March 2021 - Linux-stable-mirror

FAILED: Patch "net, bpf: Fix ip6ip6 crash with collect_md populated skbs" failed to apply to 4.14-stable tree

by Sasha Levin

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From a188bb5638d41aa99090ebf2f85d3505ab13fba5 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann <daniel(a)iogearbox.net> Date: Wed, 10 Mar 2021 01:38:10 +0100 Subject: [PATCH] net, bpf: Fix ip6ip6 crash with collect_md populated skbs I ran into a crash where setting up a ip6ip6 tunnel device which was /not/ set to collect_md mode was receiving collect_md populated skbs for xmit. The BPF prog was populating the skb via bpf_skb_set_tunnel_key() which is assigning special metadata dst entry and then redirecting the skb to the device, taking ip6_tnl_start_xmit() -> ipxip6_tnl_xmit() -> ip6_tnl_xmit() and in the latter it performs a neigh lookup based on skb_dst(skb) where we trigger a NULL pointer dereference on dst->ops->neigh_lookup() since the md_dst_ops do not populate neigh_lookup callback with a fake handler. Transform the md_dst_ops into generic dst_blackhole_ops that can also be reused elsewhere when needed, and use them for the metadata dst entries as callback ops. Also, remove the dst_md_discard{,_out}() ops and rely on dst_discard{,_out}() from dst_init() which free the skb the same way modulo the splat. Given we will be able to recover just fine from there, avoid any potential splats iff this gets ever triggered in future (or worse, panic on warns when set). Fixes: f38a9eb1f77b ("dst: Metadata destinations") Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/core/dst.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) diff --git a/net/core/dst.c b/net/core/dst.c index 5f6315601776..fb3bcba87744 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -275,37 +275,24 @@ unsigned int dst_blackhole_mtu(const struct dst_entry *dst) } EXPORT_SYMBOL_GPL(dst_blackhole_mtu); -static struct dst_ops md_dst_ops = { - .family = AF_UNSPEC, +static struct dst_ops dst_blackhole_ops = { + .family = AF_UNSPEC, + .neigh_lookup = dst_blackhole_neigh_lookup, + .check = dst_blackhole_check, + .cow_metrics = dst_blackhole_cow_metrics, + .update_pmtu = dst_blackhole_update_pmtu, + .redirect = dst_blackhole_redirect, + .mtu = dst_blackhole_mtu, }; -static int dst_md_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call output on metadata dst\n"); - kfree_skb(skb); - return 0; -} - -static int dst_md_discard(struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call input on metadata dst\n"); - kfree_skb(skb); - return 0; -} - static void __metadata_dst_init(struct metadata_dst *md_dst, enum metadata_type type, u8 optslen) - { struct dst_entry *dst; dst = &md_dst->dst; - dst_init(dst, &md_dst_ops, NULL, 1, DST_OBSOLETE_NONE, + dst_init(dst, &dst_blackhole_ops, NULL, 1, DST_OBSOLETE_NONE, DST_METADATA | DST_NOCOUNT); - - dst->input = dst_md_discard; - dst->output = dst_md_discard_out; - memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst)); md_dst->type = type; } -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "net: bridge: don't notify switchdev for local FDB addresses" failed to apply to 4.19-stable tree

by Sasha Levin

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 6ab4c3117aec4e08007d9e971fa4133e1de1082d Mon Sep 17 00:00:00 2001 From: Vladimir Oltean <vladimir.oltean(a)nxp.com> Date: Mon, 22 Mar 2021 20:21:08 +0200 Subject: [PATCH] net: bridge: don't notify switchdev for local FDB addresses As explained in this discussion: https://lore.kernel.org/netdev/20210117193009.io3nungdwuzmo5f7@skbuf/ the switchdev notifiers for FDB entries managed to have a zero-day bug. The bridge would not say that this entry is local: ip link add br0 type bridge ip link set swp0 master br0 bridge fdb add dev swp0 00:01:02:03:04:05 master local and the switchdev driver would be more than happy to offload it as a normal static FDB entry. This is despite the fact that 'local' and non-'local' entries have completely opposite directions: a local entry is locally terminated and not forwarded, whereas a static entry is forwarded and not locally terminated. So, for example, DSA would install this entry on swp0 instead of installing it on the CPU port as it should. There is an even sadder part, which is that the 'local' flag is implicit if 'static' is not specified, meaning that this command produces the same result of adding a 'local' entry: bridge fdb add dev swp0 00:01:02:03:04:05 master I've updated the man pages for 'bridge', and after reading it now, it should be pretty clear to any user that the commands above were broken and should have never resulted in the 00:01:02:03:04:05 address being forwarded (this behavior is coherent with non-switchdev interfaces): https://patchwork.kernel.org/project/netdevbpf/cover/20210211104502.2081443… If you're a user reading this and this is what you want, just use: bridge fdb add dev swp0 00:01:02:03:04:05 master static Because switchdev should have given drivers the means from day one to classify FDB entries as local/non-local, but didn't, it means that all drivers are currently broken. So we can just as well omit the switchdev notifications for local FDB entries, which is exactly what this patch does to close the bug in stable trees. For further development work where drivers might want to trap the local FDB entries to the host, we can add a 'bool is_local' to br_switchdev_fdb_call_notifiers(), and selectively make drivers act upon that bit, while all the others ignore those entries if the 'is_local' bit is set. Fixes: 6b26b51b1d13 ("net: bridge: Add support for notifying devices about FDB add/del") Signed-off-by: Vladimir Oltean <vladimir.oltean(a)nxp.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/bridge/br_switchdev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index b89503832fcc..1e24d9a2c9a7 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -128,6 +128,8 @@ br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type) { if (!fdb->dst) return; + if (test_bit(BR_FDB_LOCAL, &fdb->flags)) + return; switch (type) { case RTM_DELNEIGH: -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "selinux: vsock: Set SID for socket returned by accept()" failed to apply to 4.19-stable tree

by Sasha Levin

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 1f935e8e72ec28dddb2dc0650b3b6626a293d94b Mon Sep 17 00:00:00 2001 From: David Brazdil <dbrazdil(a)google.com> Date: Fri, 19 Mar 2021 13:05:41 +0000 Subject: [PATCH] selinux: vsock: Set SID for socket returned by accept() For AF_VSOCK, accept() currently returns sockets that are unlabelled. Other socket families derive the child's SID from the SID of the parent and the SID of the incoming packet. This is typically done as the connected socket is placed in the queue that accept() removes from. Reuse the existing 'security_sk_clone' hook to copy the SID from the parent (server) socket to the child. There is no packet SID in this case. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: David Brazdil <dbrazdil(a)google.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/vmw_vsock/af_vsock.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 5546710d8ac1..bc7fb9bf3351 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -755,6 +755,7 @@ static struct sock *__vsock_create(struct net *net, vsk->buffer_size = psk->buffer_size; vsk->buffer_min_size = psk->buffer_min_size; vsk->buffer_max_size = psk->buffer_max_size; + security_sk_clone(parent, sk); } else { vsk->trusted = ns_capable_noaudit(&init_user_ns, CAP_NET_ADMIN); vsk->owner = get_current_cred(); -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "ipv6: weaken the v4mapped source check" failed to apply to 4.19-stable tree

by Sasha Levin

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From dcc32f4f183ab8479041b23a1525d48233df1d43 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski <kuba(a)kernel.org> Date: Wed, 17 Mar 2021 09:55:15 -0700 Subject: [PATCH] ipv6: weaken the v4mapped source check This reverts commit 6af1799aaf3f1bc8defedddfa00df3192445bbf3. Commit 6af1799aaf3f ("ipv6: drop incoming packets having a v4mapped source address") introduced an input check against v4mapped addresses. Use of such addresses on the wire is indeed questionable and not allowed on public Internet. As the commit pointed out https://tools.ietf.org/html/draft-itojun-v6ops-v4mapped-harmful-02 lists potential issues. Unfortunately there are applications which use v4mapped addresses, and breaking them is a clear regression. For example v4mapped addresses (or any semi-valid addresses, really) may be used for uni-direction event streams or packet export. Since the issue which sparked the addition of the check was with TCP and request_socks in particular push the check down to TCPv6 and DCCP. This restores the ability to receive UDPv6 packets with v4mapped address as the source. Keep using the IPSTATS_MIB_INHDRERRORS statistic to minimize the user-visible changes. Fixes: 6af1799aaf3f ("ipv6: drop incoming packets having a v4mapped source address") Reported-by: Sunyi Shao <sunyishao(a)fb.com> Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Acked-by: Mat Martineau <mathew.j.martineau(a)linux.intel.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/dccp/ipv6.c | 5 +++++ net/ipv6/ip6_input.c | 10 ---------- net/ipv6/tcp_ipv6.c | 5 +++++ net/mptcp/subflow.c | 5 +++++ 4 files changed, 15 insertions(+), 10 deletions(-) diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 1f73603913f5..2be5c69824f9 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -319,6 +319,11 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) return 0; /* discard, don't send a reset here */ + if (ipv6_addr_v4mapped(&ipv6_hdr(skb)->saddr)) { + __IP6_INC_STATS(sock_net(sk), NULL, IPSTATS_MIB_INHDRERRORS); + return 0; + } + if (dccp_bad_service_code(sk, service)) { dcb->dccpd_reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; goto drop; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index e9d2a4a409aa..80256717868e 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -245,16 +245,6 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev, if (ipv6_addr_is_multicast(&hdr->saddr)) goto err; - /* While RFC4291 is not explicit about v4mapped addresses - * in IPv6 headers, it seems clear linux dual-stack - * model can not deal properly with these. - * Security models could be fooled by ::ffff:127.0.0.1 for example. - * - * https://tools.ietf.org/html/draft-itojun-v6ops-v4mapped-harmful-02 - */ - if (ipv6_addr_v4mapped(&hdr->saddr)) - goto err; - skb->transport_header = skb->network_header + sizeof(*hdr); IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index bd44ded7e50c..d0f007741e8e 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1175,6 +1175,11 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) goto drop; + if (ipv6_addr_v4mapped(&ipv6_hdr(skb)->saddr)) { + __IP6_INC_STATS(sock_net(sk), NULL, IPSTATS_MIB_INHDRERRORS); + return 0; + } + return tcp_conn_request(&tcp6_request_sock_ops, &tcp_request_sock_ipv6_ops, sk, skb); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 3d47d670e665..d17d39ccdf34 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -477,6 +477,11 @@ static int subflow_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) goto drop; + if (ipv6_addr_v4mapped(&ipv6_hdr(skb)->saddr)) { + __IP6_INC_STATS(sock_net(sk), NULL, IPSTATS_MIB_INHDRERRORS); + return 0; + } + return tcp_conn_request(&mptcp_subflow_request_sock_ops, &subflow_request_sock_ipv6_ops, sk, skb); -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "tcp: relookup sock for RST+ACK packets handled by obsolete req sock" failed to apply to 4.19-stable tree

by Sasha Levin

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 7233da86697efef41288f8b713c10c2499cffe85 Mon Sep 17 00:00:00 2001 From: Alexander Ovechkin <ovov(a)yandex-team.ru> Date: Mon, 15 Mar 2021 14:05:45 +0300 Subject: [PATCH] tcp: relookup sock for RST+ACK packets handled by obsolete req sock Currently tcp_check_req can be called with obsolete req socket for which big socket have been already created (because of CPU race or early demux assigning req socket to multiple packets in gro batch). Commit e0f9759f530bf789e984 ("tcp: try to keep packet if SYN_RCV race is lost") added retry in case when tcp_check_req is called for PSH|ACK packet. But if client sends RST+ACK immediatly after connection being established (it is performing healthcheck, for example) retry does not occur. In that case tcp_check_req tries to close req socket, leaving big socket active. Fixes: e0f9759f530 ("tcp: try to keep packet if SYN_RCV race is lost") Signed-off-by: Alexander Ovechkin <ovov(a)yandex-team.ru> Reported-by: Oleg Senin <olegsenin(a)yandex-team.ru> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- include/net/inet_connection_sock.h | 2 +- net/ipv4/inet_connection_sock.c | 7 +++++-- net/ipv4/tcp_minisocks.c | 7 +++++-- 3 files changed, 11 insertions(+), 5 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 10a625760de9..3c8c59471bc1 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -282,7 +282,7 @@ static inline int inet_csk_reqsk_queue_is_full(const struct sock *sk) return inet_csk_reqsk_queue_len(sk) >= sk->sk_max_ack_backlog; } -void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req); +bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req); void inet_csk_reqsk_queue_drop_and_put(struct sock *sk, struct request_sock *req); static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 6bd7ca09af03..fd472eae4f5c 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -705,12 +705,15 @@ static bool reqsk_queue_unlink(struct request_sock *req) return found; } -void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) { - if (reqsk_queue_unlink(req)) { + bool unlinked = reqsk_queue_unlink(req); + + if (unlinked) { reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); reqsk_put(req); } + return unlinked; } EXPORT_SYMBOL(inet_csk_reqsk_queue_drop); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0055ae0a3bf8..7513ba45553d 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -804,8 +804,11 @@ embryonic_reset: tcp_reset(sk, skb); } if (!fastopen) { - inet_csk_reqsk_queue_drop(sk, req); - __NET_INC_STATS(sock_net(sk), LINUX_MIB_EMBRYONICRSTS); + bool unlinked = inet_csk_reqsk_queue_drop(sk, req); + + if (unlinked) + __NET_INC_STATS(sock_net(sk), LINUX_MIB_EMBRYONICRSTS); + *req_stolen = !unlinked; } return NULL; } -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "net, bpf: Fix ip6ip6 crash with collect_md populated skbs" failed to apply to 4.19-stable tree

by Sasha Levin

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From a188bb5638d41aa99090ebf2f85d3505ab13fba5 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann <daniel(a)iogearbox.net> Date: Wed, 10 Mar 2021 01:38:10 +0100 Subject: [PATCH] net, bpf: Fix ip6ip6 crash with collect_md populated skbs I ran into a crash where setting up a ip6ip6 tunnel device which was /not/ set to collect_md mode was receiving collect_md populated skbs for xmit. The BPF prog was populating the skb via bpf_skb_set_tunnel_key() which is assigning special metadata dst entry and then redirecting the skb to the device, taking ip6_tnl_start_xmit() -> ipxip6_tnl_xmit() -> ip6_tnl_xmit() and in the latter it performs a neigh lookup based on skb_dst(skb) where we trigger a NULL pointer dereference on dst->ops->neigh_lookup() since the md_dst_ops do not populate neigh_lookup callback with a fake handler. Transform the md_dst_ops into generic dst_blackhole_ops that can also be reused elsewhere when needed, and use them for the metadata dst entries as callback ops. Also, remove the dst_md_discard{,_out}() ops and rely on dst_discard{,_out}() from dst_init() which free the skb the same way modulo the splat. Given we will be able to recover just fine from there, avoid any potential splats iff this gets ever triggered in future (or worse, panic on warns when set). Fixes: f38a9eb1f77b ("dst: Metadata destinations") Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/core/dst.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) diff --git a/net/core/dst.c b/net/core/dst.c index 5f6315601776..fb3bcba87744 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -275,37 +275,24 @@ unsigned int dst_blackhole_mtu(const struct dst_entry *dst) } EXPORT_SYMBOL_GPL(dst_blackhole_mtu); -static struct dst_ops md_dst_ops = { - .family = AF_UNSPEC, +static struct dst_ops dst_blackhole_ops = { + .family = AF_UNSPEC, + .neigh_lookup = dst_blackhole_neigh_lookup, + .check = dst_blackhole_check, + .cow_metrics = dst_blackhole_cow_metrics, + .update_pmtu = dst_blackhole_update_pmtu, + .redirect = dst_blackhole_redirect, + .mtu = dst_blackhole_mtu, }; -static int dst_md_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call output on metadata dst\n"); - kfree_skb(skb); - return 0; -} - -static int dst_md_discard(struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call input on metadata dst\n"); - kfree_skb(skb); - return 0; -} - static void __metadata_dst_init(struct metadata_dst *md_dst, enum metadata_type type, u8 optslen) - { struct dst_entry *dst; dst = &md_dst->dst; - dst_init(dst, &md_dst_ops, NULL, 1, DST_OBSOLETE_NONE, + dst_init(dst, &dst_blackhole_ops, NULL, 1, DST_OBSOLETE_NONE, DST_METADATA | DST_NOCOUNT); - - dst->input = dst_md_discard; - dst->output = dst_md_discard_out; - memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst)); md_dst->type = type; } -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "net: bridge: don't notify switchdev for local FDB addresses" failed to apply to 5.4-stable tree

by Sasha Levin

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 6ab4c3117aec4e08007d9e971fa4133e1de1082d Mon Sep 17 00:00:00 2001 From: Vladimir Oltean <vladimir.oltean(a)nxp.com> Date: Mon, 22 Mar 2021 20:21:08 +0200 Subject: [PATCH] net: bridge: don't notify switchdev for local FDB addresses As explained in this discussion: https://lore.kernel.org/netdev/20210117193009.io3nungdwuzmo5f7@skbuf/ the switchdev notifiers for FDB entries managed to have a zero-day bug. The bridge would not say that this entry is local: ip link add br0 type bridge ip link set swp0 master br0 bridge fdb add dev swp0 00:01:02:03:04:05 master local and the switchdev driver would be more than happy to offload it as a normal static FDB entry. This is despite the fact that 'local' and non-'local' entries have completely opposite directions: a local entry is locally terminated and not forwarded, whereas a static entry is forwarded and not locally terminated. So, for example, DSA would install this entry on swp0 instead of installing it on the CPU port as it should. There is an even sadder part, which is that the 'local' flag is implicit if 'static' is not specified, meaning that this command produces the same result of adding a 'local' entry: bridge fdb add dev swp0 00:01:02:03:04:05 master I've updated the man pages for 'bridge', and after reading it now, it should be pretty clear to any user that the commands above were broken and should have never resulted in the 00:01:02:03:04:05 address being forwarded (this behavior is coherent with non-switchdev interfaces): https://patchwork.kernel.org/project/netdevbpf/cover/20210211104502.2081443… If you're a user reading this and this is what you want, just use: bridge fdb add dev swp0 00:01:02:03:04:05 master static Because switchdev should have given drivers the means from day one to classify FDB entries as local/non-local, but didn't, it means that all drivers are currently broken. So we can just as well omit the switchdev notifications for local FDB entries, which is exactly what this patch does to close the bug in stable trees. For further development work where drivers might want to trap the local FDB entries to the host, we can add a 'bool is_local' to br_switchdev_fdb_call_notifiers(), and selectively make drivers act upon that bit, while all the others ignore those entries if the 'is_local' bit is set. Fixes: 6b26b51b1d13 ("net: bridge: Add support for notifying devices about FDB add/del") Signed-off-by: Vladimir Oltean <vladimir.oltean(a)nxp.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/bridge/br_switchdev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index b89503832fcc..1e24d9a2c9a7 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -128,6 +128,8 @@ br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type) { if (!fdb->dst) return; + if (test_bit(BR_FDB_LOCAL, &fdb->flags)) + return; switch (type) { case RTM_DELNEIGH: -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "selinux: vsock: Set SID for socket returned by accept()" failed to apply to 5.4-stable tree

by Sasha Levin

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 1f935e8e72ec28dddb2dc0650b3b6626a293d94b Mon Sep 17 00:00:00 2001 From: David Brazdil <dbrazdil(a)google.com> Date: Fri, 19 Mar 2021 13:05:41 +0000 Subject: [PATCH] selinux: vsock: Set SID for socket returned by accept() For AF_VSOCK, accept() currently returns sockets that are unlabelled. Other socket families derive the child's SID from the SID of the parent and the SID of the incoming packet. This is typically done as the connected socket is placed in the queue that accept() removes from. Reuse the existing 'security_sk_clone' hook to copy the SID from the parent (server) socket to the child. There is no packet SID in this case. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: David Brazdil <dbrazdil(a)google.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/vmw_vsock/af_vsock.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 5546710d8ac1..bc7fb9bf3351 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -755,6 +755,7 @@ static struct sock *__vsock_create(struct net *net, vsk->buffer_size = psk->buffer_size; vsk->buffer_min_size = psk->buffer_min_size; vsk->buffer_max_size = psk->buffer_max_size; + security_sk_clone(parent, sk); } else { vsk->trusted = ns_capable_noaudit(&init_user_ns, CAP_NET_ADMIN); vsk->owner = get_current_cred(); -- 2.30.1

3 years, 10 months

1
0
0 0

FAILED: Patch "drop_monitor: Perform cleanup upon probe registration failure" failed to apply to 5.4-stable tree

by Sasha Levin

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From 9398e9c0b1d44eeb700e9e766c02bcc765c82570 Mon Sep 17 00:00:00 2001 From: Ido Schimmel <idosch(a)nvidia.com> Date: Wed, 10 Mar 2021 12:28:01 +0200 Subject: [PATCH] drop_monitor: Perform cleanup upon probe registration failure In the rare case that drop_monitor fails to register its probe on the 'napi_poll' tracepoint, it will not deactivate its hysteresis timer as part of the error path. If the hysteresis timer was armed by the shortly lived 'kfree_skb' probe and user space retries to initiate tracing, a warning will be emitted for trying to initialize an active object [1]. Fix this by properly undoing all the operations that were done prior to probe registration, in both software and hardware code paths. Note that syzkaller managed to fail probe registration by injecting a slab allocation failure [2]. [1] ODEBUG: init active (active state 0) object type: timer_list hint: sched_send_work+0x0/0x60 include/linux/list.h:135 WARNING: CPU: 1 PID: 8649 at lib/debugobjects.c:505 debug_print_object+0x16e/0x250 lib/debugobjects.c:505 Modules linked in: CPU: 1 PID: 8649 Comm: syz-executor.0 Not tainted 5.11.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:debug_print_object+0x16e/0x250 lib/debugobjects.c:505 [...] Call Trace: __debug_object_init+0x524/0xd10 lib/debugobjects.c:588 debug_timer_init kernel/time/timer.c:722 [inline] debug_init kernel/time/timer.c:770 [inline] init_timer_key+0x2d/0x340 kernel/time/timer.c:814 net_dm_trace_on_set net/core/drop_monitor.c:1111 [inline] set_all_monitor_traces net/core/drop_monitor.c:1188 [inline] net_dm_monitor_start net/core/drop_monitor.c:1295 [inline] net_dm_cmd_trace+0x720/0x1220 net/core/drop_monitor.c:1339 genl_family_rcv_msg_doit+0x228/0x320 net/netlink/genetlink.c:739 genl_family_rcv_msg net/netlink/genetlink.c:783 [inline] genl_rcv_msg+0x328/0x580 net/netlink/genetlink.c:800 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2502 genl_rcv+0x24/0x40 net/netlink/genetlink.c:811 netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline] netlink_unicast+0x533/0x7d0 net/netlink/af_netlink.c:1338 netlink_sendmsg+0x856/0xd90 net/netlink/af_netlink.c:1927 sock_sendmsg_nosec net/socket.c:652 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:672 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2348 ___sys_sendmsg+0xf3/0x170 net/socket.c:2402 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2435 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xae [2] FAULT_INJECTION: forcing a failure. name failslab, interval 1, probability 0, space 0, times 1 CPU: 1 PID: 8645 Comm: syz-executor.0 Not tainted 5.11.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: dump_stack+0xfa/0x151 should_fail.cold+0x5/0xa should_failslab+0x5/0x10 __kmalloc+0x72/0x3f0 tracepoint_add_func+0x378/0x990 tracepoint_probe_register+0x9c/0xe0 net_dm_cmd_trace+0x7fc/0x1220 genl_family_rcv_msg_doit+0x228/0x320 genl_rcv_msg+0x328/0x580 netlink_rcv_skb+0x153/0x420 genl_rcv+0x24/0x40 netlink_unicast+0x533/0x7d0 netlink_sendmsg+0x856/0xd90 sock_sendmsg+0xcf/0x120 ____sys_sendmsg+0x6e8/0x810 ___sys_sendmsg+0xf3/0x170 __sys_sendmsg+0xe5/0x1b0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xae Fixes: 70c69274f354 ("drop_monitor: Initialize timer and work item upon tracing enable") Fixes: 8ee2267ad33e ("drop_monitor: Convert to using devlink tracepoint") Reported-by: syzbot+779559d6503f3a56213d(a)syzkaller.appspotmail.com Signed-off-by: Ido Schimmel <idosch(a)nvidia.com> Reviewed-by: Jiri Pirko <jiri(a)nvidia.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/core/drop_monitor.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index 571f191c06d9..db65ce62b625 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -1053,6 +1053,20 @@ static int net_dm_hw_monitor_start(struct netlink_ext_ack *extack) return 0; err_module_put: + for_each_possible_cpu(cpu) { + struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu); + struct sk_buff *skb; + + del_timer_sync(&hw_data->send_timer); + cancel_work_sync(&hw_data->dm_alert_work); + while ((skb = __skb_dequeue(&hw_data->drop_queue))) { + struct devlink_trap_metadata *hw_metadata; + + hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata; + net_dm_hw_metadata_free(hw_metadata); + consume_skb(skb); + } + } module_put(THIS_MODULE); return rc; } @@ -1134,6 +1148,15 @@ static int net_dm_trace_on_set(struct netlink_ext_ack *extack) err_unregister_trace: unregister_trace_kfree_skb(ops->kfree_skb_probe, NULL); err_module_put: + for_each_possible_cpu(cpu) { + struct per_cpu_dm_data *data = &per_cpu(dm_cpu_data, cpu); + struct sk_buff *skb; + + del_timer_sync(&data->send_timer); + cancel_work_sync(&data->dm_alert_work); + while ((skb = __skb_dequeue(&data->drop_queue))) + consume_skb(skb); + } module_put(THIS_MODULE); return rc; } -- 2.30.1

3 years, 10 months

1
0
0 0

Patch "net, bpf: Fix ip6ip6 crash with collect_md populated skbs" failed to apply to 5.4-stable tree

by Sasha Levin

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ >From a188bb5638d41aa99090ebf2f85d3505ab13fba5 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann <daniel(a)iogearbox.net> Date: Wed, 10 Mar 2021 01:38:10 +0100 Subject: [PATCH] net, bpf: Fix ip6ip6 crash with collect_md populated skbs I ran into a crash where setting up a ip6ip6 tunnel device which was /not/ set to collect_md mode was receiving collect_md populated skbs for xmit. The BPF prog was populating the skb via bpf_skb_set_tunnel_key() which is assigning special metadata dst entry and then redirecting the skb to the device, taking ip6_tnl_start_xmit() -> ipxip6_tnl_xmit() -> ip6_tnl_xmit() and in the latter it performs a neigh lookup based on skb_dst(skb) where we trigger a NULL pointer dereference on dst->ops->neigh_lookup() since the md_dst_ops do not populate neigh_lookup callback with a fake handler. Transform the md_dst_ops into generic dst_blackhole_ops that can also be reused elsewhere when needed, and use them for the metadata dst entries as callback ops. Also, remove the dst_md_discard{,_out}() ops and rely on dst_discard{,_out}() from dst_init() which free the skb the same way modulo the splat. Given we will be able to recover just fine from there, avoid any potential splats iff this gets ever triggered in future (or worse, panic on warns when set). Fixes: f38a9eb1f77b ("dst: Metadata destinations") Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: David S. Miller <davem(a)davemloft.net> --- net/core/dst.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) diff --git a/net/core/dst.c b/net/core/dst.c index 5f6315601776..fb3bcba87744 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -275,37 +275,24 @@ unsigned int dst_blackhole_mtu(const struct dst_entry *dst) } EXPORT_SYMBOL_GPL(dst_blackhole_mtu); -static struct dst_ops md_dst_ops = { - .family = AF_UNSPEC, +static struct dst_ops dst_blackhole_ops = { + .family = AF_UNSPEC, + .neigh_lookup = dst_blackhole_neigh_lookup, + .check = dst_blackhole_check, + .cow_metrics = dst_blackhole_cow_metrics, + .update_pmtu = dst_blackhole_update_pmtu, + .redirect = dst_blackhole_redirect, + .mtu = dst_blackhole_mtu, }; -static int dst_md_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call output on metadata dst\n"); - kfree_skb(skb); - return 0; -} - -static int dst_md_discard(struct sk_buff *skb) -{ - WARN_ONCE(1, "Attempting to call input on metadata dst\n"); - kfree_skb(skb); - return 0; -} - static void __metadata_dst_init(struct metadata_dst *md_dst, enum metadata_type type, u8 optslen) - { struct dst_entry *dst; dst = &md_dst->dst; - dst_init(dst, &md_dst_ops, NULL, 1, DST_OBSOLETE_NONE, + dst_init(dst, &dst_blackhole_ops, NULL, 1, DST_OBSOLETE_NONE, DST_METADATA | DST_NOCOUNT); - - dst->input = dst_md_discard; - dst->output = dst_md_discard_out; - memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst)); md_dst->type = type; } -- 2.30.1

3 years, 10 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror March 2021