr_lcl, iwr, sizeof(struct iwreq)); - iwr_lcl.u.data.pointer = compat_ptr(iwp_compat->pointer); - iwr_lcl.u.data.length = iwp_compat->length; - iwr_lcl.u.data.flags = iwp_compat->flags; - - ret = dev->netdev_ops->ndo_do_ioctl(dev, (void *) &iwr_lcl, cmd); - - iwp_compat->pointer = ptr_to_compat(iwr_lcl.u.data.pointer); - iwp_compat->length = iwr_lcl.u.data.length; - iwp_compat->flags = iwr_lcl.u.data.flags; - - return ret; - } else -#endif - return dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd); - } + if (dev->netdev_ops->ndo_do_ioctl) + return dev->netdev_ops->ndo_do_ioctl(dev, ifr, cmd); return -EOPNOTSUPP; } -- cgit v1.2.3-59-g8ed1b From 1625f4529957738be7d87cf157e107b8fb9d23b9 Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Wed, 10 Aug 2016 13:54:57 +0300 Subject: net/xfrm_input: fix possible NULL deref of tunnel.ip6->parms.i_key Running LTP 'icmp-uni-basic.sh -6 -p ipcomp -m tunnel' test over openvswitch + veth can trigger kernel panic: BUG: unable to handle kernel NULL pointer dereference at 00000000000000e0 IP: [] xfrm_input+0x82/0x750 ... [] xfrm6_rcv_spi+0x1e/0x20 [] xfrm6_tunnel_rcv+0x42/0x50 [xfrm6_tunnel] [] tunnel6_rcv+0x3e/0x8c [tunnel6] [] ip6_input_finish+0xd5/0x430 [] ip6_input+0x33/0x90 [] ip6_rcv_finish+0xa5/0xb0 ... It seems that tunnel.ip6 can have garbage values and also dereferenced without a proper check, only tunnel.ip4 is being verified. Fix it by adding one more if block for AF_INET6 and initialize tunnel.ip6 with NULL inside xfrm6_rcv_spi() (which is similar to xfrm4_rcv_spi()). Fixes: 049f8e2 ("xfrm: Override skb->mark with tunnel->parm.i_key in xfrm_input") Signed-off-by: Alexey Kodanev Signed-off-by: Steffen Klassert --- net/ipv6/xfrm6_input.c | 1 + net/xfrm/xfrm_input.c | 14 +++++++------- 2 files changed, 8 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index 0eaab1fa6be5..00a2d40677d6 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -23,6 +23,7 @@ int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb) int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi) { + XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = NULL; XFRM_SPI_SKB_CB(skb)->family = AF_INET6; XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr); return xfrm_input(skb, nexthdr, spi, 0); diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 1c4ad477ce93..6e3f0254d8a1 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -207,15 +207,15 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) family = XFRM_SPI_SKB_CB(skb)->family; /* if tunnel is present override skb->mark value with tunnel i_key */ - if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4) { - switch (family) { - case AF_INET: + switch (family) { + case AF_INET: + if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4) mark = be32_to_cpu(XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4->parms.i_key); - break; - case AF_INET6: + break; + case AF_INET6: + if (XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6) mark = be32_to_cpu(XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6->parms.i_key); - break; - } + break; } /* Allocate new secpath or COW existing one. */ -- cgit v1.2.3-59-g8ed1b From 11d7a0bb95eaaba1741bb24a7c3c169c82f09c7b Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 14 Aug 2016 19:52:56 -0700 Subject: xfrm: Only add l3mdev oif to dst lookups Subash reported that commit 42a7b32b73d6 ("xfrm: Add oif to dst lookups") broke a wifi use case that uses fib rules and xfrms. The intent of 42a7b32b73d6 was driven by VRFs with IPsec. As a compromise relax the use of oif in xfrm lookups to L3 master devices only (ie., oif is either an L3 master device or is enslaved to a master device). Fixes: 42a7b32b73d6 ("xfrm: Add oif to dst lookups") Reported-by: Subash Abhinov Kasiviswanathan Signed-off-by: David Ahern Signed-off-by: Steffen Klassert --- net/ipv4/xfrm4_policy.c | 2 +- net/ipv6/xfrm6_policy.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 7b0edb37a115..e07ed8b1deb3 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -29,7 +29,7 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, memset(fl4, 0, sizeof(*fl4)); fl4->daddr = daddr->a4; fl4->flowi4_tos = tos; - fl4->flowi4_oif = oif; + fl4->flowi4_oif = l3mdev_master_ifindex_by_index(net, oif); if (saddr) fl4->saddr = saddr->a4; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index c074771a10f7..dd84ecd1221b 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -36,7 +36,7 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif, int err; memset(&fl6, 0, sizeof(fl6)); - fl6.flowi6_oif = oif; + fl6.flowi6_oif = l3mdev_master_ifindex_by_index(net, oif); fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF; memcpy(&fl6.daddr, daddr, sizeof(fl6.daddr)); if (saddr) -- cgit v1.2.3-59-g8ed1b From 16590a228109e2f318d2cc6466221134cfab723a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 22 Aug 2016 14:57:42 -0400 Subject: SUNRPC: Silence WARN_ON when NFSv4.1 over RDMA is in use Using NFSv4.1 on RDMA should be safe, so broaden the new checks in rpc_create(). WARN_ON_ONCE is used, matching most other WARN call sites in clnt.c. Fixes: 39a9beab5acb ("rpc: share one xps between all backchannels") Fixes: d50039ea5ee6 ("nfsd4/rpc: move backchannel create logic...") Signed-off-by: Chuck Lever Reviewed-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 7f79fb7dc6a0..66f23b376fa0 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -453,7 +453,7 @@ static struct rpc_clnt *rpc_create_xprt(struct rpc_create_args *args, struct rpc_xprt_switch *xps; if (args->bc_xprt && args->bc_xprt->xpt_bc_xps) { - WARN_ON(args->protocol != XPRT_TRANSPORT_BC_TCP); + WARN_ON_ONCE(!(args->protocol & XPRT_TRANSPORT_BC)); xps = args->bc_xprt->xpt_bc_xps; xprt_switch_get(xps); } else { @@ -520,7 +520,7 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args) char servername[48]; if (args->bc_xprt) { - WARN_ON(args->protocol != XPRT_TRANSPORT_BC_TCP); + WARN_ON_ONCE(!(args->protocol & XPRT_TRANSPORT_BC)); xprt = args->bc_xprt->xpt_bc_xprt; if (xprt) { xprt_get(xprt); -- cgit v1.2.3-59-g8ed1b From 89e1f6d2b956649fbe0704d543a90b8e0cf872b0 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 01:02:18 +0800 Subject: netfilter: nft_reject: restrict to INPUT/FORWARD/OUTPUT After I add the nft rule "nft add rule filter prerouting reject with tcp reset", kernel panic happened on my system: NULL pointer dereference at ... IP: [] nf_send_reset+0xaf/0x400 Call Trace: [] ? nf_reject_ip_tcphdr_get+0x160/0x160 [] nft_reject_ipv4_eval+0x61/0xb0 [nft_reject_ipv4] [] nft_do_chain+0x1fa/0x890 [nf_tables] [] ? __nft_trace_packet+0x170/0x170 [nf_tables] [] ? nf_ct_invert_tuple+0xb0/0xc0 [nf_conntrack] [] ? nf_nat_setup_info+0x5d4/0x650 [nf_nat] [...] Because in the PREROUTING chain, routing information is not exist, then we will dereference the NULL pointer and oops happen. So we restrict reject expression to INPUT, FORWARD and OUTPUT chain. This is consistent with iptables REJECT target. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_reject.h | 4 ++++ net/ipv4/netfilter/nft_reject_ipv4.c | 1 + net/ipv6/netfilter/nft_reject_ipv6.c | 1 + net/netfilter/nft_reject.c | 16 ++++++++++++++++ net/netfilter/nft_reject_inet.c | 7 ++++++- 5 files changed, 28 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/netfilter/nft_reject.h b/include/net/netfilter/nft_reject.h index 60fa1530006b..02e28c529b29 100644 --- a/include/net/netfilter/nft_reject.h +++ b/include/net/netfilter/nft_reject.h @@ -8,6 +8,10 @@ struct nft_reject { extern const struct nla_policy nft_reject_policy[]; +int nft_reject_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data); + int nft_reject_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]); diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c index c24f41c816b3..2c2553b9026c 100644 --- a/net/ipv4/netfilter/nft_reject_ipv4.c +++ b/net/ipv4/netfilter/nft_reject_ipv4.c @@ -46,6 +46,7 @@ static const struct nft_expr_ops nft_reject_ipv4_ops = { .eval = nft_reject_ipv4_eval, .init = nft_reject_init, .dump = nft_reject_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_ipv4_type __read_mostly = { diff --git a/net/ipv6/netfilter/nft_reject_ipv6.c b/net/ipv6/netfilter/nft_reject_ipv6.c index 533cd5719c59..92bda9908bb9 100644 --- a/net/ipv6/netfilter/nft_reject_ipv6.c +++ b/net/ipv6/netfilter/nft_reject_ipv6.c @@ -47,6 +47,7 @@ static const struct nft_expr_ops nft_reject_ipv6_ops = { .eval = nft_reject_ipv6_eval, .init = nft_reject_init, .dump = nft_reject_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_ipv6_type __read_mostly = { diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c index 0522fc9bfb0a..c64de3f7379d 100644 --- a/net/netfilter/nft_reject.c +++ b/net/netfilter/nft_reject.c @@ -26,11 +26,27 @@ const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = { }; EXPORT_SYMBOL_GPL(nft_reject_policy); +int nft_reject_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data) +{ + return nft_chain_validate_hooks(ctx->chain, + (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT)); +} +EXPORT_SYMBOL_GPL(nft_reject_validate); + int nft_reject_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]) { struct nft_reject *priv = nft_expr_priv(expr); + int err; + + err = nft_reject_validate(ctx, expr, NULL); + if (err < 0) + return err; if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c index 759ca5248a3d..e79d9ca2ffee 100644 --- a/net/netfilter/nft_reject_inet.c +++ b/net/netfilter/nft_reject_inet.c @@ -66,7 +66,11 @@ static int nft_reject_inet_init(const struct nft_ctx *ctx, const struct nlattr * const tb[]) { struct nft_reject *priv = nft_expr_priv(expr); - int icmp_code; + int icmp_code, err; + + err = nft_reject_validate(ctx, expr, NULL); + if (err < 0) + return err; if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; @@ -124,6 +128,7 @@ static const struct nft_expr_ops nft_reject_inet_ops = { .eval = nft_reject_inet_eval, .init = nft_reject_inet_init, .dump = nft_reject_inet_dump, + .validate = nft_reject_validate, }; static struct nft_expr_type nft_reject_inet_type __read_mostly = { -- cgit v1.2.3-59-g8ed1b From 93fac10b99d78eb2c50a739cba2e590c7332d539 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:16 +0800 Subject: netfilter: nfnetlink: use list_for_each_entry_safe to delete all objects cttimeout and acct objects are deleted from the list while traversing it, so use list_for_each_entry is unsafe here. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_acct.c | 6 +++--- net/netfilter/nfnetlink_cttimeout.c | 5 +++-- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index 70eb2f6a3b01..d44d89b56127 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -343,12 +343,12 @@ static int nfnl_acct_del(struct net *net, struct sock *nfnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const tb[]) { - char *acct_name; - struct nf_acct *cur; + struct nf_acct *cur, *tmp; int ret = -ENOENT; + char *acct_name; if (!tb[NFACCT_NAME]) { - list_for_each_entry(cur, &net->nfnl_acct_list, head) + list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) nfnl_acct_try_del(cur); return 0; diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 68216cdc7083..f74fee1e2d0a 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -350,12 +350,13 @@ static int cttimeout_del_timeout(struct net *net, struct sock *ctnl, const struct nlmsghdr *nlh, const struct nlattr * const cda[]) { - struct ctnl_timeout *cur; + struct ctnl_timeout *cur, *tmp; int ret = -ENOENT; char *name; if (!cda[CTA_TIMEOUT_NAME]) { - list_for_each_entry(cur, &net->nfct_timeout_list, head) + list_for_each_entry_safe(cur, tmp, &net->nfct_timeout_list, + head) ctnl_timeout_try_del(net, cur); return 0; -- cgit v1.2.3-59-g8ed1b From 23aaba5ad55547db62bada5066c8fb6412d5b1c2 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:17 +0800 Subject: netfilter: cttimeout: put back l4proto when replacing timeout policy We forget to call nf_ct_l4proto_put when replacing the existing timeout policy. Acctually, there's no need to get ct l4proto before doing replace, so we can move it to a later position. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_cttimeout.c | 33 +++++++++++++++------------------ 1 file changed, 15 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index f74fee1e2d0a..6844c7af0b8f 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -98,31 +98,28 @@ static int cttimeout_new_timeout(struct net *net, struct sock *ctnl, break; } - l4proto = nf_ct_l4proto_find_get(l3num, l4num); - - /* This protocol is not supportted, skip. */ - if (l4proto->l4proto != l4num) { - ret = -EOPNOTSUPP; - goto err_proto_put; - } - if (matching) { if (nlh->nlmsg_flags & NLM_F_REPLACE) { /* You cannot replace one timeout policy by another of * different kind, sorry. */ if (matching->l3num != l3num || - matching->l4proto->l4proto != l4num) { - ret = -EINVAL; - goto err_proto_put; - } - - ret = ctnl_timeout_parse_policy(&matching->data, - l4proto, net, - cda[CTA_TIMEOUT_DATA]); - return ret; + matching->l4proto->l4proto != l4num) + return -EINVAL; + + return ctnl_timeout_parse_policy(&matching->data, + matching->l4proto, net, + cda[CTA_TIMEOUT_DATA]); } - ret = -EBUSY; + + return -EBUSY; + } + + l4proto = nf_ct_l4proto_find_get(l3num, l4num); + + /* This protocol is not supportted, skip. */ + if (l4proto->l4proto != l4num) { + ret = -EOPNOTSUPP; goto err_proto_put; } -- cgit v1.2.3-59-g8ed1b From 533e33009897c7dd1b0424c0d4b3331b222d5681 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 21:58:18 +0800 Subject: netfilter: cttimeout: unlink timeout objs in the unconfirmed ct lists KASAN reported this bug: BUG: KASAN: use-after-free in icmp_packet+0x25/0x50 [nf_conntrack_ipv4] at addr ffff880002db08c8 Read of size 4 by task lt-nf-queue/19041 Call Trace: [] dump_stack+0x63/0x88 [] kasan_report_error+0x528/0x560 [] kasan_report+0x58/0x60 [] ? icmp_packet+0x25/0x50 [nf_conntrack_ipv4] [] __asan_load4+0x61/0x80 [] icmp_packet+0x25/0x50 [nf_conntrack_ipv4] [] nf_conntrack_in+0x550/0x980 [nf_conntrack] [] ? __nf_conntrack_confirm+0xb10/0xb10 [nf_conntrack] [ ... ] The main reason is that we missed to unlink the timeout objects in the unconfirmed ct lists, so we will access the timeout objects that have already been freed. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_cttimeout.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 6844c7af0b8f..139e0867e56e 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -302,7 +302,16 @@ static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout) const struct hlist_nulls_node *nn; unsigned int last_hsize; spinlock_t *lock; - int i; + int i, cpu; + + for_each_possible_cpu(cpu) { + struct ct_pcpu *pcpu = per_cpu_ptr(net->ct.pcpu_lists, cpu); + + spin_lock_bh(&pcpu->lock); + hlist_nulls_for_each_entry(h, nn, &pcpu->unconfirmed, hnnode) + untimeout(h, timeout); + spin_unlock_bh(&pcpu->lock); + } local_bh_disable(); restart: -- cgit v1.2.3-59-g8ed1b From 960fa72f67f1be6891d63a5518860d1ae4e14b88 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 22 Aug 2016 22:57:56 +0800 Subject: netfilter: nft_meta: improve the validity check of pkttype set expr "meta pkttype set" is only supported on prerouting chain with bridge family and ingress chain with netdev family. But the validate check is incomplete, and the user can add the nft rules on input chain with bridge family, for example: # nft add table bridge filter # nft add chain bridge filter input {type filter hook input \ priority 0 \;} # nft add chain bridge filter test # nft add rule bridge filter test meta pkttype set unicast # nft add rule bridge filter input jump test This patch fixes the problem. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nft_meta.h | 4 ++++ net/bridge/netfilter/nft_meta_bridge.c | 1 + net/netfilter/nft_meta.c | 17 +++++++++++++---- 3 files changed, 18 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nft_meta.h b/include/net/netfilter/nft_meta.h index d27588c8dbd9..1139cde0fdc5 100644 --- a/include/net/netfilter/nft_meta.h +++ b/include/net/netfilter/nft_meta.h @@ -36,4 +36,8 @@ void nft_meta_set_eval(const struct nft_expr *expr, void nft_meta_set_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr); +int nft_meta_set_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data); + #endif diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c index 4b901d9f2e7c..ad47a921b701 100644 --- a/net/bridge/netfilter/nft_meta_bridge.c +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -86,6 +86,7 @@ static const struct nft_expr_ops nft_meta_bridge_set_ops = { .init = nft_meta_set_init, .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, + .validate = nft_meta_set_validate, }; static const struct nft_expr_ops * diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 2863f3493038..8a6bc7630912 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -291,10 +291,16 @@ int nft_meta_get_init(const struct nft_ctx *ctx, } EXPORT_SYMBOL_GPL(nft_meta_get_init); -static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx) +int nft_meta_set_validate(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nft_data **data) { + struct nft_meta *priv = nft_expr_priv(expr); unsigned int hooks; + if (priv->key != NFT_META_PKTTYPE) + return 0; + switch (ctx->afi->family) { case NFPROTO_BRIDGE: hooks = 1 << NF_BR_PRE_ROUTING; @@ -308,6 +314,7 @@ static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx) return nft_chain_validate_hooks(ctx->chain, hooks); } +EXPORT_SYMBOL_GPL(nft_meta_set_validate); int nft_meta_set_init(const struct nft_ctx *ctx, const struct nft_expr *expr, @@ -327,15 +334,16 @@ int nft_meta_set_init(const struct nft_ctx *ctx, len = sizeof(u8); break; case NFT_META_PKTTYPE: - err = nft_meta_set_init_pkttype(ctx); - if (err) - return err; len = sizeof(u8); break; default: return -EOPNOTSUPP; } + err = nft_meta_set_validate(ctx, expr, NULL); + if (err < 0) + return err; + priv->sreg = nft_parse_register(tb[NFTA_META_SREG]); err = nft_validate_register_load(priv->sreg, len); if (err < 0) @@ -407,6 +415,7 @@ static const struct nft_expr_ops nft_meta_set_ops = { .init = nft_meta_set_init, .destroy = nft_meta_set_destroy, .dump = nft_meta_set_dump, + .validate = nft_meta_set_validate, }; static const struct nft_expr_ops * -- cgit v1.2.3-59-g8ed1b From 4249fc1f023a2106170bbf715e2e1a0ebc2d5b1f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 23 Aug 2016 10:20:31 +0200 Subject: netfilter: ebtables: put module reference when an incorrect extension is found commit bcf493428840 ("netfilter: ebtables: Fix extension lookup with identical name") added a second lookup in case the extension that was found during the first lookup matched another extension with the same name, but didn't release the reference on the incorrect module. Fixes: bcf493428840 ("netfilter: ebtables: Fix extension lookup with identical name") Signed-off-by: Sabrina Dubroca Acked-by: Phil Sutter Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebtables.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index cceac5bb658f..0833c251aef7 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -368,6 +368,8 @@ ebt_check_match(struct ebt_entry_match *m, struct xt_mtchk_param *par, match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0); if (IS_ERR(match) || match->family != NFPROTO_BRIDGE) { + if (!IS_ERR(match)) + module_put(match->me); request_module("ebt_%s", m->u.name); match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0); } -- cgit v1.2.3-59-g8ed1b From 554d072e7bc3e56de5893c8181110a547b2062c9 Mon Sep 17 00:00:00 2001 From: Arik Nemtsov Date: Mon, 29 Aug 2016 12:37:35 +0300 Subject: mac80211: TDLS: don't require beaconing for AP BW Stop downgrading TDLS chandef when reaching the AP BW. The AP provides the necessary regulatory protection in this case. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=153961, which reported an infinite loop here. Reported-by: Kamil Toman Signed-off-by: Arik Nemtsov Signed-off-by: Johannes Berg --- net/mac80211/tdls.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index b5d28f14b9cf..afca7d103684 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -333,10 +333,11 @@ ieee80211_tdls_chandef_vht_upgrade(struct ieee80211_sub_if_data *sdata, if (!uc.center_freq1) return; - /* proceed to downgrade the chandef until usable or the same */ + /* proceed to downgrade the chandef until usable or the same as AP BW */ while (uc.width > max_width || - !cfg80211_reg_can_beacon_relax(sdata->local->hw.wiphy, &uc, - sdata->wdev.iftype)) + (uc.width > sta->tdls_chandef.width && + !cfg80211_reg_can_beacon_relax(sdata->local->hw.wiphy, &uc, + sdata->wdev.iftype))) ieee80211_chandef_downgrade(&uc); if (!cfg80211_chandef_identical(&uc, &sta->tdls_chandef)) { -- cgit v1.2.3-59-g8ed1b From c73c2484901139c28383b58eabcbf4d613e91518 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 28 Aug 2016 16:59:52 +0800 Subject: netfilter: nf_tables_netdev: remove redundant ip_hdr assignment We have already use skb_header_pointer to get the ip header pointer, so there's no need to use ip_hdr again. Moreover, in NETDEV INGRESS hook, ip header maybe not linear, so use ip_hdr is not appropriate, remove it. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_netdev.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c index 5eefe4a355c6..75d696f11045 100644 --- a/net/netfilter/nf_tables_netdev.c +++ b/net/netfilter/nf_tables_netdev.c @@ -30,7 +30,6 @@ nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt, if (!iph) return; - iph = ip_hdr(skb); if (iph->ihl < 5 || iph->version != 4) return; -- cgit v1.2.3-59-g8ed1b From 9264251ee2a55bce8fb93826b3f581fb9eb7e2c2 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Wed, 31 Aug 2016 14:16:44 +0200 Subject: bridge: re-introduce 'fix parsing of MLDv2 reports' commit bc8c20acaea1 ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") seems to have accidentally reverted commit 47cc84ce0c2f ("bridge: fix parsing of MLDv2 reports"). This commit brings back a change to br_ip6_multicast_mld2_report() where parsing of MLDv2 reports stops when the first group is successfully added to the MDB cache. Fixes: bc8c20acaea1 ("bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave") Signed-off-by: Davide Caratti Acked-by: Nikolay Aleksandrov Acked-by: Thadeu Lima de Souza Cascardo Signed-off-by: David S. Miller --- net/bridge/br_multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index a5423a1eec05..c5fea9393946 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -1138,7 +1138,7 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, } else { err = br_ip6_multicast_add_group(br, port, &grec->grec_mca, vid); - if (!err) + if (err) break; } } -- cgit v1.2.3-59-g8ed1b From c0338aff2260ea6c092806312dbb154cec07a242 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Sun, 28 Aug 2016 21:28:26 -0700 Subject: kcm: fix a socket double free Dmitry reported a double free on kcm socket, which could be easily reproduced by: #include #include int main() { int fd = syscall(SYS_socket, 0x29ul, 0x5ul, 0x0ul, 0, 0, 0); syscall(SYS_ioctl, fd, 0x89e2ul, 0x20a98000ul, 0, 0, 0); return 0; } This is because on the error path, after we install the new socket file, we call sock_release() to clean up the socket, which leaves the fd pointing to a freed socket. Fix this by calling sys_close() on that fd directly. Fixes: ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module") Reported-by: Dmitry Vyukov Cc: Tom Herbert Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/kcm/kcmsock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index cb39e05b166c..411693288648 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -2029,7 +2030,7 @@ static int kcm_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) if (copy_to_user((void __user *)arg, &info, sizeof(info))) { err = -EFAULT; - sock_release(newsock); + sys_close(info.fd); } } -- cgit v1.2.3-59-g8ed1b From d2f394dc4816b7bd1b44981d83509f18f19c53f0 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Thu, 1 Sep 2016 16:22:16 +0200 Subject: tipc: fix random link resets while adding a second bearer In a dual bearer configuration, if the second tipc link becomes active while the first link still has pending nametable "bulk" updates, it randomly leads to reset of the second link. When a link is established, the function named_distribute(), fills the skb based on node mtu (allows room for TUNNEL_PROTOCOL) with NAME_DISTRIBUTOR message for each PUBLICATION. However, the function named_distribute() allocates the buffer by increasing the node mtu by INT_H_SIZE (to insert NAME_DISTRIBUTOR). This consumes the space allocated for TUNNEL_PROTOCOL. When establishing the second link, the link shall tunnel all the messages in the first link queue including the "bulk" update. As size of the NAME_DISTRIBUTOR messages while tunnelling, exceeds the link mtu the transmission fails (-EMSGSIZE). Thus, the synch point based on the message count of the tunnel packets is never reached leading to link timeout. In this commit, we adjust the size of name distributor message so that they can be tunnelled. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/name_distr.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 6b626a64b517..a04fe9be1c60 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -62,6 +62,8 @@ static void publ_to_item(struct distr_item *i, struct publication *p) /** * named_prepare_buf - allocate & initialize a publication message + * + * The buffer returned is of size INT_H_SIZE + payload size */ static struct sk_buff *named_prepare_buf(struct net *net, u32 type, u32 size, u32 dest) @@ -141,9 +143,9 @@ static void named_distribute(struct net *net, struct sk_buff_head *list, struct publication *publ; struct sk_buff *skb = NULL; struct distr_item *item = NULL; - uint msg_dsz = (tipc_node_get_mtu(net, dnode, 0) / ITEM_SIZE) * - ITEM_SIZE; - uint msg_rem = msg_dsz; + u32 msg_dsz = ((tipc_node_get_mtu(net, dnode, 0) - INT_H_SIZE) / + ITEM_SIZE) * ITEM_SIZE; + u32 msg_rem = msg_dsz; list_for_each_entry(publ, pls, local_list) { /* Prepare next buffer: */ -- cgit v1.2.3-59-g8ed1b From d26c638c16cb54f6fb1507e27df93ede692db572 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 30 Aug 2016 10:09:21 +0200 Subject: ipv6: add missing netconf notif when 'all' is updated The 'default' value was not advertised. Fixes: f3a1bfb11ccb ("rtnl/ipv6: use netconf msg to advertise forwarding status") Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'net') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f418d2eaeddd..2a688171a188 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -778,7 +778,14 @@ static int addrconf_fixup_forwarding(struct ctl_table *table, int *p, int newf) } if (p == &net->ipv6.devconf_all->forwarding) { + int old_dflt = net->ipv6.devconf_dflt->forwarding; + net->ipv6.devconf_dflt->forwarding = newf; + if ((!newf) ^ (!old_dflt)) + inet6_netconf_notify_devconf(net, NETCONFA_FORWARDING, + NETCONFA_IFINDEX_DEFAULT, + net->ipv6.devconf_dflt); + addrconf_forward_change(net, newf); if ((!newf) ^ (!old)) inet6_netconf_notify_devconf(net, NETCONFA_FORWARDING, -- cgit v1.2.3-59-g8ed1b From 29c994e361009142ec0bca6493cc8f7b0d3c561a Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 30 Aug 2016 10:09:22 +0200 Subject: netconf: add a notif when settings are created All changes are notified, but the initial state was missing. Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 11 +++++++---- net/ipv6/addrconf.c | 9 ++++++++- 2 files changed, 15 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 415e117967c7..062a67ca9a21 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2232,7 +2232,7 @@ static struct devinet_sysctl_table { }; static int __devinet_sysctl_register(struct net *net, char *dev_name, - struct ipv4_devconf *p) + int ifindex, struct ipv4_devconf *p) { int i; struct devinet_sysctl_table *t; @@ -2255,6 +2255,8 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name, goto free; p->sysctl = t; + + inet_netconf_notify_devconf(net, NETCONFA_ALL, ifindex, p); return 0; free: @@ -2286,7 +2288,7 @@ static int devinet_sysctl_register(struct in_device *idev) if (err) return err; err = __devinet_sysctl_register(dev_net(idev->dev), idev->dev->name, - &idev->cnf); + idev->dev->ifindex, &idev->cnf); if (err) neigh_sysctl_unregister(idev->arp_parms); return err; @@ -2347,11 +2349,12 @@ static __net_init int devinet_init_net(struct net *net) } #ifdef CONFIG_SYSCTL - err = __devinet_sysctl_register(net, "all", all); + err = __devinet_sysctl_register(net, "all", NETCONFA_IFINDEX_ALL, all); if (err < 0) goto err_reg_all; - err = __devinet_sysctl_register(net, "default", dflt); + err = __devinet_sysctl_register(net, "default", + NETCONFA_IFINDEX_DEFAULT, dflt); if (err < 0) goto err_reg_dflt; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 2a688171a188..bdf368eff5ab 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6032,7 +6032,7 @@ static const struct ctl_table addrconf_sysctl[] = { static int __addrconf_sysctl_register(struct net *net, char *dev_name, struct inet6_dev *idev, struct ipv6_devconf *p) { - int i; + int i, ifindex; struct ctl_table *table; char path[sizeof("net/ipv6/conf/") + IFNAMSIZ]; @@ -6052,6 +6052,13 @@ static int __addrconf_sysctl_register(struct net *net, char *dev_name, if (!p->sysctl_header) goto free; + if (!strcmp(dev_name, "all")) + ifindex = NETCONFA_IFINDEX_ALL; + else if (!strcmp(dev_name, "default")) + ifindex = NETCONFA_IFINDEX_DEFAULT; + else + ifindex = idev->dev->ifindex; + inet6_netconf_notify_devconf(net, NETCONFA_ALL, ifindex, p); return 0; free: -- cgit v1.2.3-59-g8ed1b From 85a3d4a9356b595d5440c3f1bf07ee7cecca1567 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 30 Aug 2016 17:44:29 +0200 Subject: net: bridge: don't increment tx_dropped in br_do_proxy_arp pskb_may_pull may fail due to various reasons (e.g. alloc failure), but the skb isn't changed/dropped and processing continues so we shouldn't increment tx_dropped. CC: Kyeyoon Park CC: Roopa Prabhu CC: Stephen Hemminger CC: bridge@lists.linux-foundation.org Fixes: 958501163ddd ("bridge: Add support for IEEE 802.11 Proxy ARP") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_input.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 8e486203d133..abe11f085479 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -80,13 +80,10 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, BR_INPUT_SKB_CB(skb)->proxyarp_replied = false; - if (dev->flags & IFF_NOARP) + if ((dev->flags & IFF_NOARP) || + !pskb_may_pull(skb, arp_hdr_len(dev))) return; - if (!pskb_may_pull(skb, arp_hdr_len(dev))) { - dev->stats.tx_dropped++; - return; - } parp = arp_hdr(skb); if (parp->ar_pro != htons(ETH_P_IP) || -- cgit v1.2.3-59-g8ed1b From 28b346cbc0715ae45b2814d857f1d8a7e6817ed8 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Tue, 30 Aug 2016 11:55:23 -0400 Subject: tcp: fastopen: fix rcv_wup initialization for TFO server on SYN/data Yuchung noticed that on the first TFO server data packet sent after the (TFO) handshake, the server echoed the TCP timestamp value in the SYN/data instead of the timestamp value in the final ACK of the handshake. This problem did not happen on regular opens. The tcp_replace_ts_recent() logic that decides whether to remember an incoming TS value needs tp->rcv_wup to hold the latest receive sequence number that we have ACKed (latest tp->rcv_nxt we have ACKed). This commit fixes this issue by ensuring that a TFO server properly updates tp->rcv_wup to match tp->rcv_nxt at the time it sends a SYN/ACK for the SYN/data. Reported-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: Yuchung Cheng Signed-off-by: Eric Dumazet Signed-off-by: Soheil Hassas Yeganeh Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 54d9f9b0120f..62a5751d4fe1 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -226,6 +226,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, tcp_fastopen_add_skb(child, skb); tcp_rsk(req)->rcv_nxt = tp->rcv_nxt; + tp->rcv_wup = tp->rcv_nxt; /* tcp_conn_request() is sending the SYNACK, * and queues the child into listener accept queue. */ -- cgit v1.2.3-59-g8ed1b From 635c223cfa05af9523146b2f37e119d945f449ae Mon Sep 17 00:00:00 2001 From: Gao Feng Date: Wed, 31 Aug 2016 14:15:05 +0800 Subject: rps: flow_dissector: Fix uninitialized flow_keys used in __skb_get_hash possibly The original codes depend on that the function parameters are evaluated from left to right. But the parameter's evaluation order is not defined in C standard actually. When flow_keys_have_l4(&keys) is invoked before ___skb_get_hash(skb, &keys, hashrnd) with some compilers or environment, the keys passed to flow_keys_have_l4 is not initialized. Fixes: 6db61d79c1e1 ("flow_dissector: Ignore flow dissector return value from ___skb_get_hash") Acked-by: Eric Dumazet Signed-off-by: Gao Feng Signed-off-by: David S. Miller --- net/core/flow_dissector.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 61ad43f61c5e..52742a02814f 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -680,11 +680,13 @@ EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric); void __skb_get_hash(struct sk_buff *skb) { struct flow_keys keys; + u32 hash; __flow_hash_secret_init(); - __skb_set_sw_hash(skb, ___skb_get_hash(skb, &keys, hashrnd), - flow_keys_have_l4(&keys)); + hash = ___skb_get_hash(skb, &keys, hashrnd); + + __skb_set_sw_hash(skb, hash, flow_keys_have_l4(&keys)); } EXPORT_SYMBOL(__skb_get_hash); -- cgit v1.2.3-59-g8ed1b From ab34380162cbc9b5172afdadf5136643c687bb73 Mon Sep 17 00:00:00 2001 From: Eli Cooper Date: Fri, 26 Aug 2016 23:52:29 +0800 Subject: ipv6: Don't unset flowi6_proto in ipxip6_tnl_xmit() Commit 8eb30be0352d0916 ("ipv6: Create ip6_tnl_xmit") unsets flowi6_proto in ip4ip6_tnl_xmit() and ip6ip6_tnl_xmit(). Since xfrm_selector_match() relies on this info, IPv6 packets sent by an ip6tunnel cannot be properly selected by their protocols after removing it. This patch puts flowi6_proto back. Cc: stable@vger.kernel.org Fixes: 8eb30be0352d ("ipv6: Create ip6_tnl_xmit") Signed-off-by: Eli Cooper Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 7b0481e3738f..888543debe4e 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1174,6 +1174,7 @@ ip4ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) encap_limit = t->parms.encap_limit; memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_IPIP; dsfield = ipv4_get_dsfield(iph); @@ -1233,6 +1234,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) encap_limit = t->parms.encap_limit; memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_IPV6; dsfield = ipv6_get_dsfield(ipv6h); if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS) -- cgit v1.2.3-59-g8ed1b From 2f86953e7436c9b9a4690909c5e2db24799e173b Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 2 Sep 2016 10:22:54 +0200 Subject: l2tp: fix use-after-free during module unload Tunnel deletion is delayed by both a workqueue (l2tp_tunnel_delete -> wq -> l2tp_tunnel_del_work) and RCU (sk_destruct -> RCU -> l2tp_tunnel_destruct). By the time l2tp_tunnel_destruct() runs to destroy the tunnel and finish destroying the socket, the private data reserved via the net_generic mechanism has already been freed, but l2tp_tunnel_destruct() actually uses this data. Make sure tunnel deletion for the netns has completed before returning from l2tp_exit_net() by first flushing the tunnel removal workqueue, and then waiting for RCU callbacks to complete. Fixes: 167eb17e0b17 ("l2tp: create tunnel sockets in the right namespace") Signed-off-by: Sabrina Dubroca Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 1e40dacaa137..a2ed3bda4ddc 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1855,6 +1855,9 @@ static __net_exit void l2tp_exit_net(struct net *net) (void)l2tp_tunnel_delete(tunnel); } rcu_read_unlock_bh(); + + flush_workqueue(l2tp_wq); + rcu_barrier(); } static struct pernet_operations l2tp_net_ops = { -- cgit v1.2.3-59-g8ed1b From 24b27fc4cdf9e10c5e79e5923b6b7c2c5c95096c Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Thu, 1 Sep 2016 22:18:34 -0700 Subject: bonding: Fix bonding crash Following few steps will crash kernel - (a) Create bonding master > modprobe bonding miimon=50 (b) Create macvlan bridge on eth2 > ip link add link eth2 dev mvl0 address aa:0:0:0:0:01 \ type macvlan (c) Now try adding eth2 into the bond > echo +eth2 > /sys/class/net/bond0/bonding/slaves Bonding does lots of things before checking if the device enslaved is busy or not. In this case when the notifier call-chain sends notifications, the bond_netdev_event() assumes that the rx_handler /rx_handler_data is registered while the bond_enslave() hasn't progressed far enough to register rx_handler for the new slave. This patch adds a rx_handler check that can be performed right at the beginning of the enslave code to avoid getting into this situation. Signed-off-by: Mahesh Bandewar Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 7 ++++--- include/linux/netdevice.h | 1 + net/core/dev.c | 16 ++++++++++++++++ 3 files changed, 21 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 217e8da0628c..9599ed6f1213 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1341,9 +1341,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) slave_dev->name); } - /* already enslaved */ - if (slave_dev->flags & IFF_SLAVE) { - netdev_dbg(bond_dev, "Error: Device was already enslaved\n"); + /* already in-use? */ + if (netdev_is_rx_handler_busy(slave_dev)) { + netdev_err(bond_dev, + "Error: Device is in use and cannot be enslaved\n"); return -EBUSY; } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3a788bf0affd..e8d79d4ebcfe 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3267,6 +3267,7 @@ static inline void napi_free_frags(struct napi_struct *napi) napi->skb = NULL; } +bool netdev_is_rx_handler_busy(struct net_device *dev); int netdev_rx_handler_register(struct net_device *dev, rx_handler_func_t *rx_handler, void *rx_handler_data); diff --git a/net/core/dev.c b/net/core/dev.c index dd6ce598de89..ea6312057a71 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3974,6 +3974,22 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, return skb; } +/** + * netdev_is_rx_handler_busy - check if receive handler is registered + * @dev: device to check + * + * Check if a receive handler is already registered for a given device. + * Return true if there one. + * + * The caller must hold the rtnl_mutex. + */ +bool netdev_is_rx_handler_busy(struct net_device *dev) +{ + ASSERT_RTNL(); + return dev && rtnl_dereference(dev->rx_handler); +} +EXPORT_SYMBOL_GPL(netdev_is_rx_handler_busy); + /** * netdev_rx_handler_register - register receive handler * @dev: device to register a handler for -- cgit v1.2.3-59-g8ed1b From 38f7bd94a97b542de86a2be9229289717e33a7a4 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 1 Sep 2016 14:56:49 -0700 Subject: Revert "af_unix: Fix splice-bind deadlock" This reverts commit c845acb324aa85a39650a14e7696982ceea75dc1. It turns out that it just replaces one deadlock with another one: we can still get the wrong lock ordering with the readlock due to overlayfs calling back into the filesystem layer and still taking the vfs locks after the readlock. The proper solution ends up being to just split the readlock into two pieces: the bind lock (taken *outside* the vfs locks) and the IO lock (taken *inside* the filesystem locks). The two locks are independent anyway. Signed-off-by: Linus Torvalds Reviewed-by: Shmulik Ladkani Signed-off-by: David S. Miller --- net/unix/af_unix.c | 66 +++++++++++++++++++++--------------------------------- 1 file changed, 26 insertions(+), 40 deletions(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index f1dffe84f0d5..433ae1bbef97 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -954,20 +954,32 @@ fail: return NULL; } -static int unix_mknod(struct dentry *dentry, const struct path *path, umode_t mode, - struct path *res) +static int unix_mknod(const char *sun_path, umode_t mode, struct path *res) { - int err; + struct dentry *dentry; + struct path path; + int err = 0; + /* + * Get the parent directory, calculate the hash for last + * component. + */ + dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); + err = PTR_ERR(dentry); + if (IS_ERR(dentry)) + return err; - err = security_path_mknod(path, dentry, mode, 0); + /* + * All right, let's create it. + */ + err = security_path_mknod(&path, dentry, mode, 0); if (!err) { - err = vfs_mknod(d_inode(path->dentry), dentry, mode, 0); + err = vfs_mknod(d_inode(path.dentry), dentry, mode, 0); if (!err) { - res->mnt = mntget(path->mnt); + res->mnt = mntget(path.mnt); res->dentry = dget(dentry); } } - + done_path_create(&path, dentry); return err; } @@ -978,12 +990,10 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) struct unix_sock *u = unix_sk(sk); struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; char *sun_path = sunaddr->sun_path; - int err, name_err; + int err; unsigned int hash; struct unix_address *addr; struct hlist_head *list; - struct path path; - struct dentry *dentry; err = -EINVAL; if (sunaddr->sun_family != AF_UNIX) @@ -999,34 +1009,14 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; addr_len = err; - name_err = 0; - dentry = NULL; - if (sun_path[0]) { - /* Get the parent directory, calculate the hash for last - * component. - */ - dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); - - if (IS_ERR(dentry)) { - /* delay report until after 'already bound' check */ - name_err = PTR_ERR(dentry); - dentry = NULL; - } - } - err = mutex_lock_interruptible(&u->readlock); if (err) - goto out_path; + goto out; err = -EINVAL; if (u->addr) goto out_up; - if (name_err) { - err = name_err == -EEXIST ? -EADDRINUSE : name_err; - goto out_up; - } - err = -ENOMEM; addr = kmalloc(sizeof(*addr)+addr_len, GFP_KERNEL); if (!addr) @@ -1037,11 +1027,11 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) addr->hash = hash ^ sk->sk_type; atomic_set(&addr->refcnt, 1); - if (dentry) { - struct path u_path; + if (sun_path[0]) { + struct path path; umode_t mode = S_IFSOCK | (SOCK_INODE(sock)->i_mode & ~current_umask()); - err = unix_mknod(dentry, &path, mode, &u_path); + err = unix_mknod(sun_path, mode, &path); if (err) { if (err == -EEXIST) err = -EADDRINUSE; @@ -1049,9 +1039,9 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out_up; } addr->hash = UNIX_HASH_SIZE; - hash = d_real_inode(dentry)->i_ino & (UNIX_HASH_SIZE - 1); + hash = d_real_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE - 1); spin_lock(&unix_table_lock); - u->path = u_path; + u->path = path; list = &unix_socket_table[hash]; } else { spin_lock(&unix_table_lock); @@ -1074,10 +1064,6 @@ out_unlock: spin_unlock(&unix_table_lock); out_up: mutex_unlock(&u->readlock); -out_path: - if (dentry) - done_path_create(&path, dentry); - out: return err; } -- cgit v1.2.3-59-g8ed1b From 6e1ce3c3451291142a57c4f3f6f999a29fb5b3bc Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 1 Sep 2016 14:43:53 -0700 Subject: af_unix: split 'u->readlock' into two: 'iolock' and 'bindlock' Right now we use the 'readlock' both for protecting some of the af_unix IO path and for making the bind be single-threaded. The two are independent, but using the same lock makes for a nasty deadlock due to ordering with regards to filesystem locking. The bind locking would want to nest outside the VSF pathname locking, but the IO locking wants to nest inside some of those same locks. We tried to fix this earlier with commit c845acb324aa ("af_unix: Fix splice-bind deadlock") which moved the readlock inside the vfs locks, but that caused problems with overlayfs that will then call back into filesystem routines that take the lock in the wrong order anyway. Splitting the locks means that we can go back to having the bind lock be the outermost lock, and we don't have any deadlocks with lock ordering. Acked-by: Rainer Weikusat Acked-by: Al Viro Signed-off-by: Linus Torvalds Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/net/af_unix.h | 2 +- net/unix/af_unix.c | 45 +++++++++++++++++++++++---------------------- 2 files changed, 24 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 9b4c418bebd8..fd60eccb59a6 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -52,7 +52,7 @@ struct unix_sock { struct sock sk; struct unix_address *addr; struct path path; - struct mutex readlock; + struct mutex iolock, bindlock; struct sock *peer; struct list_head link; atomic_long_t inflight; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 433ae1bbef97..8309687a56b0 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -661,11 +661,11 @@ static int unix_set_peek_off(struct sock *sk, int val) { struct unix_sock *u = unix_sk(sk); - if (mutex_lock_interruptible(&u->readlock)) + if (mutex_lock_interruptible(&u->iolock)) return -EINTR; sk->sk_peek_off = val; - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); return 0; } @@ -779,7 +779,8 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern) spin_lock_init(&u->lock); atomic_long_set(&u->inflight, 0); INIT_LIST_HEAD(&u->link); - mutex_init(&u->readlock); /* single task reading lock */ + mutex_init(&u->iolock); /* single task reading lock */ + mutex_init(&u->bindlock); /* single task binding lock */ init_waitqueue_head(&u->peer_wait); init_waitqueue_func_entry(&u->peer_wake, unix_dgram_peer_wake_relay); unix_insert_socket(unix_sockets_unbound(sk), sk); @@ -848,7 +849,7 @@ static int unix_autobind(struct socket *sock) int err; unsigned int retries = 0; - err = mutex_lock_interruptible(&u->readlock); + err = mutex_lock_interruptible(&u->bindlock); if (err) return err; @@ -895,7 +896,7 @@ retry: spin_unlock(&unix_table_lock); err = 0; -out: mutex_unlock(&u->readlock); +out: mutex_unlock(&u->bindlock); return err; } @@ -1009,7 +1010,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; addr_len = err; - err = mutex_lock_interruptible(&u->readlock); + err = mutex_lock_interruptible(&u->bindlock); if (err) goto out; @@ -1063,7 +1064,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) out_unlock: spin_unlock(&unix_table_lock); out_up: - mutex_unlock(&u->readlock); + mutex_unlock(&u->bindlock); out: return err; } @@ -1955,17 +1956,17 @@ static ssize_t unix_stream_sendpage(struct socket *socket, struct page *page, if (false) { alloc_skb: unix_state_unlock(other); - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); newskb = sock_alloc_send_pskb(sk, 0, 0, flags & MSG_DONTWAIT, &err, 0); if (!newskb) goto err; } - /* we must acquire readlock as we modify already present + /* we must acquire iolock as we modify already present * skbs in the sk_receive_queue and mess with skb->len */ - err = mutex_lock_interruptible(&unix_sk(other)->readlock); + err = mutex_lock_interruptible(&unix_sk(other)->iolock); if (err) { err = flags & MSG_DONTWAIT ? -EAGAIN : -ERESTARTSYS; goto err; @@ -2032,7 +2033,7 @@ alloc_skb: } unix_state_unlock(other); - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); other->sk_data_ready(other); scm_destroy(&scm); @@ -2041,7 +2042,7 @@ alloc_skb: err_state_unlock: unix_state_unlock(other); err_unlock: - mutex_unlock(&unix_sk(other)->readlock); + mutex_unlock(&unix_sk(other)->iolock); err: kfree_skb(newskb); if (send_sigpipe && !(flags & MSG_NOSIGNAL)) @@ -2109,7 +2110,7 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); do { - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); skip = sk_peek_offset(sk, flags); skb = __skb_try_recv_datagram(sk, flags, &peeked, &skip, &err, @@ -2117,14 +2118,14 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, if (skb) break; - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); if (err != -EAGAIN) break; } while (timeo && !__skb_wait_for_more_packets(sk, &err, &timeo, last)); - if (!skb) { /* implies readlock unlocked */ + if (!skb) { /* implies iolock unlocked */ unix_state_lock(sk); /* Signal EOF on disconnected non-blocking SEQPACKET socket. */ if (sk->sk_type == SOCK_SEQPACKET && err == -EAGAIN && @@ -2189,7 +2190,7 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, out_free: skb_free_datagram(sk, skb); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); out: return err; } @@ -2284,7 +2285,7 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state) /* Lock the socket to prevent queue disordering * while sleeps in memcpy_tomsg */ - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); if (flags & MSG_PEEK) skip = sk_peek_offset(sk, flags); @@ -2326,7 +2327,7 @@ again: break; } - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); timeo = unix_stream_data_wait(sk, timeo, last, last_len); @@ -2337,7 +2338,7 @@ again: goto out; } - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); goto redo; unlock: unix_state_unlock(sk); @@ -2440,7 +2441,7 @@ unlock: } } while (size); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); if (state->msg) scm_recv(sock, state->msg, &scm, flags); else @@ -2481,9 +2482,9 @@ static ssize_t skb_unix_socket_splice(struct sock *sk, int ret; struct unix_sock *u = unix_sk(sk); - mutex_unlock(&u->readlock); + mutex_unlock(&u->iolock); ret = splice_to_pipe(pipe, spd); - mutex_lock(&u->readlock); + mutex_lock(&u->iolock); return ret; } -- cgit v1.2.3-59-g8ed1b From 03c2778a938aaba0893f6d6cdc29511d91a79848 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Fri, 2 Sep 2016 14:39:50 -0400 Subject: ipv6: release dst in ping_v6_sendmsg Neither the failure or success paths of ping_v6_sendmsg release the dst it acquires. This leads to a flood of warnings from "net/core/dst.c:288 dst_release" on older kernels that don't have 8bf4ada2e21378816b28205427ee6b0e1ca4c5f1 backported. That patch optimistically hoped this had been fixed post 3.10, but it seems at least one case wasn't, where I've seen this triggered a lot from machines doing unprivileged icmp sockets. Cc: Martin Lau Signed-off-by: Dave Jones Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- net/ipv6/ping.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index 0900352c924c..0e983b694ee8 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -126,8 +126,10 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) rt = (struct rt6_info *) dst; np = inet6_sk(sk); - if (!np) - return -EBADF; + if (!np) { + err = -EBADF; + goto dst_err_out; + } if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) fl6.flowi6_oif = np->mcast_oif; @@ -163,6 +165,9 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } release_sock(sk); +dst_err_out: + dst_release(dst); + if (err) return err; -- cgit v1.2.3-59-g8ed1b From 5a56a0b3a45dd0cc5b2f7bec6afd053a474ed9f5 Mon Sep 17 00:00:00 2001 From: Mark Tomlinson Date: Mon, 5 Sep 2016 10:20:20 +1200 Subject: net: Don't delete routes in different VRFs When deleting an IP address from an interface, there is a clean-up of routes which refer to this local address. However, there was no check to see that the VRF matched. This meant that deletion wasn't confined to the VRF it should have been. To solve this, a new field has been added to fib_info to hold a table id. When removing fib entries corresponding to a local ip address, this table id is also used in the comparison. The table id is populated when the fib_info is created. This was already done in some places, but not in ip_rt_ioctl(). This has now been fixed. Fixes: 021dd3b8a142 ("net: Add routes to the table associated with the device") Acked-by: David Ahern Tested-by: David Ahern Signed-off-by: Mark Tomlinson Signed-off-by: David S. Miller --- include/net/ip_fib.h | 3 ++- net/ipv4/fib_frontend.c | 3 ++- net/ipv4/fib_semantics.c | 8 ++++++-- 3 files changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 4079fc18ffe4..7d4a72e75f33 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -111,6 +111,7 @@ struct fib_info { unsigned char fib_scope; unsigned char fib_type; __be32 fib_prefsrc; + u32 fib_tb_id; u32 fib_priority; u32 *fib_metrics; #define fib_mtu fib_metrics[RTAX_MTU-1] @@ -319,7 +320,7 @@ void fib_flush_external(struct net *net); /* Exported by fib_semantics.c */ int ip_fib_check_default(__be32 gw, struct net_device *dev); int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force); -int fib_sync_down_addr(struct net *net, __be32 local); +int fib_sync_down_addr(struct net_device *dev, __be32 local); int fib_sync_up(struct net_device *dev, unsigned int nh_flags); extern u32 fib_multipath_secret __read_mostly; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index ef2ebeb89d0f..1b25daf8c7f1 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -509,6 +509,7 @@ static int rtentry_to_fib_config(struct net *net, int cmd, struct rtentry *rt, if (!dev) return -ENODEV; cfg->fc_oif = dev->ifindex; + cfg->fc_table = l3mdev_fib_table(dev); if (colon) { struct in_ifaddr *ifa; struct in_device *in_dev = __in_dev_get_rtnl(dev); @@ -1027,7 +1028,7 @@ no_promotions: * First of all, we scan fib_info list searching * for stray nexthop entries, then ignite fib_flush. */ - if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local)) + if (fib_sync_down_addr(dev, ifa->ifa_local)) fib_flush(dev_net(dev)); } } diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 539fa264e67d..e9f56225e53f 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -1057,6 +1057,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg) fi->fib_priority = cfg->fc_priority; fi->fib_prefsrc = cfg->fc_prefsrc; fi->fib_type = cfg->fc_type; + fi->fib_tb_id = cfg->fc_table; fi->fib_nhs = nhs; change_nexthops(fi) { @@ -1337,18 +1338,21 @@ nla_put_failure: * referring to it. * - device went down -> we must shutdown all nexthops going via it. */ -int fib_sync_down_addr(struct net *net, __be32 local) +int fib_sync_down_addr(struct net_device *dev, __be32 local) { int ret = 0; unsigned int hash = fib_laddr_hashfn(local); struct hlist_head *head = &fib_info_laddrhash[hash]; + struct net *net = dev_net(dev); + int tb_id = l3mdev_fib_table(dev); struct fib_info *fi; if (!fib_info_laddrhash || local == 0) return 0; hlist_for_each_entry(fi, head, fib_lhash) { - if (!net_eq(fi->fib_net, net)) + if (!net_eq(fi->fib_net, net) || + fi->fib_tb_id != tb_id) continue; if (fi->fib_prefsrc == local) { fi->fib_flags |= RTNH_F_DEAD; -- cgit v1.2.3-59-g8ed1b From 751eb6b6042a596b0080967c1a529a9fe98dac1d Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 5 Sep 2016 16:06:31 +0800 Subject: ipv6: addrconf: fix dev refcont leak when DAD failed In general, when DAD detected IPv6 duplicate address, ifp->state will be set to INET6_IFADDR_STATE_ERRDAD and DAD is stopped by a delayed work, the call tree should be like this: ndisc_recv_ns -> addrconf_dad_failure <- missing ifp put -> addrconf_mod_dad_work -> schedule addrconf_dad_work() -> addrconf_dad_stop() <- missing ifp hold before call it addrconf_dad_failure() called with ifp refcont holding but not put. addrconf_dad_work() call addrconf_dad_stop() without extra holding refcount. This will not cause any issue normally. But the race between addrconf_dad_failure() and addrconf_dad_work() may cause ifp refcount leak and netdevice can not be unregister, dmesg show the following messages: IPv6: eth0: IPv6 duplicate address fe80::XX:XXXX:XXXX:XX detected! ... unregister_netdevice: waiting for eth0 to become free. Usage count = 1 Cc: stable@vger.kernel.org Fixes: c15b1ccadb32 ("ipv6: move DAD and addrconf_verify processing to workqueue") Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index bdf368eff5ab..2f1f5d439788 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1948,6 +1948,7 @@ errdad: spin_unlock_bh(&ifp->lock); addrconf_mod_dad_work(ifp, 0); + in6_ifa_put(ifp); } /* Join to solicited addr multicast group. @@ -3857,6 +3858,7 @@ static void addrconf_dad_work(struct work_struct *w) addrconf_dad_begin(ifp); goto out; } else if (action == DAD_ABORT) { + in6_ifa_hold(ifp); addrconf_dad_stop(ifp, 1); if (disable_ipv6) addrconf_ifdown(idev->dev, 0); -- cgit v1.2.3-59-g8ed1b From 76061f631c2ea4ab9c4d66f3a96ecc5737f5aaf7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 7 Sep 2016 08:34:11 -0700 Subject: tcp: fastopen: avoid negative sk_forward_alloc When DATA and/or FIN are carried in a SYN/ACK message or SYN message, we append an skb in socket receive queue, but we forget to call sk_forced_mem_schedule(). Effect is that the socket has a negative sk->sk_forward_alloc as long as the message is not read by the application. Josh Hunt fixed a similar issue in commit d22e15371811 ("tcp: fix tcp fin memory accounting") Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: Eric Dumazet Reviewed-by: Josh Hunt Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 62a5751d4fe1..4e777a3243f9 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -150,6 +150,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb) tp->segs_in = 0; tcp_segs_in(tp, skb); __skb_pull(skb, tcp_hdrlen(skb)); + sk_forced_mem_schedule(sk, skb->truesize); skb_set_owner_r(skb, sk); TCP_SKB_CB(skb)->seq++; -- cgit v1.2.3-59-g8ed1b From db7196a0d0984b933ccf2cd6a60e26abf466e8a3 Mon Sep 17 00:00:00 2001 From: Artem Germanov Date: Wed, 7 Sep 2016 10:49:36 -0700 Subject: tcp: cwnd does not increase in TCP YeAH Commit 76174004a0f19785a328f40388e87e982bbf69b9 (tcp: do not slow start when cwnd equals ssthresh ) introduced regression in TCP YeAH. Using 100ms delay 1% loss virtual ethernet link kernel 4.2 shows bandwidth ~500KB/s for single TCP connection and kernel 4.3 and above (including 4.8-rc4) shows bandwidth ~100KB/s. That is caused by stalled cwnd when cwnd equals ssthresh. This patch fixes it by proper increasing cwnd in this case. Signed-off-by: Artem Germanov Acked-by: Dmitry Adamushko Signed-off-by: David S. Miller --- net/ipv4/tcp_yeah.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c index 028eb046ea40..9c5fc973267f 100644 --- a/net/ipv4/tcp_yeah.c +++ b/net/ipv4/tcp_yeah.c @@ -76,7 +76,7 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (!tcp_is_cwnd_limited(sk)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) + if (tcp_in_slow_start(tp)) tcp_slow_start(tp, acked); else if (!yeah->doing_reno_now) { -- cgit v1.2.3-59-g8ed1b From 7303a1475008bee5c3e82a06a282568415690d72 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 8 Sep 2016 17:54:11 +0800 Subject: sctp: identify chunks that need to be fragmented at IP level Previously, without GSO, it was easy to identify it: if the chunk didn't fit and there was no data chunk in the packet yet, we could fragment at IP level. So if there was an auth chunk and we were bundling a big data chunk, it would fragment regardless of the size of the auth chunk. This also works for the context of PMTU reductions. But with GSO, we cannot distinguish such PMTU events anymore, as the packet is allowed to exceed PMTU. So we need another check: to ensure that the chunk that we are adding, actually fits the current PMTU. If it doesn't, trigger a flush and let it be fragmented at IP level in the next round. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/output.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sctp/output.c b/net/sctp/output.c index 1f1682b9a6a8..31b7bc35895d 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -878,7 +878,7 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, struct sctp_chunk *chunk, u16 chunk_len) { - size_t psize, pmtu; + size_t psize, pmtu, maxsize; sctp_xmit_t retval = SCTP_XMIT_OK; psize = packet->size; @@ -906,6 +906,17 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, goto out; } + /* Similarly, if this chunk was built before a PMTU + * reduction, we have to fragment it at IP level now. So + * if the packet already contains something, we need to + * flush. + */ + maxsize = pmtu - packet->overhead; + if (packet->auth) + maxsize -= WORD_ROUND(packet->auth->skb->len); + if (chunk_len > maxsize) + retval = SCTP_XMIT_PMTU_FULL; + /* It is also okay to fragment if the chunk we are * adding is a control chunk, but only if current packet * is not a GSO one otherwise it causes fragmentation of -- cgit v1.2.3-59-g8ed1b