Eric Dumazet reported a use-after-free related to the per-netns ehash
series. [0]
When we create a TCP socket from userspace, the socket always holds a
refcnt of the netns. This guarantees that a reqsk timer is always fired
before netns dismantle. Each reqsk has a refcnt of its listener, so the
listener is not freed before the reqsk, and the net is not freed before
the listener as well.
OTOH, when in-kernel users create a TCP socket, it might not hold a refcnt
of its netns. Thus, a reqsk timer can be fired after the netns dismantle
and access freed per-netns ehash.
To avoid the use-after-free, we need to clean up TCP_NEW_SYN_RECV sockets
in inet_twsk_purge() if the netns uses a per-netns ehash.
[0]: https://lore.kernel.org/netdev/CANn89iLXMup0dRD_Ov79Xt8N9FM0XdhCHEN05sf3eLwxKweM6w@mail.gmail.com/
BUG: KASAN: use-after-free in tcp_or_dccp_get_hashinfo
include/net/inet_hashtables.h:181 [inline]
BUG: KASAN: use-after-free in reqsk_queue_unlink+0x320/0x350
net/ipv4/inet_connection_sock.c:913
Read of size 8 at addr ffff88807545bd80 by task syz-executor.2/8301
CPU: 1 PID: 8301 Comm: syz-executor.2 Not tainted
6.0.0-syzkaller-02757-gaf7d23f9d96a #0
Hardware name: Google Google Compute Engine/Google Compute Engine,
BIOS Google 09/22/2022
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
print_address_description mm/kasan/report.c:317 [inline]
print_report.cold+0x2ba/0x719 mm/kasan/report.c:433
kasan_report+0xb1/0x1e0 mm/kasan/report.c:495
tcp_or_dccp_get_hashinfo include/net/inet_hashtables.h:181 [inline]
reqsk_queue_unlink+0x320/0x350 net/ipv4/inet_connection_sock.c:913
inet_csk_reqsk_queue_drop net/ipv4/inet_connection_sock.c:927 [inline]
inet_csk_reqsk_queue_drop_and_put net/ipv4/inet_connection_sock.c:939 [inline]
reqsk_timer_handler+0x724/0x1160 net/ipv4/inet_connection_sock.c:1053
call_timer_fn+0x1a0/0x6b0 kernel/time/timer.c:1474
expire_timers kernel/time/timer.c:1519 [inline]
__run_timers.part.0+0x674/0xa80 kernel/time/timer.c:1790
__run_timers kernel/time/timer.c:1768 [inline]
run_timer_softirq+0xb3/0x1d0 kernel/time/timer.c:1803
__do_softirq+0x1d0/0x9c8 kernel/softirq.c:571
invoke_softirq kernel/softirq.c:445 [inline]
__irq_exit_rcu+0x123/0x180 kernel/softirq.c:650
irq_exit_rcu+0x5/0x20 kernel/softirq.c:662
sysvec_apic_timer_interrupt+0x93/0xc0 arch/x86/kernel/apic/apic.c:1107
</IRQ>
Fixes: d1e5e6408b ("tcp: Introduce optional per-netns ehash.")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Eric Dumazet <edumazet@google.com>
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20221012145036.74960-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Johannes Berg says:
====================
More wireless fixes for 6.1
This has only the fixes for the scan parsing issues.
* tag 'wireless-2022-10-13' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
wifi: cfg80211: update hidden BSSes to avoid WARN_ON
wifi: mac80211: fix crash in beacon protection for P2P-device
wifi: mac80211_hwsim: avoid mac80211 warning on bad rate
wifi: cfg80211: avoid nontransmitted BSS list corruption
wifi: cfg80211: fix BSS refcounting bugs
wifi: cfg80211: ensure length byte is present before access
wifi: mac80211: fix MBSSID parsing use-after-free
wifi: cfg80211/mac80211: reject bad MBSSID elements
wifi: cfg80211: fix u8 overflow in cfg80211_update_notlisted_nontrans()
====================
Link: https://lore.kernel.org/r/20221013100522.46346-1-johannes@sipsolutions.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
A WARN_ON call trace would be triggered when 'ct(commit, alg=helper)'
applies on a confirmed connection:
WARNING: CPU: 0 PID: 1251 at net/netfilter/nf_conntrack_extend.c:98
RIP: 0010:nf_ct_ext_add+0x12d/0x150 [nf_conntrack]
Call Trace:
<TASK>
nf_ct_helper_ext_add+0x12/0x60 [nf_conntrack]
__nf_ct_try_assign_helper+0xc4/0x160 [nf_conntrack]
__ovs_ct_lookup+0x72e/0x780 [openvswitch]
ovs_ct_execute+0x1d8/0x920 [openvswitch]
do_execute_actions+0x4e6/0xb60 [openvswitch]
ovs_execute_actions+0x60/0x140 [openvswitch]
ovs_packet_cmd_execute+0x2ad/0x310 [openvswitch]
genl_family_rcv_msg_doit.isra.15+0x113/0x150
genl_rcv_msg+0xef/0x1f0
which can be reproduced with these OVS flows:
table=0, in_port=veth1,tcp,tcp_dst=2121,ct_state=-trk
actions=ct(commit, table=1)
table=1, in_port=veth1,tcp,tcp_dst=2121,ct_state=+trk+new
actions=ct(commit, alg=ftp),normal
The issue was introduced by commit 248d45f1e1 ("openvswitch: Allow
attaching helper in later commit") where it somehow removed the check
of nf_ct_is_confirmed before asigning the helper. This patch is to fix
it by bringing it back.
Fixes: 248d45f1e1 ("openvswitch: Allow attaching helper in later commit")
Reported-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Tested-by: Aaron Conole <aconole@redhat.com>
Link: https://lore.kernel.org/r/c5c9092a22a2194650222bffaf786902613deb16.1665085502.git.lucien.xin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
setsockopt(IPV6_ADDRFORM) and tcp_v6_connect() change icsk->icsk_af_ops
under lock_sock(), but tcp_(get|set)sockopt() read it locklessly. To
avoid load/store tearing, we need to add READ_ONCE() and WRITE_ONCE()
for the reads and writes.
Thanks to Eric Dumazet for providing the syzbot report:
BUG: KCSAN: data-race in tcp_setsockopt / tcp_v6_connect
write to 0xffff88813c624518 of 8 bytes by task 23936 on cpu 0:
tcp_v6_connect+0x5b3/0xce0 net/ipv6/tcp_ipv6.c:240
__inet_stream_connect+0x159/0x6d0 net/ipv4/af_inet.c:660
inet_stream_connect+0x44/0x70 net/ipv4/af_inet.c:724
__sys_connect_file net/socket.c:1976 [inline]
__sys_connect+0x197/0x1b0 net/socket.c:1993
__do_sys_connect net/socket.c:2003 [inline]
__se_sys_connect net/socket.c:2000 [inline]
__x64_sys_connect+0x3d/0x50 net/socket.c:2000
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
read to 0xffff88813c624518 of 8 bytes by task 23937 on cpu 1:
tcp_setsockopt+0x147/0x1c80 net/ipv4/tcp.c:3789
sock_common_setsockopt+0x5d/0x70 net/core/sock.c:3585
__sys_setsockopt+0x212/0x2b0 net/socket.c:2252
__do_sys_setsockopt net/socket.c:2263 [inline]
__se_sys_setsockopt net/socket.c:2260 [inline]
__x64_sys_setsockopt+0x62/0x70 net/socket.c:2260
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
value changed: 0xffffffff8539af68 -> 0xffffffff8539aff8
Reported by Kernel Concurrency Sanitizer on:
CPU: 1 PID: 23937 Comm: syz-executor.5 Not tainted
6.0.0-rc4-syzkaller-00331-g4ed9c1e971b1-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine,
BIOS Google 08/26/2022
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 086d49058c ("ipv6: annotate some data-races around sk->sk_prot")
fixed some data-races around sk->sk_prot but it was not enough.
Some functions in inet6_(stream|dgram)_ops still access sk->sk_prot
without lock_sock() or rtnl_lock(), so they need READ_ONCE() to avoid
load tearing.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Originally, inet6_sk(sk)->XXX were changed under lock_sock(), so we were
able to clean them up by calling inet6_destroy_sock() during the IPv6 ->
IPv4 conversion by IPV6_ADDRFORM. However, commit 03485f2adc ("udpv6:
Add lockless sendmsg() support") added a lockless memory allocation path,
which could cause a memory leak:
setsockopt(IPV6_ADDRFORM) sendmsg()
+-----------------------+ +-------+
- do_ipv6_setsockopt(sk, ...) - udpv6_sendmsg(sk, ...)
- sockopt_lock_sock(sk) ^._ called via udpv6_prot
- lock_sock(sk) before WRITE_ONCE()
- WRITE_ONCE(sk->sk_prot, &tcp_prot)
- inet6_destroy_sock() - if (!corkreq)
- sockopt_release_sock(sk) - ip6_make_skb(sk, ...)
- release_sock(sk) ^._ lockless fast path for
the non-corking case
- __ip6_append_data(sk, ...)
- ipv6_local_rxpmtu(sk, ...)
- xchg(&np->rxpmtu, skb)
^._ rxpmtu is never freed.
- goto out_no_dst;
- lock_sock(sk)
For now, rxpmtu is only the case, but not to miss the future change
and a similar bug fixed in commit e27326009a ("net: ping6: Fix
memleak in ipv6_renew_options()."), let's set a new function to IPv6
sk->sk_destruct() and call inet6_cleanup_sock() there. Since the
conversion does not change sk->sk_destruct(), we can guarantee that
we can clean up IPv6 resources finally.
We can now remove all inet6_destroy_sock() calls from IPv6 protocol
specific ->destroy() functions, but such changes are invasive to
backport. So they can be posted as a follow-up later for net-next.
Fixes: 03485f2adc ("udpv6: Add lockless sendmsg() support")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 4b340ae20d ("IPv6: Complete IPV6_DONTFRAG support") forgot
to add a change to free inet6_sk(sk)->rxpmtu while converting an IPv6
socket into IPv4 with IPV6_ADDRFORM. After conversion, sk_prot is
changed to udp_prot and ->destroy() never cleans it up, resulting in
a memory leak.
This is due to the discrepancy between inet6_destroy_sock() and
IPV6_ADDRFORM, so let's call inet6_destroy_sock() from IPV6_ADDRFORM
to remove the difference.
However, this is not enough for now because rxpmtu can be changed
without lock_sock() after commit 03485f2adc ("udpv6: Add lockless
sendmsg() support"). We will fix this case in the following patch.
Note we will rename inet6_destroy_sock() to inet6_cleanup_sock() and
remove unnecessary inet6_destroy_sock() calls in sk_prot->destroy()
in the future.
Fixes: 4b340ae20d ("IPv6: Complete IPV6_DONTFRAG support")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, we have a bug where a simultaneous DROPTAG ioctl and socket
close may race, as we attempt to remove a key from lists twice, and
perform an unref for each removal operation. This may result in a uaf
when we attempt the second unref.
This change fixes the race by making __mctp_key_remove tolerant to being
called on a key that has already been removed from the socket/net lists,
and only performs the unref when we do the actual remove. We also need
to hold the list lock on the ioctl cleanup path.
This fix is based on a bug report and comprehensive analysis from
butt3rflyh4ck <butterflyhuangxx@gmail.com>, found via syzkaller.
Cc: stable@vger.kernel.org
Fixes: 63ed1aab3d ("mctp: Add SIOCMCTP{ALLOC,DROP}TAG ioctls for tag control")
Reported-by: butt3rflyh4ck <butterflyhuangxx@gmail.com>
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal says:
====================
netfilter fixes for net
This series from Phil Sutter for the *net* tree fixes a problem with a change
from the 6.1 development phase: the change to nft_fib should have used
the more recent flowic_l3mdev field. Pointed out by Guillaume Nault.
This also makes the older iptables module follow the same pattern.
Also add selftest case and avoid test failure in nft_fib.sh when the
host environment has set rp_filter=1.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
If net.ipv4.conf.all.rp_filter is set, it overrides the per-interface
setting and thus defeats the fix from bbe4c0896d ("selftests:
netfilter: disable rp_filter on router"). Unset it as well to cover that
case.
Fixes: bbe4c0896d ("selftests: netfilter: disable rp_filter on router")
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Florian Westphal <fw@strlen.de>
Use the introduced field for correct operation with VRF devices instead
of conditionally overwriting flowic_oif. This is a partial revert of
commit b575b24b8e ("netfilter: Fix rpfilter dropping vrf packets by
mistake"), implementing a simpler solution.
Signed-off-by: Phil Sutter <phil@nwl.cc>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Test reverse path (filter) matches in iptables, ip6tables and nftables.
Both with a regular interface and a VRF.
Signed-off-by: Phil Sutter <phil@nwl.cc>
Reviewed-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Eric Dumazet says:
====================
inet: ping: give ping some care
First patch fixes an ipv6 ping bug that has been there forever,
for large sizes.
Second patch fixes a recent and elusive bug, that can potentially
crash the host. This is what I mentioned privately to Paolo and
Jakub at LPC in Dublin.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
For a given ping datagram, ping_getfrag() is called once
per skb fragment.
A large datagram requiring more than one page fragment
is currently getting the checksum of the last fragment,
instead of the cumulative one.
After this patch, "ping -s 35000 ::1" is working correctly.
Fixes: 6d0bfe2261 ("net: ipv6: Add IPv6 support to the ping socket.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Lorenzo Colitti <lorenzo@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
am65_cpsw_nuss_register_ndevs() skips calling devlink_port_type_eth_set()
for ports without assigned netdev, triggering the following warning when
DEVLINK_PORT_TYPE_WARN_TIMEOUT elapses after 3600s:
Type was not set for devlink port.
WARNING: CPU: 0 PID: 129 at net/core/devlink.c:8095 devlink_port_type_warn+0x18/0x30
Fixes: 0680e20af5 ("net: ethernet: ti: am65-cpsw: Fix devlink port register sequence")
Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Kalle Valo says:
====================
wireless fixes for v6.1
First set of fixes for v6.1. Quite a lot of fixes in stack but also
for mt76.
cfg80211/mac80211
- fix locking error in mac80211's hw addr change
- fix TX queue stop for internal TXQs
- handling of very small (e.g. STP TCN) packets
- two memcpy() hardening fixes
- fix probe request 6 GHz capability warning
- fix various connection prints
- fix decapsulation offload for AP VLAN
mt76
- fix rate reporting, LLC packets and receive checksum offload on specific chipsets
iwlwifi
- fix crash due to list corruption
ath11k
- fix a compiler warning with GCC 11 and KASAN
* tag 'wireless-2022-10-11' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
wifi: ath11k: mac: fix reading 16 bytes from a region of size 0 warning
wifi: iwlwifi: mvm: fix double list_add at iwl_mvm_mac_wake_tx_queue (other cases)
wifi: mt76: fix rx checksum offload on mt7615/mt7915/mt7921
wifi: mt76: fix receiving LLC packets on mt7615/mt7915
wifi: nl80211: Split memcpy() of struct nl80211_wowlan_tcp_data_token flexible array
wifi: wext: use flex array destination for memcpy()
wifi: cfg80211: fix ieee80211_data_to_8023_exthdr handling of small packets
wifi: mac80211: netdev compatible TX stop for iTXQ drivers
wifi: mac80211: fix decap offload for stations on AP_VLAN interfaces
wifi: mac80211: unlock on error in ieee80211_can_powered_addr_change()
wifi: mac80211: remove/avoid misleading prints
wifi: mac80211: fix probe req HE capabilities access
wifi: mac80211: do not drop packets smaller than the LLC-SNAP header on fast-rx
wifi: mt76: fix rate reporting / throughput regression on mt7915 and newer
====================
Link: https://lore.kernel.org/r/20221011163123.A093CC433D6@smtp.kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Marc Kleine-Budde says:
====================
pull-request: can 2022-10-11
this is a pull request of 4 patches for net/main.
Anssi Hannula and Jimmy Assarsson contribute 4 patches for the
kvaser_usb driver. A check for actual received length of USB transfers
is added, the use of an uninitialized completion is fixed, the TX
queue is re-synced after restart, and the CAN state is fixed after
restart.
* tag 'linux-can-fixes-for-6.1-20221011' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can:
can: kvaser_usb_leaf: Fix CAN state after restart
can: kvaser_usb_leaf: Fix TX queue out of sync after restart
can: kvaser_usb: Fix use of uninitialized completion
can: kvaser_usb_leaf: Fix overread with an invalid command
====================
Link: https://lore.kernel.org/r/20221011074815.397301-1-mkl@pengutronix.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Linaro reported stringop-overread warnings in ath11k (this is one of many):
drivers/net/wireless/ath/ath11k/mac.c:2238:29: error: 'ath11k_peer_assoc_h_he_limit' reading 16 bytes from a region of size 0 [-Werror=stringop-overread]
My further investigation showed that these warnings happen on GCC 11.3 but not
with GCC 12.2, and with only the kernel config Linaro provided:
https://builds.tuxbuild.com/2F4W7nZHNx3T88RB0gaCZ9hBX6c/config
I saw the same warnings both with arm64 and x86_64 builds and KASAN seems to be
the reason triggering these warnings with GCC 11. Nobody else has reported
this so this seems to be quite rare corner case. I don't know what specific
commit started emitting this warning so I can't provide a Fixes tag. The
function hasn't been touched for a year.
I decided to workaround this by converting the pointer to a new array in stack,
and then copying the data to the new array. It's only 16 bytes anyway and this
is executed during association, so not in a hotpath.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.9
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Link: https://lore.kernel.org/all/CA+G9fYsZ_qypa=jHY_dJ=tqX4515+qrV9n2SWXVDHve826nF7Q@mail.gmail.com/
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221010160638.20152-1-kvalo@kernel.org
BUGs like this are still reproducible:
[ 31.509616] list_add corruption. prev->next should be next (ffff8f8644242300), but was ffff8f86493fd300. (prev=ffff8f86493fd300).
[ 31.521544] ------------[ cut here ]------------
[ 31.526248] kernel BUG at lib/list_debug.c:30!
[ 31.530781] invalid opcode: 0000 [#1] PREEMPT SMP PTI
[ 31.535831] CPU: 1 PID: 626 Comm: wpa_supplicant Not tainted 6.0.0+ #7
[ 31.542450] Hardware name: Dell Inc. Inspiron 660s/0478VN , BIOS A07 08/24/2012
[ 31.550484] RIP: 0010:__list_add_valid.cold+0x3a/0x5b
[ 31.555537] Code: f2 4c 89 c1 48 89 fe 48 c7 c7 28 20 69 89 e8 4c e3 fd ff 0f 0b 48 89 d1 4c 89 c6 4c 89 ca 48 c7 c7 d0 1f 69 89 e8 35 e3 fd ff <0f> 0b 4c 89 c1 48 c7 c7 78 1f 69 89 e8 24 e3 fd ff 0f 0b 48 c7 c7
[ 31.574605] RSP: 0018:ffff9f6f00dc3748 EFLAGS: 00010286
[ 31.579990] RAX: 0000000000000075 RBX: ffff8f8644242080 RCX: 0000000000000000
[ 31.587155] RDX: 0000000000000201 RSI: ffffffff8967862d RDI: 00000000ffffffff
[ 31.594482] RBP: ffff8f86493fd2e8 R08: 0000000000000000 R09: 00000000ffffdfff
[ 31.601735] R10: ffff9f6f00dc3608 R11: ffffffff89f46128 R12: ffff8f86493fd300
[ 31.608986] R13: ffff8f86493fd300 R14: ffff8f8644242300 R15: ffff8f8643dd3f2c
[ 31.616151] FS: 00007f3bb9a707c0(0000) GS:ffff8f865a300000(0000) knlGS:0000000000000000
[ 31.624447] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 31.630286] CR2: 00007fe3647d5600 CR3: 00000001125a6002 CR4: 00000000000606e0
[ 31.637539] Call Trace:
[ 31.639936] <TASK>
[ 31.642143] iwl_mvm_mac_wake_tx_queue+0x71/0x90 [iwlmvm]
[ 31.647569] ieee80211_queue_skb+0x4b6/0x720 [mac80211]
...
So, it is necessary to extend the applied solution with commit 14a3aacf51
("iwlwifi: mvm: fix double list_add at iwl_mvm_mac_wake_tx_queue")
to all other cases where the station queues are invalidated and the related
lists are not emptied. Because, otherwise as before, if some new element is
added later to the list in iwl_mvm_mac_wake_tx_queue, it can match with the
old one and produce the same commented BUG.
That is, in order to avoid this problem completely, we must also remove the
related lists for the other cases when station queues are invalidated.
Fixes: cfbc6c4c5b ("iwlwifi: mvm: support mac80211 TXQs model")
Reported-by: Petr Stourac <pstourac@redhat.com>
Tested-by: Petr Stourac <pstourac@redhat.com>
Signed-off-by: Jose Ignacio Tornos Martinez <jtornosm@redhat.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221010081611.145027-1-jtornosm@redhat.com
Checking the relevant rxd bits for the checksum information only indicates
if the checksum verification was performed by the hardware and doesn't show
actual checksum errors. Checksum errors are indicated in the info field of
the DMA descriptor. Fix packets erroneously marked as CHECKSUM_UNNECESSARY
by checking the extra bits as well.
Those bits are only passed to the driver for MMIO devices at the moment, so
limit checksum offload to those.
Fixes: 2122dfbfd0 ("mt76: mt7615: add rx checksum offload support")
Fixes: 94244d2ea5 ("mt76: mt7915: add rx checksum offload support")
Fixes: 0e75732764 ("mt76: mt7921: enable rx csum offload")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221005130824.23371-2-nbd@nbd.name
When 802.3 decap offload is enabled, the hardware indicates header translation
failure, whenever either the LLC-SNAP header was not found, or a VLAN header
with an unregcognized tag is present.
In that case, the hardware inserts a 2-byte length fields after the MAC
addresses. For VLAN packets, this tag needs to be removed. However,
for 802.3 LLC packets, the length bytes should be preserved, since there
is no separate ethertype field in the data.
This fixes an issue where the length field was omitted for LLC frames, causing
them to be malformed after hardware decap.
Fixes: 1eeff0b4c1 ("mt76: mt7915: fix decap offload corner case with 4-addr VLAN frames")
Reported-by: Chad Monroe <chad.monroe@smartrg.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221005130824.23371-1-nbd@nbd.name
The TX queue seems to be implicitly flushed by the hardware during
bus-off or bus-off recovery, but the driver does not reset the TX
bookkeeping.
Despite not resetting TX bookkeeping the driver still re-enables TX
queue unconditionally, leading to "cannot find free context" /
NETDEV_TX_BUSY errors if the TX queue was full at bus-off time.
Fix that by resetting TX bookkeeping on CAN restart.
Tested with 0bfd:0124 Kvaser Mini PCI Express 2xHS FW 4.18.778.
Cc: stable@vger.kernel.org
Fixes: 080f40a6fa ("can: kvaser_usb: Add support for Kvaser CAN/USB devices")
Tested-by: Jimmy Assarsson <extja@kvaser.com>
Signed-off-by: Anssi Hannula <anssi.hannula@bitwise.fi>
Signed-off-by: Jimmy Assarsson <extja@kvaser.com>
Link: https://lore.kernel.org/all/20221010150829.199676-4-extja@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
flush_comp is initialized when CMD_FLUSH_QUEUE is sent to the device and
completed when the device sends CMD_FLUSH_QUEUE_RESP.
This causes completion of uninitialized completion if the device sends
CMD_FLUSH_QUEUE_RESP before CMD_FLUSH_QUEUE is ever sent (e.g. as a
response to a flush by a previously bound driver, or a misbehaving
device).
Fix that by initializing flush_comp in kvaser_usb_init_one() like the
other completions.
This issue is only triggerable after RX URBs have been set up, i.e. the
interface has been opened at least once.
Cc: stable@vger.kernel.org
Fixes: aec5fb2268 ("can: kvaser_usb: Add support for Kvaser USB hydra family")
Tested-by: Jimmy Assarsson <extja@kvaser.com>
Signed-off-by: Anssi Hannula <anssi.hannula@bitwise.fi>
Signed-off-by: Jimmy Assarsson <extja@kvaser.com>
Link: https://lore.kernel.org/all/20221010150829.199676-3-extja@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
For command events read from the device,
kvaser_usb_leaf_read_bulk_callback() verifies that cmd->len does not
exceed the size of the received data, but the actual kvaser_cmd handlers
will happily read any kvaser_cmd fields without checking for cmd->len.
This can cause an overread if the last cmd in the buffer is shorter than
expected for the command type (with cmd->len showing the actual short
size).
Maximum overread seems to be 22 bytes (CMD_LEAF_LOG_MESSAGE), some of
which are delivered to userspace as-is.
Fix that by verifying the length of command before handling it.
This issue can only occur after RX URBs have been set up, i.e. the
interface has been opened at least once.
Cc: stable@vger.kernel.org
Fixes: 080f40a6fa ("can: kvaser_usb: Add support for Kvaser CAN/USB devices")
Tested-by: Jimmy Assarsson <extja@kvaser.com>
Signed-off-by: Anssi Hannula <anssi.hannula@bitwise.fi>
Signed-off-by: Jimmy Assarsson <extja@kvaser.com>
Link: https://lore.kernel.org/all/20221010150829.199676-2-extja@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The original commit that added support for the SYSTEMPORT Lite variant
halved the number of RX descriptors due to a confusion between the
number of descriptors and the number of descriptor words. There are 512
descriptor *words* which means 256 descriptors total.
Fixes: 44a4524c54 ("net: systemport: Add support for SYSTEMPORT Lite")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20221007034201.4126054-1-f.fainelli@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When updating beacon elements in a non-transmitted BSS,
also update the hidden sub-entries to the same beacon
elements, so that a future update through other paths
won't trigger a WARN_ON().
The warning is triggered because the beacon elements in
the hidden BSSes that are children of the BSS should
always be the same as in the parent.
Reported-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Tested-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Fixes: 0b8fb8235b ("cfg80211: Parsing of Multiple BSSID information in scanning")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
If beacon protection is active but the beacon cannot be
decrypted or is otherwise malformed, we call the cfg80211
API to report this to userspace, but that uses a netdev
pointer, which isn't present for P2P-Device. Fix this to
call it only conditionally to ensure cfg80211 won't crash
in the case of P2P-Device.
This fixes CVE-2022-42722.
Reported-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Fixes: 9eaf183af7 ("mac80211: Report beacon protection failures to user space")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
If the tool on the other side (e.g. wmediumd) gets confused
about the rate, we hit a warning in mac80211. Silence that
by effectively duplicating the check here and dropping the
frame silently (in mac80211 it's dropped with the warning).
Reported-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Tested-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
If a non-transmitted BSS shares enough information (both
SSID and BSSID!) with another non-transmitted BSS of a
different AP, then we can find and update it, and then
try to add it to the non-transmitted BSS list. We do a
search for it on the transmitted BSS, but if it's not
there (but belongs to another transmitted BSS), the list
gets corrupted.
Since this is an erroneous situation, simply fail the
list insertion in this case and free the non-transmitted
BSS.
This fixes CVE-2022-42721.
Reported-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Tested-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Fixes: 0b8fb8235b ("cfg80211: Parsing of Multiple BSSID information in scanning")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
There are multiple refcounting bugs related to multi-BSSID:
- In bss_ref_get(), if the BSS has a hidden_beacon_bss, then
the bss pointer is overwritten before checking for the
transmitted BSS, which is clearly wrong. Fix this by using
the bss_from_pub() macro.
- In cfg80211_bss_update() we copy the transmitted_bss pointer
from tmp into new, but then if we release new, we'll unref
it erroneously. We already set the pointer and ref it, but
need to NULL it since it was copied from the tmp data.
- In cfg80211_inform_single_bss_data(), if adding to the non-
transmitted list fails, we unlink the BSS and yet still we
return it, but this results in returning an entry without
a reference. We shouldn't return it anyway if it was broken
enough to not get added there.
This fixes CVE-2022-42720.
Reported-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Tested-by: Sönke Huster <shuster@seemoo.tu-darmstadt.de>
Fixes: a3584f56de ("cfg80211: Properly track transmitting and non-transmitting BSS")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
When iterating the elements here, ensure the length byte is
present before checking it to see if the entire element will
fit into the buffer.
Longer term, we should rewrite this code using the type-safe
element iteration macros that check all of this.
Fixes: 0b8fb8235b ("cfg80211: Parsing of Multiple BSSID information in scanning")
Reported-by: Soenke Huster <shuster@seemoo.tu-darmstadt.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
When we parse a multi-BSSID element, we might point some
element pointers into the allocated nontransmitted_profile.
However, we free this before returning, causing UAF when the
relevant pointers in the parsed elements are accessed.
Fix this by not allocating the scratch buffer separately but
as part of the returned structure instead, that way, there
are no lifetime issues with it.
The scratch buffer introduction as part of the returned data
here is taken from MLO feature work done by Ilan.
This fixes CVE-2022-42719.
Fixes: 5023b14cf4 ("mac80211: support profile split between elements")
Co-developed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Per spec, the maximum value for the MaxBSSID ('n') indicator is 8,
and the minimum is 1 since a multiple BSSID set with just one BSSID
doesn't make sense (the # of BSSIDs is limited by 2^n).
Limit this in the parsing in both cfg80211 and mac80211, rejecting
any elements with an invalid value.
This fixes potentially bad shifts in the processing of these inside
the cfg80211_gen_new_bssid() function later.
I found this during the investigation of CVE-2022-41674 fixed by the
previous patch.
Fixes: 0b8fb8235b ("cfg80211: Parsing of Multiple BSSID information in scanning")
Fixes: 78ac51f815 ("mac80211: support multi-bssid")
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
In the copy code of the elements, we do the following calculation
to reach the end of the MBSSID element:
/* copy the IEs after MBSSID */
cpy_len = mbssid[1] + 2;
This looks fine, however, cpy_len is a u8, the same as mbssid[1],
so the addition of two can overflow. In this case the subsequent
memcpy() will overflow the allocated buffer, since it copies 256
bytes too much due to the way the allocation and memcpy() sizes
are calculated.
Fix this by using size_t for the cpy_len variable.
This fixes CVE-2022-41674.
Reported-by: Soenke Huster <shuster@seemoo.tu-darmstadt.de>
Tested-by: Soenke Huster <shuster@seemoo.tu-darmstadt.de>
Fixes: 0b8fb8235b ("cfg80211: Parsing of Multiple BSSID information in scanning")
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
In error path after calling cn10k_mcs_init(), cn10k_mcs_free() need
be called to avoid memory leak.
Fixes: c54ffc7360 ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If alloc_mem() fails in mcs_register_interrupts(), it should return error
code.
Fixes: 6c635f78c4 ("octeontx2-af: cn10k: mcs: Handle MCS block interrupts")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix wrong pointer passed to PTR_ERR() in dsa_port_phylink_create() to print
error message.
Fixes: cf5ca4ddc3 ("net: dsa: don't leave dangling pointers in dp->pl when failing")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
