platform/kernel/linux-rpi.git
2 years agorxrpc: Split the receive code
David Howells [Thu, 23 Jan 2020 13:01:33 +0000 (13:01 +0000)]
rxrpc: Split the receive code

Split the code that handles packet reception in softirq mode as a prelude
to moving all the packet processing beyond routing to the appropriate call
and setting up of a new call out into process context.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Don't hold a ref for connection workqueue
David Howells [Fri, 25 Nov 2022 12:43:50 +0000 (12:43 +0000)]
rxrpc: Don't hold a ref for connection workqueue

Currently, rxrpc gives the connection's work item a ref on the connection
when it queues it - and this is called from the timer expiration function.
The problem comes when queue_work() fails (ie. the work item is already
queued): the timer routine must put the ref - but this may cause the
cleanup code to run.

This has the unfortunate effect that the cleanup code may then be run in
softirq context - which means that any spinlocks it might need to touch
have to be guarded to disable softirqs (ie. they need a "_bh" suffix).

 (1) Don't give a ref to the work item.

 (2) Simplify handling of service connections by adding a separate active
     count so that the refcount isn't also used for this.

 (3) Connection destruction for both client and service connections can
     then be cleaned up by putting rxrpc_put_connection() out of line and
     making a tidy progression through the destruction code (offloaded to a
     workqueue if put from softirq or processor function context).  The RCU
     part of the cleanup then only deals with the freeing at the end.

 (4) Make rxrpc_queue_conn() return immediately if it sees the active count
     is -1 rather then queuing the connection.

 (5) Make sure that the cleanup routine waits for the work item to
     complete.

 (6) Stash the rxrpc_net pointer in the conn struct so that the rcu free
     routine can use it, even if the local endpoint has been freed.

Unfortunately, neither the timer nor the work item can simply get around
the problem by just using refcount_inc_not_zero() as the waits would still
have to be done, and there would still be the possibility of having to put
the ref in the expiration function.

Note the connection work item is mostly going to go away with the main
event work being transferred to the I/O thread, so the wait in (6) will
become obsolete.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Don't hold a ref for call timer or workqueue
David Howells [Fri, 25 Nov 2022 09:00:55 +0000 (09:00 +0000)]
rxrpc: Don't hold a ref for call timer or workqueue

Currently, rxrpc gives the call timer a ref on the call when it starts it
and this is passed along to the workqueue by the timer expiration function.
The problem comes when queue_work() fails (ie. the work item is already
queued): the timer routine must put the ref - but this may cause the
cleanup code to run.

This has the unfortunate effect that the cleanup code may then be run in
softirq context - which means that any spinlocks it might need to touch
have to be guarded to disable softirqs (ie. they need a "_bh" suffix).

Fix this by:

 (1) Don't give a ref to the timer.

 (2) Making the expiration function not do anything if the refcount is 0.
     Note that this is more of an optimisation.

 (3) Make sure that the cleanup routine waits for timer to complete.

However, this has a consequence that timer cannot give a ref to the work
item.  Therefore the following fixes are also necessary:

 (4) Don't give a ref to the work item.

 (5) Make the work item return asap if it sees the ref count is 0.

 (6) Make sure that the cleanup routine waits for the work item to
     complete.

Unfortunately, neither the timer nor the work item can simply get around
the problem by just using refcount_inc_not_zero() as the waits would still
have to be done, and there would still be the possibility of having to put
the ref in the expiration function.

Note the call work item is going to go away with the work being transferred
to the I/O thread, so the wait in (6) will become obsolete.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: trace: Don't use __builtin_return_address for sk_buff tracing
David Howells [Fri, 21 Oct 2022 14:31:21 +0000 (15:31 +0100)]
rxrpc: trace: Don't use __builtin_return_address for sk_buff tracing

In rxrpc tracing, use enums to generate lists of points of interest rather
than __builtin_return_address() for the sk_buff tracepoint.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Trace rxrpc_bundle refcount
David Howells [Wed, 2 Nov 2022 21:54:46 +0000 (21:54 +0000)]
rxrpc: Trace rxrpc_bundle refcount

Add a tracepoint for the rxrpc_bundle refcounting.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: trace: Don't use __builtin_return_address for rxrpc_call tracing
David Howells [Fri, 21 Oct 2022 13:39:26 +0000 (14:39 +0100)]
rxrpc: trace: Don't use __builtin_return_address for rxrpc_call tracing

In rxrpc tracing, use enums to generate lists of points of interest rather
than __builtin_return_address() for the rxrpc_call tracepoint

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: trace: Don't use __builtin_return_address for rxrpc_conn tracing
David Howells [Fri, 21 Oct 2022 13:06:16 +0000 (14:06 +0100)]
rxrpc: trace: Don't use __builtin_return_address for rxrpc_conn tracing

In rxrpc tracing, use enums to generate lists of points of interest rather
than __builtin_return_address() for the rxrpc_conn tracepoint

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: trace: Don't use __builtin_return_address for rxrpc_peer tracing
David Howells [Fri, 21 Oct 2022 12:39:34 +0000 (13:39 +0100)]
rxrpc: trace: Don't use __builtin_return_address for rxrpc_peer tracing

In rxrpc tracing, use enums to generate lists of points of interest rather
than __builtin_return_address() for the rxrpc_peer tracepoint

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: trace: Don't use __builtin_return_address for rxrpc_local tracing
David Howells [Fri, 21 Oct 2022 12:00:34 +0000 (13:00 +0100)]
rxrpc: trace: Don't use __builtin_return_address for rxrpc_local tracing

In rxrpc tracing, use enums to generate lists of points of interest rather
than __builtin_return_address() for the rxrpc_local tracepoint

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Extract the code from a received ABORT packet much earlier
David Howells [Mon, 14 Nov 2022 12:21:32 +0000 (12:21 +0000)]
rxrpc: Extract the code from a received ABORT packet much earlier

Extract the code from a received rx ABORT packet much earlier and in a
single place and harmonise the responses to malformed ABORT packets.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Drop rxrpc_conn_parameters from rxrpc_connection and rxrpc_bundle
David Howells [Wed, 19 Oct 2022 12:49:02 +0000 (13:49 +0100)]
rxrpc: Drop rxrpc_conn_parameters from rxrpc_connection and rxrpc_bundle

Remove the rxrpc_conn_parameters struct from the rxrpc_connection and
rxrpc_bundle structs and emplace the members directly.  These are going to
get filled in from the rxrpc_call struct in future.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Remove the [_k]net() debugging macros
David Howells [Thu, 20 Oct 2022 11:04:20 +0000 (12:04 +0100)]
rxrpc: Remove the [_k]net() debugging macros

Remove the _net() and knet() debugging macros in favour of tracepoints.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Remove the [k_]proto() debugging macros
David Howells [Thu, 20 Oct 2022 10:51:06 +0000 (11:51 +0100)]
rxrpc: Remove the [k_]proto() debugging macros

Remove the kproto() and _proto() debugging macros in preference to using
tracepoints for this.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Remove handling of duplicate packets in recvmsg_queue
David Howells [Fri, 11 Nov 2022 09:41:34 +0000 (09:41 +0000)]
rxrpc: Remove handling of duplicate packets in recvmsg_queue

We should not now see duplicate packets in the recvmsg_queue.  At one
point, jumbo packets that overlapped with already queued data would be
added to the queue and dealt with in recvmsg rather than in the softirq
input code, but now jumbo packets are split/cloned before being processed
by the input code and the subpackets can be discarded individually.

So remove the recvmsg-side code for handling this.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Remove decl for rxrpc_kernel_call_is_complete()
David Howells [Thu, 17 Nov 2022 10:04:15 +0000 (10:04 +0000)]
rxrpc: Remove decl for rxrpc_kernel_call_is_complete()

rxrpc_kernel_call_is_complete() has been removed, so remove its declaration
too.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Fix call leak
David Howells [Mon, 28 Nov 2022 09:24:06 +0000 (09:24 +0000)]
rxrpc: Fix call leak

When retransmitting a packet, rxrpc_resend() shouldn't be attaching a ref
to the call to the txbuf as that pins the call and prevents the call from
clearing the packet buffer.

Signed-off-by: David Howells <dhowells@redhat.com>
Fixes: d57a3a151660 ("rxrpc: Save last ACK's SACK table rather than marking txbufs")
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agorxrpc: Implement an in-kernel rxperf server for testing purposes
David Howells [Thu, 3 Nov 2022 22:27:52 +0000 (22:27 +0000)]
rxrpc: Implement an in-kernel rxperf server for testing purposes

Implement an in-kernel rxperf server to allow kernel-based rxrpc services
to be tested directly, unlike with AFS where they're accessed by the
fileserver when the latter decides it wants to.

This is implemented as a module that, if loaded, opens UDP port 7009
(afs3-rmtsys) and listens on it for incoming calls.  Calls can be generated
using the rxperf command shipped with OpenAFS, for example.

Changes
=======
ver #2)
 - Use min_t() instead of min().

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: Jakub Kicinski <kuba@kernel.org>

2 years agorxrpc: Fix checker warning
David Howells [Thu, 1 Dec 2022 12:24:36 +0000 (12:24 +0000)]
rxrpc: Fix checker warning

Fix the following checker warning:

../net/rxrpc/key.c:692:9: error: subtraction of different types can't work (different address spaces)

Checker is wrong in this case, but cast the pointers to unsigned long to
avoid the warning.

Whilst we're at it, reduce the assertions to WARN_ON() and return an error.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org

2 years agoselftests/net: add csum offload test
Willem de Bruijn [Mon, 28 Nov 2022 14:02:10 +0000 (09:02 -0500)]
selftests/net: add csum offload test

Test NIC hardware checksum offload:

- Rx + Tx
- IPv4 + IPv6
- TCP + UDP

Optional features:

- zero checksum 0xFFFF
- checksum disable 0x0000
- transport encap headers
- randomization

See file header for detailed comments.

Expected results differ depending on NIC features:

- CHECKSUM_UNNECESSARY vs CHECKSUM_COMPLETE
- NETIF_F_HW_CSUM (csum_start/csum_off) vs NETIF_F_IP(V6)_CSUM

Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20221128140210.553391-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec...
Jakub Kicinski [Wed, 30 Nov 2022 04:50:50 +0000 (20:50 -0800)]
Merge branch 'master' of git://git./linux/kernel/git/klassert/ipsec-next

Steffen Klassert says:

====================
ipsec-next 2022-11-26

1) Remove redundant variable in esp6.
   From Colin Ian King.

2) Update x->lastused for every packet. It was used only for
   outgoing mobile IPv6 packets, but showed to be usefull
   to check if the a SA is still in use in general.
   From Antony Antony.

3) Remove unused variable in xfrm_byidx_resize.
   From Leon Romanovsky.

4) Finalize extack support for xfrm.
   From Sabrina Dubroca.

* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next:
  xfrm: add extack to xfrm_set_spdinfo
  xfrm: add extack to xfrm_alloc_userspi
  xfrm: add extack to xfrm_do_migrate
  xfrm: add extack to xfrm_new_ae and xfrm_replay_verify_len
  xfrm: add extack to xfrm_del_sa
  xfrm: add extack to xfrm_add_sa_expire
  xfrm: a few coding style clean ups
  xfrm: Remove not-used total variable
  xfrm: update x->lastused for every packet
  esp6: remove redundant variable err
====================

Link: https://lore.kernel.org/r/20221126110303.1859238-1-steffen.klassert@secunet.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge branch 'net-pcs-altera-tse-simplify-and-clean-up-the-driver'
Jakub Kicinski [Wed, 30 Nov 2022 04:29:58 +0000 (20:29 -0800)]
Merge branch 'net-pcs-altera-tse-simplify-and-clean-up-the-driver'

Maxime Chevallier says:

====================
net: pcs: altera-tse: simplify and clean-up the driver

This small series does a bit of code cleanup in the altera TSE pcs
driver, removing unused register definitions, handling 1000BaseX speed
configuration correctly according to the datasheet, and making use of
proper poll_timeout helpers.

No functional change is introduced.
====================

Link: https://lore.kernel.org/r/20221125131801.64234-1-maxime.chevallier@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: pcs: altera-tse: remove unnecessary register definitions
Maxime Chevallier [Fri, 25 Nov 2022 13:18:01 +0000 (14:18 +0100)]
net: pcs: altera-tse: remove unnecessary register definitions

remove unused register definitions, left from the split with the
altera-tse mac driver.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: pcs: altera-tse: don't set the speed for 1000BaseX
Maxime Chevallier [Fri, 25 Nov 2022 13:18:00 +0000 (14:18 +0100)]
net: pcs: altera-tse: don't set the speed for 1000BaseX

When disabling the SGMII mode bit, the PCS defaults to 1000BaseX mode.
In that mode, we don't need to set the speed since it's always 1000Mbps.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: pcs: altera-tse: use read_poll_timeout to wait for reset
Maxime Chevallier [Fri, 25 Nov 2022 13:17:59 +0000 (14:17 +0100)]
net: pcs: altera-tse: use read_poll_timeout to wait for reset

Software resets on the TSE PCS don't clear registers, but rather reset
all internal state machines regarding AN, comma detection and
encoding/decoding. Use read_poll_timeout to wait for the reset to clear
instead of manually polling the register.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge branch 'mptcp-msg_fastopen-and-tfo-listener-side-support'
Jakub Kicinski [Wed, 30 Nov 2022 04:24:31 +0000 (20:24 -0800)]
Merge branch 'mptcp-msg_fastopen-and-tfo-listener-side-support'

Matthieu Baerts says:

====================
mptcp: MSG_FASTOPEN and TFO listener side support

Before this series, only the initiator of a connection was able to combine
both TCP FastOpen and MPTCP when using TCP_FASTOPEN_CONNECT socket option.

These new patches here add (in theory) the full support of TFO with MPTCP,
which means:

 - MSG_FASTOPEN sendmsg flag support (patch 1/8)
 - TFO support for the listener side (patches 2-5/8)
 - TCP_FASTOPEN socket option (patch 6/8)
 - TCP_FASTOPEN_KEY socket option (patch 7/8)

To support TFO for the server side, a few preparation patches are needed
(patches 2 to 5/8). Some of them were inspired by a previous work from
Benjamin Hesmans.

Note that TFO support with MPTCP has been validated with selftests
(patch 8/8) but also with Packetdrill tests running with a modified
but still very WIP version supporting MPTCP. Both the modified tool
and the tests are available online:

  https://github.com/multipath-tcp/packetdrill/
====================

Link: https://lore.kernel.org/r/20221125222958.958636-1-matthieu.baerts@tessares.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoselftests: mptcp: mptfo Initiator/Listener
Dmytro Shytyi [Fri, 25 Nov 2022 22:29:54 +0000 (23:29 +0100)]
selftests: mptcp: mptfo Initiator/Listener

This patch first adds TFO support in mptcp_connect.c.

This can be enabled via a new option: -o MPTFO.

Once enabled, the TCP_FASTOPEN socket option is enabled for the server
side and a sendto() with MSG_FASTOPEN is used instead of a connect() for
the client side.

Note that the first SYN has a limit of bytes it can carry. In other
words, it is allowed to send less data than the provided one. We then
need to track more status info to properly allow the next sendmsg()
starting from the next part of the data to send the rest.

Also in TFO scenarios, we need to completely spool the partially xmitted
buffer -- and account for that -- before starting sendfile/mmap xmit,
otherwise the relevant tests will fail.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Dmytro Shytyi <dmytro@shytyi.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: add support for TCP_FASTOPEN_KEY sockopt
Matthieu Baerts [Fri, 25 Nov 2022 22:29:53 +0000 (23:29 +0100)]
mptcp: add support for TCP_FASTOPEN_KEY sockopt

The goal of this socket option is to set different keys per listener,
see commit 1fba70e5b6be ("tcp: socket option to set TCP fast open key")
for more details about this socket option.

The only thing to do here with MPTCP is to relay the request to the
first subflow like it is already done for the other TCP_FASTOPEN* socket
options.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: add TCP_FASTOPEN sock option
Dmytro Shytyi [Fri, 25 Nov 2022 22:29:52 +0000 (23:29 +0100)]
mptcp: add TCP_FASTOPEN sock option

The TCP_FASTOPEN socket option is one way for the application to tell
the kernel TFO support has to be enabled for the listener socket.

The only thing to do here with MPTCP is to relay the request to the
first subflow like it is already done for the other TCP_FASTOPEN* socket
options.

Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Dmytro Shytyi <dmytro@shytyi.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: add subflow_v(4,6)_send_synack()
Dmytro Shytyi [Fri, 25 Nov 2022 22:29:51 +0000 (23:29 +0100)]
mptcp: add subflow_v(4,6)_send_synack()

The send_synack() needs to be overridden for MPTCP to support TFO for
two reasons:

- There is not be enough space in the TCP options if the TFO cookie has
  to be added in the SYN+ACK with other options: MSS (4), SACK OK (2),
  Timestamps (10), Window Scale (3+1), TFO (10+2), MP_CAPABLE (12).
  MPTCPv1 specs -- RFC 8684, section B.1 [1] -- suggest to drop the TCP
  timestamps option in this case.

- The data received in the SYN has to be handled: the SKB can be
  dequeued from the subflow sk and transferred to the MPTCP sk. Counters
  need to be updated accordingly and the application can be notified at
  the end because some bytes have been received.

[1] https://www.rfc-editor.org/rfc/rfc8684.html#section-b.1

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Dmytro Shytyi <dmytro@shytyi.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: implement delayed seq generation for passive fastopen
Dmytro Shytyi [Fri, 25 Nov 2022 22:29:50 +0000 (23:29 +0100)]
mptcp: implement delayed seq generation for passive fastopen

With fastopen in place, the first subflow socket is created before the
MPC handshake completes, and we need to properly initialize the sequence
numbers at MPC ACK reception.

Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Dmytro Shytyi <dmytro@shytyi.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: consolidate initial ack seq generation
Paolo Abeni [Fri, 25 Nov 2022 22:29:49 +0000 (23:29 +0100)]
mptcp: consolidate initial ack seq generation

Currently the initial ack sequence is generated on demand whenever
it's requested and the remote key is handy. The relevant code is
scattered in different places and can lead to multiple, unneeded,
crypto operations.

This change consolidates the ack sequence generation code in a single
helper, storing the sequence number at the subflow level.

The above additionally saves a few conditional in fast-path and will
simplify the upcoming fast-open implementation.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: track accurately the incoming MPC suboption type
Paolo Abeni [Fri, 25 Nov 2022 22:29:48 +0000 (23:29 +0100)]
mptcp: track accurately the incoming MPC suboption type

Currently in the receive path we don't need to discriminate
between MPC SYN, MPC SYN-ACK and MPC ACK, but soon the fastopen
code will need that info to properly track the fully established
status.

Track the exact MPC suboption type into the receive opt bitmap.
No functional change intended.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: add MSG_FASTOPEN sendmsg flag support
Dmytro Shytyi [Fri, 25 Nov 2022 22:29:47 +0000 (23:29 +0100)]
mptcp: add MSG_FASTOPEN sendmsg flag support

Since commit 54f1944ed6d2 ("mptcp: factor out mptcp_connect()"), all the
infrastructure is now in place to support the MSG_FASTOPEN flag, we
just need to call into the fastopen path in mptcp_sendmsg().

Co-developed-by: Benjamin Hesmans <benjamin.hesmans@tessares.net>
Signed-off-by: Benjamin Hesmans <benjamin.hesmans@tessares.net>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Dmytro Shytyi <dmytro@shytyi.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Jakub Kicinski [Tue, 29 Nov 2022 21:04:52 +0000 (13:04 -0800)]
Merge git://git./linux/kernel/git/netdev/net

tools/lib/bpf/ringbuf.c
  927cbb478adf ("libbpf: Handle size overflow for ringbuf mmap")
  b486d19a0ab0 ("libbpf: checkpatch: Fixed code alignments in ringbuf.c")
https://lore.kernel.org/all/20221121122707.44d1446a@canb.auug.org.au/

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge tag 'net-6.1-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Tue, 29 Nov 2022 17:52:10 +0000 (09:52 -0800)]
Merge tag 'net-6.1-rc8-2' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from bpf, can and wifi.

  Current release - new code bugs:

   - eth: mlx5e:
      - use kvfree() in mlx5e_accel_fs_tcp_create()
      - MACsec, fix RX data path 16 RX security channel limit
      - MACsec, fix memory leak when MACsec device is deleted
      - MACsec, fix update Rx secure channel active field
      - MACsec, fix add Rx security association (SA) rule memory leak

  Previous releases - regressions:

   - wifi: cfg80211: don't allow multi-BSSID in S1G

   - stmmac: set MAC's flow control register to reflect current settings

   - eth: mlx5:
      - E-switch, fix duplicate lag creation
      - fix use-after-free when reverting termination table

  Previous releases - always broken:

   - ipv4: fix route deletion when nexthop info is not specified

   - bpf: fix a local storage BPF map bug where the value's spin lock
     field can get initialized incorrectly

   - tipc: re-fetch skb cb after tipc_msg_validate

   - wifi: wilc1000: fix Information Element parsing

   - packet: do not set TP_STATUS_CSUM_VALID on CHECKSUM_COMPLETE

   - sctp: fix memory leak in sctp_stream_outq_migrate()

   - can: can327: fix potential skb leak when netdev is down

   - can: add number of missing netdev freeing on error paths

   - aquantia: do not purge addresses when setting the number of rings

   - wwan: iosm:
      - fix incorrect skb length leading to truncated packet
      - fix crash in peek throughput test due to skb UAF"

* tag 'net-6.1-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (79 commits)
  net: ethernet: renesas: ravb: Fix promiscuous mode after system resumed
  MAINTAINERS: Update maintainer list for chelsio drivers
  ionic: update MAINTAINERS entry
  sctp: fix memory leak in sctp_stream_outq_migrate()
  packet: do not set TP_STATUS_CSUM_VALID on CHECKSUM_COMPLETE
  net/mlx5: Lag, Fix for loop when checking lag
  Revert "net/mlx5e: MACsec, remove replay window size limitation in offload path"
  net: marvell: prestera: Fix a NULL vs IS_ERR() check in some functions
  net: tun: Fix use-after-free in tun_detach()
  net: mdiobus: fix unbalanced node reference count
  net: hsr: Fix potential use-after-free
  tipc: re-fetch skb cb after tipc_msg_validate
  mptcp: fix sleep in atomic at close time
  mptcp: don't orphan ssk in mptcp_close()
  dsa: lan9303: Correct stat name
  ipv4: Fix route deletion when nexthop info is not specified
  net: wwan: iosm: fix incorrect skb length
  net: wwan: iosm: fix crash in peek throughput test
  net: wwan: iosm: fix dma_alloc_coherent incompatible pointer type
  net: wwan: iosm: fix kernel test robot reported error
  ...

2 years agoudp_tunnel: Add checks for nla_nest_start() in __udp_tunnel_nic_dump_write()
Yuan Can [Tue, 29 Nov 2022 01:39:34 +0000 (01:39 +0000)]
udp_tunnel: Add checks for nla_nest_start() in __udp_tunnel_nic_dump_write()

As the nla_nest_start() may fail with NULL returned, the return value
should be checked.

Note that this is not a real bug, nothing will break here.
The next nla_put() will fail as well and we'll bail (and
nla_nest_cancel() can handle NULL). But we keep getting
those "fixes" so whatever.

Signed-off-by: Yuan Can <yuancan@huawei.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20221129013934.55184-1-yuancan@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: ethernet: renesas: ravb: Fix promiscuous mode after system resumed
Yoshihiro Shimoda [Mon, 28 Nov 2022 06:56:04 +0000 (15:56 +0900)]
net: ethernet: renesas: ravb: Fix promiscuous mode after system resumed

After system resumed on some environment board, the promiscuous mode
is disabled because the SoC turned off. So, call ravb_set_rx_mode() in
the ravb_resume() to fix the issue.

Reported-by: Tho Vu <tho.vu.wh@renesas.com>
Fixes: 0184165b2f42 ("ravb: add sleep PM suspend/resume support")
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://lore.kernel.org/r/20221128065604.1864391-1-yoshihiro.shimoda.uh@renesas.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMAINTAINERS: Update maintainer list for chelsio drivers
Ayush Sawal [Mon, 28 Nov 2022 23:13:48 +0000 (04:43 +0530)]
MAINTAINERS: Update maintainer list for chelsio drivers

This updates the maintainers for chelsio inline crypto drivers and
chelsio crypto drivers.

Signed-off-by: Ayush Sawal <ayush.sawal@chelsio.com>
Link: https://lore.kernel.org/r/20221128231348.8225-1-ayush.sawal@chelsio.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoionic: update MAINTAINERS entry
Shannon Nelson [Tue, 29 Nov 2022 01:17:34 +0000 (17:17 -0800)]
ionic: update MAINTAINERS entry

Now that Pensando is a part of AMD we need to update
a couple of addresses.  We're keeping the mailing list
address for the moment, but that will likely change in
the near future.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Link: https://lore.kernel.org/r/20221129011734.20849-1-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agosctp: fix memory leak in sctp_stream_outq_migrate()
Zhengchao Shao [Sat, 26 Nov 2022 03:17:20 +0000 (11:17 +0800)]
sctp: fix memory leak in sctp_stream_outq_migrate()

When sctp_stream_outq_migrate() is called to release stream out resources,
the memory pointed to by prio_head in stream out is not released.

The memory leak information is as follows:
 unreferenced object 0xffff88801fe79f80 (size 64):
   comm "sctp_repo", pid 7957, jiffies 4294951704 (age 36.480s)
   hex dump (first 32 bytes):
     80 9f e7 1f 80 88 ff ff 80 9f e7 1f 80 88 ff ff  ................
     90 9f e7 1f 80 88 ff ff 90 9f e7 1f 80 88 ff ff  ................
   backtrace:
     [<ffffffff81b215c6>] kmalloc_trace+0x26/0x60
     [<ffffffff88ae517c>] sctp_sched_prio_set+0x4cc/0x770
     [<ffffffff88ad64f2>] sctp_stream_init_ext+0xd2/0x1b0
     [<ffffffff88aa2604>] sctp_sendmsg_to_asoc+0x1614/0x1a30
     [<ffffffff88ab7ff1>] sctp_sendmsg+0xda1/0x1ef0
     [<ffffffff87f765ed>] inet_sendmsg+0x9d/0xe0
     [<ffffffff8754b5b3>] sock_sendmsg+0xd3/0x120
     [<ffffffff8755446a>] __sys_sendto+0x23a/0x340
     [<ffffffff87554651>] __x64_sys_sendto+0xe1/0x1b0
     [<ffffffff89978b49>] do_syscall_64+0x39/0xb0
     [<ffffffff89a0008b>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

Link: https://syzkaller.appspot.com/bug?exrid=29c402e56c4760763cc0
Fixes: 637784ade221 ("sctp: introduce priority based stream scheduler")
Reported-by: syzbot+29c402e56c4760763cc0@syzkaller.appspotmail.com
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
Link: https://lore.kernel.org/r/20221126031720.378562-1-shaozhengchao@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agopacket: do not set TP_STATUS_CSUM_VALID on CHECKSUM_COMPLETE
Willem de Bruijn [Mon, 28 Nov 2022 16:18:12 +0000 (11:18 -0500)]
packet: do not set TP_STATUS_CSUM_VALID on CHECKSUM_COMPLETE

CHECKSUM_COMPLETE signals that skb->csum stores the sum over the
entire packet. It does not imply that an embedded l4 checksum
field has been validated.

Fixes: 682f048bd494 ("af_packet: pass checksum validation status to the user")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20221128161812.640098-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet/mlx5: Lag, Fix for loop when checking lag
Chris Mi [Tue, 29 Nov 2022 09:30:06 +0000 (01:30 -0800)]
net/mlx5: Lag, Fix for loop when checking lag

The cited commit adds a for loop to check if each port supports lag
or not. But dev is not initialized correctly. Fix it by initializing
dev for each iteration.

Fixes: e87c6a832f88 ("net/mlx5: E-switch, Fix duplicate lag creation")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reported-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://lore.kernel.org/r/20221129093006.378840-2-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoRevert "net/mlx5e: MACsec, remove replay window size limitation in offload path"
Saeed Mahameed [Tue, 29 Nov 2022 09:30:05 +0000 (01:30 -0800)]
Revert "net/mlx5e: MACsec, remove replay window size limitation in offload path"

This reverts commit c0071be0e16c461680d87b763ba1ee5e46548fde.

The cited commit removed the validity checks which initialized the
window_sz and never removed the use of the now uninitialized variable,
so now we are left with wrong value in the window size and the following
clang warning: [-Wuninitialized]
drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c:232:45:
       warning: variable 'window_sz' is uninitialized when used here
       MLX5_SET(macsec_aso, aso_ctx, window_size, window_sz);

Revet at this time to address the clang issue due to lack of time to
test the proper solution.

Fixes: c0071be0e16c ("net/mlx5e: MACsec, remove replay window size limitation in offload path")
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reported-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20221129093006.378840-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonfp: ethtool: support reporting link modes
Yu Xiao [Fri, 25 Nov 2022 11:30:30 +0000 (12:30 +0100)]
nfp: ethtool: support reporting link modes

Add support for reporting link modes,
including `Supported link modes` and `Advertised link modes`,
via ethtool $DEV.

A new command `SPCODE_READ_MEDIA` is added to read info from
management firmware. Also, the mapping table `nfp_eth_media_table`
associates the link modes between NFP and kernel. Both of them
help to support this ability.

Signed-off-by: Yu Xiao <yu.xiao@corigine.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/r/20221125113030.141642-1-simon.horman@corigine.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: devlink: add WARN_ON_ONCE to check return value of unregister_netdevice_notifier...
Jiri Pirko [Fri, 25 Nov 2022 10:02:55 +0000 (11:02 +0100)]
net: devlink: add WARN_ON_ONCE to check return value of unregister_netdevice_notifier_net() call

As the return value is not 0 only in case there is no such notifier
block registered, add a WARN_ON_ONCE() to yell about it.

Suggested-by: Ido Schimmel <idosch@idosch.org>
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://lore.kernel.org/r/20221125100255.1786741-1-jiri@resnulli.us
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoMerge branch 'add-support-for-lan966x-is2-vcap'
Paolo Abeni [Tue, 29 Nov 2022 12:08:26 +0000 (13:08 +0100)]
Merge branch 'add-support-for-lan966x-is2-vcap'

Horatiu Vultur says:

====================
Add support for lan966x IS2 VCAP

This provides initial support for lan966x for 'tc' traffic control
userspace tool and its flower filter. For this is required to use
the VCAP library.

Currently supported flower filter keys and actions are:
- source and destination MAC address keys
- trap action
====================

Link: https://lore.kernel.org/r/20221125095010.124458-1-horatiu.vultur@microchip.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: microchip: vcap: Implement w32be
Horatiu Vultur [Fri, 25 Nov 2022 09:50:10 +0000 (10:50 +0100)]
net: microchip: vcap: Implement w32be

On lan966x the layout of the vcap memory is different than on sparx5.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: Add port keyset config and callback interface
Horatiu Vultur [Fri, 25 Nov 2022 09:50:09 +0000 (10:50 +0100)]
net: lan966x: Add port keyset config and callback interface

Implement vcap_operations and enable default port keyset configuration
for each port. Now it is possible actually write/read/move entries in
the VCAP.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: add tc matchall goto action
Horatiu Vultur [Fri, 25 Nov 2022 09:50:08 +0000 (10:50 +0100)]
net: lan966x: add tc matchall goto action

Extend matchall with action goto. This is needed to enable the lookup in
the VCAP. It is needed to connect chain 0 to a chain that is recognized
by the HW.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: add tc flower support for VCAP API
Horatiu Vultur [Fri, 25 Nov 2022 09:50:07 +0000 (10:50 +0100)]
net: lan966x: add tc flower support for VCAP API

Currently the only supported action is ACTION_TRAP and the only
dissector is ETH_ADDRS. Others will be added in future patches.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: add vcap registers
Horatiu Vultur [Fri, 25 Nov 2022 09:50:06 +0000 (10:50 +0100)]
net: lan966x: add vcap registers

Add registers used to access vcap controller.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: Add is2 vcap model to vcap API.
Horatiu Vultur [Fri, 25 Nov 2022 09:50:05 +0000 (10:50 +0100)]
net: lan966x: Add is2 vcap model to vcap API.

This provides the lan966x is2 model and adds it to the vcap control
instance that will be provided to the vcap API.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: lan966x: Add initial VCAP
Horatiu Vultur [Fri, 25 Nov 2022 09:50:04 +0000 (10:50 +0100)]
net: lan966x: Add initial VCAP

When lan966x driver is initialized, initialize also the VCAP module for
lan966x.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: microchip: vcap: Extend vcap with lan966x
Horatiu Vultur [Fri, 25 Nov 2022 09:50:03 +0000 (10:50 +0100)]
net: microchip: vcap: Extend vcap with lan966x

Add the keysets, keys, actionsets and actions used by lan966x in IS2.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: microchip: vcap: Merge the vcap_ag_api_kunit.h into vcap_ag_api.h
Horatiu Vultur [Fri, 25 Nov 2022 09:50:02 +0000 (10:50 +0100)]
net: microchip: vcap: Merge the vcap_ag_api_kunit.h into vcap_ag_api.h

Currently there are 2 files that contain the keyfields, keys,
actionfields and actions. First file is used by the kunit while the
second one is used by VCAP api.
The header file that is used by kunit is just a super set of the of the
header file used by VCAP api.
Therefore not to have duplicate information in different files which is
also harder to maintain, create a single file that is used both by API
and by kunit.

Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: marvell: prestera: Fix a NULL vs IS_ERR() check in some functions
Shang XiaoJing [Fri, 25 Nov 2022 01:27:51 +0000 (09:27 +0800)]
net: marvell: prestera: Fix a NULL vs IS_ERR() check in some functions

rhashtable_lookup_fast() returns NULL when failed instead of error
pointer.

Fixes: 396b80cb5cc8 ("net: marvell: prestera: Add neighbour cache accounting")
Fixes: 0a23ae237171 ("net: marvell: prestera: Add router nexthops ABI")
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Link: https://lore.kernel.org/r/20221125012751.23249-1-shangxiaojing@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: tun: Fix use-after-free in tun_detach()
Shigeru Yoshida [Thu, 24 Nov 2022 17:51:34 +0000 (02:51 +0900)]
net: tun: Fix use-after-free in tun_detach()

syzbot reported use-after-free in tun_detach() [1].  This causes call
trace like below:

==================================================================
BUG: KASAN: use-after-free in notifier_call_chain+0x1ee/0x200 kernel/notifier.c:75
Read of size 8 at addr ffff88807324e2a8 by task syz-executor.0/3673

CPU: 0 PID: 3673 Comm: syz-executor.0 Not tainted 6.1.0-rc5-syzkaller-00044-gcc675d22e422 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0xd1/0x138 lib/dump_stack.c:106
 print_address_description mm/kasan/report.c:284 [inline]
 print_report+0x15e/0x461 mm/kasan/report.c:395
 kasan_report+0xbf/0x1f0 mm/kasan/report.c:495
 notifier_call_chain+0x1ee/0x200 kernel/notifier.c:75
 call_netdevice_notifiers_info+0x86/0x130 net/core/dev.c:1942
 call_netdevice_notifiers_extack net/core/dev.c:1983 [inline]
 call_netdevice_notifiers net/core/dev.c:1997 [inline]
 netdev_wait_allrefs_any net/core/dev.c:10237 [inline]
 netdev_run_todo+0xbc6/0x1100 net/core/dev.c:10351
 tun_detach drivers/net/tun.c:704 [inline]
 tun_chr_close+0xe4/0x190 drivers/net/tun.c:3467
 __fput+0x27c/0xa90 fs/file_table.c:320
 task_work_run+0x16f/0x270 kernel/task_work.c:179
 exit_task_work include/linux/task_work.h:38 [inline]
 do_exit+0xb3d/0x2a30 kernel/exit.c:820
 do_group_exit+0xd4/0x2a0 kernel/exit.c:950
 get_signal+0x21b1/0x2440 kernel/signal.c:2858
 arch_do_signal_or_restart+0x86/0x2300 arch/x86/kernel/signal.c:869
 exit_to_user_mode_loop kernel/entry/common.c:168 [inline]
 exit_to_user_mode_prepare+0x15f/0x250 kernel/entry/common.c:203
 __syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline]
 syscall_exit_to_user_mode+0x1d/0x50 kernel/entry/common.c:296
 do_syscall_64+0x46/0xb0 arch/x86/entry/common.c:86
 entry_SYSCALL_64_after_hwframe+0x63/0xcd

The cause of the issue is that sock_put() from __tun_detach() drops
last reference count for struct net, and then notifier_call_chain()
from netdev_state_change() accesses that struct net.

This patch fixes the issue by calling sock_put() from tun_detach()
after all necessary accesses for the struct net has done.

Fixes: 83c1f36f9880 ("tun: send netlink notification when the device is modified")
Reported-by: syzbot+106f9b687cd64ee70cd1@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?id=96eb7f1ce75ef933697f24eeab928c4a716edefe
Signed-off-by: Shigeru Yoshida <syoshida@redhat.com>
Link: https://lore.kernel.org/r/20221124175134.1589053-1-syoshida@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoMerge branch 'refactor-mtk_wed-code-to-introduce-ser-support'
Paolo Abeni [Tue, 29 Nov 2022 10:40:27 +0000 (11:40 +0100)]
Merge branch 'refactor-mtk_wed-code-to-introduce-ser-support'

Lorenzo Bianconi says:

====================
refactor mtk_wed code to introduce SER support

Refactor mtk_wed support in order to introduce proper integration for hw reset
between mtk_eth_soc/mtk_wed and mt76 drivers.
====================

Link: https://lore.kernel.org/r/cover.1669303154.git.lorenzo@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: ethernet: mtk_wed: add reset to tx_ring_setup callback
Lorenzo Bianconi [Thu, 24 Nov 2022 15:22:55 +0000 (16:22 +0100)]
net: ethernet: mtk_wed: add reset to tx_ring_setup callback

Introduce reset parameter to mtk_wed_tx_ring_setup signature.
This is a preliminary patch to add Wireless Ethernet Dispatcher reset
support.

Co-developed-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: ethernet: mtk_wed: add mtk_wed_rx_reset routine
Lorenzo Bianconi [Thu, 24 Nov 2022 15:22:54 +0000 (16:22 +0100)]
net: ethernet: mtk_wed: add mtk_wed_rx_reset routine

Introduce mtk_wed_rx_reset routine in order to reset rx DMA for Wireless
Ethernet Dispatcher available on MT7986 SoC.

Co-developed-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: ethernet: mtk_wed: update mtk_wed_stop
Lorenzo Bianconi [Thu, 24 Nov 2022 15:22:53 +0000 (16:22 +0100)]
net: ethernet: mtk_wed: update mtk_wed_stop

Update mtk_wed_stop routine and rename old mtk_wed_stop() to
mtk_wed_deinit(). This is a preliminary patch to add Wireless Ethernet
Dispatcher reset support.

Co-developed-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: ethernet: mtk_wed: move MTK_WDMA_RESET_IDX_TX configuration in mtk_wdma_tx_reset
Lorenzo Bianconi [Thu, 24 Nov 2022 15:22:52 +0000 (16:22 +0100)]
net: ethernet: mtk_wed: move MTK_WDMA_RESET_IDX_TX configuration in mtk_wdma_tx_reset

Remove duplicated code. Increase poll timeout to 10ms in order to be
aligned with vendor sdk.
This is a preliminary patch to add Wireless Ethernet Dispatcher reset
support.

Co-developed-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: ethernet: mtk_wed: return status value in mtk_wdma_rx_reset
Lorenzo Bianconi [Thu, 24 Nov 2022 15:22:51 +0000 (16:22 +0100)]
net: ethernet: mtk_wed: return status value in mtk_wdma_rx_reset

Move MTK_WDMA_RESET_IDX configuration in mtk_wdma_rx_reset routine.
Increase poll timeout to 10ms in order to be aligned with vendor sdk.
This is a preliminary patch to add Wireless Ethernet Dispatcher reset
support.

Co-developed-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoMerge branch 'marvell-nvmem-mac-addresses-support'
Paolo Abeni [Tue, 29 Nov 2022 09:46:41 +0000 (10:46 +0100)]
Merge branch 'marvell-nvmem-mac-addresses-support'

Miquel Raynal says:

====================
Marvell nvmem mac addresses support

Now that we are aligned on how to make information available from static
storage media to drivers like Ethernet controller drivers or switch
drivers by using nvmem cells and going through the whole nvmem
infrastructure, here are two driver updates to reflect these changes.

Prior to the driver updates, I propose:
* Reverting binding changes which should have never been accepted like
  that.
* A conversion of the (old) Prestera and DFX server bindings (optional,
  can be dropped if not considered necessary).
* A better description of the more recent Prestera PCI switch.

Please mind that this series cannot break anything since retrieving the
MAC address Prestera driver has never worked upstream, because the (ONIE
tlv) driver supposed to export the MAC address has not been accepted in
its original form and has been updated to the nvmem-layout
infrastructure (bindings have been merged, the code remains to be
applied).
====================

Link: https://lore.kernel.org/r/20221124111556.264647-1-miquel.raynal@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: mvpp2: Consider NVMEM cells as possible MAC address source
Miquel Raynal [Thu, 24 Nov 2022 11:15:56 +0000 (12:15 +0100)]
net: mvpp2: Consider NVMEM cells as possible MAC address source

The ONIE standard describes the organization of tlv (type-length-value)
arrays commonly stored within NVMEM devices on common networking
hardware.

Several drivers already make use of NVMEM cells for purposes like
retrieving a default MAC address provided by the manufacturer.

What made ONIE tables unusable so far was the fact that the information
where "dynamically" located within the table depending on the
manufacturer wishes, while Linux NVMEM support only allowed statically
defined NVMEM cells. Fortunately, this limitation was eventually tackled
with the introduction of discoverable cells through the use of NVMEM
layouts, making it possible to extract and consistently use the content
of tables like ONIE's tlv arrays.

Parsing this table at runtime in order to get various information is now
possible. So, because many Marvell networking switches already follow
this standard, let's consider using NVMEM cells as a new valid source of
information when looking for a base MAC address, which is one of the
primary uses of these new fields. Indeed, manufacturers following the
ONIE standard are encouraged to provide a default MAC address there, so
let's eventually use it if no other MAC address has been found using the
existing methods.

Link: https://opencomputeproject.github.io/onie/design-spec/hw_requirements.html
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agonet: marvell: prestera: Avoid unnecessary DT lookups
Miquel Raynal [Thu, 24 Nov 2022 11:15:55 +0000 (12:15 +0100)]
net: marvell: prestera: Avoid unnecessary DT lookups

This driver fist makes an expensive DT lookup to retrieve its DT node
(this is a PCI driver) in order to later search for the
base-mac-provider property. This property has no reality upstream and
this code should not have been accepted like this in the first
place. Instead, there is a proper nvmem interface that should be
used. Let's avoid these extra lookups and rely on the nvmem internal
logic.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoof: net: export of_get_mac_address_nvmem()
Miquel Raynal [Thu, 24 Nov 2022 11:15:54 +0000 (12:15 +0100)]
of: net: export of_get_mac_address_nvmem()

Export

of_get_mac_addr_nvmem()

and rename it to

of_get_mac_address_nvmem()

in order to fit the convention followed by the existing exported helpers
of the same kind.

This way, OF compatible drivers using eg. fwnode_get_mac_address() can
do a direct call to it instead of calling of_get_mac_address() just for
the nvmem step, avoiding to repeat an expensive DT lookup which has
already been done once.

Eventually, fwnode_get_mac_address() should probably be updated to
perform the nvmem lookup directly, but as of today, nvmem cells seem not
to be supported by ACPI yet which would defeat this kind of extension.

Suggested-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agodt-bindings: net: marvell,prestera: Describe PCI devices of the prestera family
Miquel Raynal [Thu, 24 Nov 2022 11:15:53 +0000 (12:15 +0100)]
dt-bindings: net: marvell,prestera: Describe PCI devices of the prestera family

Even though the devices have very little in common beside the name and
the main "switch" feature, Marvell Prestera switch family is also
composed of PCI-only devices which can receive additional static
properties, like nvmem cells to point at MAC addresses, for
instance. Let's describe them.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agodt-bindings: net: marvell,prestera: Convert to yaml
Miquel Raynal [Thu, 24 Nov 2022 11:15:52 +0000 (12:15 +0100)]
dt-bindings: net: marvell,prestera: Convert to yaml

The currently described switch family is named AlleyCat3, it is a memory
mapped switch found on Armada XP boards.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agodt-bindings: net: marvell,dfx-server: Convert to yaml
Miquel Raynal [Thu, 24 Nov 2022 11:15:51 +0000 (12:15 +0100)]
dt-bindings: net: marvell,dfx-server: Convert to yaml

Even though this description is not used anywhere upstream (no matching
driver), while on this file I decided I would try a conversion to yaml
in order to clarify the prestera family description.

I cannot keep the nodename dfx-server@xxxx so I switched to dfx-bus@xxxx
which matches simple-bus.yaml. Otherwise I took the example context from
the only user of this compatible: armada-xp-98dx3236.dtsi, which is a
rather old and not perfect DT.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoRevert "dt-bindings: marvell,prestera: Add description for device-tree bindings"
Miquel Raynal [Thu, 24 Nov 2022 11:15:50 +0000 (12:15 +0100)]
Revert "dt-bindings: marvell,prestera: Add description for device-tree bindings"

This reverts commit 40acc05271abc2852c32622edbebd75698736b9b.

marvell,prestera.txt is an old file describing the old Alleycat3
standalone switches. The commit mentioned above actually hacked these
bindings to add support for a device tree property for a more modern
version of the IP connected over PCI, using only the generic compatible
in order to retrieve the device node from the prestera driver to read
one static property.

The problematic property discussed here is "base-mac-provider". The
original intent was to point to a nvmem device which could produce the
relevant nvmem-cell. This property has never been acked by DT
maintainers and fails all the layering that has been brought with the nvmem
bindings by pointing at a nvmem producer, bypassing the existing nvmem
bindings, rather than a nvmem cell directly. Furthermore, the property
cannot even be used upstream because it expected the ONIE tlv driver to
produce a specific cell, driver which used nacked bindings and thus was
never merged, replaced by a more integrated concept: the nvmem-layout.

So let's forget about this temporary addition, safely avoiding the need
for any backward compatibility handling. A new (yaml) binding file will
be brought with the prestera bindings, and there we will actually
include a description of the modern IP over PCI, including the right way
to point to a nvmem cell.

Cc: Vadym Kochan <vadym.kochan@plvision.eu>
Cc: Taras Chornyi <tchornyi@marvell.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Acked-by: Rob Herring <robh@kernel.org>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2 years agoDaniel Borkmann says:
Jakub Kicinski [Tue, 29 Nov 2022 01:14:01 +0000 (17:14 -0800)]
Daniel Borkmann says:

====================
bpf-next 2022-11-25

We've added 101 non-merge commits during the last 11 day(s) which contain
a total of 109 files changed, 8827 insertions(+), 1129 deletions(-).

The main changes are:

1) Support for user defined BPF objects: the use case is to allocate own
   objects, build own object hierarchies and use the building blocks to
   build own data structures flexibly, for example, linked lists in BPF,
   from Kumar Kartikeya Dwivedi.

2) Add bpf_rcu_read_{,un}lock() support for sleepable programs,
   from Yonghong Song.

3) Add support storing struct task_struct objects as kptrs in maps,
   from David Vernet.

4) Batch of BPF map documentation improvements, from Maryam Tahhan
   and Donald Hunter.

5) Improve BPF verifier to propagate nullness information for branches
   of register to register comparisons, from Eduard Zingerman.

6) Fix cgroup BPF iter infra to hold reference on the start cgroup,
   from Hou Tao.

7) Fix BPF verifier to not mark fentry/fexit program arguments as trusted
   given it is not the case for them, from Alexei Starovoitov.

8) Improve BPF verifier's realloc handling to better play along with dynamic
   runtime analysis tools like KASAN and friends, from Kees Cook.

9) Remove legacy libbpf mode support from bpftool,
   from Sahid Orentino Ferdjaoui.

10) Rework zero-len skb redirection checks to avoid potentially breaking
    existing BPF test infra users, from Stanislav Fomichev.

11) Two small refactorings which are independent and have been split out
    of the XDP queueing RFC series, from Toke Høiland-Jørgensen.

12) Fix a memory leak in LSM cgroup BPF selftest, from Wang Yufen.

13) Documentation on how to run BPF CI without patch submission,
    from Daniel Müller.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================

Link: https://lore.kernel.org/r/20221125012450.441-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoafs: Fix fileserver probe RTT handling
David Howells [Mon, 28 Nov 2022 22:02:56 +0000 (22:02 +0000)]
afs: Fix fileserver probe RTT handling

The fileserver probing code attempts to work out the best fileserver to
use for a volume by retrieving the RTT calculated by AF_RXRPC for the
probe call sent to each server and comparing them.  Sometimes, however,
no RTT estimate is available and rxrpc_kernel_get_srtt() returns false,
leading good fileservers to be given an RTT of UINT_MAX and thus causing
the rotation algorithm to ignore them.

Fix afs_select_fileserver() to ignore rxrpc_kernel_get_srtt()'s return
value and just take the estimated RTT it provides - which will be capped
at 1 second.

Fixes: 1d4adfaf6574 ("rxrpc: Make rxrpc_kernel_get_srtt() indicate validity")
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Marc Dionne <marc.dionne@auristor.com>
Tested-by: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Link: https://lore.kernel.org/r/166965503999.3392585.13954054113218099395.stgit@warthog.procyon.org.uk/
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agonet: mdiobus: fix unbalanced node reference count
Yang Yingliang [Thu, 24 Nov 2022 15:01:30 +0000 (23:01 +0800)]
net: mdiobus: fix unbalanced node reference count

I got the following report while doing device(mscc-miim) load test
with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:

  OF: ERROR: memory leak, expected refcount 1 instead of 2,
  of_node_get()/of_node_put() unbalanced - destroy cset entry:
  attach overlay node /spi/soc@0/mdio@7107009c/ethernet-phy@0

If the 'fwnode' is not an acpi node, the refcount is get in
fwnode_mdiobus_phy_device_register(), but it has never been
put when the device is freed in the normal path. So call
fwnode_handle_put() in phy_device_release() to avoid leak.

If it's an acpi node, it has never been get, but it's put
in the error path, so call fwnode_handle_get() before
phy_device_register() to keep get/put operation balanced.

Fixes: bc1bee3b87ee ("net: mdiobus: Introduce fwnode_mdiobus_register_phy()")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20221124150130.609420-1-yangyingliang@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: hsr: Fix potential use-after-free
YueHaibing [Fri, 25 Nov 2022 07:57:24 +0000 (15:57 +0800)]
net: hsr: Fix potential use-after-free

The skb is delivered to netif_rx() which may free it, after calling this,
dereferencing skb may trigger use-after-free.

Fixes: f421436a591d ("net/hsr: Add support for the High-availability Seamless Redundancy protocol (HSRv0)")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20221125075724.27912-1-yuehaibing@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agotipc: re-fetch skb cb after tipc_msg_validate
Xin Long [Fri, 25 Nov 2022 17:46:43 +0000 (12:46 -0500)]
tipc: re-fetch skb cb after tipc_msg_validate

As the call trace shows, the original skb was freed in tipc_msg_validate(),
and dereferencing the old skb cb would cause an use-after-free crash.

  BUG: KASAN: use-after-free in tipc_crypto_rcv_complete+0x1835/0x2240 [tipc]
  Call Trace:
   <IRQ>
   tipc_crypto_rcv_complete+0x1835/0x2240 [tipc]
   tipc_crypto_rcv+0xd32/0x1ec0 [tipc]
   tipc_rcv+0x744/0x1150 [tipc]
  ...
  Allocated by task 47078:
   kmem_cache_alloc_node+0x158/0x4d0
   __alloc_skb+0x1c1/0x270
   tipc_buf_acquire+0x1e/0xe0 [tipc]
   tipc_msg_create+0x33/0x1c0 [tipc]
   tipc_link_build_proto_msg+0x38a/0x2100 [tipc]
   tipc_link_timeout+0x8b8/0xef0 [tipc]
   tipc_node_timeout+0x2a1/0x960 [tipc]
   call_timer_fn+0x2d/0x1c0
  ...
  Freed by task 47078:
   tipc_msg_validate+0x7b/0x440 [tipc]
   tipc_crypto_rcv_complete+0x4b5/0x2240 [tipc]
   tipc_crypto_rcv+0xd32/0x1ec0 [tipc]
   tipc_rcv+0x744/0x1150 [tipc]

This patch fixes it by re-fetching the skb cb from the new allocated skb
after calling tipc_msg_validate().

Fixes: fc1b6d6de220 ("tipc: introduce TIPC encryption & authentication")
Reported-by: Shuang Li <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Link: https://lore.kernel.org/r/1b1cdba762915325bd8ef9a98d0276eb673df2a5.1669398403.git.lucien.xin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge branch 'mptcp-more-fixes-for-6-1'
Jakub Kicinski [Tue, 29 Nov 2022 02:03:11 +0000 (18:03 -0800)]
Merge branch 'mptcp-more-fixes-for-6-1'

Matthieu Baerts says:

====================
mptcp: More fixes for 6.1

Patch 1 makes sure data received after a close will still be processed
and acked as exepected. This is a regression for a commit introduced
in v5.11.

Patch 2 fixes a kernel deadlock found when working on validating TFO
with a listener MPTCP socket. This is not directly linked to TFO but
it is easier to reproduce the issue with it. This fixes a bug introduced
by a commit from v6.0.
====================

Link: https://lore.kernel.org/r/20221128154239.1999234-1-matthieu.baerts@tessares.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: fix sleep in atomic at close time
Paolo Abeni [Mon, 28 Nov 2022 15:42:38 +0000 (16:42 +0100)]
mptcp: fix sleep in atomic at close time

Matt reported a splat at msk close time:

    BUG: sleeping function called from invalid context at net/mptcp/protocol.c:2877
    in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 155, name: packetdrill
    preempt_count: 201, expected: 0
    RCU nest depth: 0, expected: 0
    4 locks held by packetdrill/155:
    #0: ffff888001536990 (&sb->s_type->i_mutex_key#6){+.+.}-{3:3}, at: __sock_release (net/socket.c:650)
    #1: ffff88800b498130 (sk_lock-AF_INET){+.+.}-{0:0}, at: mptcp_close (net/mptcp/protocol.c:2973)
    #2: ffff88800b49a130 (sk_lock-AF_INET/1){+.+.}-{0:0}, at: __mptcp_close_ssk (net/mptcp/protocol.c:2363)
    #3: ffff88800b49a0b0 (slock-AF_INET){+...}-{2:2}, at: __lock_sock_fast (include/net/sock.h:1820)
    Preemption disabled at:
    0x0
    CPU: 1 PID: 155 Comm: packetdrill Not tainted 6.1.0-rc5 #365
    Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
    Call Trace:
    <TASK>
    dump_stack_lvl (lib/dump_stack.c:107 (discriminator 4))
    __might_resched.cold (kernel/sched/core.c:9891)
    __mptcp_destroy_sock (include/linux/kernel.h:110)
    __mptcp_close (net/mptcp/protocol.c:2959)
    mptcp_subflow_queue_clean (include/net/sock.h:1777)
    __mptcp_close_ssk (net/mptcp/protocol.c:2363)
    mptcp_destroy_common (net/mptcp/protocol.c:3170)
    mptcp_destroy (include/net/sock.h:1495)
    __mptcp_destroy_sock (net/mptcp/protocol.c:2886)
    __mptcp_close (net/mptcp/protocol.c:2959)
    mptcp_close (net/mptcp/protocol.c:2974)
    inet_release (net/ipv4/af_inet.c:432)
    __sock_release (net/socket.c:651)
    sock_close (net/socket.c:1367)
    __fput (fs/file_table.c:320)
    task_work_run (kernel/task_work.c:181 (discriminator 1))
    exit_to_user_mode_prepare (include/linux/resume_user_mode.h:49)
    syscall_exit_to_user_mode (kernel/entry/common.c:130)
    do_syscall_64 (arch/x86/entry/common.c:87)
    entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120)

We can't call mptcp_close under the 'fast' socket lock variant, replace
it with a sock_lock_nested() as the relevant code is already under the
listening msk socket lock protection.

Reported-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/316
Fixes: 30e51b923e43 ("mptcp: fix unreleased socket in accept queue")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agomptcp: don't orphan ssk in mptcp_close()
Menglong Dong [Mon, 28 Nov 2022 15:42:37 +0000 (16:42 +0100)]
mptcp: don't orphan ssk in mptcp_close()

All of the subflows of a msk will be orphaned in mptcp_close(), which
means the subflows are in DEAD state. After then, DATA_FIN will be sent,
and the other side will response with a DATA_ACK for this DATA_FIN.

However, if the other side still has pending data, the data that received
on these subflows will not be passed to the msk, as they are DEAD and
subflow_data_ready() will not be called in tcp_data_ready(). Therefore,
these data can't be acked, and they will be retransmitted again and again,
until timeout.

Fix this by setting ssk->sk_socket and ssk->sk_wq to 'NULL', instead of
orphaning the subflows in __mptcp_close(), as Paolo suggested.

Fixes: e16163b6e2b7 ("mptcp: refactor shutdown and close")
Reviewed-by: Biao Jiang <benbjiang@tencent.com>
Reviewed-by: Mengen Sun <mengensun@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agodsa: lan9303: Correct stat name
Jerry Ray [Mon, 28 Nov 2022 19:35:59 +0000 (13:35 -0600)]
dsa: lan9303: Correct stat name

This patch changes the reported ethtool statistics for the lan9303
family of parts covered by this driver.

The TxUnderRun statistic label is renamed to RxShort to accurately
reflect what stat the device is reporting.  I did not reorder the
statistics as that might cause problems with existing user code that
are expecting the stats at a certain offset.

Fixes: a1292595e006 ("net: dsa: add new DSA switch driver for the SMSC-LAN9303")
Signed-off-by: Jerry Ray <jerry.ray@microchip.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20221128193559.6572-1-jerry.ray@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge tag 'wireless-2022-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git...
Jakub Kicinski [Tue, 29 Nov 2022 01:58:45 +0000 (17:58 -0800)]
Merge tag 'wireless-2022-11-28' of git://git./linux/kernel/git/wireless/wireless

Kalle Valo says:

====================
wireless fixes for v6.1

Third, and hopefully final, set of fixes for v6.1. We are marking the
rsi driver as orphan, have some Information Element parsing fixes to
wilc1000 driver and three small fixes to the stack.

* tag 'wireless-2022-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
  wifi: mac8021: fix possible oob access in ieee80211_get_rate_duration
  wifi: cfg80211: don't allow multi-BSSID in S1G
  wifi: cfg80211: fix buffer overflow in elem comparison
  wifi: wilc1000: validate number of channels
  wifi: wilc1000: validate length of IEEE80211_P2P_ATTR_CHANNEL_LIST attribute
  wifi: wilc1000: validate length of IEEE80211_P2P_ATTR_OPER_CHANNEL attribute
  wifi: wilc1000: validate pairwise and authentication suite offsets
  MAINTAINERS: mark rsi wifi driver as orphan
====================

Link: https://lore.kernel.org/r/20221128113513.6F459C433C1@smtp.kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Jakub Kicinski [Tue, 29 Nov 2022 01:06:51 +0000 (17:06 -0800)]
Merge tag 'for-netdev' of https://git./linux/kernel/git/bpf/bpf

Daniel Borkmann says:

====================
bpf 2022-11-25

We've added 10 non-merge commits during the last 8 day(s) which contain
a total of 7 files changed, 48 insertions(+), 30 deletions(-).

The main changes are:

1) Several libbpf ringbuf fixes related to probing for its availability,
   size overflows when mmaping a 2G ringbuf and rejection of invalid
   reservationsizes, from Hou Tao.

2) Fix a buggy return pointer in libbpf for attach_raw_tp function,
   from Jiri Olsa.

3) Fix a local storage BPF map bug where the value's spin lock field
   can get initialized incorrectly, from Xu Kuohai.

4) Two follow-up fixes in kprobe_multi BPF selftests for BPF CI,
   from Jiri Olsa.

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  selftests/bpf: Make test_bench_attach serial
  selftests/bpf: Filter out default_idle from kprobe_multi bench
  bpf: Set and check spin lock value in sk_storage_map_test
  bpf: Do not copy spin lock field from user in bpf_selem_alloc
  libbpf: Check the validity of size in user_ring_buffer__reserve()
  libbpf: Handle size overflow for user ringbuf mmap
  libbpf: Handle size overflow for ringbuf mmap
  libbpf: Use page size as max_entries when probing ring buffer map
  bpf, perf: Use subprog name when reporting subprog ksymbol
  libbpf: Use correct return pointer in attach_raw_tp
====================

Link: https://lore.kernel.org/r/20221125001034.29473-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoRevert "net: stmmac: use sysfs_streq() instead of strncmp()"
Vladimir Oltean [Fri, 25 Nov 2022 10:53:04 +0000 (12:53 +0200)]
Revert "net: stmmac: use sysfs_streq() instead of strncmp()"

This reverts commit f72cd76b05ea1ce9258484e8127932d0ea928f22.
This patch is so broken, it hurts. Apparently no one reviewed it and it
passed the build testing (because the code was compiled out), but it was
obviously never compile-tested, since it produces the following build
error, due to an incomplete conversion where an extra argument was left,
although the function being called was left:

stmmac_main.c: In function ‘stmmac_cmdline_opt’:
stmmac_main.c:7586:28: error: too many arguments to function ‘sysfs_streq’
 7586 |                 } else if (sysfs_streq(opt, "pause:", 6)) {
      |                            ^~~~~~~~~~~
In file included from ../include/linux/bitmap.h:11,
                 from ../include/linux/cpumask.h:12,
                 from ../include/linux/smp.h:13,
                 from ../include/linux/lockdep.h:14,
                 from ../include/linux/mutex.h:17,
                 from ../include/linux/notifier.h:14,
                 from ../include/linux/clk.h:14,
                 from ../drivers/net/ethernet/stmicro/stmmac/stmmac_main.c:17:
../include/linux/string.h:185:13: note: declared here
  185 | extern bool sysfs_streq(const char *s1, const char *s2);
      |             ^~~~~~~~~~~

What's even worse is that the patch is flat out wrong. The stmmac_cmdline_opt()
function does not parse sysfs input, but cmdline input such as
"stmmaceth=tc:1,pause:1". The pattern of using strsep() followed by
strncmp() for such strings is not unique to stmmac, it can also be found
mainly in drivers under drivers/video/fbdev/.

With strncmp("tc:", 3), the code matches on the "tc:1" token properly.
With sysfs_streq("tc:"), it doesn't.

Fixes: f72cd76b05ea ("net: stmmac: use sysfs_streq() instead of strncmp()")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20221125105304.3012153-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agonet: usb: cdc_ether: add u-blox 0x1343 composition
Davide Tronchin [Thu, 24 Nov 2022 11:28:11 +0000 (12:28 +0100)]
net: usb: cdc_ether: add u-blox 0x1343 composition

Add CDC-ECM support for LARA-L6.

LARA-L6 module can be configured (by AT interface) in three different
USB modes:
* Default mode (Vendor ID: 0x1546 Product ID: 0x1341) with 4 serial
interfaces
* RmNet mode (Vendor ID: 0x1546 Product ID: 0x1342) with 4 serial
interfaces and 1 RmNet virtual network interface
* CDC-ECM mode (Vendor ID: 0x1546 Product ID: 0x1343) with 4 serial
interface and 1 CDC-ECM virtual network interface

In CDC-ECM mode LARA-L6 exposes the following interfaces:
If 0: Diagnostic
If 1: AT parser
If 2: AT parser
If 3: AT parset/alternative functions
If 4: CDC-ECM interface

Signed-off-by: Davide Tronchin <davide.tronchin.94@gmail.com>
Link: https://lore.kernel.org/r/20221124112811.3548-1-davide.tronchin.94@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoipv4: Fix route deletion when nexthop info is not specified
Ido Schimmel [Thu, 24 Nov 2022 21:09:32 +0000 (23:09 +0200)]
ipv4: Fix route deletion when nexthop info is not specified

When the kernel receives a route deletion request from user space it
tries to delete a route that matches the route attributes specified in
the request.

If only prefix information is specified in the request, the kernel
should delete the first matching FIB alias regardless of its associated
FIB info. However, an error is currently returned when the FIB info is
backed by a nexthop object:

 # ip nexthop add id 1 via 192.0.2.2 dev dummy10
 # ip route add 198.51.100.0/24 nhid 1
 # ip route del 198.51.100.0/24
 RTNETLINK answers: No such process

Fix by matching on such a FIB info when legacy nexthop attributes are
not specified in the request. An earlier check already covers the case
where a nexthop ID is specified in the request.

Add tests that cover these flows. Before the fix:

 # ./fib_nexthops.sh -t ipv4_fcnal
 ...
 TEST: Delete route when not specifying nexthop attributes           [FAIL]

 Tests passed:  11
 Tests failed:   1

After the fix:

 # ./fib_nexthops.sh -t ipv4_fcnal
 ...
 TEST: Delete route when not specifying nexthop attributes           [ OK ]

 Tests passed:  12
 Tests failed:   0

No regressions in other tests:

 # ./fib_nexthops.sh
 ...
 Tests passed: 228
 Tests failed:   0

 # ./fib_tests.sh
 ...
 Tests passed: 186
 Tests failed:   0

Cc: stable@vger.kernel.org
Reported-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Jonas Gorski <jonas.gorski@gmail.com>
Fixes: 493ced1ac47c ("ipv4: Allow routes to use nexthop objects")
Fixes: 6bf92d70e690 ("net: ipv4: fix route with nexthop object delete warning")
Fixes: 61b91eb33a69 ("ipv4: Handle attempt to delete multipath route when fib_info contains an nh reference")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20221124210932.2470010-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2 years agoMerge tag 'devicetree-fixes-for-6.1-2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 28 Nov 2022 22:59:17 +0000 (14:59 -0800)]
Merge tag 'devicetree-fixes-for-6.1-2' of git://git./linux/kernel/git/robh/linux

Pull devicetree fix from Rob Herring:

 - Fix DT node reference counting for of_fwnode_get_reference_args()
   exit path

* tag 'devicetree-fixes-for-6.1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  of: property: decrement node refcount in of_fwnode_get_reference_args()

2 years agoMerge tag 'trace-v6.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/trace...
Linus Torvalds [Mon, 28 Nov 2022 22:42:29 +0000 (14:42 -0800)]
Merge tag 'trace-v6.1-rc6' of git://git./linux/kernel/git/trace/linux-trace

Pull tracing fixes from Steven Rostedt:

 - Fix osnoise duration type to 64bit not 32bit

 - Have histogram triggers be able to handle an unexpected NULL pointer
   for the record event, which can happen when the histogram first
   starts up

 - Clear out ring buffers when dynamic events are removed, as the type
   that is saved in the ring buffer is used to read the event, and a
   stale type that is reused by another event could cause use after free
   issues

 - Trivial comment fix

 - Fix memory leak in user_event_create()

* tag 'trace-v6.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  tracing: Free buffers when a used dynamic event is removed
  tracing: Add tracing_reset_all_online_cpus_unlocked() function
  tracing: Fix race where histograms can be called before the event
  tracing/osnoise: Fix duration type
  tracing/user_events: Fix memory leak in user_event_create()
  tracing/hist: add in missing * in comment blocks

2 years agoMerge tag 'fuse-fixes-6.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/mszer...
Linus Torvalds [Mon, 28 Nov 2022 21:08:46 +0000 (13:08 -0800)]
Merge tag 'fuse-fixes-6.1-rc8' of git://git./linux/kernel/git/mszeredi/fuse

Pull fuse fix from Miklos Szeredi:
 "Fix a regression introduced in -rc4"

* tag 'fuse-fixes-6.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: lock inode unconditionally in fuse_fallocate()

2 years agoMerge branch 'wwan-iosm-fixes'
David S. Miller [Mon, 28 Nov 2022 11:31:59 +0000 (11:31 +0000)]
Merge branch 'wwan-iosm-fixes'

M Chetan Kumar says:

====================
net: wwan: iosm: fix build errors & bugs

This patch series fixes iosm driver bugs & build errors.

PATCH1: Fix kernel build robot reported error.
PATCH2: Fix build error reported on armhf while preparing
        6.1-rc5 for Debian.
PATCH3: Fix UL throughput crash.
PATCH4: Fix incorrect skb length.

Refer to commit message for details.

Changes since v1:
* PATCH4: Fix sparse warning.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: wwan: iosm: fix incorrect skb length
M Chetan Kumar [Thu, 24 Nov 2022 10:38:32 +0000 (16:08 +0530)]
net: wwan: iosm: fix incorrect skb length

skb passed to network layer contains incorrect length.

In mux aggregation protocol, the datagram block received
from device contains block signature, packet & datagram
header. The right skb len to be calculated by subracting
datagram pad len from datagram length.

Whereas in mux lite protocol, the skb contains single
datagram so skb len is calculated by subtracting the
packet offset from datagram header.

Fixes: 1f52d7b62285 ("net: wwan: iosm: Enable M.2 7360 WWAN card support")
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: wwan: iosm: fix crash in peek throughput test
M Chetan Kumar [Thu, 24 Nov 2022 10:38:17 +0000 (16:08 +0530)]
net: wwan: iosm: fix crash in peek throughput test

Peek throughput UL test is resulting in crash. If the UL
transfer block free list is exhaust, the peeked skb is freed.
In the next transfer freed skb is referred from UL list which
results in crash.

Don't free the skb if UL transfer blocks are unavailable. The
pending skb will be picked for transfer on UL transfer block
available.

Fixes: 1f52d7b62285 ("net: wwan: iosm: Enable M.2 7360 WWAN card support")
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: wwan: iosm: fix dma_alloc_coherent incompatible pointer type
M Chetan Kumar [Thu, 24 Nov 2022 10:38:03 +0000 (16:08 +0530)]
net: wwan: iosm: fix dma_alloc_coherent incompatible pointer type

Fix build error reported on armhf while preparing 6.1-rc5
for Debian.

iosm_ipc_protocol.c:244:36: error: passing argument 3 of
'dma_alloc_coherent' from incompatible pointer type.

Change phy_ap_shm type from phys_addr_t to dma_addr_t.

Fixes: faed4c6f6f48 ("net: iosm: shared memory protocol")
Reported-by: Bonaccorso Salvatore <carnil@debian.org>
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: wwan: iosm: fix kernel test robot reported error
M Chetan Kumar [Thu, 24 Nov 2022 10:37:46 +0000 (16:07 +0530)]
net: wwan: iosm: fix kernel test robot reported error

sparse warnings - iosm_ipc_mux_codec.c:1474 using plain
integer as NULL pointer.

Use skb_trim() to reset skb tail & len.

Fixes: 9413491e20e1 ("net: iosm: encode or decode datagram")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: phylink: fix PHY validation with rate adaption
Russell King (Oracle) [Thu, 24 Nov 2022 09:06:48 +0000 (09:06 +0000)]
net: phylink: fix PHY validation with rate adaption

Tim Harvey reports that link modes which he does not expect to be
supported are being advertised, and this is because of the workaround
we have for PHYs that switch interface modes.

Fix this up by checking whether rate matching will be used for the
requested interface mode, and if rate matching will be used, perform
validation only with the requested interface mode, rather than invoking
this workaround.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: ethernet: nixge: fix NULL dereference
Yuri Karpov [Thu, 24 Nov 2022 08:43:03 +0000 (11:43 +0300)]
net: ethernet: nixge: fix NULL dereference

In function nixge_hw_dma_bd_release() dereference of NULL pointer
priv->rx_bd_v is possible for the case of its allocation failure in
nixge_hw_dma_bd_init().

Move for() loop with priv->rx_bd_v dereference under the check for
its validity.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: 492caffa8a1a ("net: ethernet: nixge: Add support for National Instruments XGE netdev")
Signed-off-by: Yuri Karpov <YKarpov@ispras.ru>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet/9p: Fix a potential socket leak in p9_socket_open
Wang Hai [Thu, 24 Nov 2022 08:10:05 +0000 (16:10 +0800)]
net/9p: Fix a potential socket leak in p9_socket_open

Both p9_fd_create_tcp() and p9_fd_create_unix() will call
p9_socket_open(). If the creation of p9_trans_fd fails,
p9_fd_create_tcp() and p9_fd_create_unix() will return an
error directly instead of releasing the cscoket, which will
result in a socket leak.

This patch adds sock_release() to fix the leak issue.

Fixes: 6b18662e239a ("9p connect fixes")
Signed-off-by: Wang Hai <wanghai38@huawei.com>
ACKed-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: net_netdev: Fix error handling in ntb_netdev_init_module()
Yuan Can [Thu, 24 Nov 2022 07:09:17 +0000 (07:09 +0000)]
net: net_netdev: Fix error handling in ntb_netdev_init_module()

The ntb_netdev_init_module() returns the ntb_transport_register_client()
directly without checking its return value, if
ntb_transport_register_client() failed, the NTB client device is not
unregistered.

Fix by unregister NTB client device when ntb_transport_register_client()
failed.

Fixes: 548c237c0a99 ("net: Add support for NTB virtual ethernet device")
Signed-off-by: Yuan Can <yuancan@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agoocteontx2-pf: Add support to filter packet based on IP fragment
Suman Ghosh [Thu, 24 Nov 2022 06:35:48 +0000 (12:05 +0530)]
octeontx2-pf: Add support to filter packet based on IP fragment

1. Added support to filter packets based on IP fragment.
For IPv4 packets check for ip_flag == 0x20 (more fragment bit set).
For IPv6 packets check for next_header == 0x2c (next_header set to
'fragment header for IPv6')
2. Added configuration support from both "ethtool ntuple" and "tc flower".

Signed-off-by: Suman Ghosh <sumang@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: ethernet: mtk_wed: add wcid overwritten support for wed v1
Sujuan Chen [Thu, 24 Nov 2022 03:18:14 +0000 (11:18 +0800)]
net: ethernet: mtk_wed: add wcid overwritten support for wed v1

All wed versions should enable the wcid overwritten feature,
since the wcid size is controlled by the wlan driver.

Tested-by: Sujuan Chen <sujuan.chen@mediatek.com>
Co-developed-by: Bo Jiao <bo.jiao@mediatek.com>
Signed-off-by: Bo Jiao <bo.jiao@mediatek.com>
Signed-off-by: Sujuan Chen <sujuan.chen@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2 years agonet: ethernet: ti: am65-cpsw: fix error handling in am65_cpsw_nuss_probe()
Zhang Changzhong [Thu, 24 Nov 2022 03:03:08 +0000 (11:03 +0800)]
net: ethernet: ti: am65-cpsw: fix error handling in am65_cpsw_nuss_probe()

The am65_cpsw_nuss_cleanup_ndev() function calls unregister_netdev()
even if register_netdev() fails, which triggers WARN_ON(1) in
unregister_netdevice_many(). To fix it, make sure that
unregister_netdev() is called only on registered netdev.

Compile tested only.

Fixes: 84b4aa493249 ("net: ethernet: ti: am65-cpsw: add multi port support in mac-only mode")
Signed-off-by: Zhang Changzhong <zhangchangzhong@huawei.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>