Merge branch 'linus'

Merge tag 'net-6.0-rc3' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from ipsec and netfilter (with one broken Fixes tag).

  Current release - new code bugs:

   - dsa: don't dereference NULL extack in dsa_slave_changeupper()

   - dpaa: fix <1G ethernet on LS1046ARDB

   - neigh: don't call kfree_skb() under spin_lock_irqsave()

  Previous releases - regressions:

   - r8152: fix the RX FIFO settings when suspending

   - dsa: microchip: keep compatibility with device tree blobs with no
     phy-mode

   - Revert "net: macsec: update SCI upon MAC address change."

   - Revert "xfrm: update SA curlft.use_time", comply with RFC 2367

  Previous releases - always broken:

   - netfilter: conntrack: work around exceeded TCP receive window

   - ipsec: fix a null pointer dereference of dst->dev on a metadata dst
     in xfrm_lookup_with_ifid

   - moxa: get rid of asymmetry in DMA mapping/unmapping

   - dsa: microchip: make learning configurable and keep it off while
     standalone

   - ice: xsk: prohibit usage of non-balanced queue id

   - rxrpc: fix locking in rxrpc's sendmsg

  Misc:

   - another chunk of sysctl data race silencing"

* tag 'net-6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (87 commits)
  net: lantiq_xrx200: restore buffer if memory allocation failed
  net: lantiq_xrx200: fix lock under memory pressure
  net: lantiq_xrx200: confirm skb is allocated before using
  net: stmmac: work around sporadic tx issue on link-up
  ionic: VF initial random MAC address if no assigned mac
  ionic: fix up issues with handling EAGAIN on FW cmds
  ionic: clear broken state on generation change
  rxrpc: Fix locking in rxrpc's sendmsg
  net: ethernet: mtk_eth_soc: fix hw hash reporting for MTK_NETSYS_V2
  MAINTAINERS: rectify file entry in BONDING DRIVER
  i40e: Fix incorrect address type for IPv6 flow rules
  ixgbe: stop resetting SYSTIME in ixgbe_ptp_start_cyclecounter
  net: Fix a data-race around sysctl_somaxconn.
  net: Fix a data-race around netdev_unregister_timeout_secs.
  net: Fix a data-race around gro_normal_batch.
  net: Fix data-races around sysctl_devconf_inherit_init_net.
  net: Fix data-races around sysctl_fb_tunnels_only_for_init_net.
  net: Fix a data-race around netdev_budget_usecs.
  net: Fix data-races around sysctl_max_skb_frags.
  net: Fix a data-race around netdev_budget.
  ...

Merge branch 'net-lantiq_xrx200-fix-errors-under-memory-pressure'

Aleksander Jan Bajkowski says:

====================
net: lantiq_xrx200: fix errors under memory pressure

This series fixes issues that can occur in the driver under memory pressure.
Situations when the system cannot allocate memory are rare, so the mentioned
bugs have been fixed recently. The patches have been tested on a BT Home
router with the Lantiq xRX200 chipset.

Changelog:
  v3: - removed netdev_err() log from the first patch
  v2:
   - the second patch has been changed, so that under memory pressure situation
     the driver will not receive packets indefinitely regardless of the NAPI budget,
   - the third patch has been added.
====================

Link: https://lore.kernel.org/r/20220824215408.4695-1-olek2@wp.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: lantiq_xrx200: restore buffer if memory allocation failed

In a situation where memory allocation fails, an invalid buffer address
is stored. When this descriptor is used again, the system panics in the
build_skb() function when accessing memory.

Fixes: 7ea6cd16f159 ("lantiq: net: fix duplicated skb in rx descriptor ring")
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: lantiq_xrx200: fix lock under memory pressure

When the xrx200_hw_receive() function returns -ENOMEM, the NAPI poll
function immediately returns an error.
This is incorrect for two reasons:
* the function terminates without enabling interrupts or scheduling NAPI,
* the error code (-ENOMEM) is returned instead of the number of received
packets.

After the first memory allocation failure occurs, packet reception is
locked due to disabled interrupts from DMA..

Fixes: fe1a56420cf2 ("net: lantiq: Add Lantiq / Intel VRX200 Ethernet driver")
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: lantiq_xrx200: confirm skb is allocated before using

xrx200_hw_receive() assumes build_skb() always works and goes straight
to skb_reserve(). However, build_skb() can fail under memory pressure.

Add a check in case build_skb() failed to allocate and return NULL.

Fixes: e015593573b3 ("net: lantiq_xrx200: convert to build_skb")
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: stmmac: work around sporadic tx issue on link-up

This is a follow-up to the discussion in [0]. It seems to me that
at least the IP version used on Amlogic SoC's sometimes has a problem
if register MAC_CTRL_REG is written whilst the chip is still processing
a previous write. But that's just a guess.
Adding a delay between two writes to this register helps, but we can
also simply omit the offending second write. This patch uses the second
approach and is based on a suggestion from Qi Duan.
Benefit of this approach is that we can save few register writes, also
on not affected chip versions.

[0] https://www.spinics.net/lists/netdev/msg831526.html

Fixes: bfab27a146ed ("stmmac: add the experimental PCI support")
Suggested-by: Qi Duan <qi.duan@amlogic.com>
Suggested-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/e99857ce-bd90-5093-ca8c-8cd480b5a0a2@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch '10GbE' of git://git./linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2022-08-24 (ixgbe, i40e)

This series contains updates to ixgbe and i40e drivers.

Jake stops incorrect resetting of SYSTIME registers when starting
cyclecounter for ixgbe.

Sylwester corrects a check on source IP address when validating destination
for i40e.

* '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
  i40e: Fix incorrect address type for IPv6 flow rules
  ixgbe: stop resetting SYSTIME in ixgbe_ptp_start_cyclecounter
====================

Link: https://lore.kernel.org/r/20220824193748.874343-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'ionic-bug-fixes'

Shannon Nelson says:

====================
ionic: bug fixes

These are a couple of maintenance bug fixes for the Pensando ionic
networking driver.

Mohamed takes care of a "plays well with others" issue where the
VF spec is a bit vague on VF mac addresses, but certain customers
have come to expect behavior based on other vendor drivers.

Shannon addresses a couple of corner cases seen in internal
stress testing.
====================

Link: https://lore.kernel.org/r/20220824165051.6185-1-snelson@pensando.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ionic: VF initial random MAC address if no assigned mac

Assign a random mac address to the VF interface station
address if it boots with a zero mac address in order to match
similar behavior seen in other VF drivers.  Handle the errors
where the older firmware does not allow the VF to set its own
station address.

Newer firmware will allow the VF to set the station mac address
if it hasn't already been set administratively through the PF.
Setting it will also be allowed if the VF has trust.

Fixes: fbb39807e9ae ("ionic: support sr-iov operations")
Signed-off-by: R Mohamed Shah <mohamed@pensando.io>
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ionic: fix up issues with handling EAGAIN on FW cmds

In looping on FW update tests we occasionally see the
FW_ACTIVATE_STATUS command fail while it is in its EAGAIN loop
waiting for the FW activate step to finsh inside the FW.  The
firmware is complaining that the done bit is set when a new
dev_cmd is going to be processed.

Doing a clean on the cmd registers and doorbell before exiting
the wait-for-done and cleaning the done bit before the sleep
prevents this from occurring.

Fixes: fbfb8031533c ("ionic: Add hardware init and device commands")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

ionic: clear broken state on generation change

There is a case found in heavy testing where a link flap happens just
before a firmware Recovery event and the driver gets stuck in the
BROKEN state.  This comes from the driver getting interrupted by a FW
generation change when coming back up from the link flap, and the call
to ionic_start_queues() in ionic_link_status_check() fails.  This can be
addressed by having the fw_up code clear the BROKEN bit if seen, rather
than waiting for a user to manually force the interface down and then
back up.

Fixes: 9e8eaf8427b6 ("ionic: stop watchdog when in broken state")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

rxrpc: Fix locking in rxrpc's sendmsg

Fix three bugs in the rxrpc's sendmsg implementation:

 (1) rxrpc_new_client_call() should release the socket lock when returning
     an error from rxrpc_get_call_slot().

 (2) rxrpc_wait_for_tx_window_intr() will return without the call mutex
     held in the event that we're interrupted by a signal whilst waiting
     for tx space on the socket or relocking the call mutex afterwards.

     Fix this by: (a) moving the unlock/lock of the call mutex up to
     rxrpc_send_data() such that the lock is not held around all of
     rxrpc_wait_for_tx_window*() and (b) indicating to higher callers
     whether we're return with the lock dropped.  Note that this means
     recvmsg() will not block on this call whilst we're waiting.

 (3) After dropping and regaining the call mutex, rxrpc_send_data() needs
     to go and recheck the state of the tx_pending buffer and the
     tx_total_len check in case we raced with another sendmsg() on the same
     call.

Thinking on this some more, it might make sense to have different locks for
sendmsg() and recvmsg().  There's probably no need to make recvmsg() wait
for sendmsg().  It does mean that recvmsg() can return MSG_EOR indicating
that a call is dead before a sendmsg() to that call returns - but that can
currently happen anyway.

Without fix (2), something like the following can be induced:

WARNING: bad unlock balance detected!
5.16.0-rc6-syzkaller #0 Not tainted
-------------------------------------
syz-executor011/3597 is trying to release lock (&call->user_mutex) at:
[<ffffffff885163a3>] rxrpc_do_sendmsg+0xc13/0x1350 net/rxrpc/sendmsg.c:748
but there are no more locks to release!

other info that might help us debug this:
no locks held by syz-executor011/3597.
...
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
 print_unlock_imbalance_bug include/trace/events/lock.h:58 [inline]
 __lock_release kernel/locking/lockdep.c:5306 [inline]
 lock_release.cold+0x49/0x4e kernel/locking/lockdep.c:5657
 __mutex_unlock_slowpath+0x99/0x5e0 kernel/locking/mutex.c:900
 rxrpc_do_sendmsg+0xc13/0x1350 net/rxrpc/sendmsg.c:748
 rxrpc_sendmsg+0x420/0x630 net/rxrpc/af_rxrpc.c:561
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:724
 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2409
 ___sys_sendmsg+0xf3/0x170 net/socket.c:2463
 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2492
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae

[Thanks to Hawkins Jiawei and Khalid Masum for their attempts to fix this]

Fixes: bc5e3a546d55 ("rxrpc: Use MSG_WAITALL to tell sendmsg() to temporarily ignore signals")
Reported-by: syzbot+7f0483225d0c94cb3441@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Marc Dionne <marc.dionne@auristor.com>
Tested-by: syzbot+7f0483225d0c94cb3441@syzkaller.appspotmail.com
cc: Hawkins Jiawei <yin31149@gmail.com>
cc: Khalid Masum <khalid.masum.92@gmail.com>
cc: Dan Carpenter <dan.carpenter@oracle.com>
cc: linux-afs@lists.infradead.org
Link: https://lore.kernel.org/r/166135894583.600315.7170979436768124075.stgit@warthog.procyon.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge tag 'cgroup-for-6.0-rc2-fixes-2' of git://git./linux/kernel/git/tj/cgroup

Pull another cgroup fix from Tejun Heo:
 "Commit 4f7e7236435c ("cgroup: Fix threadgroup_rwsem <->
  cpus_read_lock() deadlock") required the cgroup
  core to grab cpus_read_lock() before invoking ->attach().

  Unfortunately, it missed adding cpus_read_lock() in
  cgroup_attach_task_all(). Fix it"

* tag 'cgroup-for-6.0-rc2-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: Add missing cpus_read_lock() to cgroup_attach_task_all()

cgroup: Add missing cpus_read_lock() to cgroup_attach_task_all()

syzbot is hitting percpu_rwsem_assert_held(&cpu_hotplug_lock) warning at
cpuset_attach() [1], for commit 4f7e7236435ca0ab ("cgroup: Fix
threadgroup_rwsem <-> cpus_read_lock() deadlock") missed that
cpuset_attach() is also called from cgroup_attach_task_all().
Add cpus_read_lock() like what cgroup_procs_write_start() does.

Link: https://syzkaller.appspot.com/bug?extid=29d3a3b4d86c8136ad9e
Reported-by: syzbot <syzbot+29d3a3b4d86c8136ad9e@syzkaller.appspotmail.com>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Fixes: 4f7e7236435ca0ab ("cgroup: Fix threadgroup_rwsem <-> cpus_read_lock() deadlock")
Signed-off-by: Tejun Heo <tj@kernel.org>

net: ethernet: mtk_eth_soc: fix hw hash reporting for MTK_NETSYS_V2

Properly report hw rx hash for mt7986 chipset accroding to the new dma
descriptor layout.

Fixes: 197c9e9b17b11 ("net: ethernet: mtk_eth_soc: introduce support for mt7986 chipset")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://lore.kernel.org/r/091394ea4e705fbb35f828011d98d0ba33808f69.1661257293.git.lorenzo@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Merge git://git./linux/kernel/git/netfilter/nf

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

1) Fix crash with malformed ebtables blob which do not provide all
   entry points, from Florian Westphal.

2) Fix possible TCP connection clogging up with default 5-days
   timeout in conntrack, from Florian.

3) Fix crash in nf_tables tproxy with unsupported chains, also from Florian.

4) Do not allow to update implicit chains.

5) Make table handle allocation per-netns to fix data race.

6) Do not truncated payload length and offset, and checksum offset.
   Instead report EINVAl.

7) Enable chain stats update via static key iff no error occurs.

8) Restrict osf expression to ip, ip6 and inet families.

9) Restrict tunnel expression to netdev family.

10) Fix crash when trying to bind again an already bound chain.

11) Flowtable garbage collector might leave behind pending work to
    delete entries. This patch comes with a previous preparation patch
    as dependency.

12) Allow net.netfilter.nf_conntrack_frag6_high_thresh to be lowered,
    from Eric Dumazet.

* git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: nf_defrag_ipv6: allow nf_conntrack_frag6_high_thresh increases
  netfilter: flowtable: fix stuck flows on cleanup due to pending work
  netfilter: flowtable: add function to invoke garbage collection immediately
  netfilter: nf_tables: disallow binding to already bound chain
  netfilter: nft_tunnel: restrict it to netdev family
  netfilter: nft_osf: restrict osf to ipv4, ipv6 and inet families
  netfilter: nf_tables: do not leave chain stats enabled on error
  netfilter: nft_payload: do not truncate csum_offset and csum_type
  netfilter: nft_payload: report ERANGE for too long offset and length
  netfilter: nf_tables: make table handle allocation per-netns friendly
  netfilter: nf_tables: disallow updates of implicit chain
  netfilter: nft_tproxy: restrict to prerouting hook
  netfilter: conntrack: work around exceeded receive window
  netfilter: ebtables: reject blobs that don't provide all entry points
====================

Link: https://lore.kernel.org/r/20220824220330.64283-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

MAINTAINERS: rectify file entry in BONDING DRIVER

Commit c078290a2b76 ("selftests: include bonding tests into the kselftest
infra") adds the bonding tests in the directory:

  tools/testing/selftests/drivers/net/bonding/

The file entry in MAINTAINERS for the BONDING DRIVER however refers to:

  tools/testing/selftests/net/bonding/

Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about a
broken file pattern.

Repair this file entry in BONDING DRIVER.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Acked-by: Jonathan Toppins <jtoppins@redhat.com>
Link: https://lore.kernel.org/r/20220824072945.28606-1-lukas.bulwahn@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

i40e: Fix incorrect address type for IPv6 flow rules

It was not possible to create 1-tuple flow director
rule for IPv6 flow type. It was caused by incorrectly
checking for source IP address when validating user provided
destination IP address.

Fix this by changing ip6src to correct ip6dst address
in destination IP address validation for IPv6 flow type.

Fixes: efca91e89b67 ("i40e: Add flow director support for IPv6")
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

ixgbe: stop resetting SYSTIME in ixgbe_ptp_start_cyclecounter

The ixgbe_ptp_start_cyclecounter is intended to be called whenever the
cyclecounter parameters need to be changed.

Since commit a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x
devices"), this function has cleared the SYSTIME registers and reset the
TSAUXC DISABLE_SYSTIME bit.

While these need to be cleared during ixgbe_ptp_reset, it is wrong to clear
them during ixgbe_ptp_start_cyclecounter. This function may be called
during both reset and link status change. When link changes, the SYSTIME
counter is still operating normally, but the cyclecounter should be updated
to account for the possibly changed parameters.

Clearing SYSTIME when link changes causes the timecounter to jump because
the cycle counter now reads zero.

Extract the SYSTIME initialization out to a new function and call this
during ixgbe_ptp_reset. This prevents the timecounter adjustment and avoids
an unnecessary reset of the current time.

This also restores the original SYSTIME clearing that occurred during
ixgbe_ptp_reset before the commit above.

Reported-by: Steve Payne <spayne@aurora.tech>
Reported-by: Ilya Evenbach <ievenbach@aurora.tech>
Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

Merge tag 'trace-v6.0-rc2' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fix from Steven Rostedt:

 - Fix build warning for when MODULES and FTRACE_WITH_DIRECT_CALLS are
   not set. A warning happens with ops_references_rec() defined but not
   used.

* tag 'trace-v6.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  ftrace: Fix build warning for ops_references_rec() not used

Merge branch 'dmi-for-linus' of git://git./linux/kernel/git/jdelvare/staging

Pull dmi update from Jean Delvare.

Tiny cleanup.

* 'dmi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
  firmware: dmi: Use the proper accessor for the version field

Merge branch 'sysctl-data-races'

Kuniyuki Iwashima says:

====================
net: sysctl: Fix data-races around net.core.XXX

This series fixes data-races around all knobs in net_core_table and
netns_core_table except for bpf stuff.

These knobs are skipped:

  - 4 bpf knobs
  - netdev_rss_key: Written only once by net_get_random_once() and
                    read-only knob
  - rps_sock_flow_entries: Protected with sock_flow_mutex
  - flow_limit_cpu_bitmap: Protected with flow_limit_update_mutex
  - flow_limit_table_len: Protected with flow_limit_update_mutex
  - default_qdisc: Protected with qdisc_mod_lock
  - warnings: Unused
  - high_order_alloc_disable: Protected with static_key_mutex
  - skb_defer_max: Already using READ_ONCE()
  - sysctl_txrehash: Already using READ_ONCE()

Note 5th patch fixes net.core.message_cost and net.core.message_burst,
and lib/ratelimit.c does not have an explicit maintainer.

Changes:
  v3:
    * Fix build failures of CONFIG_SYSCTL=n case in 13th & 14th patches

  v2: https://lore.kernel.org/netdev/20220818035227.81567-1-kuniyu@amazon.com/
    * Remove 4 bpf knobs and added 6 knobs

  v1: https://lore.kernel.org/netdev/20220816052347.70042-1-kuniyu@amazon.com/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around sysctl_somaxconn.

While reading sysctl_somaxconn, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around netdev_unregister_timeout_secs.

While reading netdev_unregister_timeout_secs, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its reader.

Fixes: 5aa3afe107d9 ("net: make unregister netdev warning timeout configurable")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around gro_normal_batch.

While reading gro_normal_batch, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 323ebb61e32b ("net: use listified RX for handling GRO_NORMAL skbs")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around sysctl_devconf_inherit_init_net.

While reading sysctl_devconf_inherit_init_net, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its readers.

Fixes: 856c395cfa63 ("net: introduce a knob to control whether to inherit devconf config")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around sysctl_fb_tunnels_only_for_init_net.

While reading sysctl_fb_tunnels_only_for_init_net, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its readers.

Fixes: 79134e6ce2c9 ("net: do not create fallback tunnels for non-default namespaces")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around netdev_budget_usecs.

While reading netdev_budget_usecs, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 7acf8a1e8a28 ("Replace 2 jiffies with sysctl netdev_budget_usecs to enable softirq tuning")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around sysctl_max_skb_frags.

While reading sysctl_max_skb_frags, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its readers.

Fixes: 5f74f82ea34c ("net:Add sysctl_max_skb_frags")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around netdev_budget.

While reading netdev_budget, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 51b0bdedb8e7 ("[NET]: Separate two usages of netdev_max_backlog.")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around sysctl_net_busy_read.

While reading sysctl_net_busy_read, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 2d48d67fa8cd ("net: poll/select low latency socket support")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around sysctl_net_busy_poll.

While reading sysctl_net_busy_poll, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its reader.

Fixes: 060212928670 ("net: add low latency socket poll")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix a data-race around sysctl_tstamp_allow_data.

While reading sysctl_tstamp_allow_data, it can be changed
concurrently.  Thus, we need to add READ_ONCE() to its reader.

Fixes: b245be1f4db1 ("net-timestamp: no-payload only sysctl")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around sysctl_optmem_max.

While reading sysctl_optmem_max, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its readers.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ratelimit: Fix data-races in ___ratelimit().

While reading rs->interval and rs->burst, they can be changed
concurrently via sysctl (e.g. net_ratelimit_state).  Thus, we
need to add READ_ONCE() to their readers.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around netdev_tstamp_prequeue.

While reading netdev_tstamp_prequeue, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its readers.

Fixes: 3b098e2d7c69 ("net: Consistent skb timestamping")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around netdev_max_backlog.

While reading netdev_max_backlog, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its readers.

While at it, we remove the unnecessary spaces in the doc.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around weight_p and dev_weight_[rt]x_bias.

While reading weight_p, it can be changed concurrently.  Thus, we need
to add READ_ONCE() to its reader.

Also, dev_[rt]x_weight can be read/written at the same time.  So, we
need to use READ_ONCE() and WRITE_ONCE() for its access.  Moreover, to
use the same weight_p while changing dev_[rt]x_weight, we add a mutex
in proc_do_dev_weight().

Fixes: 3d48b53fb2ae ("net: dev_weight: TX/RX orthogonality")
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: Fix data-races around sysctl_[rw]mem_(max|default).

While reading sysctl_[rw]mem_(max|default), they can be changed
concurrently.  Thus, we need to add READ_ONCE() to its readers.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/core/skbuff: Check the return value of skb_copy_bits()

skb_copy_bits() could fail, which requires a check on the return
value.

Signed-off-by: Li Zhong <floridsleeves@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'master' of git://git./linux/kernel/git/klassert/ipsec

Steffen Klassert says:

====================
pull request (net): ipsec 2022-08-24

1) Fix a refcount leak in __xfrm_policy_check.
   From Xin Xiong.

2) Revert "xfrm: update SA curlft.use_time". This
   violates RFC 2367. From Antony Antony.

3) Fix a comment on XFRMA_LASTUSED.
   From Antony Antony.

4) x->lastused is not cloned in xfrm_do_migrate.
   Fix from Antony Antony.

5) Serialize the calls to xfrm_probe_algs.
   From Herbert Xu.

6) Fix a null pointer dereference of dst->dev on a metadata
   dst in xfrm_lookup_with_ifid. From Nikolay Aleksandrov.

Please pull or let me know if there are problems.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

fec: Restart PPS after link state change

On link state change, the controller gets reset,
causing PPS to drop out and the PHC to lose its
time and calibration. So we restart it if needed,
restoring calibration and time registers.

Changes since v2:
* Add `fec_ptp_save_state()`/`fec_ptp_restore_state()`
* Use `ktime_get_real_ns()`
* Use `BIT()` macro
Changes since v1:
* More ECR #define's
* Stop PPS in `fec_ptp_stop()`

Signed-off-by: Csókás Bence <csokas.bence@prolan.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>

net: neigh: don't call kfree_skb() under spin_lock_irqsave()

It is not allowed to call kfree_skb() from hardware interrupt
context or with interrupts being disabled. So add all skb to
a tmp list, then free them after spin_unlock_irqrestore() at
once.

Fixes: 66ba215cb513 ("neigh: fix possible DoS due to net iface start/stop loop")
Suggested-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

netfilter: nf_defrag_ipv6: allow nf_conntrack_frag6_high_thresh increases

Currently, net.netfilter.nf_conntrack_frag6_high_thresh can only be lowered.

I found this issue while investigating a probable kernel issue
causing flakes in tools/testing/selftests/net/ip_defrag.sh

In particular, these sysctl changes were ignored:
ip netns exec "${NETNS}" sysctl -w net.netfilter.nf_conntrack_frag6_high_thresh=9000000 >/dev/null 2>&1
ip netns exec "${NETNS}" sysctl -w net.netfilter.nf_conntrack_frag6_low_thresh=7000000  >/dev/null 2>&1

This change is inline with commit 836196239298 ("net/ipfrag: let ip[6]frag_high_thresh
in ns be higher than in init_net")

Fixes: 8db3d41569bb ("netfilter: nf_defrag_ipv6: use net_generic infra")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: flowtable: fix stuck flows on cleanup due to pending work

To clear the flow table on flow table free, the following sequence
normally happens in order:

  1) gc_step work is stopped to disable any further stats/del requests.
  2) All flow table entries are set to teardown state.
  3) Run gc_step which will queue HW del work for each flow table entry.
  4) Waiting for the above del work to finish (flush).
  5) Run gc_step again, deleting all entries from the flow table.
  6) Flow table is freed.

But if a flow table entry already has pending HW stats or HW add work
step 3 will not queue HW del work (it will be skipped), step 4 will wait
for the pending add/stats to finish, and step 5 will queue HW del work
which might execute after freeing of the flow table.

To fix the above, this patch flushes the pending work, then it sets the
teardown flag to all flows in the flowtable and it forces a garbage
collector run to queue work to remove the flows from hardware, then it
flushes this new pending work and (finally) it forces another garbage
collector run to remove the entry from the software flowtable.

Stack trace:
[47773.882335] BUG: KASAN: use-after-free in down_read+0x99/0x460
[47773.883634] Write of size 8 at addr ffff888103b45aa8 by task kworker/u20:6/543704
[47773.885634] CPU: 3 PID: 543704 Comm: kworker/u20:6 Not tainted 5.12.0-rc7+ #2
[47773.886745] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009)
[47773.888438] Workqueue: nf_ft_offload_del flow_offload_work_handler [nf_flow_table]
[47773.889727] Call Trace:
[47773.890214]  dump_stack+0xbb/0x107
[47773.890818]  print_address_description.constprop.0+0x18/0x140
[47773.892990]  kasan_report.cold+0x7c/0xd8
[47773.894459]  kasan_check_range+0x145/0x1a0
[47773.895174]  down_read+0x99/0x460
[47773.899706]  nf_flow_offload_tuple+0x24f/0x3c0 [nf_flow_table]
[47773.907137]  flow_offload_work_handler+0x72d/0xbe0 [nf_flow_table]
[47773.913372]  process_one_work+0x8ac/0x14e0
[47773.921325]
[47773.921325] Allocated by task 592159:
[47773.922031]  kasan_save_stack+0x1b/0x40
[47773.922730]  __kasan_kmalloc+0x7a/0x90
[47773.923411]  tcf_ct_flow_table_get+0x3cb/0x1230 [act_ct]
[47773.924363]  tcf_ct_init+0x71c/0x1156 [act_ct]
[47773.925207]  tcf_action_init_1+0x45b/0x700
[47773.925987]  tcf_action_init+0x453/0x6b0
[47773.926692]  tcf_exts_validate+0x3d0/0x600
[47773.927419]  fl_change+0x757/0x4a51 [cls_flower]
[47773.928227]  tc_new_tfilter+0x89a/0x2070
[47773.936652]
[47773.936652] Freed by task 543704:
[47773.937303]  kasan_save_stack+0x1b/0x40
[47773.938039]  kasan_set_track+0x1c/0x30
[47773.938731]  kasan_set_free_info+0x20/0x30
[47773.939467]  __kasan_slab_free+0xe7/0x120
[47773.940194]  slab_free_freelist_hook+0x86/0x190
[47773.941038]  kfree+0xce/0x3a0
[47773.941644]  tcf_ct_flow_table_cleanup_work

Original patch description and stack trace by Paul Blakey.

Fixes: c29f74e0df7a ("netfilter: nf_flow_table: hardware offload support")
Reported-by: Paul Blakey <paulb@nvidia.com>
Tested-by: Paul Blakey <paulb@nvidia.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: flowtable: add function to invoke garbage collection immediately

Expose nf_flow_table_gc_run() to force a garbage collector run from the
offload infrastructure.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nf_tables: disallow binding to already bound chain

Update nft_data_init() to report EINVAL if chain is already bound.

Fixes: d0e2c7de92c7 ("netfilter: nf_tables: add NFT_CHAIN_BINDING")
Reported-by: Gwangun Jung <exsociety@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nft_tunnel: restrict it to netdev family

Only allow to use this expression from NFPROTO_NETDEV family.

Fixes: af308b94a2a4 ("netfilter: nf_tables: add tunnel support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nft_osf: restrict osf to ipv4, ipv6 and inet families

As it was originally intended, restrict extension to supported families.

Fixes: b96af92d6eaf ("netfilter: nf_tables: implement Passive OS fingerprint module in nft_osf")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nf_tables: do not leave chain stats enabled on error

Error might occur later in the nf_tables_addchain() codepath, enable
static key only after transaction has been created.

Fixes: 9f08ea848117 ("netfilter: nf_tables: keep chain counters away from hot path")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nft_payload: do not truncate csum_offset and csum_type

Instead report ERANGE if csum_offset is too long, and EOPNOTSUPP if type
is not support.

Fixes: 7ec3f7b47b8d ("netfilter: nft_payload: add packet mangling support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nft_payload: report ERANGE for too long offset and length

Instead of offset and length are truncation to u8, report ERANGE.

Fixes: 96518518cc41 ("netfilter: add nftables")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nf_tables: make table handle allocation per-netns friendly

mutex is per-netns, move table_netns to the pernet area.

*read-write* to 0xffffffff883a01e8 of 8 bytes by task 6542 on cpu 0:
 nf_tables_newtable+0x6dc/0xc00 net/netfilter/nf_tables_api.c:1221
 nfnetlink_rcv_batch net/netfilter/nfnetlink.c:513 [inline]
 nfnetlink_rcv_skb_batch net/netfilter/nfnetlink.c:634 [inline]
 nfnetlink_rcv+0xa6a/0x13a0 net/netfilter/nfnetlink.c:652
 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline]
 netlink_unicast+0x652/0x730 net/netlink/af_netlink.c:1345
 netlink_sendmsg+0x643/0x740 net/netlink/af_netlink.c:1921

Fixes: f102d66b335a ("netfilter: nf_tables: use dedicated mutex to guard transactions")
Reported-by: Abhishek Shah <abhishek.shah@columbia.edu>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

netfilter: nf_tables: disallow updates of implicit chain

Updates on existing implicit chain make no sense, disallow this.

Fixes: d0e2c7de92c7 ("netfilter: nf_tables: add NFT_CHAIN_BINDING")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

Merge tag 'cgroup-for-6.0-rc2-fixes' of git://git./linux/kernel/git/tj/cgroup

Pull cgroup fixes from Tejun Heo:

 - The psi data structure was changed to be allocated dynamically but
   it wasn't being cleared leading to it reporting garbage values and
   triggering spurious oom kills.

 - A deadlock involving cpuset and cpu hotplug.

 - When a controller is moved across cgroup hierarchies,
   css->rstat_css_node didn't get RCU drained properly from the previous
   list.

* tag 'cgroup-for-6.0-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: Fix race condition at rebind_subsystems()
  cgroup: Fix threadgroup_rwsem <-> cpus_read_lock() deadlock
  sched/psi: Remove redundant cgroup_psi() when !CONFIG_CGROUPS
  sched/psi: Remove unused parameter nbytes of psi_trigger_create()
  sched/psi: Zero the memory of struct psi_group

Merge tag 'audit-pr-20220823' of git://git./linux/kernel/git/pcmoore/audit

Pull audit fix from Paul Moore:
 "A single fix for a potential double-free on a fsnotify error path"

* tag 'audit-pr-20220823' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:
  audit: fix potential double free on error path from fsnotify_add_inode_mark

Merge tag 'fs.fixes.v6.0-rc3' of git://git./linux/kernel/git/vfs/idmapping

Pull file_remove_privs() fix from Christian Brauner:
 "As part of Stefan's and Jens' work to add async buffered write
  support to xfs we refactored file_remove_privs() and added
  __file_remove_privs() to avoid calling __remove_privs() when
  IOCB_NOWAIT is passed.

  While debugging a recent performance regression report I found that
  during review we missed that commit faf99b563558 ("fs: add
  __remove_file_privs() with flags parameter") accidently changed
  behavior when dentry_needs_remove_privs() returns zero.

  Before the commit it would still call inode_has_no_xattr() setting
  the S_NOSEC bit and thereby avoiding even calling into
  dentry_needs_remove_privs() the next time this function is called.
  After that commit inode_has_no_xattr() would only be called if
  __remove_privs() had to be called.

  Restore the old behavior. This is likely the cause of the performance
  regression"

* tag 'fs.fixes.v6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/idmapping:
  fs: __file_remove_privs(): restore call to inode_has_no_xattr()

Merge tag 'mlx5-fixes-2022-08-22' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5 fixes 2022-08-22

This series provides bug fixes to mlx5 driver.

* tag 'mlx5-fixes-2022-08-22' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5: Unlock on error in mlx5_sriov_enable()
  net/mlx5e: Fix use after free in mlx5e_fs_init()
  net/mlx5e: kTLS, Use _safe() iterator in mlx5e_tls_priv_tx_list_cleanup()
  net/mlx5: unlock on error path in esw_vfs_changed_event_handler()
  net/mlx5e: Fix wrong tc flag used when set hw-tc-offload off
  net/mlx5e: TC, Add missing policer validation
  net/mlx5e: Fix wrong application of the LRO state
  net/mlx5: Avoid false positive lockdep warning by adding lock_class_key
  net/mlx5: Fix cmd error logging for manage pages cmd
  net/mlx5: Disable irq when locking lag_lock
  net/mlx5: Eswitch, Fix forwarding decision to uplink
  net/mlx5: LAG, fix logic over MLX5_LAG_FLAG_NDEVS_READY
  net/mlx5e: Properly disable vlan strip on non-UL reps
====================

Link: https://lore.kernel.org/r/20220822195917.216025-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch '100GbE' of git://git./linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
ice: xsk: reduced queue count fixes

Maciej Fijalkowski says:

this small series is supposed to fix the issues around AF_XDP usage with
reduced queue count on interface. Due to the XDP rings setup, some
configurations can result in sockets not seeing traffic flowing. More
about this in description of patch 2.

* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
  ice: xsk: use Rx ring's XDP ring when picking NAPI context
  ice: xsk: prohibit usage of non-balanced queue id
====================

Link: https://lore.kernel.org/r/20220822163257.2382487-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'linus'

Merge branch 'bnxt_en-bug-fixes'

Michael Chan says:

====================
bnxt_en: Bug fixes

This series includes 2 fixes for regressions introduced by the XDP
multi-buffer feature, 1 devlink reload bug fix, and 1 SRIOV resource
accounting bug fix.
====================

Link: https://lore.kernel.org/r/1661180814-19350-1-git-send-email-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bnxt_en: fix LRO/GRO_HW features in ndo_fix_features callback

LRO/GRO_HW should be disabled if there is an attached XDP program.
BNXT_FLAG_TPA is the current setting of the LRO/GRO_HW.  Using
BNXT_FLAG_TPA to disable LRO/GRO_HW will cause these features to be
permanently disabled once they are disabled.

Fixes: 1dc4c557bfed ("bnxt: adding bnxt_xdp_build_skb to build skb from multibuffer xdp_buff")
Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bnxt_en: fix NQ resource accounting during vf creation on 57500 chips

There are 2 issues:

1. We should decrement hw_resc->max_nqs instead of hw_resc->max_irqs
   with the number of NQs assigned to the VFs.  The IRQs are fixed
   on each function and cannot be re-assigned.  Only the NQs are being
   assigned to the VFs.

2. vf_msix is the total number of NQs to be assigned to the VFs.  So
   we should decrement vf_msix from hw_resc->max_nqs.

Fixes: b16b68918674 ("bnxt_en: Add SR-IOV support for 57500 chips.")
Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bnxt_en: set missing reload flag in devlink features

Add missing devlink_set_features() API for callbacks reload_down
and reload_up to function.

Fixes: 228ea8c187d8 ("bnxt_en: implement devlink dev reload driver_reinit")
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bnxt_en: Use PAGE_SIZE to init buffer when multi buffer XDP is not in use

Using BNXT_PAGE_MODE_BUF_SIZE + offset as buffer length value is not
sufficient when running single buffer XDP programs doing redirect
operations. The stack will complain on missing skb tail room. Fix it
by using PAGE_SIZE when calling xdp_init_buff() for single buffer
programs.

Fixes: b231c3f3414c ("bnxt: refactor bnxt_rx_xdp to separate xdp_init_buff/xdp_prepare_buff")
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: microchip: make learning configurable and keep it off while standalone

Address learning should initially be turned off by the driver for port
operation in standalone mode, then the DSA core handles changes to it
via ds->ops->port_bridge_flags().

Leaving address learning enabled while ports are standalone breaks any
kind of communication which involves port B receiving what port A has
sent. Notably it breaks the ksz9477 driver used with a (non offloaded,
ports act as if standalone) bonding interface in active-backup mode,
when the ports are connected together through external switches, for
redundancy purposes.

This fixes a major design flaw in the ksz9477 and ksz8795 drivers, which
unconditionally leave address learning enabled even while ports operate
as standalone.

Fixes: b987e98e50ab ("dsa: add DSA switch driver for Microchip KSZ9477")
Link: https://lore.kernel.org/netdev/CAFZh4h-JVWt80CrQWkFji7tZJahMfOToUJQgKS5s0_=9zzpvYQ@mail.gmail.com/
Reported-by: Brian Hutchinson <b.hutchman@gmail.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20220818164809.3198039-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge tag 'parisc-for-6.0-2' of git://git./linux/kernel/git/deller/parisc-linux

Pull parisc fixes from Helge Deller:
 "Some interesting background to the current patchset:

  It turned out that the fldw instruction (which loads a 32-bit word
  from memory into one half of a FP register) failed on unaligned
  addresses and even trashed some other random FP register instead. It's
  a trivial one-liner fix in the exception handler but this failure
  dates back to the very beginnings of the parisc-port. It's strange
  that it was never noticed before.

  Another patch fixes an annoyance noticed by Randy Dunlap. Running
  "make ARCH=parisc64 randconfig" always returned a 32-bit config,
  although one would expect a 64-bit config. Masahiro Yamada suggested
  to mimik sparc Kconfig code, which fixed the issue nicely. This
  allowed to drop some compiler build checks too.

  Third, it's possible to build an optimized 32-bit kernel for PA8X00
  (64-bit) CPUs, which then wouldn't start on 32-bit-only (PA1.x)
  machines. I've added a bootup check which prevents that and which
  prints a message to the console. This can be tested with qemu, which
  currently only supports 32-bit emulation.

  The other patches are usual clean-up stuff like added return value
  checks and typo fixes in comments.

  Summary:

   - Fix emulation of fldw instruction on unaligned addresses

   - Fix "make ARCH=parisc64 randconfig" to return a 64-bit config

   - Prevent boot if trying to boot a 32-bit kernel compiled for PA8X00
     CPUs on 32-bit only machines

   - ccio-dma: Handle kmalloc failure in ccio_init_resources()"

* tag 'parisc-for-6.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Add runtime check to prevent PA2.0 kernels on PA1.x machines
  parisc: ccio-dma: Handle kmalloc failure in ccio_init_resources()
  parisc: led: Move from strlcpy with unused retval to strscpy
  parisc: ccio-dma: Fix typo in comment
  Revert "parisc: Show error if wrong 32/64-bit compiler is being used"
  parisc: Make CONFIG_64BIT available for ARCH=parisc64 only
  parisc: Fix exception handler for fldw and fstw instructions

Merge tag 'mm-hotfixes-stable-2022-08-22' of git://git./linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "Thirteen fixes, almost all for MM.

  Seven of these are cc:stable and the remainder fix up the changes
  which went into this -rc cycle"

* tag 'mm-hotfixes-stable-2022-08-22' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  kprobes: don't call disarm_kprobe() for disabled kprobes
  mm/shmem: shmem_replace_page() remember NR_SHMEM
  mm/shmem: tmpfs fallocate use file_modified()
  mm/shmem: fix chattr fsflags support in tmpfs
  mm/hugetlb: support write-faults in shared mappings
  mm/hugetlb: fix hugetlb not supporting softdirty tracking
  mm/uffd: reset write protection when unregister with wp-mode
  mm/smaps: don't access young/dirty bit if pte unpresent
  mm: add DEVICE_ZONE to FOR_ALL_ZONES
  kernel/sys_ni: add compat entry for fadvise64_64
  mm/gup: fix FOLL_FORCE COW security issue and remove FOLL_COW
  Revert "zram: remove double compression logic"
  get_maintainer: add Alan to .get_maintainer.ignore

Merge tag 'linux-kselftest-kunit-fixes-6.0-rc3' of git://git./linux/kernel/git/shuah/linux-kselftest

Pull KUnit fixes from Shuah Khan:
 "Fix for a mmc test and to load .kunit_test_suites section when
  CONFIG_KUNIT=m, and not just when KUnit is built-in"

* tag 'linux-kselftest-kunit-fixes-6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
  module: kunit: Load .kunit_test_suites section when CONFIG_KUNIT=m
  mmc: sdhci-of-aspeed: test: Fix dependencies when KUNIT=m

Merge tag 'linux-kselftest-fixes-6.0-rc3' of git://git./linux/kernel/git/shuah/linux-kselftest

Pull Kselftest fixes from Shuah Khan:
 "Fixes to vm and sgx test builds"

* tag 'linux-kselftest-fixes-6.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
  selftests/vm: fix inability to build any vm tests
  selftests/sgx: Ignore OpenSSL 3.0 deprecated functions warning

netfilter: nft_tproxy: restrict to prerouting hook

TPROXY is only allowed from prerouting, but nft_tproxy doesn't check this.
This fixes a crash (null dereference) when using tproxy from e.g. output.

Fixes: 4ed8eb6570a4 ("netfilter: nf_tables: Add native tproxy support")
Reported-by: Shell Chen <xierch@gmail.com>
Signed-off-by: Florian Westphal <fw@strlen.de>

cgroup: Fix race condition at rebind_subsystems()

Root cause:
The rebind_subsystems() is no lock held when move css object from A
list to B list,then let B's head be treated as css node at
list_for_each_entry_rcu().

Solution:
Add grace period before invalidating the removed rstat_css_node.

Reported-by: Jing-Ting Wu <jing-ting.wu@mediatek.com>
Suggested-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Jing-Ting Wu <jing-ting.wu@mediatek.com>
Tested-by: Jing-Ting Wu <jing-ting.wu@mediatek.com>
Link: https://lore.kernel.org/linux-arm-kernel/d8f0bc5e2fb6ed259f9334c83279b4c011283c41.camel@mediatek.com/T/
Acked-by: Mukesh Ojha <quic_mojha@quicinc.com>
Fixes: a7df69b81aac ("cgroup: rstat: support cgroup1")
Cc: stable@vger.kernel.org # v5.13+
Signed-off-by: Tejun Heo <tj@kernel.org>

netfilter: conntrack: work around exceeded receive window

When a TCP sends more bytes than allowed by the receive window, all future
packets can be marked as invalid.
This can clog up the conntrack table because of 5-day default timeout.

Sequence of packets:
 01 initiator > responder: [S], seq 171, win 5840, options [mss 1330,sackOK,TS val 63 ecr 0,nop,wscale 1]
 02 responder > initiator: [S.], seq 33211, ack 172, win 65535, options [mss 1460,sackOK,TS val 010 ecr 63,nop,wscale 8]
 03 initiator > responder: [.], ack 33212, win 2920, options [nop,nop,TS val 068 ecr 010], length 0
 04 initiator > responder: [P.], seq 172:240, ack 33212, win 2920, options [nop,nop,TS val 279 ecr 010], length 68

Window is 5840 starting from 33212 -> 39052.

 05 responder > initiator: [.], ack 240, win 256, options [nop,nop,TS val 872 ecr 279], length 0
 06 responder > initiator: [.], seq 33212:34530, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 1318

This is fine, conntrack will flag the connection as having outstanding
data (UNACKED), which lowers the conntrack timeout to 300s.

 07 responder > initiator: [.], seq 34530:35848, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 1318
 08 responder > initiator: [.], seq 35848:37166, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 1318
 09 responder > initiator: [.], seq 37166:38484, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 1318
 10 responder > initiator: [.], seq 38484:39802, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 1318

Packet 10 is already sending more than permitted, but conntrack doesn't
validate this (only seq is tested vs. maxend, not 'seq+len').

38484 is acceptable, but only up to 39052, so this packet should
not have been sent (or only 568 bytes, not 1318).

At this point, connection is still in '300s' mode.

Next packet however will get flagged:
 11 responder > initiator: [P.], seq 39802:40128, ack 240, win 256, options [nop,nop,TS val 892 ecr 279], length 326

nf_ct_proto_6: SEQ is over the upper bound (over the window of the receiver) .. LEN=378 .. SEQ=39802 ACK=240 ACK PSH ..

Now, a couple of replies/acks comes in:

 12 initiator > responder: [.], ack 34530, win 4368,
[.. irrelevant acks removed ]
 16 initiator > responder: [.], ack 39802, win 8712, options [nop,nop,TS val 296201291 ecr 2982371892], length 0

This ack is significant -- this acks the last packet send by the
responder that conntrack considered valid.

This means that ack == td_end.  This will withdraw the
'unacked data' flag, the connection moves back to the 5-day timeout
of established conntracks.

 17 initiator > responder: ack 40128, win 10030, ...

This packet is also flagged as invalid.

Because conntrack only updates state based on packets that are
considered valid, packet 11 'did not exist' and that gets us:

nf_ct_proto_6: ACK is over upper bound 39803 (ACKed data not seen yet) .. SEQ=240 ACK=40128 WINDOW=10030 RES=0x00 ACK URG

Because this received and processed by the endpoints, the conntrack entry
remains in a bad state, no packets will ever be considered valid again:

 30 responder > initiator: [F.], seq 40432, ack 2045, win 391, ..
 31 initiator > responder: [.], ack 40433, win 11348, ..
 32 initiator > responder: [F.], seq 2045, ack 40433, win 11348 ..

... all trigger 'ACK is over bound' test and we end up with
non-early-evictable 5-day default timeout.

NB: This patch triggers a bunch of checkpatch warnings because of silly
indent.  I will resend the cleanup series linked below to reduce the
indent level once this change has propagated to net-next.

I could route the cleanup via nf but that causes extra backport work for
stable maintainers.

Link: https://lore.kernel.org/netfilter-devel/20220720175228.17880-1-fw@strlen.de/T/#mb1d7147d36294573cc4f81d00f9f8dadfdd06cd8
Signed-off-by: Florian Westphal <fw@strlen.de>

netfilter: ebtables: reject blobs that don't provide all entry points

Harshit Mogalapalli says:
 In ebt_do_table() function dereferencing 'private->hook_entry[hook]'
 can lead to NULL pointer dereference. [..] Kernel panic:

general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
[..]
RIP: 0010:ebt_do_table+0x1dc/0x1ce0
Code: 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 5c 16 00 00 48 b8 00 00 00 00 00 fc ff df 49 8b 6c df 08 48 8d 7d 2c 48 89 fa 48 c1 ea 03 <0f> b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 88
[..]
Call Trace:
 nf_hook_slow+0xb1/0x170
 __br_forward+0x289/0x730
 maybe_deliver+0x24b/0x380
 br_flood+0xc6/0x390
 br_dev_xmit+0xa2e/0x12c0

For some reason ebtables rejects blobs that provide entry points that are
not supported by the table, but what it should instead reject is the
opposite: blobs that DO NOT provide an entry point supported by the table.

t->valid_hooks is the bitmask of hooks (input, forward ...) that will see
packets.  Providing an entry point that is not support is harmless
(never called/used), but the inverse isn't: it results in a crash
because the ebtables traverser doesn't expect a NULL blob for a location
its receiving packets for.

Instead of fixing all the individual checks, do what iptables is doing and
reject all blobs that differ from the expected hooks.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Reported-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Reported-by: syzkaller <syzkaller@googlegroups.com>
Signed-off-by: Florian Westphal <fw@strlen.de>

net: dsa: don't dereference NULL extack in dsa_slave_changeupper()

When a driver returns -EOPNOTSUPP in dsa_port_bridge_join() but failed
to provide a reason for it, DSA attempts to set the extack to say that
software fallback will kick in.

The problem is, when we use brctl and the legacy bridge ioctls, the
extack will be NULL, and DSA dereferences it in the process of setting
it.

Sergei Antonov proves this using the following stack trace:

Unable to handle kernel NULL pointer dereference at virtual address 00000000
PC is at dsa_slave_changeupper+0x5c/0x158

 dsa_slave_changeupper from raw_notifier_call_chain+0x38/0x6c
 raw_notifier_call_chain from __netdev_upper_dev_link+0x198/0x3b4
 __netdev_upper_dev_link from netdev_master_upper_dev_link+0x50/0x78
 netdev_master_upper_dev_link from br_add_if+0x430/0x7f4
 br_add_if from br_ioctl_stub+0x170/0x530
 br_ioctl_stub from br_ioctl_call+0x54/0x7c
 br_ioctl_call from dev_ifsioc+0x4e0/0x6bc
 dev_ifsioc from dev_ioctl+0x2f8/0x758
 dev_ioctl from sock_ioctl+0x5f0/0x674
 sock_ioctl from sys_ioctl+0x518/0xe40
 sys_ioctl from ret_fast_syscall+0x0/0x1c

Fix the problem by only overriding the extack if non-NULL.

Fixes: 1c6e8088d9a7 ("net: dsa: allow port_bridge_join() to override extack message")
Link: https://lore.kernel.org/netdev/CABikg9wx7vB5eRDAYtvAm7fprJ09Ta27a4ZazC=NX5K4wn6pWA@mail.gmail.com/
Reported-by: Sergei Antonov <saproj@gmail.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Sergei Antonov <saproj@gmail.com>
Link: https://lore.kernel.org/r/20220819173925.3581871-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ipvtap - add __init/__exit annotations to module init/exit funcs

Looks to have been left out in an oversight.

Cc: Mahesh Bandewar <maheshb@google.com>
Cc: Sainath Grandhi <sainath.grandhi@intel.com>
Fixes: 235a9d89da97 ('ipvtap: IP-VLAN based tap driver')
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Link: https://lore.kernel.org/r/20220821130808.12143-1-zenczykowski@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

Merge branch 'bonding-802-3ad-fix-no-transmission-of-lacpdus'

Jonathan Toppins says:

====================
bonding: 802.3ad: fix no transmission of LACPDUs

Configuring a bond in a specific order can leave the bond in a state
where it never transmits LACPDUs.

The first patch adds some kselftest infrastructure and the reproducer
that demonstrates the problem. The second patch fixes the issue. The
new third patch makes ad_ticks_per_sec a static const and removes the
passing of this variable via the stack.
====================

Link: https://lore.kernel.org/r/cover.1660919940.git.jtoppins@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bonding: 3ad: make ad_ticks_per_sec a const

The value is only ever set once in bond_3ad_initialize and only ever
read otherwise. There seems to be no reason to set the variable via
bond_3ad_initialize when setting the global variable will do. Change
ad_ticks_per_sec to a const to enforce its read-only usage.

Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

bonding: 802.3ad: fix no transmission of LACPDUs

This is caused by the global variable ad_ticks_per_sec being zero as
demonstrated by the reproducer script discussed below. This causes
all timer values in __ad_timer_to_ticks to be zero, resulting
in the periodic timer to never fire.

To reproduce:
Run the script in
`tools/testing/selftests/drivers/net/bonding/bond-break-lacpdu-tx.sh` which
puts bonding into a state where it never transmits LACPDUs.

line 44: ip link add fbond type bond mode 4 miimon 200 \
            xmit_hash_policy 1 ad_actor_sys_prio 65535 lacp_rate fast
setting bond param: ad_actor_sys_prio
given:
    params.ad_actor_system = 0
call stack:
    bond_option_ad_actor_sys_prio()
    -> bond_3ad_update_ad_actor_settings()
       -> set ad.system.sys_priority = bond->params.ad_actor_sys_prio
       -> ad.system.sys_mac_addr = bond->dev->dev_addr; because
            params.ad_actor_system == 0
results:
     ad.system.sys_mac_addr = bond->dev->dev_addr

line 48: ip link set fbond address 52:54:00:3B:7C:A6
setting bond MAC addr
call stack:
    bond->dev->dev_addr = new_mac

line 52: ip link set fbond type bond ad_actor_sys_prio 65535
setting bond param: ad_actor_sys_prio
given:
    params.ad_actor_system = 0
call stack:
    bond_option_ad_actor_sys_prio()
    -> bond_3ad_update_ad_actor_settings()
       -> set ad.system.sys_priority = bond->params.ad_actor_sys_prio
       -> ad.system.sys_mac_addr = bond->dev->dev_addr; because
            params.ad_actor_system == 0
results:
     ad.system.sys_mac_addr = bond->dev->dev_addr

line 60: ip link set veth1-bond down master fbond
given:
    params.ad_actor_system = 0
    params.mode = BOND_MODE_8023AD
    ad.system.sys_mac_addr == bond->dev->dev_addr
call stack:
    bond_enslave
    -> bond_3ad_initialize(); because first slave
       -> if ad.system.sys_mac_addr != bond->dev->dev_addr
          return
results:
     Nothing is run in bond_3ad_initialize() because dev_addr equals
     sys_mac_addr leaving the global ad_ticks_per_sec zero as it is
     never initialized anywhere else.

The if check around the contents of bond_3ad_initialize() is no longer
needed due to commit 5ee14e6d336f ("bonding: 3ad: apply ad_actor settings
changes immediately") which sets ad.system.sys_mac_addr if any one of
the bonding parameters whos set function calls
bond_3ad_update_ad_actor_settings(). This is because if
ad.system.sys_mac_addr is zero it will be set to the current bond mac
address, this causes the if check to never be true.

Fixes: 5ee14e6d336f ("bonding: 3ad: apply ad_actor settings changes immediately")
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: include bonding tests into the kselftest infra

This creates a test collection in drivers/net/bonding for bonding
specific kernel selftests.

The first test is a reproducer that provisions a bond and given the
specific order in how the ip-link(8) commands are issued the bond never
transmits an LACPDU frame on any of its slaves.

Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: moxa: get rid of asymmetry in DMA mapping/unmapping

Since priv->rx_mapping[i] is maped in moxart_mac_open(), we
should unmap it from moxart_mac_stop(). Fixes 2 warnings.

1. During error unwinding in moxart_mac_probe(): "goto init_fail;",
then moxart_mac_free_memory() calls dma_unmap_single() with
priv->rx_mapping[i] pointers zeroed.

WARNING: CPU: 0 PID: 1 at kernel/dma/debug.c:963 check_unmap+0x704/0x980
DMA-API: moxart-ethernet 92000000.mac: device driver tries to free DMA memory it has not allocated [device address=0x0000000000000000] [size=1600 bytes]
CPU: 0 PID: 1 Comm: swapper Not tainted 5.19.0+ #60
Hardware name: Generic DT based system
 unwind_backtrace from show_stack+0x10/0x14
 show_stack from dump_stack_lvl+0x34/0x44
 dump_stack_lvl from __warn+0xbc/0x1f0
 __warn from warn_slowpath_fmt+0x94/0xc8
 warn_slowpath_fmt from check_unmap+0x704/0x980
 check_unmap from debug_dma_unmap_page+0x8c/0x9c
 debug_dma_unmap_page from moxart_mac_free_memory+0x3c/0xa8
 moxart_mac_free_memory from moxart_mac_probe+0x190/0x218
 moxart_mac_probe from platform_probe+0x48/0x88
 platform_probe from really_probe+0xc0/0x2e4

2. After commands:
 ip link set dev eth0 down
 ip link set dev eth0 up

WARNING: CPU: 0 PID: 55 at kernel/dma/debug.c:570 add_dma_entry+0x204/0x2ec
DMA-API: moxart-ethernet 92000000.mac: cacheline tracking EEXIST, overlapping mappings aren't supported
CPU: 0 PID: 55 Comm: ip Not tainted 5.19.0+ #57
Hardware name: Generic DT based system
 unwind_backtrace from show_stack+0x10/0x14
 show_stack from dump_stack_lvl+0x34/0x44
 dump_stack_lvl from __warn+0xbc/0x1f0
 __warn from warn_slowpath_fmt+0x94/0xc8
 warn_slowpath_fmt from add_dma_entry+0x204/0x2ec
 add_dma_entry from dma_map_page_attrs+0x110/0x328
 dma_map_page_attrs from moxart_mac_open+0x134/0x320
 moxart_mac_open from __dev_open+0x11c/0x1ec
 __dev_open from __dev_change_flags+0x194/0x22c
 __dev_change_flags from dev_change_flags+0x14/0x44
 dev_change_flags from devinet_ioctl+0x6d4/0x93c
 devinet_ioctl from inet_ioctl+0x1ac/0x25c

v1 -> v2:
Extraneous change removed.

Fixes: 6c821bd9edc9 ("net: Add MOXA ART SoCs ethernet driver")
Signed-off-by: Sergei Antonov <saproj@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220819110519.1230877-1-saproj@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: Don't WARN for PHY_READY state in mdio_bus_phy_resume()

For some MAC drivers, they set the mac_managed_pm to true in its
->ndo_open() callback. So before the mac_managed_pm is set to true,
we still want to leverage the mdio_bus_phy_suspend()/resume() for
the phy device suspend and resume. In this case, the phy device is
in PHY_READY, and we shouldn't warn about this. It also seems that
the check of mac_managed_pm in WARN_ON is redundant since we already
check this in the entry of mdio_bus_phy_resume(), so drop it.

Fixes: 744d23c71af3 ("net: phy: Warn about incorrect mdio_bus_phy_resume() state")
Signed-off-by: Xiaolei Wang <xiaolei.wang@windriver.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20220819082451.1992102-1-xiaolei.wang@windriver.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ipa: don't assume SMEM is page-aligned

In ipa_smem_init(), a Qualcomm SMEM region is allocated (if needed)
and then its virtual address is fetched using qcom_smem_get().  The
physical address associated with that region is also fetched.

The physical address is adjusted so that it is page-aligned, and an
attempt is made to update the size of the region to compensate for
any non-zero adjustment.

But that adjustment isn't done properly.  The physical address is
aligned twice, and as a result the size is never actually adjusted.

Fix this by *not* aligning the "addr" local variable, and instead
making the "phys" local variable be the adjusted "addr" value.

Fixes: a0036bb413d5b ("net: ipa: define SMEM memory region for IPA")
Signed-off-by: Alex Elder <elder@linaro.org>
Link: https://lore.kernel.org/r/20220818134206.567618-1-elder@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: microchip: keep compatibility with device tree blobs with no phy-mode

DSA has multiple ways of specifying a MAC connection to an internal PHY.
One requires a DT description like this:

port@0 {
reg = <0>;
phy-handle = <&internal_phy>;
phy-mode = "internal";
};

(which is IMO the recommended approach, as it is the clearest
description)

but it is also possible to leave the specification as just:

port@0 {
reg = <0>;
}

and if the driver implements ds->ops->phy_read and ds->ops->phy_write,
the DSA framework "knows" it should create a ds->slave_mii_bus, and it
should connect to a non-OF-based internal PHY on this MDIO bus, at an
MDIO address equal to the port address.

There is also an intermediary way of describing things:

port@0 {
reg = <0>;
phy-handle = <&internal_phy>;
};

In case 2, DSA calls phylink_connect_phy() and in case 3, it calls
phylink_of_phy_connect(). In both cases, phylink_create() has been
called with a phy_interface_t of PHY_INTERFACE_MODE_NA, and in both
cases, PHY_INTERFACE_MODE_NA is translated into phy->interface.

It is important to note that phy_device_create() initializes
dev->interface = PHY_INTERFACE_MODE_GMII, and so, when we use
phylink_create(PHY_INTERFACE_MODE_NA), no one will override this, and we
will end up with a PHY_INTERFACE_MODE_GMII interface inherited from the
PHY.

All this means that in order to maintain compatibility with device tree
blobs where the phy-mode property is missing, we need to allow the
"gmii" phy-mode and treat it as "internal".

Fixes: 2c709e0bdad4 ("net: dsa: microchip: ksz8795: add phylink support")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216320
Reported-by: Craig McQueen <craig@mcqueen.id.au>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk>
Tested-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Link: https://lore.kernel.org/r/20220818143250.2797111-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'mm-hotfixes-stable'

audit: fix potential double free on error path from fsnotify_add_inode_mark

Audit_alloc_mark() assign pathname to audit_mark->path, on error path
from fsnotify_add_inode_mark(), fsnotify_put_mark will free memory
of audit_mark->path, but the caller of audit_alloc_mark will free
the pathname again, so there will be double free problem.

Fix this by resetting audit_mark->path to NULL pointer on error path
from fsnotify_add_inode_mark().

Cc: stable@vger.kernel.org
Fixes: 7b1293234084d ("fsnotify: Add group pointer in fsnotify_init_mark()")
Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Paul Moore <paul@paul-moore.com>

net/mlx5: Unlock on error in mlx5_sriov_enable()

Unlock before returning if mlx5_device_enable_sriov() fails.

Fixes: 84a433a40d0e ("net/mlx5: Lock mlx5 devlink reload callbacks")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: Fix use after free in mlx5e_fs_init()

Call mlx5e_fs_vlan_free(fs) before kvfree(fs).

Fixes: af8bbf730068 ("net/mlx5e: Convert mlx5e_flow_steering member of mlx5e_priv to pointer")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: kTLS, Use _safe() iterator in mlx5e_tls_priv_tx_list_cleanup()

Use the list_for_each_entry_safe() macro to prevent dereferencing "obj"
after it has been freed.

Fixes: c4dfe704f53f ("net/mlx5e: kTLS, Recycle objects of device-offloaded TLS TX connections")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: unlock on error path in esw_vfs_changed_event_handler()

Unlock before returning on this error path.

Fixes: f1bc646c9a06 ("net/mlx5: Use devl_ API in mlx5_esw_offloads_devlink_port_register")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: Fix wrong tc flag used when set hw-tc-offload off

The cited commit reintroduced the ability to set hw-tc-offload
in switchdev mode by reusing NIC mode calls without modifying it
to support both modes, this can cause an illegal memory access
when trying to turn hw-tc-offload off.

Fix this by using the right TC_FLAG when checking if tc rules
are installed while disabling hw-tc-offload.

Fixes: d3cbd4254df8 ("net/mlx5e: Add ndo_set_feature for uplink representor")
Signed-off-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: TC, Add missing policer validation

There is a missing policer validation when offloading police action
with tc action api. Add it.

Fixes: 7d1a5ce46e47 ("net/mlx5e: TC, Support tc action api for police")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: Fix wrong application of the LRO state

Driver caches packet merge type in mlx5e_params instance which must be
in perfect sync with the netdev_feature's bit.
Prior to this patch, in certain conditions (*) LRO state was set in
mlx5e_params, while netdev_feature's bit was off. Causing the LRO to
be applied on the RQs (HW level).

(*) This can happen only on profile init (mlx5e_build_nic_params()),
when RQ expect non-linear SKB and PCI is fast enough in comparison to
link width.

Solution: remove setting of packet merge type from
mlx5e_build_nic_params() as netdev features are not updated.

Fixes: 619a8f2a42f1 ("net/mlx5e: Use linear SKB in Striding RQ")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: Avoid false positive lockdep warning by adding lock_class_key

Add a lock_class_key per mlx5 device to avoid a false positive
"possible circular locking dependency" warning by lockdep, on flows
which lock more than one mlx5 device, such as adding SF.

kernel log:
 ======================================================
 WARNING: possible circular locking dependency detected
 5.19.0-rc8+ #2 Not tainted
 ------------------------------------------------------
 kworker/u20:0/8 is trying to acquire lock:
 ffff88812dfe0d98 (&dev->intf_state_mutex){+.+.}-{3:3}, at: mlx5_init_one+0x2e/0x490 [mlx5_core]

 but task is already holding lock:
 ffff888101aa7898 (&(&notifier->n_head)->rwsem){++++}-{3:3}, at: blocking_notifier_call_chain+0x5a/0x130

 which lock already depends on the new lock.

 the existing dependency chain (in reverse order) is:

 -> #1 (&(&notifier->n_head)->rwsem){++++}-{3:3}:
        down_write+0x90/0x150
        blocking_notifier_chain_register+0x53/0xa0
        mlx5_sf_table_init+0x369/0x4a0 [mlx5_core]
        mlx5_init_one+0x261/0x490 [mlx5_core]
        probe_one+0x430/0x680 [mlx5_core]
        local_pci_probe+0xd6/0x170
        work_for_cpu_fn+0x4e/0xa0
        process_one_work+0x7c2/0x1340
        worker_thread+0x6f6/0xec0
        kthread+0x28f/0x330
        ret_from_fork+0x1f/0x30

 -> #0 (&dev->intf_state_mutex){+.+.}-{3:3}:
        __lock_acquire+0x2fc7/0x6720
        lock_acquire+0x1c1/0x550
        __mutex_lock+0x12c/0x14b0
        mlx5_init_one+0x2e/0x490 [mlx5_core]
        mlx5_sf_dev_probe+0x29c/0x370 [mlx5_core]
        auxiliary_bus_probe+0x9d/0xe0
        really_probe+0x1e0/0xaa0
        __driver_probe_device+0x219/0x480
        driver_probe_device+0x49/0x130
        __device_attach_driver+0x1b8/0x280
        bus_for_each_drv+0x123/0x1a0
        __device_attach+0x1a3/0x460
        bus_probe_device+0x1a2/0x260
        device_add+0x9b1/0x1b40
        __auxiliary_device_add+0x88/0xc0
        mlx5_sf_dev_state_change_handler+0x67e/0x9d0 [mlx5_core]
        blocking_notifier_call_chain+0xd5/0x130
        mlx5_vhca_state_work_handler+0x2b0/0x3f0 [mlx5_core]
        process_one_work+0x7c2/0x1340
        worker_thread+0x59d/0xec0
        kthread+0x28f/0x330
        ret_from_fork+0x1f/0x30

  other info that might help us debug this:

  Possible unsafe locking scenario:

        CPU0                    CPU1
        ----                    ----
   lock(&(&notifier->n_head)->rwsem);
                                lock(&dev->intf_state_mutex);
                                lock(&(&notifier->n_head)->rwsem);
   lock(&dev->intf_state_mutex);

  *** DEADLOCK ***

 4 locks held by kworker/u20:0/8:
  #0: ffff888150612938 ((wq_completion)mlx5_events){+.+.}-{0:0}, at: process_one_work+0x6e2/0x1340
  #1: ffff888100cafdb8 ((work_completion)(&work->work)#3){+.+.}-{0:0}, at: process_one_work+0x70f/0x1340
  #2: ffff888101aa7898 (&(&notifier->n_head)->rwsem){++++}-{3:3}, at: blocking_notifier_call_chain+0x5a/0x130
  #3: ffff88813682d0e8 (&dev->mutex){....}-{3:3}, at:__device_attach+0x76/0x460

 stack backtrace:
 CPU: 6 PID: 8 Comm: kworker/u20:0 Not tainted 5.19.0-rc8+
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
 Workqueue: mlx5_events mlx5_vhca_state_work_handler [mlx5_core]
 Call Trace:
  <TASK>
  dump_stack_lvl+0x57/0x7d
  check_noncircular+0x278/0x300
  ? print_circular_bug+0x460/0x460
  ? lock_chain_count+0x20/0x20
  ? register_lock_class+0x1880/0x1880
  __lock_acquire+0x2fc7/0x6720
  ? register_lock_class+0x1880/0x1880
  ? register_lock_class+0x1880/0x1880
  lock_acquire+0x1c1/0x550
  ? mlx5_init_one+0x2e/0x490 [mlx5_core]
  ? lockdep_hardirqs_on_prepare+0x400/0x400
  __mutex_lock+0x12c/0x14b0
  ? mlx5_init_one+0x2e/0x490 [mlx5_core]
  ? mlx5_init_one+0x2e/0x490 [mlx5_core]
  ? _raw_read_unlock+0x1f/0x30
  ? mutex_lock_io_nested+0x1320/0x1320
  ? __ioremap_caller.constprop.0+0x306/0x490
  ? mlx5_sf_dev_probe+0x269/0x370 [mlx5_core]
  ? iounmap+0x160/0x160
  mlx5_init_one+0x2e/0x490 [mlx5_core]
  mlx5_sf_dev_probe+0x29c/0x370 [mlx5_core]
  ? mlx5_sf_dev_remove+0x130/0x130 [mlx5_core]
  auxiliary_bus_probe+0x9d/0xe0
  really_probe+0x1e0/0xaa0
  __driver_probe_device+0x219/0x480
  ? auxiliary_match_id+0xe9/0x140
  driver_probe_device+0x49/0x130
  __device_attach_driver+0x1b8/0x280
  ? driver_allows_async_probing+0x140/0x140
  bus_for_each_drv+0x123/0x1a0
  ? bus_for_each_dev+0x1a0/0x1a0
  ? lockdep_hardirqs_on_prepare+0x286/0x400
  ? trace_hardirqs_on+0x2d/0x100
  __device_attach+0x1a3/0x460
  ? device_driver_attach+0x1e0/0x1e0
  ? kobject_uevent_env+0x22d/0xf10
  bus_probe_device+0x1a2/0x260
  device_add+0x9b1/0x1b40
  ? dev_set_name+0xab/0xe0
  ? __fw_devlink_link_to_suppliers+0x260/0x260
  ? memset+0x20/0x40
  ? lockdep_init_map_type+0x21a/0x7d0
  __auxiliary_device_add+0x88/0xc0
  ? auxiliary_device_init+0x86/0xa0
  mlx5_sf_dev_state_change_handler+0x67e/0x9d0 [mlx5_core]
  blocking_notifier_call_chain+0xd5/0x130
  mlx5_vhca_state_work_handler+0x2b0/0x3f0 [mlx5_core]
  ? mlx5_vhca_event_arm+0x100/0x100 [mlx5_core]
  ? lock_downgrade+0x6e0/0x6e0
  ? lockdep_hardirqs_on_prepare+0x286/0x400
  process_one_work+0x7c2/0x1340
  ? lockdep_hardirqs_on_prepare+0x400/0x400
  ? pwq_dec_nr_in_flight+0x230/0x230
  ? rwlock_bug.part.0+0x90/0x90
  worker_thread+0x59d/0xec0
  ? process_one_work+0x1340/0x1340
  kthread+0x28f/0x330
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x1f/0x30
  </TASK>

Fixes: 6a3273217469 ("net/mlx5: SF, Port function state change support")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: Fix cmd error logging for manage pages cmd

When the driver unloads, give/reclaim_pages may fail as PF driver in
teardown flow, current code will lead to the following kernel log print
'failed reclaiming pages: err 0'.

Fix it to get same behavior as before the cited commits,
by calling mlx5_cmd_check before handling error state.
mlx5_cmd_check will verify if the returned error is an actual error
needed to be handled by the driver or not and will return an
appropriate value.

Fixes: 8d564292a166 ("net/mlx5: Remove redundant error on reclaim pages")
Fixes: 4dac2f10ada0 ("net/mlx5: Remove redundant notify fail on give pages")
Signed-off-by: Roy Novich <royno@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: Disable irq when locking lag_lock

The lag_lock is taken from both process and softirq contexts which results
lockdep warning[0] about potential deadlock. However, just disabling
softirqs by using *_bh spinlock API is not enough since it will cause
warning in some contexts where the lock is obtained with hard irqs
disabled. To fix the issue save current irq state, disable them before
obtaining the lock an re-enable irqs from saved state after releasing it.

[0]:

[Sun Aug  7 13:12:29 2022] ================================
[Sun Aug  7 13:12:29 2022] WARNING: inconsistent lock state
[Sun Aug  7 13:12:29 2022] 5.19.0_for_upstream_debug_2022_08_04_16_06 #1 Not tainted
[Sun Aug  7 13:12:29 2022] --------------------------------
[Sun Aug  7 13:12:29 2022] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
[Sun Aug  7 13:12:29 2022] swapper/0/0 [HC0[0]:SC1[1]:HE1:SE0] takes:
[Sun Aug  7 13:12:29 2022] ffffffffa06dc0d8 (lag_lock){+.?.}-{2:2}, at: mlx5_lag_is_shared_fdb+0x1f/0x120 [mlx5_core]
[Sun Aug  7 13:12:29 2022] {SOFTIRQ-ON-W} state was registered at:
[Sun Aug  7 13:12:29 2022]   lock_acquire+0x1c1/0x550
[Sun Aug  7 13:12:29 2022]   _raw_spin_lock+0x2c/0x40
[Sun Aug  7 13:12:29 2022]   mlx5_lag_add_netdev+0x13b/0x480 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   mlx5e_nic_enable+0x114/0x470 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   mlx5e_attach_netdev+0x30e/0x6a0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   mlx5e_resume+0x105/0x160 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   mlx5e_probe+0xac3/0x14f0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   auxiliary_bus_probe+0x9d/0xe0
[Sun Aug  7 13:12:29 2022]   really_probe+0x1e0/0xaa0
[Sun Aug  7 13:12:29 2022]   __driver_probe_device+0x219/0x480
[Sun Aug  7 13:12:29 2022]   driver_probe_device+0x49/0x130
[Sun Aug  7 13:12:29 2022]   __driver_attach+0x1e4/0x4d0
[Sun Aug  7 13:12:29 2022]   bus_for_each_dev+0x11e/0x1a0
[Sun Aug  7 13:12:29 2022]   bus_add_driver+0x3f4/0x5a0
[Sun Aug  7 13:12:29 2022]   driver_register+0x20f/0x390
[Sun Aug  7 13:12:29 2022]   __auxiliary_driver_register+0x14e/0x260
[Sun Aug  7 13:12:29 2022]   mlx5e_init+0x38/0x90 [mlx5_core]
[Sun Aug  7 13:12:29 2022]   vhost_iotlb_itree_augment_rotate+0xcb/0x180 [vhost_iotlb]
[Sun Aug  7 13:12:29 2022]   do_one_initcall+0xc4/0x400
[Sun Aug  7 13:12:29 2022]   do_init_module+0x18a/0x620
[Sun Aug  7 13:12:29 2022]   load_module+0x563a/0x7040
[Sun Aug  7 13:12:29 2022]   __do_sys_finit_module+0x122/0x1d0
[Sun Aug  7 13:12:29 2022]   do_syscall_64+0x3d/0x90
[Sun Aug  7 13:12:29 2022]   entry_SYSCALL_64_after_hwframe+0x46/0xb0
[Sun Aug  7 13:12:29 2022] irq event stamp: 3596508
[Sun Aug  7 13:12:29 2022] hardirqs last  enabled at (3596508): [<ffffffff813687c2>] __local_bh_enable_ip+0xa2/0x100
[Sun Aug  7 13:12:29 2022] hardirqs last disabled at (3596507): [<ffffffff813687da>] __local_bh_enable_ip+0xba/0x100
[Sun Aug  7 13:12:29 2022] softirqs last  enabled at (3596488): [<ffffffff81368a2a>] irq_exit_rcu+0x11a/0x170
[Sun Aug  7 13:12:29 2022] softirqs last disabled at (3596495): [<ffffffff81368a2a>] irq_exit_rcu+0x11a/0x170
[Sun Aug  7 13:12:29 2022]
                           other info that might help us debug this:
[Sun Aug  7 13:12:29 2022]  Possible unsafe locking scenario:

[Sun Aug  7 13:12:29 2022]        CPU0
[Sun Aug  7 13:12:29 2022]        ----
[Sun Aug  7 13:12:29 2022]   lock(lag_lock);
[Sun Aug  7 13:12:29 2022]   <Interrupt>
[Sun Aug  7 13:12:29 2022]     lock(lag_lock);
[Sun Aug  7 13:12:29 2022]
                            *** DEADLOCK ***

[Sun Aug  7 13:12:29 2022] 4 locks held by swapper/0/0:
[Sun Aug  7 13:12:29 2022]  #0: ffffffff84643260 (rcu_read_lock){....}-{1:2}, at: mlx5e_napi_poll+0x43/0x20a0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  #1: ffffffff84643260 (rcu_read_lock){....}-{1:2}, at: netif_receive_skb_list_internal+0x2d7/0xd60
[Sun Aug  7 13:12:29 2022]  #2: ffff888144a18b58 (&br->hash_lock){+.-.}-{2:2}, at: br_fdb_update+0x301/0x570
[Sun Aug  7 13:12:29 2022]  #3: ffffffff84643260 (rcu_read_lock){....}-{1:2}, at: atomic_notifier_call_chain+0x5/0x1d0
[Sun Aug  7 13:12:29 2022]
                           stack backtrace:
[Sun Aug  7 13:12:29 2022] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.19.0_for_upstream_debug_2022_08_04_16_06 #1
[Sun Aug  7 13:12:29 2022] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[Sun Aug  7 13:12:29 2022] Call Trace:
[Sun Aug  7 13:12:29 2022]  <IRQ>
[Sun Aug  7 13:12:29 2022]  dump_stack_lvl+0x57/0x7d
[Sun Aug  7 13:12:29 2022]  mark_lock.part.0.cold+0x5f/0x92
[Sun Aug  7 13:12:29 2022]  ? lock_chain_count+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? unwind_next_frame+0x1c4/0x1b50
[Sun Aug  7 13:12:29 2022]  ? secondary_startup_64_no_verify+0xcd/0xdb
[Sun Aug  7 13:12:29 2022]  ? mlx5e_napi_poll+0x4e9/0x20a0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? mlx5e_napi_poll+0x4e9/0x20a0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? stack_access_ok+0x1d0/0x1d0
[Sun Aug  7 13:12:29 2022]  ? start_kernel+0x3a7/0x3c5
[Sun Aug  7 13:12:29 2022]  __lock_acquire+0x1260/0x6720
[Sun Aug  7 13:12:29 2022]  ? lock_chain_count+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? lock_chain_count+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? register_lock_class+0x1880/0x1880
[Sun Aug  7 13:12:29 2022]  ? mark_lock.part.0+0xed/0x3060
[Sun Aug  7 13:12:29 2022]  ? stack_trace_save+0x91/0xc0
[Sun Aug  7 13:12:29 2022]  lock_acquire+0x1c1/0x550
[Sun Aug  7 13:12:29 2022]  ? mlx5_lag_is_shared_fdb+0x1f/0x120 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? lockdep_hardirqs_on_prepare+0x400/0x400
[Sun Aug  7 13:12:29 2022]  ? __lock_acquire+0xd6f/0x6720
[Sun Aug  7 13:12:29 2022]  _raw_spin_lock+0x2c/0x40
[Sun Aug  7 13:12:29 2022]  ? mlx5_lag_is_shared_fdb+0x1f/0x120 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  mlx5_lag_is_shared_fdb+0x1f/0x120 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  mlx5_esw_bridge_rep_vport_num_vhca_id_get+0x1a0/0x600 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? mlx5_esw_bridge_update_work+0x90/0x90 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? lock_acquire+0x1c1/0x550
[Sun Aug  7 13:12:29 2022]  mlx5_esw_bridge_switchdev_event+0x185/0x8f0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? mlx5_esw_bridge_port_obj_attr_set+0x3e0/0x3e0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? check_chain_key+0x24a/0x580
[Sun Aug  7 13:12:29 2022]  atomic_notifier_call_chain+0xd7/0x1d0
[Sun Aug  7 13:12:29 2022]  br_switchdev_fdb_notify+0xea/0x100
[Sun Aug  7 13:12:29 2022]  ? br_switchdev_set_port_flag+0x310/0x310
[Sun Aug  7 13:12:29 2022]  fdb_notify+0x11b/0x150
[Sun Aug  7 13:12:29 2022]  br_fdb_update+0x34c/0x570
[Sun Aug  7 13:12:29 2022]  ? lock_chain_count+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? br_fdb_add_local+0x50/0x50
[Sun Aug  7 13:12:29 2022]  ? br_allowed_ingress+0x5f/0x1070
[Sun Aug  7 13:12:29 2022]  ? check_chain_key+0x24a/0x580
[Sun Aug  7 13:12:29 2022]  br_handle_frame_finish+0x786/0x18e0
[Sun Aug  7 13:12:29 2022]  ? check_chain_key+0x24a/0x580
[Sun Aug  7 13:12:29 2022]  ? br_handle_local_finish+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? __lock_acquire+0xd6f/0x6720
[Sun Aug  7 13:12:29 2022]  ? sctp_inet_bind_verify+0x4d/0x190
[Sun Aug  7 13:12:29 2022]  ? xlog_unpack_data+0x2e0/0x310
[Sun Aug  7 13:12:29 2022]  ? br_handle_local_finish+0x20/0x20
[Sun Aug  7 13:12:29 2022]  br_nf_hook_thresh+0x227/0x380 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? setup_pre_routing+0x460/0x460 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? br_handle_local_finish+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? br_nf_pre_routing_ipv6+0x48b/0x69c [br_netfilter]
[Sun Aug  7 13:12:29 2022]  br_nf_pre_routing_finish_ipv6+0x5c2/0xbf0 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? br_handle_local_finish+0x20/0x20
[Sun Aug  7 13:12:29 2022]  br_nf_pre_routing_ipv6+0x4c6/0x69c [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? br_validate_ipv6+0x9e0/0x9e0 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? br_nf_forward_arp+0xb70/0xb70 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  ? br_nf_pre_routing+0xacf/0x1160 [br_netfilter]
[Sun Aug  7 13:12:29 2022]  br_handle_frame+0x8a9/0x1270
[Sun Aug  7 13:12:29 2022]  ? br_handle_frame_finish+0x18e0/0x18e0
[Sun Aug  7 13:12:29 2022]  ? register_lock_class+0x1880/0x1880
[Sun Aug  7 13:12:29 2022]  ? br_handle_local_finish+0x20/0x20
[Sun Aug  7 13:12:29 2022]  ? bond_handle_frame+0xf9/0xac0 [bonding]
[Sun Aug  7 13:12:29 2022]  ? br_handle_frame_finish+0x18e0/0x18e0
[Sun Aug  7 13:12:29 2022]  __netif_receive_skb_core+0x7c0/0x2c70
[Sun Aug  7 13:12:29 2022]  ? check_chain_key+0x24a/0x580
[Sun Aug  7 13:12:29 2022]  ? generic_xdp_tx+0x5b0/0x5b0
[Sun Aug  7 13:12:29 2022]  ? __lock_acquire+0xd6f/0x6720
[Sun Aug  7 13:12:29 2022]  ? register_lock_class+0x1880/0x1880
[Sun Aug  7 13:12:29 2022]  ? check_chain_key+0x24a/0x580
[Sun Aug  7 13:12:29 2022]  __netif_receive_skb_list_core+0x2d7/0x8a0
[Sun Aug  7 13:12:29 2022]  ? lock_acquire+0x1c1/0x550
[Sun Aug  7 13:12:29 2022]  ? process_backlog+0x960/0x960
[Sun Aug  7 13:12:29 2022]  ? lockdep_hardirqs_on_prepare+0x129/0x400
[Sun Aug  7 13:12:29 2022]  ? kvm_clock_get_cycles+0x14/0x20
[Sun Aug  7 13:12:29 2022]  netif_receive_skb_list_internal+0x5f4/0xd60
[Sun Aug  7 13:12:29 2022]  ? do_xdp_generic+0x150/0x150
[Sun Aug  7 13:12:29 2022]  ? mlx5e_poll_rx_cq+0xf6b/0x2960 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? mlx5e_poll_ico_cq+0x3d/0x1590 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  napi_complete_done+0x188/0x710
[Sun Aug  7 13:12:29 2022]  mlx5e_napi_poll+0x4e9/0x20a0 [mlx5_core]
[Sun Aug  7 13:12:29 2022]  ? __queue_work+0x53c/0xeb0
[Sun Aug  7 13:12:29 2022]  __napi_poll+0x9f/0x540
[Sun Aug  7 13:12:29 2022]  net_rx_action+0x420/0xb70
[Sun Aug  7 13:12:29 2022]  ? napi_threaded_poll+0x470/0x470
[Sun Aug  7 13:12:29 2022]  ? __common_interrupt+0x79/0x1a0
[Sun Aug  7 13:12:29 2022]  __do_softirq+0x271/0x92c
[Sun Aug  7 13:12:29 2022]  irq_exit_rcu+0x11a/0x170
[Sun Aug  7 13:12:29 2022]  common_interrupt+0x7d/0xa0
[Sun Aug  7 13:12:29 2022]  </IRQ>
[Sun Aug  7 13:12:29 2022]  <TASK>
[Sun Aug  7 13:12:29 2022]  asm_common_interrupt+0x22/0x40
[Sun Aug  7 13:12:29 2022] RIP: 0010:default_idle+0x42/0x60
[Sun Aug  7 13:12:29 2022] Code: c1 83 e0 07 48 c1 e9 03 83 c0 03 0f b6 14 11 38 d0 7c 04 84 d2 75 14 8b 05 6b f1 22 02 85 c0 7e 07 0f 00 2d 80 3b 4a 00 fb f4 <c3> 48 c7 c7 e0 07 7e 85 e8 21 bd 40 fe eb de 66 66 2e 0f 1f 84 00
[Sun Aug  7 13:12:29 2022] RSP: 0018:ffffffff84407e18 EFLAGS: 00000242
[Sun Aug  7 13:12:29 2022] RAX: 0000000000000001 RBX: ffffffff84ec4a68 RCX: 1ffffffff0afc0fc
[Sun Aug  7 13:12:29 2022] RDX: 0000000000000004 RSI: 0000000000000000 RDI: ffffffff835b1fac
[Sun Aug  7 13:12:29 2022] RBP: 0000000000000000 R08: 0000000000000001 R09: ffff8884d2c44ac3
[Sun Aug  7 13:12:29 2022] R10: ffffed109a588958 R11: 00000000ffffffff R12: 0000000000000000
[Sun Aug  7 13:12:29 2022] R13: ffffffff84efac20 R14: 0000000000000000 R15: dffffc0000000000
[Sun Aug  7 13:12:29 2022]  ? default_idle_call+0xcc/0x460
[Sun Aug  7 13:12:29 2022]  default_idle_call+0xec/0x460
[Sun Aug  7 13:12:29 2022]  do_idle+0x394/0x450
[Sun Aug  7 13:12:29 2022]  ? arch_cpu_idle_exit+0x40/0x40
[Sun Aug  7 13:12:29 2022]  cpu_startup_entry+0x19/0x20
[Sun Aug  7 13:12:29 2022]  rest_init+0x156/0x250
[Sun Aug  7 13:12:29 2022]  arch_call_rest_init+0xf/0x15
[Sun Aug  7 13:12:29 2022]  start_kernel+0x3a7/0x3c5
[Sun Aug  7 13:12:29 2022]  secondary_startup_64_no_verify+0xcd/0xdb
[Sun Aug  7 13:12:29 2022]  </TASK>

Fixes: ff9b7521468b ("net/mlx5: Bridge, support LAG")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: Eswitch, Fix forwarding decision to uplink

Make sure to modify the rule for uplink forwarding only for the case
where destination vport number is MLX5_VPORT_UPLINK.

Fixes: 94db33177819 ("net/mlx5: Support multiport eswitch mode")
Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5: LAG, fix logic over MLX5_LAG_FLAG_NDEVS_READY

Only set MLX5_LAG_FLAG_NDEVS_READY if both netdevices are registered.
Doing so guarantees that both ldev->pf[MLX5_LAG_P0].dev and
ldev->pf[MLX5_LAG_P1].dev have valid pointers when
MLX5_LAG_FLAG_NDEVS_READY is set.

The core issue is asymmetry in setting MLX5_LAG_FLAG_NDEVS_READY and
clearing it. Setting it is done wrongly when both
ldev->pf[MLX5_LAG_P0].dev and ldev->pf[MLX5_LAG_P1].dev are set;
clearing it is done right when either of ldev->pf[i].netdev is cleared.

Consider the following scenario:
1. PF0 loads and sets ldev->pf[MLX5_LAG_P0].dev to a valid pointer
2. PF1 loads and sets both ldev->pf[MLX5_LAG_P1].dev and
   ldev->pf[MLX5_LAG_P1].netdev with valid pointers. This results in
   MLX5_LAG_FLAG_NDEVS_READY is set.
3. PF0 is unloaded before setting dev->pf[MLX5_LAG_P0].netdev.
   MLX5_LAG_FLAG_NDEVS_READY remains set.

Further execution of mlx5_do_bond() will result in null pointer
dereference when calling mlx5_lag_is_multipath()

This patch fixes the following call trace actually encountered:

[ 1293.475195] BUG: kernel NULL pointer dereference, address: 00000000000009a8
[ 1293.478756] #PF: supervisor read access in kernel mode
[ 1293.481320] #PF: error_code(0x0000) - not-present page
[ 1293.483686] PGD 0 P4D 0
[ 1293.484434] Oops: 0000 [#1] SMP PTI
[ 1293.485377] CPU: 1 PID: 23690 Comm: kworker/u16:2 Not tainted 5.18.0-rc5_for_upstream_min_debug_2022_05_05_10_13 #1
[ 1293.488039] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[ 1293.490836] Workqueue: mlx5_lag mlx5_do_bond_work [mlx5_core]
[ 1293.492448] RIP: 0010:mlx5_lag_is_multipath+0x5/0x50 [mlx5_core]
[ 1293.494044] Code: e8 70 40 ff e0 48 8b 14 24 48 83 05 5c 1a 1b 00 01 e9 19 ff ff ff 48 83 05 47 1a 1b 00 01 eb d7 0f 1f 44 00 00 0f 1f 44 00 00 <48> 8b 87 a8 09 00 00 48 85 c0 74 26 48 83 05 a7 1b 1b 00 01 41 b8
[ 1293.498673] RSP: 0018:ffff88811b2fbe40 EFLAGS: 00010202
[ 1293.500152] RAX: ffff88818a94e1c0 RBX: ffff888165eca6c0 RCX: 0000000000000000
[ 1293.501841] RDX: 0000000000000001 RSI: ffff88818a94e1c0 RDI: 0000000000000000
[ 1293.503585] RBP: 0000000000000000 R08: ffff888119886740 R09: ffff888165eca73c
[ 1293.505286] R10: 0000000000000018 R11: 0000000000000018 R12: ffff88818a94e1c0
[ 1293.506979] R13: ffff888112729800 R14: 0000000000000000 R15: ffff888112729858
[ 1293.508753] FS:  0000000000000000(0000) GS:ffff88852cc40000(0000) knlGS:0000000000000000
[ 1293.510782] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1293.512265] CR2: 00000000000009a8 CR3: 00000001032d4002 CR4: 0000000000370ea0
[ 1293.514001] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1293.515806] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

Fixes: 8a66e4585979 ("net/mlx5: Change ownership model for lag")
Signed-off-by: Eli Cohen <elic@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

net/mlx5e: Properly disable vlan strip on non-UL reps

When querying mlx5 non-uplink representors capabilities with ethtool
rx-vlan-offload is marked as "off [fixed]". However, it is actually always
enabled because mlx5e_params->vlan_strip_disable is 0 by default when
initializing struct mlx5e_params instance. Fix the issue by explicitly
setting the vlan_strip_disable to 'true' for non-uplink representors.

Fixes: cb67b832921c ("net/mlx5e: Introduce SRIOV VF representors")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>