commit 8590541473 upstream.
Since we're setting the CRYPTO_TFM_REQ_MAY_BACKLOG flag on our
requests to the crypto API, crypto_aead_{encrypt,decrypt} can return
-EBUSY instead of -EINPROGRESS in valid situations. For example, when
the cryptd queue for AESNI is full (easy to trigger with an
artificially low cryptd.cryptd_max_cpu_qlen), requests will be enqueued
to the backlog but still processed. In that case, the async callback
will also be called twice: first with err == -EINPROGRESS, which it
seems we can just ignore, then with err == 0.
Compared to Sabrina's original patch this version uses the new
tls_*crypt_async_wait() helpers and converts the EBUSY to
EINPROGRESS to avoid having to modify all the error handling
paths. The handling is identical.
Fixes: a54667f672 ("tls: Add support for encryption using async offload accelerator")
Fixes: 94524d8fc9 ("net/tls: Add support for async decryption of tls records")
Co-developed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://lore.kernel.org/netdev/9681d1febfec295449a62300938ed2ae66983f28.1694018970.git.sd@queasysnail.net/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
[v5.15: fixed contextual merge-conflicts in tls_decrypt_done and tls_encrypt_done]
Cc: <stable@vger.kernel.org> # 5.15
Signed-off-by: Shaoying Xu <shaoyi@amazon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit aec7961916 upstream.
The submitting thread (one which called recvmsg/sendmsg)
may exit as soon as the async crypto handler calls complete()
so any code past that point risks touching already freed data.
Try to avoid the locking and extra flags altogether.
Have the main thread hold an extra reference, this way
we can depend solely on the atomic ref counter for
synchronization.
Don't futz with reiniting the completion, either, we are now
tightly controlling when completion fires.
Reported-by: valis <sec@valis.email>
Fixes: 0cada33241 ("net/tls: fix race condition causing kernel panic")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
[v5.15: fixed contextual conflicts in struct tls_sw_context_rx and func
init_ctx_rx; replaced DEBUG_NET_WARN_ON_ONCE with BUILD_BUG_ON_INVALID
since they're equivalent when DEBUG_NET is not defined]
Cc: <stable@vger.kernel.org> # 5.15
Signed-off-by: Shaoying Xu <shaoyi@amazon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c57ca512f3 upstream.
Factor out waiting for async encrypt and decrypt to finish.
There are already multiple copies and a subsequent fix will
need more. No functional changes.
Note that crypto_wait_req() returns wait->err
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: aec7961916 ("tls: fix race between async notify and socket close")
[v5.15: removed changes in tls_sw_splice_eof and adjusted waiting factor out for
async descrypt in tls_sw_recvmsg]
Cc: <stable@vger.kernel.org> # 5.15
Signed-off-by: Shaoying Xu <shaoyi@amazon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 37943f047b upstream.
Since we are protected from async completions by decrypt_compl_lock
we can drop the async_notify and reinit the completion before we
start waiting.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: aec7961916 ("tls: fix race between async notify and socket close")
Signed-off-by: Shaoying Xu <shaoyi@amazon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 3903902401 upstream.
The original implementation of nfsd used signals to stop threads during
shutdown.
In Linux 2.3.46pre5 nfsd gained the ability to shutdown threads
internally it if was asked to run "0" threads. After this user-space
transitioned to using "rpc.nfsd 0" to stop nfsd and sending signals to
threads was no longer an important part of the API.
In commit 3ebdbe5203 ("SUNRPC: discard svo_setup and rename
svc_set_num_threads_sync()") (v5.17-rc1~75^2~41) we finally removed the
use of signals for stopping threads, using kthread_stop() instead.
This patch makes the "obvious" next step and removes the ability to
signal nfsd threads - or any svc threads. nfsd stops allowing signals
and we don't check for their delivery any more.
This will allow for some simplification in later patches.
A change worth noting is in nfsd4_ssc_setup_dul(). There was previously
a signal_pending() check which would only succeed when the thread was
being shut down. It should really have tested kthread_should_stop() as
well. Now it just does the latter, not the former.
Signed-off-by: NeilBrown <neilb@suse.de>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a26ff37e62 upstream.
net_alloc_generic is called by net_alloc, which is called without any
locking. It reads max_gen_ptrs, which is changed under pernet_ops_rwsem. It
is read twice, first to allocate an array, then to set s.len, which is
later used to limit the bounds of the array access.
It is possible that the array is allocated and another thread is
registering a new pernet ops, increments max_gen_ptrs, which is then used
to set s.len with a larger than allocated length for the variable array.
Fix it by reading max_gen_ptrs only once in net_alloc_generic. If
max_gen_ptrs is later incremented, it will be caught in net_assign_generic.
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@igalia.com>
Fixes: 073862ba5d ("netns: fix net_alloc_generic()")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20240502132006.3430840-1-cascardo@igalia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 86b29d830a ]
The change from skb_copy to pskb_copy unfortunately changed the data
copying to omit the ethernet header, since it was pulled before reaching
this point. Fix this by calling __skb_push/pull around pskb_copy.
Fixes: 59c878cbcd ("net: bridge: fix multicast-to-unicast with fraglist GSO")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1aec77b2bb ]
Each attribute inside a nested IFLA_VF_VLAN_LIST is assumed to be a
struct ifla_vf_vlan_info so the size of such attribute needs to be at least
of sizeof(struct ifla_vf_vlan_info) which is 14 bytes.
The current size validation in do_setvfinfo is against NLA_HDRLEN (4 bytes)
which is less than sizeof(struct ifla_vf_vlan_info) so this validation
is not enough and a too small attribute might be cast to a
struct ifla_vf_vlan_info, this might result in an out of bands
read access when accessing the saved (casted) entry in ivvl.
Fixes: 79aab093a0 ("net: Update API for VF vlan protocol 802.1ad support")
Signed-off-by: Roded Zats <rzats@paloaltonetworks.com>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/r/20240502155751.75705-1-rzats@paloaltonetworks.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 58fbfecab9 ]
The software GRO path for esp transport mode uses skb_mac_header_rebuild
prior to re-injecting the packet via the xfrm_napi_dev. This only
copies skb->mac_len bytes of header which may not be sufficient if the
packet contains 802.1Q tags or other VLAN tags. Worse copying only the
initial header will leave a packet marked as being VLAN tagged but
without the corresponding tag leading to mangling when it is later
untagged.
The VLAN tags are important when receiving the decrypted esp transport
mode packet after GRO processing to ensure it is received on the correct
interface.
Therefore record the full mac header length in xfrm*_transport_input for
later use in corresponding xfrm*_transport_finish to copy the entire mac
header when rebuilding the mac header for GRO. The skb->data pointer is
left pointing skb->mac_header bytes after the start of the mac header as
is expected by the network stack and network and transport header
offsets reset to this location.
Fixes: 7785bba299 ("esp: Add a software GRO codepath")
Signed-off-by: Paul Davey <paul.davey@alliedtelesis.co.nz>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 405df89dd5 ]
We noticed some rare sk_buffs were stepping past the queue when system was
under memory pressure. The general theory is to skip enqueueing
sk_buffs when its not necessary which is the normal case with a system
that is properly provisioned for the task, no memory pressure and enough
cpu assigned.
But, if we can't allocate memory due to an ENOMEM error when enqueueing
the sk_buff into the sockmap receive queue we push it onto a delayed
workqueue to retry later. When a new sk_buff is received we then check
if that queue is empty. However, there is a problem with simply checking
the queue length. When a sk_buff is being processed from the ingress queue
but not yet on the sockmap msg receive queue its possible to also recv
a sk_buff through normal path. It will check the ingress queue which is
zero and then skip ahead of the pkt being processed.
Previously we used sock lock from both contexts which made the problem
harder to hit, but not impossible.
To fix instead of popping the skb from the queue entirely we peek the
skb from the queue and do the copy there. This ensures checks to the
queue length are non-zero while skb is being processed. Then finally
when the entire skb has been copied to user space queue or another
socket we pop it off the queue. This way the queue length check allows
bypassing the queue only after the list has been completely processed.
To reproduce issue we run NGINX compliance test with sockmap running and
observe some flakes in our testing that we attributed to this issue.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Suggested-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: William Findlay <will@isovalent.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-5-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit bce22552f9 ]
Now that the backlog manages the reschedule() logic correctly we can drop
the partial fix to reschedule from recvmsg hook.
Rescheduling on recvmsg hook was added to address a corner case where we
still had data in the backlog state but had nothing to kick it and
reschedule the backlog worker to run and finish copying data out of the
state. This had a couple limitations, first it required user space to
kick it introducing an unnecessary EBUSY and retry. Second it only
handled the ingress case and egress redirects would still be hung.
With the correct fix, pushing the reschedule logic down to where the
enomem error occurs we can drop this fix.
Fixes: bec217197b ("skmsg: Schedule psock work if the cached skb exists on the psock")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-4-john.fastabend@gmail.com
Stable-dep-of: 405df89dd5 ("bpf, sockmap: Improved check for empty queue")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 29173d07f7 ]
Sk_buffs are fed into sockmap verdict programs either from a strparser
(when the user might want to decide how framing of skb is done by attaching
another parser program) or directly through tcp_read_sock. The
tcp_read_sock is the preferred method for performance when the BPF logic is
a stream parser.
The flow for Cilium's common use case with a stream parser is,
tcp_read_sock()
sk_psock_verdict_recv
ret = bpf_prog_run_pin_on_cpu()
sk_psock_verdict_apply(sock, skb, ret)
// if system is under memory pressure or app is slow we may
// need to queue skb. Do this queuing through ingress_skb and
// then kick timer to wake up handler
skb_queue_tail(ingress_skb, skb)
schedule_work(work);
The work queue is wired up to sk_psock_backlog(). This will then walk the
ingress_skb skb list that holds our sk_buffs that could not be handled,
but should be OK to run at some later point. However, its possible that
the workqueue doing this work still hits an error when sending the skb.
When this happens the skbuff is requeued on a temporary 'state' struct
kept with the workqueue. This is necessary because its possible to
partially send an skbuff before hitting an error and we need to know how
and where to restart when the workqueue runs next.
Now for the trouble, we don't rekick the workqueue. This can cause a
stall where the skbuff we just cached on the state variable might never
be sent. This happens when its the last packet in a flow and no further
packets come along that would cause the system to kick the workqueue from
that side.
To fix we could do simple schedule_work(), but while under memory pressure
it makes sense to back off some instead of continue to retry repeatedly. So
instead to fix convert schedule_work to schedule_delayed_work and add
backoff logic to reschedule from backlog queue on errors. Its not obvious
though what a good backoff is so use '1'.
To test we observed some flakes whil running NGINX compliance test with
sockmap we attributed these failed test to this bug and subsequent issue.
>From on list discussion. This commit
bec217197b41("skmsg: Schedule psock work if the cached skb exists on the psock")
was intended to address similar race, but had a couple cases it missed.
Most obvious it only accounted for receiving traffic on the local socket
so if redirecting into another socket we could still get an sk_buff stuck
here. Next it missed the case where copied=0 in the recv() handler and
then we wouldn't kick the scheduler. Also its sub-optimal to require
userspace to kick the internal mechanisms of sockmap to wake it up and
copy data to user. It results in an extra syscall and requires the app
to actual handle the EAGAIN correctly.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: William Findlay <will@isovalent.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-3-john.fastabend@gmail.com
Stable-dep-of: 405df89dd5 ("bpf, sockmap: Improved check for empty queue")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 901546fd8f ]
The sockmap code is returning EAGAIN after a FIN packet is received and no
more data is on the receive queue. Correct behavior is to return 0 to the
user and the user can then close the socket. The EAGAIN causes many apps
to retry which masks the problem. Eventually the socket is evicted from
the sockmap because its released from sockmap sock free handling. The
issue creates a delay and can cause some errors on application side.
To fix this check on sk_msg_recvmsg side if length is zero and FIN flag
is set then set return to zero. A selftest will be added to check this
condition.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: William Findlay <will@isovalent.com>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-6-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ea444185a6 ]
A common mechanism to put a TCP socket into the sockmap is to hook the
BPF_SOCK_OPS_{ACTIVE_PASSIVE}_ESTABLISHED_CB event with a BPF program
that can map the socket info to the correct BPF verdict parser. When
the user adds the socket to the map the psock is created and the new
ops are assigned to ensure the verdict program will 'see' the sk_buffs
as they arrive.
Part of this process hooks the sk_data_ready op with a BPF specific
handler to wake up the BPF verdict program when data is ready to read.
The logic is simple enough (posted here for easy reading)
static void sk_psock_verdict_data_ready(struct sock *sk)
{
struct socket *sock = sk->sk_socket;
if (unlikely(!sock || !sock->ops || !sock->ops->read_skb))
return;
sock->ops->read_skb(sk, sk_psock_verdict_recv);
}
The oversight here is sk->sk_socket is not assigned until the application
accepts() the new socket. However, its entirely ok for the peer application
to do a connect() followed immediately by sends. The socket on the receiver
is sitting on the backlog queue of the listening socket until its accepted
and the data is queued up. If the peer never accepts the socket or is slow
it will eventually hit data limits and rate limit the session. But,
important for BPF sockmap hooks when this data is received TCP stack does
the sk_data_ready() call but the read_skb() for this data is never called
because sk_socket is missing. The data sits on the sk_receive_queue.
Then once the socket is accepted if we never receive more data from the
peer there will be no further sk_data_ready calls and all the data
is still on the sk_receive_queue(). Then user calls recvmsg after accept()
and for TCP sockets in sockmap we use the tcp_bpf_recvmsg_parser() handler.
The handler checks for data in the sk_msg ingress queue expecting that
the BPF program has already run from the sk_data_ready hook and enqueued
the data as needed. So we are stuck.
To fix do an unlikely check in recvmsg handler for data on the
sk_receive_queue and if it exists wake up data_ready. We have the sock
locked in both read_skb and recvmsg so should avoid having multiple
runners.
Fixes: 04919bed94 ("tcp: Introduce tcp_read_skb()")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20230523025618.113937-7-john.fastabend@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c2deb2e971 ]
sk->sk_rcvbuf in __sock_queue_rcv_skb() and __sk_receive_skb() can be
changed by other threads. Mark this as benign using READ_ONCE().
This patch is aimed at reducing the number of benign races reported by
KCSAN in order to focus future debugging effort on harmful races.
Signed-off-by: linke li <lilinke99@qq.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ec50f3114e ]
Fix the order of arguments in the TP_ARGS macro
for the rdev_dump_mpp tracepoint event.
Found by Linux Verification Center (linuxtesting.org).
Signed-off-by: Igor Artemiev <Igor.A.Artemiev@mcst.ru>
Link: https://msgid.link/20240311164519.118398-1-Igor.A.Artemiev@mcst.ru
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 774f8841f5 ]
Running kernel-doc on ieee80211_i.h flagged the following:
net/mac80211/ieee80211_i.h:145: warning: expecting prototype for enum ieee80211_corrupt_data_flags. Prototype was for enum ieee80211_bss_corrupt_data_flags instead
net/mac80211/ieee80211_i.h:162: warning: expecting prototype for enum ieee80211_valid_data_flags. Prototype was for enum ieee80211_bss_valid_data_flags instead
Fix these warnings.
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://msgid.link/20240314-kdoc-ieee80211_i-v1-1-72b91b55b257@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5babae777c ]
GRO-GSO path is supposed to be transparent and as such L3 flush checks are
relevant to all UDP flows merging in GRO. This patch uses the same logic
and code from tcp_gro_receive, terminating merge if flush is non zero.
Fixes: e20cf8d3f1 ("udp: implement GRO for plain UDP sockets.")
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 97bf6f81b2 ]
__skb_linearize() doesn't free the skb when it fails, so move
'*buf = NULL' after __skb_linearize(), so that the skb can be
freed on the err path.
Fixes: b7df21cf1b ("tipc: skb_linearize the head skb when reassembling msgs")
Reported-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Link: https://lore.kernel.org/r/90710748c29a1521efac4f75ea01b3b7e61414cf.1714485818.git.lucien.xin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d091e579b8 ]
SKB_GSO_FRAGLIST skbs must not be linearized, otherwise they become
invalid. Return NULL if such an skb is passed to skb_copy or
skb_copy_expand, in order to prevent a crash on a potential later
call to skb_gso_segment.
Fixes: 3a1296a38d ("net: Support GRO/GSO fraglist chaining.")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 59c878cbcd ]
Calling skb_copy on a SKB_GSO_FRAGLIST skb is not valid, since it returns
an invalid linearized skb. This code only needs to change the ethernet
header, so pskb_copy is the right function to call here.
Fixes: 6db6f0eae6 ("bridge: multicast to unicast")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 42f853b428 ]
Drop the flow-hash of the skb when forwarding to the L2TP netdev.
This avoids the L2TP qdisc from using the flow-hash from the outer
packet, which is identical for every flow within the tunnel.
This does not affect every platform but is specific for the ethernet
driver. It depends on the platform including L4 information in the
flow-hash.
One such example is the Mediatek Filogic MT798x family of networking
processors.
Fixes: d9e31d17ce ("l2tp: Add L2TP ethernet pseudowire support")
Acked-by: James Chapman <jchapman@katalix.com>
Signed-off-by: David Bauer <mail@david-bauer.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240424171110.13701-1-mail@david-bauer.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4b911a9690 ]
syzbot triggered various splats (see [0] and links) by a crafted GSO
packet of VIRTIO_NET_HDR_GSO_UDP layering the following protocols:
ETH_P_8021AD + ETH_P_NSH + ETH_P_IPV6 + IPPROTO_UDP
NSH can encapsulate IPv4, IPv6, Ethernet, NSH, and MPLS. As the inner
protocol can be Ethernet, NSH GSO handler, nsh_gso_segment(), calls
skb_mac_gso_segment() to invoke inner protocol GSO handlers.
nsh_gso_segment() does the following for the original skb before
calling skb_mac_gso_segment()
1. reset skb->network_header
2. save the original skb->{mac_heaeder,mac_len} in a local variable
3. pull the NSH header
4. resets skb->mac_header
5. set up skb->mac_len and skb->protocol for the inner protocol.
and does the following for the segmented skb
6. set ntohs(ETH_P_NSH) to skb->protocol
7. push the NSH header
8. restore skb->mac_header
9. set skb->mac_header + mac_len to skb->network_header
10. restore skb->mac_len
There are two problems in 6-7 and 8-9.
(a)
After 6 & 7, skb->data points to the NSH header, so the outer header
(ETH_P_8021AD in this case) is stripped when skb is sent out of netdev.
Also, if NSH is encapsulated by NSH + Ethernet (so NSH-Ethernet-NSH),
skb_pull() in the first nsh_gso_segment() will make skb->data point
to the middle of the outer NSH or Ethernet header because the Ethernet
header is not pulled by the second nsh_gso_segment().
(b)
While restoring skb->{mac_header,network_header} in 8 & 9,
nsh_gso_segment() does not assume that the data in the linear
buffer is shifted.
However, udp6_ufo_fragment() could shift the data and change
skb->mac_header accordingly as demonstrated by syzbot.
If this happens, even the restored skb->mac_header points to
the middle of the outer header.
It seems nsh_gso_segment() has never worked with outer headers so far.
At the end of nsh_gso_segment(), the outer header must be restored for
the segmented skb, instead of the NSH header.
To do that, let's calculate the outer header position relatively from
the inner header and set skb->{data,mac_header,protocol} properly.
[0]:
BUG: KMSAN: uninit-value in ipvlan_process_outbound drivers/net/ipvlan/ipvlan_core.c:524 [inline]
BUG: KMSAN: uninit-value in ipvlan_xmit_mode_l3 drivers/net/ipvlan/ipvlan_core.c:602 [inline]
BUG: KMSAN: uninit-value in ipvlan_queue_xmit+0xf44/0x16b0 drivers/net/ipvlan/ipvlan_core.c:668
ipvlan_process_outbound drivers/net/ipvlan/ipvlan_core.c:524 [inline]
ipvlan_xmit_mode_l3 drivers/net/ipvlan/ipvlan_core.c:602 [inline]
ipvlan_queue_xmit+0xf44/0x16b0 drivers/net/ipvlan/ipvlan_core.c:668
ipvlan_start_xmit+0x5c/0x1a0 drivers/net/ipvlan/ipvlan_main.c:222
__netdev_start_xmit include/linux/netdevice.h:4989 [inline]
netdev_start_xmit include/linux/netdevice.h:5003 [inline]
xmit_one net/core/dev.c:3547 [inline]
dev_hard_start_xmit+0x244/0xa10 net/core/dev.c:3563
__dev_queue_xmit+0x33ed/0x51c0 net/core/dev.c:4351
dev_queue_xmit include/linux/netdevice.h:3171 [inline]
packet_xmit+0x9c/0x6b0 net/packet/af_packet.c:276
packet_snd net/packet/af_packet.c:3081 [inline]
packet_sendmsg+0x8aef/0x9f10 net/packet/af_packet.c:3113
sock_sendmsg_nosec net/socket.c:730 [inline]
__sock_sendmsg net/socket.c:745 [inline]
__sys_sendto+0x735/0xa10 net/socket.c:2191
__do_sys_sendto net/socket.c:2203 [inline]
__se_sys_sendto net/socket.c:2199 [inline]
__x64_sys_sendto+0x125/0x1c0 net/socket.c:2199
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b
Uninit was created at:
slab_post_alloc_hook mm/slub.c:3819 [inline]
slab_alloc_node mm/slub.c:3860 [inline]
__do_kmalloc_node mm/slub.c:3980 [inline]
__kmalloc_node_track_caller+0x705/0x1000 mm/slub.c:4001
kmalloc_reserve+0x249/0x4a0 net/core/skbuff.c:582
__alloc_skb+0x352/0x790 net/core/skbuff.c:651
skb_segment+0x20aa/0x7080 net/core/skbuff.c:4647
udp6_ufo_fragment+0xcab/0x1150 net/ipv6/udp_offload.c:109
ipv6_gso_segment+0x14be/0x2ca0 net/ipv6/ip6_offload.c:152
skb_mac_gso_segment+0x3e8/0x760 net/core/gso.c:53
nsh_gso_segment+0x6f4/0xf70 net/nsh/nsh.c:108
skb_mac_gso_segment+0x3e8/0x760 net/core/gso.c:53
__skb_gso_segment+0x4b0/0x730 net/core/gso.c:124
skb_gso_segment include/net/gso.h:83 [inline]
validate_xmit_skb+0x107f/0x1930 net/core/dev.c:3628
__dev_queue_xmit+0x1f28/0x51c0 net/core/dev.c:4343
dev_queue_xmit include/linux/netdevice.h:3171 [inline]
packet_xmit+0x9c/0x6b0 net/packet/af_packet.c:276
packet_snd net/packet/af_packet.c:3081 [inline]
packet_sendmsg+0x8aef/0x9f10 net/packet/af_packet.c:3113
sock_sendmsg_nosec net/socket.c:730 [inline]
__sock_sendmsg net/socket.c:745 [inline]
__sys_sendto+0x735/0xa10 net/socket.c:2191
__do_sys_sendto net/socket.c:2203 [inline]
__se_sys_sendto net/socket.c:2199 [inline]
__x64_sys_sendto+0x125/0x1c0 net/socket.c:2199
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b
CPU: 1 PID: 5101 Comm: syz-executor421 Not tainted 6.8.0-rc5-syzkaller-00297-gf2e367d6ad3b #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/25/2024
Fixes: c411ed8545 ("nsh: add GSO support")
Reported-and-tested-by: syzbot+42a0dc856239de4de60e@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=42a0dc856239de4de60e
Reported-and-tested-by: syzbot+c298c9f0e46a3c86332b@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=c298c9f0e46a3c86332b
Link: https://lore.kernel.org/netdev/20240415222041.18537-1-kuniyu@amazon.com/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240424023549.21862-1-kuniyu@amazon.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5bcf0dcbf9 ]
When redirecting a packet using XDP, the bpf_redirect_map() helper will set
up the redirect destination information in struct bpf_redirect_info (using
the __bpf_xdp_redirect_map() helper function), and the xdp_do_redirect()
function will read this information after the XDP program returns and pass
the frame on to the right redirect destination.
When using the BPF_F_BROADCAST flag to do multicast redirect to a whole
map, __bpf_xdp_redirect_map() sets the 'map' pointer in struct
bpf_redirect_info to point to the destination map to be broadcast. And
xdp_do_redirect() reacts to the value of this map pointer to decide whether
it's dealing with a broadcast or a single-value redirect. However, if the
destination map is being destroyed before xdp_do_redirect() is called, the
map pointer will be cleared out (by bpf_clear_redirect_map()) without
waiting for any XDP programs to stop running. This causes xdp_do_redirect()
to think that the redirect was to a single target, but the target pointer
is also NULL (since broadcast redirects don't have a single target), so
this causes a crash when a NULL pointer is passed to dev_map_enqueue().
To fix this, change xdp_do_redirect() to react directly to the presence of
the BPF_F_BROADCAST flag in the 'flags' value in struct bpf_redirect_info
to disambiguate between a single-target and a broadcast redirect. And only
read the 'map' pointer if the broadcast flag is set, aborting if that has
been cleared out in the meantime. This prevents the crash, while keeping
the atomic (cmpxchg-based) clearing of the map pointer itself, and without
adding any more checks in the non-broadcast fast path.
Fixes: e624d4ed4a ("xdp: Extend xdp_redirect_map with broadcast support")
Reported-and-tested-by: syzbot+af9492708df9797198d6@syzkaller.appspotmail.com
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Link: https://lore.kernel.org/r/20240418071840.156411-1-toke@redhat.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1372d34ccf ]
Add an xdp_do_redirect_frame() variant which supports pre-computed
xdp_frame structures. This will be used in bpf_prog_run() to avoid having
to write to the xdp_frame structure when the XDP program doesn't modify the
frame boundaries.
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20220103150812.87914-6-toke@redhat.com
Stable-dep-of: 5bcf0dcbf9 ("xdp: use flags field to disambiguate broadcast redirect")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d53ad5d8b2 ]
All map redirect functions except XSK maps convert xdp_buff to xdp_frame
before enqueueing it. So move this conversion of out the map functions
and into xdp_do_redirect(). This removes a bit of duplicated code, but more
importantly it makes it possible to support caller-allocated xdp_frame
structures, which will be added in a subsequent commit.
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20220103150812.87914-5-toke@redhat.com
Stable-dep-of: 5bcf0dcbf9 ("xdp: use flags field to disambiguate broadcast redirect")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2057a48d0d ]
We want to be able to have our rpc stats handled in a per network
namespace manner, so add an option to rpc_create_args to specify a
different rpc_stats struct instead of using the one on the rpc_program.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Stable-dep-of: 24457f1be2 ("nfs: Handle error of rpc_proc_register() in nfs_net_init().")
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 680d11f6e5 upstream.
If "udp_cmsg_send()" returned 0 (i.e. only UDP cmsg),
"connected" should not be set to 0. Otherwise it stops
the connected socket from using the cached route.
Fixes: 2e8de85763 ("udp: add gso segment cmsg")
Signed-off-by: Yick Xie <yick.xie@gmail.com>
Cc: stable@vger.kernel.org
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240418170610.867084-1-yick.xie@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Yick Xie <yick.xie@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1c4e97dd2d upstream.
inet_twsk_purge() uses rcu to find TIME_WAIT and NEW_SYN_RECV
objects to purge.
These objects use SLAB_TYPESAFE_BY_RCU semantic and need special
care. We need to use refcount_inc_not_zero(&sk->sk_refcnt).
Reuse the existing correct logic I wrote for TIME_WAIT,
because both structures have common locations for
sk_state, sk_family, and netns pointer.
If after the refcount_inc_not_zero() the object fields longer match
the keys, use sock_gen_put(sk) to release the refcount.
Then we can call inet_twsk_deschedule_put() for TIME_WAIT,
inet_csk_reqsk_queue_drop_and_put() for NEW_SYN_RECV sockets,
with BH disabled.
Then we need to restart the loop because we had drop rcu_read_lock().
Fixes: 740ea3c4a0 ("tcp: Clean up kernel listener's reqsk in inet_twsk_purge()")
Link: https://lore.kernel.org/netdev/CANn89iLvFuuihCtt9PME2uS1WJATnf5fKjDToa1WzVnRzHnPfg@mail.gmail.com/T/#u
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240308200122.64357-2-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 740ea3c4a0 upstream.
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>
[shaozhengchao: resolved conflicts in 5.15]
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9bf4e919cc upstream.
After an innocuous optimization change in LLVM main (19.0.0), x86_64
allmodconfig (which enables CONFIG_KCSAN / -fsanitize=thread) fails to
build due to the checks in check_copy_size():
In file included from net/bluetooth/sco.c:27:
In file included from include/linux/module.h:13:
In file included from include/linux/stat.h:19:
In file included from include/linux/time.h:60:
In file included from include/linux/time32.h:13:
In file included from include/linux/timex.h:67:
In file included from arch/x86/include/asm/timex.h:6:
In file included from arch/x86/include/asm/tsc.h:10:
In file included from arch/x86/include/asm/msr.h:15:
In file included from include/linux/percpu.h:7:
In file included from include/linux/smp.h:118:
include/linux/thread_info.h:244:4: error: call to '__bad_copy_from'
declared with 'error' attribute: copy source size is too small
244 | __bad_copy_from();
| ^
The same exact error occurs in l2cap_sock.c. The copy_to_user()
statements that are failing come from l2cap_sock_getsockopt_old() and
sco_sock_getsockopt_old(). This does not occur with GCC with or without
KCSAN or Clang without KCSAN enabled.
len is defined as an 'int' because it is assigned from
'__user int *optlen'. However, it is clamped against the result of
sizeof(), which has a type of 'size_t' ('unsigned long' for 64-bit
platforms). This is done with min_t() because min() requires compatible
types, which results in both len and the result of sizeof() being casted
to 'unsigned int', meaning len changes signs and the result of sizeof()
is truncated. From there, len is passed to copy_to_user(), which has a
third parameter type of 'unsigned long', so it is widened and changes
signs again. This excessive casting in combination with the KCSAN
instrumentation causes LLVM to fail to eliminate the __bad_copy_from()
call, failing the build.
The official recommendation from LLVM developers is to consistently use
long types for all size variables to avoid the unnecessary casting in
the first place. Change the type of len to size_t in both
l2cap_sock_getsockopt_old() and sco_sock_getsockopt_old(). This clears
up the error while allowing min_t() to be replaced with min(), resulting
in simpler code with no casts and fewer implicit conversions. While len
is a different type than optlen now, it should result in no functional
change because the result of sizeof() will clamp all values of optlen in
the same manner as before.
Cc: stable@vger.kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/2007
Link: https://github.com/llvm/llvm-project/issues/85647
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Justin Stitt <justinstitt@google.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 5ea7b72d4f ]
Since kfree_rcu, which is called in the hlist_for_each_entry_rcu traversal
of ovs_ct_limit_exit, is not part of the RCU read critical section, it
is possible that the RCU grace period will pass during the traversal and
the key will be free.
To prevent this, it should be changed to hlist_for_each_entry_safe.
Fixes: 11efd5cb04 ("openvswitch: Support conntrack zone limit")
Signed-off-by: Hyunwoo Kim <v4bel@theori.io>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Link: https://lore.kernel.org/r/ZiYvzQN/Ry5oeFQW@v4bel-B760M-AORUS-ELITE-AX
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e10d3ba4d4 ]
It was observed in the wild that pairs of consecutive packets would leave
the IPVS with the same wrong checksum, and the issue only went away when
disabling GSO.
IPVS needs to avoid computing the SCTP checksum when using GSO.
Fixes: 90017accff ("sctp: Add GSO support")
Co-developed-by: Firo Yang <firo.yang@suse.com>
Signed-off-by: Ismael Luceno <iluceno@suse.de>
Tested-by: Andreas Taschner <andreas.taschner@suse.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4fd1edcdf1 ]
br_info_notify is a void function. There is no need to return.
Fixes: b6d0425b81 ("bridge: cfm: Netlink Notifications.")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Eric Dumazet