sfrench/cifs-2.6.git
3 months agoionic: Add XDP_TX support
Shannon Nelson [Wed, 14 Feb 2024 17:59:06 +0000 (09:59 -0800)]
ionic: Add XDP_TX support

The XDP_TX packets get fed back into the Rx queue's partnered
Tx queue as an xdp_frame.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoionic: Add XDP packet headroom
Shannon Nelson [Wed, 14 Feb 2024 17:59:05 +0000 (09:59 -0800)]
ionic: Add XDP packet headroom

If an xdp program is loaded, add headroom at the beginning
of the frame to allow for editing and insertions that an XDP
program might need room for, and tailroom used later for XDP
frame tracking.  These are only needed in the first Rx buffer
in a packet, not for any trailing frags.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoionic: add initial framework for XDP support
Shannon Nelson [Wed, 14 Feb 2024 17:59:04 +0000 (09:59 -0800)]
ionic: add initial framework for XDP support

Set up the basics for running Rx packets through XDP programs.
Add new queue setup and teardown steps for adding/removing an
XDP program, and add the call to run the XDP on a packet.

The XDP frame size needs to be the MTU plus standard ethernet
header, plus head room for XDP scribblings and tail room for a
struct skb_shared_info.  Also, at this point, we don't support
XDP frags, only a single contiguous Rx buffer.  This means
that our page splitting is not very useful, so when XDP is in
use we need to use the full Rx buffer size and not do sharing.

Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoionic: use dma range APIs
Shannon Nelson [Wed, 14 Feb 2024 17:59:03 +0000 (09:59 -0800)]
ionic: use dma range APIs

Convert Rx datapath handling to use the DMA range APIs
in preparation for adding XDP handling.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoionic: add helpers for accessing buffer info
Shannon Nelson [Wed, 14 Feb 2024 17:59:02 +0000 (09:59 -0800)]
ionic: add helpers for accessing buffer info

These helpers clean up some of the code around DMA mapping
and other buffer references, and will be used in the next
few patches for the XDP support.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoionic: set adminq irq affinity
Shannon Nelson [Wed, 14 Feb 2024 17:59:01 +0000 (09:59 -0800)]
ionic: set adminq irq affinity

We claim to have the AdminQ on our irq0 and thus cpu id 0,
but we need to be sure we set the affinity hint to try to
keep it there.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet/iucv: fix virtual vs physical address confusion
Alexander Gordeev [Wed, 14 Feb 2024 08:47:07 +0000 (09:47 +0100)]
net/iucv: fix virtual vs physical address confusion

Fix virtual vs physical address confusion. This does not fix a bug
since virtual and physical address spaces are currently the same.

Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch 'ravb-rutime-PM-support'
David S. Miller [Fri, 16 Feb 2024 08:32:05 +0000 (08:32 +0000)]
Merge branch 'ravb-rutime-PM-support'

Claudiu Beznea says:

====================
net: ravb: Add runtime PM support (part 2)

Series adds runtime PM support for the ravb driver. This is a continuation
of [1].

There are 5 more preparation patches (patches 1-5) and patch 6
adds runtime PM support.

Patches in this series were part of [2].

Changes in v4:
- remove unnecessary code from patch 4/6
- improve the code in patch 5/6

Changes in v3:
- fixed typos
- added patch "net: ravb: Move the update of ndev->features to
  ravb_set_features()"
- changes title of patch "net: ravb: Do not apply RX checksum
  settings to hardware if the interface is down" from v2 into
  "net: ravb: Do not apply features to hardware if the interface
  is down", changed patch description and updated the patch
- collected tags

Changes in v2:
- address review comments
- in patch 4/5 take into account the latest changes introduced
  in ravb_set_features_gbeth()

Changes since [2]:
- patch 1/5 is new
- use pm_runtime_get_noresume() and pm_runtime_active() in patches
  3/5, 4/5
- fixed higlighted typos in patch 4/5

[1] https://lore.kernel.org/all/20240202084136.3426492-1-claudiu.beznea.uj@bp.renesas.com/
[2] https://lore.kernel.org/all/20240105082339.1468817-1-claudiu.beznea.uj@bp.renesas.com/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Add runtime PM support
Claudiu Beznea [Wed, 14 Feb 2024 13:58:00 +0000 (15:58 +0200)]
net: ravb: Add runtime PM support

Add runtime PM support for the ravb driver. As the driver is used by
different IP variants, with different behaviors, to be able to have the
runtime PM support available for all devices, the preparatory commits
moved all the resources parsing and allocations in the driver's probe
function and kept the settings for ravb_open(). This is due to the fact
that on some IP variants-platforms tuples disabling/enabling the clocks
will switch the IP to the reset operation mode where register contents is
lost and reconfiguration needs to be done. For this the rabv_open()
function enables the clocks, switches the IP to configuration mode, applies
all the register settings and switches the IP to the operational mode. At
the end of ravb_open() IP is ready to send/receive data.

In ravb_close() necessary reverts are done (compared with ravb_open()), the
IP is switched to reset mode and clocks are disabled.

The ethtool APIs or IOCTLs that might execute while the interface is down
are either cached (and applied in ravb_open()) or rejected (as at that time
the IP is in reset mode). Keeping the IP in the reset mode also increases
the power saved (according to the hardware manual).

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Do not apply features to hardware if the interface is down
Claudiu Beznea [Wed, 14 Feb 2024 13:57:59 +0000 (15:57 +0200)]
net: ravb: Do not apply features to hardware if the interface is down

Do not apply features to hardware if the interface is down. In case runtime
PM is enabled, and while the interface is down, the IP will be in reset
mode (as for some platforms disabling the clocks will switch the IP to
reset mode, which will lead to losing register contents) and applying
settings in reset mode is not an option. Instead, cache the features and
apply them in ravb_open() through ravb_emac_init().

To avoid accessing the hardware while the interface is down
pm_runtime_active() check was introduced. Along with it the device runtime
PM usage counter has been incremented to avoid disabling the device clocks
while the check is in progress (if any).

Commit prepares for the addition of runtime PM.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Move the update of ndev->features to ravb_set_features()
Claudiu Beznea [Wed, 14 Feb 2024 13:57:58 +0000 (15:57 +0200)]
net: ravb: Move the update of ndev->features to ravb_set_features()

Commit c2da9408579d ("ravb: Add Rx checksum offload support for GbEth")
introduced support for setting GbEth features. With this the IP-specific
features update functions update the ndev->features individually.

Next commits add runtime PM support for the ravb driver. The runtime PM
implementation will enable/disable the IP clocks on
the ravb_open()/ravb_close() functions. Accessing the IP registers with
clocks disabled blocks the system.

The ravb_set_features() function could be executed when the Ethernet
interface is closed so we need to ensure we don't access IP registers while
the interface is down when runtime PM support will be in place.

For these, move the update of ndev->features to ravb_set_features(). In
this way we update the ndev->features only when the IP-specific features
set function returns success and we can avoid code duplication when
introducing runtime PM registers protection.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Return cached statistics if the interface is down
Claudiu Beznea [Wed, 14 Feb 2024 13:57:57 +0000 (15:57 +0200)]
net: ravb: Return cached statistics if the interface is down

Return the cached statistics in case the interface is down. There should be
no drawback to this, as cached statistics are updated in ravb_close().

In order to avoid accessing the IP registers while the IP is runtime
suspended pm_runtime_active() check was introduced. The device runtime
PM usage counter has been incremented to avoid disabling the device clocks
while the check is in progress (if any).

The commit prepares the code for the addition of runtime PM support.

Suggested-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Keep the reverse order of operations in ravb_close()
Claudiu Beznea [Wed, 14 Feb 2024 13:57:56 +0000 (15:57 +0200)]
net: ravb: Keep the reverse order of operations in ravb_close()

Keep the reverse order of operations in ravb_close() when compared with
ravb_open(). This is the recommended configuration sequence.

Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ravb: Get rid of the temporary variable irq
Claudiu Beznea [Wed, 14 Feb 2024 13:57:55 +0000 (15:57 +0200)]
net: ravb: Get rid of the temporary variable irq

The 4th argument of ravb_setup_irq() is used to save the IRQ number that
will be further used by the driver code. Not all ravb_setup_irqs() calls
need to save the IRQ number. The previous code used to pass a dummy
variable as the 4th argument in case the IRQ is not needed for further
usage. That is not necessary as the code from ravb_setup_irq() can detect
by itself if the IRQ needs to be saved. Thus, get rid of the code that is
not needed.

Reported-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoselftests: bonding: make sure new active is not null
Hangbin Liu [Wed, 14 Feb 2024 09:21:28 +0000 (17:21 +0800)]
selftests: bonding: make sure new active is not null

One of Jakub's tests[1] shows that there may be period all ports
are down and no active slave. This makes the new_active_slave null
and the test fails. Add a check to make sure the new active is not null.

 [  189.051966] br0: port 2(s1) entered disabled state
 [  189.317881] bond0: (slave eth1): link status definitely down, disabling slave
 [  189.318487] bond0: (slave eth2): making interface the new active one
 [  190.435430] br0: port 4(s2) entered disabled state
 [  190.773786] bond0: (slave eth0): link status definitely down, disabling slave
 [  190.774204] bond0: (slave eth2): link status definitely down, disabling slave
 [  190.774715] bond0: now running without any active interface!
 [  190.877760] bond0: (slave eth0): link status definitely up
 [  190.878098] bond0: (slave eth0): making interface the new active one
 [  190.878495] bond0: active interface up!
 [  191.802872] br0: port 4(s2) entered blocking state
 [  191.803157] br0: port 4(s2) entered forwarding state
 [  191.813756] bond0: (slave eth2): link status definitely up
 [  192.847095] br0: port 2(s1) entered blocking state
 [  192.847396] br0: port 2(s1) entered forwarding state
 [  192.853740] bond0: (slave eth1): link status definitely up
 # TEST: prio (active-backup ns_ip6_target primary_reselect 1)         [FAIL]
 # Current active slave is null but not eth0

[1] https://netdev-3.bots.linux.dev/vmksft-bonding/results/464481/1-bond-options-sh/stdout

Fixes: 45bf79bc56c4 ("selftests: bonding: reduce garp_test/arp_validate test time")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: dsa: remove OF-based MDIO bus registration from DSA core
Arınç ÜNAL [Tue, 13 Feb 2024 07:29:05 +0000 (10:29 +0300)]
net: dsa: remove OF-based MDIO bus registration from DSA core

The code block under the "!ds->user_mii_bus && ds->ops->phy_read" check
under dsa_switch_setup() populates ds->user_mii_bus. The use of
ds->user_mii_bus is inappropriate when the MDIO bus of the switch is
described on the device tree [1].

For this reason, use this code block only for switches [with MDIO bus]
probed on platform_data, and OF which the switch MDIO bus isn't described
on the device tree. Therefore, remove OF-based MDIO bus registration as
it's useless for these cases.

These subdrivers which control switches [with MDIO bus] probed on OF, will
lose the ability to register the MDIO bus OF-based:

drivers/net/dsa/b53/b53_common.c
drivers/net/dsa/lan9303-core.c
drivers/net/dsa/vitesse-vsc73xx-core.c

These subdrivers let the DSA core driver register the bus:
- ds->ops->phy_read() and ds->ops->phy_write() are present.
- ds->user_mii_bus is not populated.

The commit fe7324b93222 ("net: dsa: OF-ware slave_mii_bus") which brought
OF-based MDIO bus registration on the DSA core driver is reasonably recent
and, in this time frame, there have been no device trees in the Linux
repository that started describing the MDIO bus, or dt-bindings defining
the MDIO bus for the switches these subdrivers control. So I don't expect
any devices to be affected.

The logic we encourage is that all subdrivers should register the switch
MDIO bus on their own [2]. And, for subdrivers which control switches [with
MDIO bus] probed on OF, this logic must be followed to support all cases
properly:

No switch MDIO bus defined: Populate ds->user_mii_bus, register the MDIO
bus, set the interrupts for PHYs if "interrupt-controller" is defined at
the switch node. This case should only be covered for the switches which
their dt-bindings documentation didn't document the MDIO bus from the
start. This is to keep supporting the device trees that do not describe the
MDIO bus on the device tree but the MDIO bus is being used nonetheless.

Switch MDIO bus defined: Don't populate ds->user_mii_bus, register the MDIO
bus, set the interrupts for PHYs if ["interrupt-controller" is defined at
the switch node and "interrupts" is defined at the PHY nodes under the
switch MDIO bus node].

Switch MDIO bus defined but explicitly disabled: If the device tree says
status = "disabled" for the MDIO bus, we shouldn't need an MDIO bus at all.
Instead, just exit as early as possible and do not call any MDIO API.

After all subdrivers that control switches with MDIO buses are made to
register the MDIO buses on their own, we will be able to get rid of
dsa_switch_ops :: phy_read() and :: phy_write(), and the code block for
registering the MDIO bus on the DSA core driver.

Link: https://lore.kernel.org/netdev/20231213120656.x46fyad6ls7sqyzv@skbuf/
Link: https://lore.kernel.org/netdev/20240103184459.dcbh57wdnlox6w7d@skbuf/
Suggested-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Acked-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240213-for-netnext-dsa-mdio-bus-v2-1-0ff6f4823a9e@arinc9.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge branch 'for-thermal-genetlink-family-bind-unbind-callbacks'
Jakub Kicinski [Fri, 16 Feb 2024 01:49:24 +0000 (17:49 -0800)]
Merge branch 'for-thermal-genetlink-family-bind-unbind-callbacks'

Stanislaw Gruszka says:

====================
thermal/netlink/intel_hfi: Enable HFI feature only when required

The patchset introduces a new genetlink family bind/unbind callbacks
and thermal/netlink notifications, which allow drivers to send netlink
multicast events based on the presence of actual user-space consumers.
This functionality optimizes resource usage by allowing disabling
of features when not needed.

v1: https://lore.kernel.org/linux-pm/20240131120535.933424-1-stanislaw.gruszka@linux.intel.com//
v2: https://lore.kernel.org/linux-pm/20240206133605.1518373-1-stanislaw.gruszka@linux.intel.com/
v3: https://lore.kernel.org/linux-pm/20240209120625.1775017-1-stanislaw.gruszka@linux.intel.com/
====================

Link: https://lore.kernel.org/r/20240212161615.161935-1-stanislaw.gruszka@linux.intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agogenetlink: Add per family bind/unbind callbacks
Stanislaw Gruszka [Mon, 12 Feb 2024 16:16:13 +0000 (17:16 +0100)]
genetlink: Add per family bind/unbind callbacks

Add genetlink family bind()/unbind() callbacks when adding/removing
multicast group to/from netlink client socket via setsockopt() or
bind() syscall.

They can be used to track if consumers of netlink multicast messages
emerge or disappear. Thus, a client implementing callbacks, can now
send events only when there are active consumers, preventing unnecessary
work when none exist.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/20240212161615.161935-2-stanislaw.gruszka@linux.intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoconfigs/debug: add NET debug config
Matthieu Baerts (NGI0) [Mon, 12 Feb 2024 10:47:14 +0000 (11:47 +0100)]
configs/debug: add NET debug config

The debug.config file is really great to easily enable a bunch of
general debugging features on a CI-like setup. But it would be great to
also include core networking debugging config.

A few CI's validating features from the Net tree also enable a few other
debugging options on top of debug.config. A small selection is quite
generic for the whole net tree. They validate some assumptions in
different parts of the core net tree. As suggested by Jakub Kicinski in
[1], having them added to this debug.config file would help other CIs
using network features to find bugs in this area.

Note that the two REFCNT configs also select REF_TRACKER, which doesn't
seem to be an issue.

Link: https://lore.kernel.org/netdev/20240202093148.33bd2b14@kernel.org/T/
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20240212-kconfig-debug-enable-net-v1-1-fb026de8174c@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Jakub Kicinski [Thu, 15 Feb 2024 22:01:43 +0000 (14:01 -0800)]
Merge git://git./linux/kernel/git/netdev/net

Cross-merge networking fixes after downstream PR.

No conflicts.

Adjacent changes:

net/core/dev.c
  9f30831390ed ("net: add rcu safety to rtnl_prop_list_size()")
  723de3ebef03 ("net: free altname using an RCU callback")

net/unix/garbage.c
  11498715f266 ("af_unix: Remove io_uring code for GC.")
  25236c91b5ab ("af_unix: Fix task hung while purging oob_skb in GC.")

drivers/net/ethernet/renesas/ravb_main.c
  ed4adc07207d ("net: ravb: Count packets instead of descriptors in GbEth RX path"
)
  c2da9408579d ("ravb: Add Rx checksum offload support for GbEth")

net/mptcp/protocol.c
  bdd70eb68913 ("mptcp: drop the push_pending field")
  28e5c1380506 ("mptcp: annotate lockless accesses around read-mostly fields")

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge tag 'net-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 15 Feb 2024 19:39:27 +0000 (11:39 -0800)]
Merge tag 'net-6.8-rc5' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from can, wireless and netfilter.

  Current release - regressions:

   - af_unix: fix task hung while purging oob_skb in GC

   - pds_core: do not try to run health-thread in VF path

  Current release - new code bugs:

   - sched: act_mirred: don't zero blockid when net device is being
     deleted

  Previous releases - regressions:

   - netfilter:
      - nat: restore default DNAT behavior
      - nf_tables: fix bidirectional offload, broken when unidirectional
        offload support was added

   - openvswitch: limit the number of recursions from action sets

   - eth: i40e: do not allow untrusted VF to remove administratively set
     MAC address

  Previous releases - always broken:

   - tls: fix races and bugs in use of async crypto

   - mptcp: prevent data races on some of the main socket fields, fix
     races in fastopen handling

   - dpll: fix possible deadlock during netlink dump operation

   - dsa: lan966x: fix crash when adding interface under a lag when some
     of the ports are disabled

   - can: j1939: prevent deadlock by changing j1939_socks_lock to rwlock

  Misc:

   - a handful of fixes and reliability improvements for selftests

   - fix sysfs documentation missing net/ in paths

   - finish the work of squashing the missing MODULE_DESCRIPTION()
     warnings in networking"

* tag 'net-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (92 commits)
  net: fill in MODULE_DESCRIPTION()s for missing arcnet
  net: fill in MODULE_DESCRIPTION()s for mdio_devres
  net: fill in MODULE_DESCRIPTION()s for ppp
  net: fill in MODULE_DESCRIPTION()s for fddik/skfp
  net: fill in MODULE_DESCRIPTION()s for plip
  net: fill in MODULE_DESCRIPTION()s for ieee802154/fakelb
  net: fill in MODULE_DESCRIPTION()s for xen-netback
  net: ravb: Count packets instead of descriptors in GbEth RX path
  pppoe: Fix memory leak in pppoe_sendmsg()
  net: sctp: fix skb leak in sctp_inq_free()
  net: bcmasp: Handle RX buffer allocation failure
  net-timestamp: make sk_tskey more predictable in error path
  selftests: tls: increase the wait in poll_partial_rec_async
  ice: Add check for lport extraction to LAG init
  netfilter: nf_tables: fix bidirectional offload regression
  netfilter: nat: restore default DNAT behavior
  netfilter: nft_set_pipapo: fix missing : in kdoc
  igc: Remove temporary workaround
  igb: Fix string truncation warnings in igb_set_fw_version
  can: netlink: Fix TDCO calculation using the old data bittiming
  ...

3 months agoMerge tag 'for-linus-6.8a-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 15 Feb 2024 19:33:35 +0000 (11:33 -0800)]
Merge tag 'for-linus-6.8a-rc5-tag' of git://git./linux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:
 "Fixes and simple cleanups:

   - use a proper flexible array instead of a one-element array in order
     to avoid array-bounds sanitizer errors

   - add NULL pointer checks after allocating memory

   - use memdup_array_user() instead of open-coding it

   - fix a rare race condition in Xen event channel allocation code

   - make struct bus_type instances const

   - make kerneldoc inline comments match reality"

* tag 'for-linus-6.8a-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/events: close evtchn after mapping cleanup
  xen/gntalloc: Replace UAPI 1-element array
  xen: balloon: make balloon_subsys const
  xen: pcpu: make xen_pcpu_subsys const
  xen/privcmd: Use memdup_array_user() in alloc_ioreq()
  x86/xen: Add some null pointer checking to smp.c
  xen/xenbus: document will_handle argument for xenbus_watch_path()

3 months agoupdate workarounds for gcc "asm goto" issue
Linus Torvalds [Thu, 15 Feb 2024 19:14:33 +0000 (11:14 -0800)]
update workarounds for gcc "asm goto" issue

In commit 4356e9f841f7 ("work around gcc bugs with 'asm goto' with
outputs") I did the gcc workaround unconditionally, because the cause of
the bad code generation wasn't entirely clear.

In the meantime, Jakub Jelinek debugged the issue, and has come up with
a fix in gcc [2], which also got backported to the still maintained
branches of gcc-11, gcc-12 and gcc-13.

Note that while the fix technically wasn't in the original gcc-14
branch, Jakub says:

 "while it is true that no GCC 14 snapshots until today (or whenever the
  fix will be committed) have the fix, for GCC trunk it is up to the
  distros to use the latest snapshot if they use it at all and would
  allow better testing of the kernel code without the workaround, so
  that if there are other issues they won't be discovered years later.
  Most userland code doesn't actually use asm goto with outputs..."

so we will consider gcc-14 to be fixed - if somebody is using gcc
snapshots of the gcc-14 before the fix, they should upgrade.

Note that while the bug goes back to gcc-11, in practice other gcc
changes seem to have effectively hidden it since gcc-12.1 as per a
bisect by Jakub.  So even a gcc-14 snapshot without the fix likely
doesn't show actual problems.

Also, make the default 'asm_goto_output()' macro mark the asm as
volatile by hand, because of an unrelated gcc issue [1] where it doesn't
match the documented behavior ("asm goto is always volatile").

Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103979
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113921
Link: https://lore.kernel.org/all/20240208220604.140859-1-seanjc@google.com/
Requested-by: Jakub Jelinek <jakub@redhat.com>
Cc: Uros Bizjak <ubizjak@gmail.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Andrew Pinski <quic_apinski@quicinc.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 months agoMerge tag 'devicetree-fixes-for-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 15 Feb 2024 18:19:55 +0000 (10:19 -0800)]
Merge tag 'devicetree-fixes-for-6.8-1' of git://git./linux/kernel/git/robh/linux

Pull devicetree fixes from Rob Herring:

 - Improve devlink dependency parsing for DT graphs

 - Fix devlink handling of io-channels dependencies

 - Fix PCI addressing in marvell,prestera example

 - A few schema fixes for property constraints

 - Improve performance of DT unprobed devices kselftest

 - Fix regression in DT_SCHEMA_FILES handling

 - Fix compile error in unittest for !OF_DYNAMIC

* tag 'devicetree-fixes-for-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  dt-bindings: ufs: samsung,exynos-ufs: Add size constraints on "samsung,sysreg"
  of: property: Add in-ports/out-ports support to of_graph_get_port_parent()
  of: property: Improve finding the supplier of a remote-endpoint property
  of: property: Improve finding the consumer of a remote-endpoint property
  net: marvell,prestera: Fix example PCI bus addressing
  of: unittest: Fix compile in the non-dynamic case
  of: property: fix typo in io-channels
  dt-bindings: tpm: Drop type from "resets"
  dt-bindings: display: nxp,tda998x: Fix 'audio-ports' constraints
  dt-bindings: xilinx: replace Piyush Mehta maintainership
  kselftest: dt: Stop relying on dirname to improve performance
  dt-bindings: don't anchor DT_SCHEMA_FILES to bindings directory

3 months agoMerge tag 'spi-fix-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Linus Torvalds [Thu, 15 Feb 2024 17:13:12 +0000 (09:13 -0800)]
Merge tag 'spi-fix-v6.8-rc4' of git://git./linux/kernel/git/broonie/spi

Pull spi fixes from Mark Brown:
 "A smallish collection of fixes for SPI, all driver specific, plus one
  device ID addition for a new Intel part.

  The ppc4xx isn't routinely covered by most of the automated testing so
  there were some errors that were missed in some of the recent API
  conversions, otherwise there's nothing super remarkable here"

* tag 'spi-fix-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi-mxs: Fix chipselect glitch
  spi: intel-pci: Add support for Lunar Lake-M SPI serial flash
  spi: omap2-mcspi: Revert FIFO support without DMA
  spi: ppc4xx: Drop write-only variable
  spi: ppc4xx: Fix fallout from rename in struct spi_bitbang
  spi: ppc4xx: Fix fallout from include cleanup
  spi: spi-ppc4xx: include missing platform_device.h
  spi: imx: fix the burst length at DMA mode and CPU mode

3 months agoMerge tag 'regmap-fix-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 15 Feb 2024 17:11:06 +0000 (09:11 -0800)]
Merge tag 'regmap-fix-v6.8-rc4' of git://git./linux/kernel/git/broonie/regmap

Pull regmap test fixes from Mark Brown:
 "Guenter runs a lot of KUnit tests so noticed that there were a couple
  of the regmap tests, including the newly added noinc test, which could
  show spurious failures due to the use of randomly generated test
  values. These changes handle the randomly generated data properly"

* tag 'regmap-fix-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
  regmap: kunit: Ensure that changed bytes are actually different
  regmap: kunit: fix raw noinc write test wrapping

3 months agoMerge tag 'hid-for-linus-2024021501' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 15 Feb 2024 17:08:19 +0000 (09:08 -0800)]
Merge tag 'hid-for-linus-2024021501' of git://git./linux/kernel/git/hid/hid

Pull HID fixes from Jiri Kosina:

 - fix for 'MSC_SERIAL = 0' corner case handling in wacom driver (Jason
   Gerecke)

 - ACPI S3 suspend/resume fix for intel-ish-hid (Even Xu)

 - race condition fix preventing Wacom driver from losing events shortly
   after initialization (Jason Gerecke)

 - fix preventing certain Logitech HID++ devices from spamming kernel
   log (Oleksandr Natalenko)

* tag 'hid-for-linus-2024021501' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:
  HID: wacom: generic: Avoid reporting a serial of '0' to userspace
  HID: Intel-ish-hid: Ishtp: Fix sensor reads after ACPI S3 suspend
  HID: multitouch: Add required quirk for Synaptics 0xcddc device
  HID: wacom: Do not register input devices until after hid_hw_start
  HID: logitech-hidpp: Do not flood kernel log

3 months agoMerge branch '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Jakub Kicinski [Thu, 15 Feb 2024 16:06:50 +0000 (08:06 -0800)]
Merge branch '1GbE' of git://git./linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2024-02-06 (igb, igc)

This series contains updates to igb and igc drivers.

Kunwu Chan adjusts firmware version string implementation to resolve
possible NULL pointer issue for igb.

Sasha removes workaround on igc.

* '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
  igc: Remove temporary workaround
  igb: Fix string truncation warnings in igb_set_fw_version
====================

Link: https://lore.kernel.org/r/20240214180347.3219650-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge branch 'fix-module_description-for-net-p6'
Jakub Kicinski [Thu, 15 Feb 2024 16:03:49 +0000 (08:03 -0800)]
Merge branch 'fix-module_description-for-net-p6'

Breno Leitao says:

====================
Fix MODULE_DESCRIPTION() for net (p6)

There are a few network modules left that misses MODULE_DESCRIPTION(),
causing a warnning when compiling with W=1. Example:

        WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/arcnet/....

This last patchset solves the problem for all the missing driver. It is
not expect to see any warning for the driver/net and net/ directory once
all these patches have landed.

v1: https://lore.kernel.org/all/20240213112122.404045-1-leitao@debian.org/
====================

Link: https://lore.kernel.org/r/20240214152741.670178-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for missing arcnet
Breno Leitao [Wed, 14 Feb 2024 15:27:41 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for missing arcnet

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the ARC modules.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240214152741.670178-8-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for mdio_devres
Breno Leitao [Wed, 14 Feb 2024 15:27:40 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for mdio_devres

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the PHY MDIO helpers.

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240214152741.670178-7-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for ppp
Breno Leitao [Wed, 14 Feb 2024 15:27:39 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for ppp

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the PPP modules.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240214152741.670178-6-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for fddik/skfp
Breno Leitao [Wed, 14 Feb 2024 15:27:38 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for fddik/skfp

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the SysKonnect FDDI PCI module.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240214152741.670178-5-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for plip
Breno Leitao [Wed, 14 Feb 2024 15:27:37 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for plip

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the PLIP (parallel port) network module

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240214152741.670178-4-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for ieee802154/fakelb
Breno Leitao [Wed, 14 Feb 2024 15:27:36 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for ieee802154/fakelb

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the IEEE 802.15.4 loopback driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240214152741.670178-3-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: fill in MODULE_DESCRIPTION()s for xen-netback
Breno Leitao [Wed, 14 Feb 2024 15:27:35 +0000 (07:27 -0800)]
net: fill in MODULE_DESCRIPTION()s for xen-netback

W=1 builds now warn if module is built without a MODULE_DESCRIPTION().
Add descriptions to the Xen backend network module.

Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Paul Durrant <paul@xen.org>
Link: https://lore.kernel.org/r/20240214152741.670178-2-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: ravb: Count packets instead of descriptors in GbEth RX path
Paul Barker [Wed, 14 Feb 2024 15:12:04 +0000 (15:12 +0000)]
net: ravb: Count packets instead of descriptors in GbEth RX path

The units of "work done" in the RX path should be packets instead of
descriptors, as large packets can be spread over multiple descriptors.

Fixes: 1c59eb678cbd ("ravb: Fillup ravb_rx_gbeth() stub")
Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://lore.kernel.org/r/20240214151204.2976-1-paul.barker.ct@bp.renesas.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agopppoe: Fix memory leak in pppoe_sendmsg()
Gavrilov Ilia [Wed, 14 Feb 2024 09:01:50 +0000 (09:01 +0000)]
pppoe: Fix memory leak in pppoe_sendmsg()

syzbot reports a memory leak in pppoe_sendmsg [1].

The problem is in the pppoe_recvmsg() function that handles errors
in the wrong order. For the skb_recv_datagram() function, check
the pointer to skb for NULL first, and then check the 'error' variable,
because the skb_recv_datagram() function can set 'error'
to -EAGAIN in a loop but return a correct pointer to socket buffer
after a number of attempts, though 'error' remains set to -EAGAIN.

skb_recv_datagram
      __skb_recv_datagram          // Loop. if (err == -EAGAIN) then
                                   // go to the next loop iteration
          __skb_try_recv_datagram  // if (skb != NULL) then return 'skb'
                                   // else if a signal is received then
                                   // return -EAGAIN

Found by InfoTeCS on behalf of Linux Verification Center
(linuxtesting.org) with Syzkaller.

Link: https://syzkaller.appspot.com/bug?extid=6bdfd184eac7709e5cc9
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Reported-by: syzbot+6bdfd184eac7709e5cc9@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=6bdfd184eac7709e5cc9
Signed-off-by: Gavrilov Ilia <Ilia.Gavrilov@infotecs.ru>
Reviewed-by: Guillaume Nault <gnault@redhat.com>
Link: https://lore.kernel.org/r/20240214085814.3894917-1-Ilia.Gavrilov@infotecs.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agonet: sctp: fix skb leak in sctp_inq_free()
Dmitry Antipov [Wed, 14 Feb 2024 08:22:24 +0000 (11:22 +0300)]
net: sctp: fix skb leak in sctp_inq_free()

In case of GSO, 'chunk->skb' pointer may point to an entry from
fraglist created in 'sctp_packet_gso_append()'. To avoid freeing
random fraglist entry (and so undefined behavior and/or memory
leak), introduce 'sctp_inq_chunk_free()' helper to ensure that
'chunk->skb' is set to 'chunk->head_skb' (i.e. fraglist head)
before calling 'sctp_chunk_free()', and use the aforementioned
helper in 'sctp_inq_pop()' as well.

Reported-by: syzbot+8bb053b5d63595ab47db@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?id=0d8351bbe54fd04a492c2daab0164138db008042
Fixes: 90017accff61 ("sctp: Add GSO support")
Suggested-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Acked-by: Xin Long <lucien.xin@gmail.com>
Link: https://lore.kernel.org/r/20240214082224.10168-1-dmantipov@yandex.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge branch 'net-ipv6-addrconf-ensure-that-temporary-addresses-preferred-lifetimes...
Paolo Abeni [Thu, 15 Feb 2024 14:34:42 +0000 (15:34 +0100)]
Merge branch 'net-ipv6-addrconf-ensure-that-temporary-addresses-preferred-lifetimes-are-long-enough'

Alex Henrie says:

====================
net: ipv6/addrconf: ensure that temporary addresses' preferred lifetimes are long enough

v2 corrects and updates the documentation for these features.

Changes from v1:
- Update the typical minimum lifetime stated in the documentation, and
  make it a range to emphasize the variability
- Fix spelling of "determine" in the documentation
- Mention RFC 8981's requirements in the documentation
- Arrange variables in "reverse Christmas tree"
- Update documentation of what happens if temp_prefered_lft is less
  than the minimum required lifetime

Thanks to David, Paolo, and Dan for your feedback.
====================

Link: https://lore.kernel.org/r/20240214062711.608363-1-alexhenrie24@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: ipv6/addrconf: clamp preferred_lft to the minimum required
Alex Henrie [Wed, 14 Feb 2024 06:26:32 +0000 (23:26 -0700)]
net: ipv6/addrconf: clamp preferred_lft to the minimum required

If the preferred lifetime was less than the minimum required lifetime,
ipv6_create_tempaddr would error out without creating any new address.
On my machine and network, this error happened immediately with the
preferred lifetime set to 5 seconds or less, after a few minutes with
the preferred lifetime set to 6 seconds, and not at all with the
preferred lifetime set to 7 seconds. During my investigation, I found a
Stack Exchange post from another person who seems to have had the same
problem: They stopped getting new addresses if they lowered the
preferred lifetime below 3 seconds, and they didn't really know why.

The preferred lifetime is a preference, not a hard requirement. The
kernel does not strictly forbid new connections on a deprecated address,
nor does it guarantee that the address will be disposed of the instant
its total valid lifetime expires. So rather than disable IPv6 privacy
extensions altogether if the minimum required lifetime swells above the
preferred lifetime, it is more in keeping with the user's intent to
increase the temporary address's lifetime to the minimum necessary for
the current network conditions.

With these fixes, setting the preferred lifetime to 5 or 6 seconds "just
works" because the extra fraction of a second is practically
unnoticeable. It's even possible to reduce the time before deprecation
to 1 or 2 seconds by setting /proc/sys/net/ipv6/conf/*/regen_min_advance
and /proc/sys/net/ipv6/conf/*/dad_transmits to 0. I realize that that is
a pretty niche use case, but I know at least one person who would gladly
sacrifice performance and convenience to be sure that they are getting
the maximum possible level of privacy.

Link: https://serverfault.com/a/1031168/310447
Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: ipv6/addrconf: introduce a regen_min_advance sysctl
Alex Henrie [Wed, 14 Feb 2024 06:26:31 +0000 (23:26 -0700)]
net: ipv6/addrconf: introduce a regen_min_advance sysctl

In RFC 8981, REGEN_ADVANCE cannot be less than 2 seconds, and the RFC
does not permit the creation of temporary addresses with lifetimes
shorter than that:

> When processing a Router Advertisement with a
> Prefix Information option carrying a prefix for the purposes of
> address autoconfiguration (i.e., the A bit is set), the host MUST
> perform the following steps:

> 5.  A temporary address is created only if this calculated preferred
>     lifetime is greater than REGEN_ADVANCE time units.

However, some users want to change their IPv6 address as frequently as
possible regardless of the RFC's arbitrary minimum lifetime. For the
benefit of those users, add a regen_min_advance sysctl parameter that
can be set to below or above 2 seconds.

Link: https://datatracker.ietf.org/doc/html/rfc8981
Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: ipv6/addrconf: ensure that regen_advance is at least 2 seconds
Alex Henrie [Wed, 14 Feb 2024 06:26:30 +0000 (23:26 -0700)]
net: ipv6/addrconf: ensure that regen_advance is at least 2 seconds

RFC 8981 defines REGEN_ADVANCE as follows:

REGEN_ADVANCE = 2 + (TEMP_IDGEN_RETRIES * DupAddrDetectTransmits * RetransTimer / 1000)

Thus, allowing it to be less than 2 seconds is technically a protocol
violation.

Link: https://datatracker.ietf.org/doc/html/rfc8981#name-defined-protocol-parameters
Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agoigc: Add support for LEDs on i225/i226
Kurt Kanzenbach [Tue, 13 Feb 2024 18:41:37 +0000 (10:41 -0800)]
igc: Add support for LEDs on i225/i226

Add support for LEDs on i225/i226. The LEDs can be controlled via sysfs
from user space using the netdev trigger. The LEDs are named as
igc-<bus><device>-<led> to be easily identified.

Offloading link speed and activity are supported. Other modes are simulated
in software by using on/off. Tested on Intel i225.

Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20240213184138.1483968-1-anthony.l.nguyen@intel.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: bcmasp: Handle RX buffer allocation failure
Florian Fainelli [Tue, 13 Feb 2024 17:33:39 +0000 (09:33 -0800)]
net: bcmasp: Handle RX buffer allocation failure

The buffer_pg variable needs to hold an order-5 allocation (32 x
PAGE_SIZE) which, under memory pressure may fail to be allocated. Deal
with that error condition properly to avoid doing a NULL pointer
de-reference in the subsequent call to dma_map_page().

In addition, the err_reclaim_tx error label in bcmasp_netif_init() needs
to ensure that the TX NAPI object is properly deleted, otherwise
unregister_netdev() will spin forever attempting to test and clear
the NAPI_STATE_HASHED bit.

Fixes: 490cb412007d ("net: bcmasp: Add support for ASP2.0 Ethernet controller")
Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Justin Chen <justin.chen@broadcom.com>
Link: https://lore.kernel.org/r/20240213173339.3438713-1-florian.fainelli@broadcom.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: ena: Remove unlikely() from IS_ERR() condition
Kamal Heib [Tue, 13 Feb 2024 16:15:02 +0000 (11:15 -0500)]
net: ena: Remove unlikely() from IS_ERR() condition

IS_ERR() is already using unlikely internally.

Signed-off-by: Kamal Heib <kheib@redhat.com>
Acked-by: Arthur Kiyanovski <akiyano@amazon.com>
Link: https://lore.kernel.org/r/20240213161502.2297048-1-kheib@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: mdio_bus: make mdio_bus_type const
Ricardo B. Marliere [Tue, 13 Feb 2024 14:48:00 +0000 (11:48 -0300)]
net: mdio_bus: make mdio_bus_type const

Since commit d492cc2573a0 ("driver core: device.h: make struct
bus_type a const *"), the driver core can properly handle constant
struct bus_type, move the mdio_bus_type variable to be a constant
structure as well, placing it into read-only memory which can not be
modified at runtime.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20240213-bus_cleanup-mdio-v1-1-f9e799da7fda@marliere.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agotipc: Cleanup tipc_nl_bearer_add() error paths
Shigeru Yoshida [Tue, 13 Feb 2024 13:40:58 +0000 (22:40 +0900)]
tipc: Cleanup tipc_nl_bearer_add() error paths

Consolidate the error paths of tipc_nl_bearer_add() under the common label
if the function holds rtnl_lock.

Signed-off-by: Shigeru Yoshida <syoshida@redhat.com>
Reviewed-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Link: https://lore.kernel.org/r/20240213134058.386123-1-syoshida@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: phy: aquantia: add AQR111 and AQR111B0 PHY ID
Christian Marangi [Tue, 13 Feb 2024 13:35:51 +0000 (14:35 +0100)]
net: phy: aquantia: add AQR111 and AQR111B0 PHY ID

Add Aquantia AQR111 and AQR111B0 PHY ID. These PHY advertise 10G speed
but actually supports up to 5G speed, hence some manual fixup is needed.

The Aquantia AQR111B0 PHY is just a variant of the AQR111 with smaller
chip size.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240213133558.1836-1-ansuelsmth@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agotcp: no need to use acceptable for conn_request
Jason Xing [Tue, 13 Feb 2024 13:12:05 +0000 (21:12 +0800)]
tcp: no need to use acceptable for conn_request

Since tcp_conn_request() always returns zero, there is no need to
keep the dead code. Remove it then.

Link: https://lore.kernel.org/netdev/CANn89iJwx9b2dUGUKFSV3PF=kN5o+kxz3A_fHZZsOS4AnXhBNw@mail.gmail.com/
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240213131205.4309-1-kerneljasonxing@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agoMerge tag 'nf-24-02-15' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Paolo Abeni [Thu, 15 Feb 2024 11:48:56 +0000 (12:48 +0100)]
Merge tag 'nf-24-02-15' of git://git./linux/kernel/git/netfilter/nf

Pablo Neira Ayuso says:

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

The following batch contains Netfilter fixes for net:

1) Missing : in kdoc field in nft_set_pipapo.

2) Restore default DNAT behavior When a DNAT rule is configured via
   iptables with different port ranges, from Kyle Swenson.

3) Restore flowtable hardware offload for bidirectional flows
   by setting NF_FLOW_HW_BIDIRECTIONAL flag, from Felix Fietkau.

netfilter pull request 24-02-15

* tag 'nf-24-02-15' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: nf_tables: fix bidirectional offload regression
  netfilter: nat: restore default DNAT behavior
  netfilter: nft_set_pipapo: fix missing : in kdoc
====================

Link: https://lore.kernel.org/r/20240214233818.7946-1-pablo@netfilter.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agoMerge tag 'linux-can-fixes-for-6.8-20240214' of git://git.kernel.org/pub/scm/linux...
Paolo Abeni [Thu, 15 Feb 2024 11:31:22 +0000 (12:31 +0100)]
Merge tag 'linux-can-fixes-for-6.8-20240214' of git://git./linux/kernel/git/mkl/linux-can

Marc Kleine-Budde says:

====================
pull-request: can 2024-02-14

this is a pull request of 3 patches for net/master.

the first patch is by Ziqi Zhao and targets the CAN J1939 protocol, it
fixes a potential deadlock by replacing the spinlock by an rwlock.

Oleksij Rempel's patch adds a missing spin_lock_bh() to prevent a
potential Use-After-Free in the CAN J1939's
setsockopt(SO_J1939_FILTER).

Maxime Jayat contributes a patch to fix the transceiver delay
compensation (TDCO) calculation, which is needed for higher CAN-FD bit
rates (usually 2Mbit/s).

* tag 'linux-can-fixes-for-6.8-20240214' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can:
  can: netlink: Fix TDCO calculation using the old data bittiming
  can: j1939: Fix UAF in j1939_sk_match_filter during setsockopt(SO_J1939_FILTER)
  can: j1939: prevent deadlock by changing j1939_socks_lock to rwlock
====================

Link: https://lore.kernel.org/r/20240214140348.2412776-1-mkl@pengutronix.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet-timestamp: make sk_tskey more predictable in error path
Vadim Fedorenko [Tue, 13 Feb 2024 11:04:28 +0000 (03:04 -0800)]
net-timestamp: make sk_tskey more predictable in error path

When SOF_TIMESTAMPING_OPT_ID is used to ambiguate timestamped datagrams,
the sk_tskey can become unpredictable in case of any error happened
during sendmsg(). Move increment later in the code and make decrement of
sk_tskey in error path. This solution is still racy in case of multiple
threads doing snedmsg() over the very same socket in parallel, but still
makes error path much more predictable.

Fixes: 09c2d251b707 ("net-timestamp: add key to disambiguate concurrent datagrams")
Reported-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240213110428.1681540-1-vadfed@meta.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agonet: phy: qca807x: move interface mode check to .config_init_once
Robert Marko [Mon, 12 Feb 2024 11:49:34 +0000 (12:49 +0100)]
net: phy: qca807x: move interface mode check to .config_init_once

Currently, we are checking whether the PHY package mode matches the
individual PHY interface modes at PHY package probe time, but at that time
we only know the PHY package mode and not the individual PHY interface
modes as of_get_phy_mode() that populates it will only get called once the
netdev to which PHY-s are attached to is being probed and thus this check
will always fail and return -EINVAL.

So, lets move this check to .config_init_once as at that point individual
PHY interface modes should be populated.

Fixes: d1cb613efbd3 ("net: phy: qcom: add support for QCA807x PHY Family")
Signed-off-by: Robert Marko <robimarko@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240212115043.1725918-1-robimarko@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 months agoselftests: tls: increase the wait in poll_partial_rec_async
Jakub Kicinski [Tue, 13 Feb 2024 14:20:55 +0000 (06:20 -0800)]
selftests: tls: increase the wait in poll_partial_rec_async

Test runners on debug kernels occasionally fail with:

 # #  RUN           tls_err.13_aes_gcm.poll_partial_rec_async ...
 # # tls.c:1883:poll_partial_rec_async:Expected poll(&pfd, 1, 5) (0) == 1 (1)
 # # tls.c:1870:poll_partial_rec_async:Expected status (256) == 0 (0)
 # # poll_partial_rec_async: Test failed at step #17
 # #          FAIL  tls_err.13_aes_gcm.poll_partial_rec_async
 # not ok 699 tls_err.13_aes_gcm.poll_partial_rec_async
 # # FAILED: 698 / 699 tests passed.

This points to the second poll() in the test which is expected
to wait for the sender to send the rest of the data.
Apparently under some conditions that doesn't happen within 5ms,
bump the timeout to 20ms.

Fixes: 23fcb62bc19c ("selftests: tls: add tests for poll behavior")
Link: https://lore.kernel.org/r/20240213142055.395564-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoice: Add check for lport extraction to LAG init
Dave Ertman [Tue, 13 Feb 2024 18:39:55 +0000 (10:39 -0800)]
ice: Add check for lport extraction to LAG init

To fully support initializing the LAG support code, a DDP package that
extracts the logical port from the metadata is required.  If such a
package is not present, there could be difficulties in supporting some
bond types.

Add a check into the initialization flow that will bypass the new paths
if any of the support pieces are missing.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Fixes: df006dd4b1dc ("ice: Add initial support framework for LAG")
Signed-off-by: Dave Ertman <david.m.ertman@intel.com>
Reviewed-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20240213183957.1483857-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge tag 'wireless-2024-02-14' of git://git.kernel.org/pub/scm/linux/kernel/git...
Jakub Kicinski [Thu, 15 Feb 2024 01:32:37 +0000 (17:32 -0800)]
Merge tag 'wireless-2024-02-14' of git://git./linux/kernel/git/wireless/wireless

Johannes Berg says:

====================
Valentine's day edition, with just few fixes because
that's how we love it ;-)

iwlwifi:
 - correct A3 in A-MSDUs
 - fix crash when operating as AP and running out of station
   slots to use
 - clear link ID to correct some later checks against it
 - fix error codes in SAR table loading
 - fix error path in PPAG table read

mac80211:
 - reload a pointer after SKB may have changed
   (only in certain monitor inject mode scenarios)

* tag 'wireless-2024-02-14' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
  wifi: iwlwifi: mvm: fix a crash when we run out of stations
  wifi: iwlwifi: uninitialized variable in iwl_acpi_get_ppag_table()
  wifi: iwlwifi: Fix some error codes
  wifi: iwlwifi: clear link_id in time_event
  wifi: iwlwifi: mvm: use correct address 3 in A-MSDU
  wifi: mac80211: reload info pointer in ieee80211_tx_dequeue()
====================

Link: https://lore.kernel.org/r/20240214184326.132813-3-johannes@sipsolutions.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 months agoMerge tag 'mips-fixes_6.8_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips...
Linus Torvalds [Thu, 15 Feb 2024 00:06:31 +0000 (16:06 -0800)]
Merge tag 'mips-fixes_6.8_2' of git://git./linux/kernel/git/mips/linux

Pull MIPS fixes from Thomas Bogendoerfer:

 - Fix for broken ipv6 checksums

 - Fix handling of exceptions in delay slots

* tag 'mips-fixes_6.8_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  mm/memory: Use exception ip to search exception tables
  MIPS: Clear Cause.BD in instruction_pointer_set
  ptrace: Introduce exception_ip arch hook
  MIPS: Add 'memory' clobber to csum_ipv6_magic() inline assembler

3 months agoMerge tag 'landlock-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/mic...
Linus Torvalds [Thu, 15 Feb 2024 00:02:36 +0000 (16:02 -0800)]
Merge tag 'landlock-6.8-rc5' of git://git./linux/kernel/git/mic/linux

Pull landlock test fixes from Mickaël Salaün:
 "Fix build issues for tests, and improve test compatibility"

* tag 'landlock-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/mic/linux:
  selftests/landlock: Fix capability for net_test
  selftests/landlock: Fix fs_test build with old libc
  selftests/landlock: Fix net_test build with old libc

3 months agoMerge tag 'for-6.8-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave...
Linus Torvalds [Wed, 14 Feb 2024 23:47:02 +0000 (15:47 -0800)]
Merge tag 'for-6.8-rc4-tag' of git://git./linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "A few regular fixes and one fix for space reservation regression since
  6.7 that users have been reporting:

   - fix over-reservation of metadata chunks due to not keeping proper
     balance between global block reserve and delayed refs reserve; in
     practice this leaves behind empty metadata block groups, the
     workaround is to reclaim them by using the '-musage=1' balance
     filter

   - other space reservation fixes:
      - do not delete unused block group if it may be used soon
      - do not reserve space for checksums for NOCOW files

   - fix extent map assertion failure when writing out free space inode

   - reject encoded write if inode has nodatasum flag set

   - fix chunk map leak when loading block group zone info"

* tag 'for-6.8-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: don't refill whole delayed refs block reserve when starting transaction
  btrfs: zoned: fix chunk map leak when loading block group zone info
  btrfs: reject encoded write if inode has nodatasum flag set
  btrfs: don't reserve space for checksums when writing to nocow files
  btrfs: add new unused block groups to the list of unused block groups
  btrfs: do not delete unused block group if it may be used soon
  btrfs: add and use helper to check if block group is used
  btrfs: don't drop extent_map for free space inode on write error

3 months agoMerge tag 'linux_kselftest-kunit-fixes-6.8-rc5' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Wed, 14 Feb 2024 23:34:03 +0000 (15:34 -0800)]
Merge tag 'linux_kselftest-kunit-fixes-6.8-rc5' of git://git./linux/kernel/git/shuah/linux-kselftest

Pull KUnit fix from Shuah Khan:
 "One important fix to unregister kunit_bus when KUnit module is
  unloaded.

  Not doing so causes an error when KUnit module tries to re-register
  the bus when it gets reloaded"

* tag 'linux_kselftest-kunit-fixes-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
  kunit: device: Unregister the kunit_bus on shutdown

3 months agonetfilter: nf_tables: fix bidirectional offload regression
Felix Fietkau [Wed, 14 Feb 2024 14:42:35 +0000 (15:42 +0100)]
netfilter: nf_tables: fix bidirectional offload regression

Commit 8f84780b84d6 ("netfilter: flowtable: allow unidirectional rules")
made unidirectional flow offload possible, while completely ignoring (and
breaking) bidirectional flow offload for nftables.
Add the missing flag that was left out as an exercise for the reader :)

Cc: Vlad Buslov <vladbu@nvidia.com>
Fixes: 8f84780b84d6 ("netfilter: flowtable: allow unidirectional rules")
Reported-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
3 months agonetfilter: nat: restore default DNAT behavior
Kyle Swenson [Thu, 8 Feb 2024 23:56:31 +0000 (23:56 +0000)]
netfilter: nat: restore default DNAT behavior

When a DNAT rule is configured via iptables with different port ranges,

iptables -t nat -A PREROUTING -p tcp -d 10.0.0.2 -m tcp --dport 32000:32010
-j DNAT --to-destination 192.168.0.10:21000-21010

we seem to be DNATing to some random port on the LAN side. While this is
expected if --random is passed to the iptables command, it is not
expected without passing --random.  The expected behavior (and the
observed behavior prior to the commit in the "Fixes" tag) is the traffic
will be DNAT'd to 192.168.0.10:21000 unless there is a tuple collision
with that destination.  In that case, we expect the traffic to be
instead DNAT'd to 192.168.0.10:21001, so on so forth until the end of
the range.

This patch intends to restore the behavior observed prior to the "Fixes"
tag.

Fixes: 6ed5943f8735 ("netfilter: nat: remove l4 protocol port rovers")
Signed-off-by: Kyle Swenson <kyle.swenson@est.tech>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
3 months agonetfilter: nft_set_pipapo: fix missing : in kdoc
Pablo Neira Ayuso [Thu, 8 Feb 2024 14:46:03 +0000 (15:46 +0100)]
netfilter: nft_set_pipapo: fix missing : in kdoc

Add missing : in kdoc field names.

Fixes: 8683f4b9950d ("nft_set_pipapo: Prepare for vectorised implementation: helpers")
Reported-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
3 months agoigc: Remove temporary workaround
Sasha Neftin [Wed, 24 Jan 2024 05:57:00 +0000 (07:57 +0200)]
igc: Remove temporary workaround

PHY_CONTROL register works as defined in the IEEE 802.3 specification
(IEEE 802.3-2008 22.2.4.1). Tidy up the temporary workaround.

User impact: PHY can now be powered down when the ethernet link is down.

Testing hints: ip link set down <device> (or just disconnect the
ethernet cable).

Oldest tested NVM version is: 1045:740.

Fixes: 5586838fe9ce ("igc: Add code for PHY support")
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 months agoigb: Fix string truncation warnings in igb_set_fw_version
Kunwu Chan [Mon, 15 Jan 2024 08:28:25 +0000 (16:28 +0800)]
igb: Fix string truncation warnings in igb_set_fw_version

Commit 1978d3ead82c ("intel: fix string truncation warnings")
fixes '-Wformat-truncation=' warnings in igb_main.c by using kasprintf.

drivers/net/ethernet/intel/igb/igb_main.c:3092:53: warning:‘%d’ directive output may be truncated writing between 1 and 5 bytes into a region of size between 1 and 13 [-Wformat-truncation=]
 3092 |                                  "%d.%d, 0x%08x, %d.%d.%d",
      |                                                     ^~
drivers/net/ethernet/intel/igb/igb_main.c:3092:34: note:directive argument in the range [0, 65535]
 3092 |                                  "%d.%d, 0x%08x, %d.%d.%d",
      |                                  ^~~~~~~~~~~~~~~~~~~~~~~~~
drivers/net/ethernet/intel/igb/igb_main.c:3092:34: note:directive argument in the range [0, 65535]
drivers/net/ethernet/intel/igb/igb_main.c:3090:25: note:‘snprintf’ output between 23 and 43 bytes into a destination of size 32

kasprintf() returns a pointer to dynamically allocated memory
which can be NULL upon failure.

Fix this warning by using a larger space for adapter->fw_version,
and then fall back and continue to use snprintf.

Fixes: 1978d3ead82c ("intel: fix string truncation warnings")
Signed-off-by: Kunwu Chan <chentao@kylinos.cn>
Cc: Kunwu Chan <kunwu.chan@hotmail.com>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 months agocan: netlink: Fix TDCO calculation using the old data bittiming
Maxime Jayat [Mon, 6 Nov 2023 18:01:58 +0000 (19:01 +0100)]
can: netlink: Fix TDCO calculation using the old data bittiming

The TDCO calculation was done using the currently applied data bittiming,
instead of the newly computed data bittiming, which means that the TDCO
had an invalid value unless setting the same data bittiming twice.

Fixes: d99755f71a80 ("can: netlink: add interface for CAN-FD Transmitter Delay Compensation (TDC)")
Signed-off-by: Maxime Jayat <maxime.jayat@mobile-devices.fr>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/40579c18-63c0-43a4-8d4c-f3a6c1c0b417@munic.io
Cc: stable@vger.kernel.org
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
3 months agocan: j1939: Fix UAF in j1939_sk_match_filter during setsockopt(SO_J1939_FILTER)
Oleksij Rempel [Fri, 20 Oct 2023 13:38:14 +0000 (15:38 +0200)]
can: j1939: Fix UAF in j1939_sk_match_filter during setsockopt(SO_J1939_FILTER)

Lock jsk->sk to prevent UAF when setsockopt(..., SO_J1939_FILTER, ...)
modifies jsk->filters while receiving packets.

Following trace was seen on affected system:
 ==================================================================
 BUG: KASAN: slab-use-after-free in j1939_sk_recv_match_one+0x1af/0x2d0 [can_j1939]
 Read of size 4 at addr ffff888012144014 by task j1939/350

 CPU: 0 PID: 350 Comm: j1939 Tainted: G        W  OE      6.5.0-rc5 #1
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
 Call Trace:
  print_report+0xd3/0x620
  ? kasan_complete_mode_report_info+0x7d/0x200
  ? j1939_sk_recv_match_one+0x1af/0x2d0 [can_j1939]
  kasan_report+0xc2/0x100
  ? j1939_sk_recv_match_one+0x1af/0x2d0 [can_j1939]
  __asan_load4+0x84/0xb0
  j1939_sk_recv_match_one+0x1af/0x2d0 [can_j1939]
  j1939_sk_recv+0x20b/0x320 [can_j1939]
  ? __kasan_check_write+0x18/0x20
  ? __pfx_j1939_sk_recv+0x10/0x10 [can_j1939]
  ? j1939_simple_recv+0x69/0x280 [can_j1939]
  ? j1939_ac_recv+0x5e/0x310 [can_j1939]
  j1939_can_recv+0x43f/0x580 [can_j1939]
  ? __pfx_j1939_can_recv+0x10/0x10 [can_j1939]
  ? raw_rcv+0x42/0x3c0 [can_raw]
  ? __pfx_j1939_can_recv+0x10/0x10 [can_j1939]
  can_rcv_filter+0x11f/0x350 [can]
  can_receive+0x12f/0x190 [can]
  ? __pfx_can_rcv+0x10/0x10 [can]
  can_rcv+0xdd/0x130 [can]
  ? __pfx_can_rcv+0x10/0x10 [can]
  __netif_receive_skb_one_core+0x13d/0x150
  ? __pfx___netif_receive_skb_one_core+0x10/0x10
  ? __kasan_check_write+0x18/0x20
  ? _raw_spin_lock_irq+0x8c/0xe0
  __netif_receive_skb+0x23/0xb0
  process_backlog+0x107/0x260
  __napi_poll+0x69/0x310
  net_rx_action+0x2a1/0x580
  ? __pfx_net_rx_action+0x10/0x10
  ? __pfx__raw_spin_lock+0x10/0x10
  ? handle_irq_event+0x7d/0xa0
  __do_softirq+0xf3/0x3f8
  do_softirq+0x53/0x80
  </IRQ>
  <TASK>
  __local_bh_enable_ip+0x6e/0x70
  netif_rx+0x16b/0x180
  can_send+0x32b/0x520 [can]
  ? __pfx_can_send+0x10/0x10 [can]
  ? __check_object_size+0x299/0x410
  raw_sendmsg+0x572/0x6d0 [can_raw]
  ? __pfx_raw_sendmsg+0x10/0x10 [can_raw]
  ? apparmor_socket_sendmsg+0x2f/0x40
  ? __pfx_raw_sendmsg+0x10/0x10 [can_raw]
  sock_sendmsg+0xef/0x100
  sock_write_iter+0x162/0x220
  ? __pfx_sock_write_iter+0x10/0x10
  ? __rtnl_unlock+0x47/0x80
  ? security_file_permission+0x54/0x320
  vfs_write+0x6ba/0x750
  ? __pfx_vfs_write+0x10/0x10
  ? __fget_light+0x1ca/0x1f0
  ? __rcu_read_unlock+0x5b/0x280
  ksys_write+0x143/0x170
  ? __pfx_ksys_write+0x10/0x10
  ? __kasan_check_read+0x15/0x20
  ? fpregs_assert_state_consistent+0x62/0x70
  __x64_sys_write+0x47/0x60
  do_syscall_64+0x60/0x90
  ? do_syscall_64+0x6d/0x90
  ? irqentry_exit+0x3f/0x50
  ? exc_page_fault+0x79/0xf0
  entry_SYSCALL_64_after_hwframe+0x6e/0xd8

 Allocated by task 348:
  kasan_save_stack+0x2a/0x50
  kasan_set_track+0x29/0x40
  kasan_save_alloc_info+0x1f/0x30
  __kasan_kmalloc+0xb5/0xc0
  __kmalloc_node_track_caller+0x67/0x160
  j1939_sk_setsockopt+0x284/0x450 [can_j1939]
  __sys_setsockopt+0x15c/0x2f0
  __x64_sys_setsockopt+0x6b/0x80
  do_syscall_64+0x60/0x90
  entry_SYSCALL_64_after_hwframe+0x6e/0xd8

 Freed by task 349:
  kasan_save_stack+0x2a/0x50
  kasan_set_track+0x29/0x40
  kasan_save_free_info+0x2f/0x50
  __kasan_slab_free+0x12e/0x1c0
  __kmem_cache_free+0x1b9/0x380
  kfree+0x7a/0x120
  j1939_sk_setsockopt+0x3b2/0x450 [can_j1939]
  __sys_setsockopt+0x15c/0x2f0
  __x64_sys_setsockopt+0x6b/0x80
  do_syscall_64+0x60/0x90
  entry_SYSCALL_64_after_hwframe+0x6e/0xd8

Fixes: 9d71dd0c70099 ("can: add support of SAE J1939 protocol")
Reported-by: Sili Luo <rootlab@huawei.com>
Suggested-by: Sili Luo <rootlab@huawei.com>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: stable@vger.kernel.org
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://lore.kernel.org/all/20231020133814.383996-1-o.rempel@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
3 months agocan: j1939: prevent deadlock by changing j1939_socks_lock to rwlock
Ziqi Zhao [Fri, 21 Jul 2023 16:22:26 +0000 (09:22 -0700)]
can: j1939: prevent deadlock by changing j1939_socks_lock to rwlock

The following 3 locks would race against each other, causing the
deadlock situation in the Syzbot bug report:

- j1939_socks_lock
- active_session_list_lock
- sk_session_queue_lock

A reasonable fix is to change j1939_socks_lock to an rwlock, since in
the rare situations where a write lock is required for the linked list
that j1939_socks_lock is protecting, the code does not attempt to
acquire any more locks. This would break the circular lock dependency,
where, for example, the current thread already locks j1939_socks_lock
and attempts to acquire sk_session_queue_lock, and at the same time,
another thread attempts to acquire j1939_socks_lock while holding
sk_session_queue_lock.

NOTE: This patch along does not fix the unregister_netdevice bug
reported by Syzbot; instead, it solves a deadlock situation to prepare
for one or more further patches to actually fix the Syzbot bug, which
appears to be a reference counting problem within the j1939 codebase.

Reported-by: <syzbot+1591462f226d9cbf0564@syzkaller.appspotmail.com>
Signed-off-by: Ziqi Zhao <astrajoan@yahoo.com>
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://lore.kernel.org/all/20230721162226.8639-1-astrajoan@yahoo.com
[mkl: remove unrelated newline change]
Cc: stable@vger.kernel.org
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
3 months agoethernet: cpts: fix function pointer cast warnings
Arnd Bergmann [Tue, 13 Feb 2024 10:16:34 +0000 (11:16 +0100)]
ethernet: cpts: fix function pointer cast warnings

clang-16 warns about the mismatched prototypes for the devm_* callbacks:

drivers/net/ethernet/ti/cpts.c:691:12: error: cast from 'void (*)(struct clk_hw *)' to 'void (*)(void *)' converts to incompatible function type [-Werror,-Wcast-function-type-strict]
  691 |                                        (void(*)(void *))clk_hw_unregister_mux,
      |                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/device.h:406:34: note: expanded from macro 'devm_add_action_or_reset'
  406 |         __devm_add_action_or_reset(dev, action, data, #action)
      |                                         ^~~~~~
drivers/net/ethernet/ti/cpts.c:703:12: error: cast from 'void (*)(struct device_node *)' to 'void (*)(void *)' converts to incompatible function type [-Werror,-Wcast-function-type-strict]
  703 |                                        (void(*)(void *))of_clk_del_provider,
      |                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/device.h:406:34: note: expanded from macro 'devm_add_action_or_reset'
  406 |         __devm_add_action_or_reset(dev, action, data, #action)

Use separate helper functions for this instead, using the expected prototypes
with a void* argument.

Fixes: a3047a81ba13 ("net: ethernet: ti: cpts: add support for ext rftclk selection")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agobnad: fix work_queue type mismatch
Arnd Bergmann [Tue, 13 Feb 2024 10:07:50 +0000 (11:07 +0100)]
bnad: fix work_queue type mismatch

clang-16 warns about a function pointer cast:

drivers/net/ethernet/brocade/bna/bnad.c:1995:4: error: cast from 'void (*)(struct delayed_work *)' to 'work_func_t' (aka 'void (*)(struct work_struct *)') converts to incompatible function type [-Werror,-Wcast-function-type-strict]
 1995 |                         (work_func_t)bnad_tx_cleanup);
drivers/net/ethernet/brocade/bna/bnad.c:2252:4: error: cast from 'void (*)(void *)' to 'work_func_t' (aka 'void (*)(struct work_struct *)') converts to incompatible function type [-Werror,-Wcast-function-type-strict]
 2252 |                         (work_func_t)(bnad_rx_cleanup));

The problem here is mixing up work_struct and delayed_work, which relies
the former being the first member of the latter.

Change the code to use consistent types here to address the warning and
make it more robust against workqueue interface changes.

Side note: the use of a delayed workqueue for cleaning up TX descriptors
is probably a bad idea since this introduces a noticeable delay. The
driver currently does not appear to use BQL, but if one wanted to add
that, this would have to be changed as well.

Fixes: 01b54b145185 ("bna: tx rx cleanup fix")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: phy: dp83826: support TX data voltage tuning
Catalin Popescu [Tue, 13 Feb 2024 08:07:05 +0000 (09:07 +0100)]
net: phy: dp83826: support TX data voltage tuning

DP83826 offers the possibility to tune the voltage of logical
levels of the MLT-3 encoded TX data. This is useful when there
is a voltage drop in between the PHY and the connector and we
want to increase the voltage levels to compensate for that drop.

Prior to PHY configuration, the driver SW resets the PHY which has
the same effect as the HW reset pin according to the datasheet.
Hence, there's no need to force update the VOD_CFG registers to make
sure they hold their reset values. VOD_CFG registers need to be
updated only if the DT has been configured with values other than
the reset ones.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Catalin Popescu <catalin.popescu@leica-geosystems.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agodt-bindings: net: dp83826: support TX data voltage tuning
Catalin Popescu [Tue, 13 Feb 2024 08:07:04 +0000 (09:07 +0100)]
dt-bindings: net: dp83826: support TX data voltage tuning

Add properties ti,cfg-dac-minus-one-bp/ti,cfg-dac-plus-one-bp
to support voltage tuning of logical levels -1/+1 of the MLT-3
encoded TX data.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Catalin Popescu <catalin.popescu@leica-geosystems.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch 'dev_base_lock-remove'
David S. Miller [Wed, 14 Feb 2024 11:20:14 +0000 (11:20 +0000)]
Merge branch 'dev_base_lock-remove'

Eric Dumazet says:

====================
net: complete dev_base_lock removal

Back in 2009 we started an effort to get rid of dev_base_lock
in favor of RCU.

It is time to finish this work.

Say goodbye to dev_base_lock !

v4: rebase, and move dev_addr_sem to net/core/dev.h in patch 06/13 (Jakub)

v3: I misread kbot reports, the issue was with dev->operstate (patch 10/13)
    So dev->reg_state is back to u8, and dev->operstate becomes an u32.
    Sorry for the noise.

v2: dev->reg_state must be a standard enum, some arches
    do not support cmpxchg() on u8.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: remove dev_base_lock
Eric Dumazet [Tue, 13 Feb 2024 06:32:45 +0000 (06:32 +0000)]
net: remove dev_base_lock

dev_base_lock is not needed anymore, all remaining users also hold RTNL.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: remove dev_base_lock from register_netdevice() and friends.
Eric Dumazet [Tue, 13 Feb 2024 06:32:44 +0000 (06:32 +0000)]
net: remove dev_base_lock from register_netdevice() and friends.

RTNL already protects writes to dev->reg_state, we no longer need to hold
dev_base_lock to protect the readers.

unlist_netdevice() second argument can be removed.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: remove dev_base_lock from do_setlink()
Eric Dumazet [Tue, 13 Feb 2024 06:32:43 +0000 (06:32 +0000)]
net: remove dev_base_lock from do_setlink()

We hold RTNL here, and dev->link_mode readers already
are using READ_ONCE().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: add netdev_set_operstate() helper
Eric Dumazet [Tue, 13 Feb 2024 06:32:42 +0000 (06:32 +0000)]
net: add netdev_set_operstate() helper

dev_base_lock is going away, add netdev_set_operstate() helper
so that hsr does not have to know core internals.

Remove dev_base_lock acquisition from rfc2863_policy()

v3: use an "unsigned int" for dev->operstate,
    so that try_cmpxchg() can work on all arches.
        ( https://lore.kernel.org/oe-kbuild-all/202402081918.OLyGaea3-lkp@intel.com/ )

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: remove stale mentions of dev_base_lock in comments
Eric Dumazet [Tue, 13 Feb 2024 06:32:41 +0000 (06:32 +0000)]
net: remove stale mentions of dev_base_lock in comments

Change comments incorrectly mentioning dev_base_lock.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet-sysfs: convert netstat_show() to RCU
Eric Dumazet [Tue, 13 Feb 2024 06:32:40 +0000 (06:32 +0000)]
net-sysfs: convert netstat_show() to RCU

dev_get_stats() can be called from RCU, there is no need
to acquire dev_base_lock.

Change dev_isalive() comment to reflect we no longer use
dev_base_lock from net/core/net-sysfs.c

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet-sysfs: convert dev->operstate reads to lockless ones
Eric Dumazet [Tue, 13 Feb 2024 06:32:39 +0000 (06:32 +0000)]
net-sysfs: convert dev->operstate reads to lockless ones

operstate_show() can omit dev_base_lock acquisition only
to read dev->operstate.

Annotate accesses to dev->operstate.

Writers still acquire dev_base_lock for mutual exclusion.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet-sysfs: use dev_addr_sem to remove races in address_show()
Eric Dumazet [Tue, 13 Feb 2024 06:32:38 +0000 (06:32 +0000)]
net-sysfs: use dev_addr_sem to remove races in address_show()

Using dev_base_lock is not preventing from reading garbage.

Use dev_addr_sem instead.

v4: place dev_addr_sem extern in net/core/dev.h (Jakub Kicinski)
Link: https://lore.kernel.org/netdev/20240212175845.10f6680a@kernel.org/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet-sysfs: convert netdev_show() to RCU
Eric Dumazet [Tue, 13 Feb 2024 06:32:37 +0000 (06:32 +0000)]
net-sysfs: convert netdev_show() to RCU

Make clear dev_isalive() can be called with RCU protection.

Then convert netdev_show() to RCU, to remove dev_base_lock
dependency.

Also add RCU to broadcast_show().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: convert dev->reg_state to u8
Eric Dumazet [Tue, 13 Feb 2024 06:32:36 +0000 (06:32 +0000)]
net: convert dev->reg_state to u8

Prepares things so that dev->reg_state reads can be lockless,
by adding WRITE_ONCE() on write side.

READ_ONCE()/WRITE_ONCE() do not support bitfields.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agodev: annotate accesses to dev->link
Eric Dumazet [Tue, 13 Feb 2024 06:32:35 +0000 (06:32 +0000)]
dev: annotate accesses to dev->link

Following patch will read dev->link locklessly,
annotate the write from do_setlink().

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoip_tunnel: annotate data-races around t->parms.link
Eric Dumazet [Tue, 13 Feb 2024 06:32:34 +0000 (06:32 +0000)]
ip_tunnel: annotate data-races around t->parms.link

t->parms.link is read locklessly, annotate these reads
and opposite writes accordingly.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: annotate data-races around dev->name_assign_type
Eric Dumazet [Tue, 13 Feb 2024 06:32:33 +0000 (06:32 +0000)]
net: annotate data-races around dev->name_assign_type

name_assign_type_show() runs locklessly, we should annotate
accesses to dev->name_assign_type.

Alternative would be to grab devnet_rename_sem semaphore
from name_assign_type_show(), but this would not bring
more accuracy.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch 'per-epoll-context-busy-poll'
David S. Miller [Wed, 14 Feb 2024 11:01:01 +0000 (11:01 +0000)]
Merge branch 'per-epoll-context-busy-poll'

Joe Damato says:

====================
Per epoll context busy poll support

Greetings:

Welcome to v8.

TL;DR This builds on commit bf3b9f6372c4 ("epoll: Add busy poll support to
epoll with socket fds.") by allowing user applications to enable
epoll-based busy polling, set a busy poll packet budget, and enable or
disable prefer busy poll on a per epoll context basis.

This makes epoll-based busy polling much more usable for user
applications than the current system-wide sysctl and hardcoded budget.

To allow for this, two ioctls have been added for epoll contexts for
getting and setting a new struct, struct epoll_params.

ioctl was chosen vs a new syscall after reviewing a suggestion by Willem
de Bruijn [1]. I am open to using a new syscall instead of an ioctl, but it
seemed that:
  - Busy poll affects all existing epoll_wait and epoll_pwait variants in
    the same way, so new verions of many syscalls might be needed. It
    seems much simpler for users to use the correct
    epoll_wait/epoll_pwait for their app and add a call to ioctl to enable
    or disable busy poll as needed. This also probably means less work to
    get an existing epoll app using busy poll.

  - previously added epoll_pwait2 helped to bring epoll closer to
    existing syscalls (like pselect and ppoll) and this busy poll change
    reflected as a new syscall would not have the same effect.

Note: patch 1/4 as of v4 uses an or (||) instead of an xor. I thought about
it some more and I realized that if the user enables both the per-epoll
context setting and the system wide sysctl, then busy poll should be
enabled and not disabled. Using xor doesn't seem to make much sense after
thinking through this a bit.

Longer explanation:

Presently epoll has support for a very useful form of busy poll based on
the incoming NAPI ID (see also: SO_INCOMING_NAPI_ID [2]).

This form of busy poll allows epoll_wait to drive NAPI packet processing
which allows for a few interesting user application designs which can
reduce latency and also potentially improve L2/L3 cache hit rates by
deferring NAPI until userland has finished its work.

The documentation available on this is, IMHO, a bit confusing so please
allow me to explain how one might use this:

1. Ensure each application thread has its own epoll instance mapping
1-to-1 with NIC RX queues. An n-tuple filter would likely be used to
direct connections with specific dest ports to these queues.

2. Optionally: Setup IRQ coalescing for the NIC RX queues where busy
polling will occur. This can help avoid the userland app from being
pre-empted by a hard IRQ while userland is running. Note this means that
userland must take care to call epoll_wait and not take too long in
userland since it now drives NAPI via epoll_wait.

3. Optionally: Consider using napi_defer_hard_irqs and gro_flush_timeout to
further restrict IRQ generation from the NIC. These settings are
system-wide so their impact must be carefully weighed against the running
applications.

4. Ensure that all incoming connections added to an epoll instance
have the same NAPI ID. This can be done with a BPF filter when
SO_REUSEPORT is used or getsockopt + SO_INCOMING_NAPI_ID when a single
accept thread is used which dispatches incoming connections to threads.

5. Lastly, busy poll must be enabled via a sysctl
(/proc/sys/net/core/busy_poll).

Please see Eric Dumazet's paper about busy polling [3] and a recent
academic paper about measured performance improvements of busy polling [4]
(albeit with a modification that is not currently present in the kernel)
for additional context.

The unfortunate part about step 5 above is that this enables busy poll
system-wide which affects all user applications on the system,
including epoll-based network applications which were not intended to
be used this way or applications where increased CPU usage for lower
latency network processing is unnecessary or not desirable.

If the user wants to run one low latency epoll-based server application
with epoll-based busy poll, but would like to run the rest of the
applications on the system (which may also use epoll) without busy poll,
this system-wide sysctl presents a significant problem.

This change preserves the system-wide sysctl, but adds a mechanism (via
ioctl) to enable or disable busy poll for epoll contexts as needed by
individual applications, making epoll-based busy poll more usable.

Note that this change includes an or (as of v4) instead of an xor. If the
user has enabled both the system-wide sysctl and also the per epoll-context
busy poll settings, then epoll should probably busy poll (vs being
disabled).

Thanks,
Joe

v7 -> v8:
   - Reviewed-by tag from Eric Dumazet applied to commit message of patch
     1/4.

   - patch 4/4:
     - EPIOCSPARAMS and EPIOCGPARAMS updated to use WRITE_ONCE and
       READ_ONCE, as requested by Eric Dumazet
     - Wrapped a long line (via netdev/checkpatch)

v6 -> v7:
   - Acked-by tags from Stanislav Fomichev applied to commit messages of
     all patches.
   - Reviewed-by tags from Jakub Kicinski, Eric Dumazet applied to commit
     messages of patches 2 and 3. Jiri Slaby's Reviewed-by applied to patch
     4.

   - patch 1/4:
     - busy_poll_usecs reduced from u64 to u32.
     - Unnecessary parens removed (via netdev/checkpatch)
     - Wrapped long line (via netdev/checkpatch)
     - Remove inline from busy_loop_ep_timeout as objdump suggests the
       function is already inlined
     - Moved struct eventpoll assignment to declaration
     - busy_loop_ep_timeout is moved within CONFIG_NET_RX_BUSY_POLL and the
       ifdefs internally have been removed as per Eric Dumazet's review
     - Removed ep_busy_loop_on from the !defined CONFIG_NET_RX_BUSY_POLL
       section as it is only called when CONFIG_NET_RX_BUSY_POLL is
       defined

   - patch 3/4:
     - Fix whitespace alignment issue (via netdev/checkpatch)

   - patch 4/4:
     - epoll_params.busy_poll_usecs has been reduced to u32
     - epoll_params.busy_poll_usecs is now checked to ensure it is <=
       S32_MAX
     - __pad has been reduced to a single u8
     - memchr_inv has been dropped and replaced with a simple check for the
       single __pad byte
     - Removed space after cast (via netdev/checkpatch)
     - Wrap long line (via netdev/checkpatch)
     - Move struct eventpoll *ep assignment to declaration as per Jiri
       Slaby's review
     - Remove unnecessary !! as per Jiri Slaby's review
     - Reorganized variables to be reverse christmas tree order

v5 -> v6:
  - patch 1/3 no functional change, but commit message corrected to explain
    that an or (||) is being used instead of xor.

  - patch 3/4 is a new patch which adds support for per epoll context
    prefer busy poll setting.

  - patch 4/4 updated to allow getting/setting per epoll context prefer
    busy poll setting; this setting is limited to either 0 or 1.

v4 -> v5:
  - patch 3/3 updated to use memchr_inv to ensure that __pad is zero for
    the EPIOCSPARAMS ioctl. Recommended by Greg K-H [5], Dave Chinner [6],
    and Jiri Slaby [7].

v3 -> v4:
  - patch 1/3 was updated to include an important functional change:
    ep_busy_loop_on was updated to use or (||) instead of xor (^). After
    thinking about it a bit more, I thought xor didn't make much sense.
    Enabling both the per-epoll context and the system-wide sysctl should
    probably enable busy poll, not disable it. So, or (||) makes more
    sense, I think.

  - patch 3/3 was updated:
    - to change the epoll_params fields to be __u64, __u16, and __u8 and
      to pad the struct to a multiple of 64bits. Suggested by Greg K-H [8]
      and Arnd Bergmann [9].
    - remove an unused pr_fmt, left over from the previous revision.
    - ioctl now returns -EINVAL when epoll_params.busy_poll_usecs >
      U32_MAX.

v2 -> v3:
  - cover letter updated to mention why ioctl seems (to me) like a better
    choice vs a new syscall.

  - patch 3/4 was modified in 3 ways:
    - when an unknown ioctl is received, -ENOIOCTLCMD is returned instead
      of -EINVAL as the ioctl documentation requires.
    - epoll_params.busy_poll_budget can only be set to a value larger than
      NAPI_POLL_WEIGHT if code is run by privileged (CAP_NET_ADMIN) users.
      Otherwise, -EPERM is returned.
    - busy poll specific ioctl code moved out to its own function. On
      kernels without busy poll support, -EOPNOTSUPP is returned. This also
      makes the kernel build robot happier without littering the code with
      more #ifdefs.

  - dropped patch 4/4 after Eric Dumazet's review of it when it was sent
    independently to the list [10].

v1 -> v2:
  - cover letter updated to make a mention of napi_defer_hard_irqs and
    gro_flush_timeout as an added step 3 and to cite both Eric Dumazet's
    busy polling paper and a paper from University of Waterloo for
    additional context. Specifically calling out the xor in patch 1/4
    incase it is missed by reviewers.

  - Patch 2/4 has its commit message updated, but no functional changes.
    Commit message now describes that allowing for a settable budget helps
    to improve throughput and is more consistent with other busy poll
    mechanisms that allow a settable budget via SO_BUSY_POLL_BUDGET.

  - Patch 3/4 was modified to check if the epoll_params.busy_poll_budget
    exceeds NAPI_POLL_WEIGHT. The larger value is allowed, but an error is
    printed. This was done for consistency with netif_napi_add_weight,
    which does the same.

  - Patch 3/4 the struct epoll_params was updated to fix the type of the
    data field; it was uint8_t and was changed to u8.

  - Patch 4/4 added to check if SO_BUSY_POLL_BUDGET exceeds
    NAPI_POLL_WEIGHT. The larger value is allowed, but an error is
    printed. This was done for consistency with netif_napi_add_weight,
    which does the same.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoeventpoll: Add epoll ioctl for epoll_params
Joe Damato [Tue, 13 Feb 2024 06:16:45 +0000 (06:16 +0000)]
eventpoll: Add epoll ioctl for epoll_params

Add an ioctl for getting and setting epoll_params. User programs can use
this ioctl to get and set the busy poll usec time, packet budget, and
prefer busy poll params for a specific epoll context.

Parameters are limited:
  - busy_poll_usecs is limited to <= s32_max
  - busy_poll_budget is limited to <= NAPI_POLL_WEIGHT by unprivileged
    users (!capable(CAP_NET_ADMIN))
  - prefer_busy_poll must be 0 or 1
  - __pad must be 0

Signed-off-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Jiri Slaby <jirislaby@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoeventpoll: Add per-epoll prefer busy poll option
Joe Damato [Tue, 13 Feb 2024 06:16:44 +0000 (06:16 +0000)]
eventpoll: Add per-epoll prefer busy poll option

When using epoll-based busy poll, the prefer_busy_poll option is hardcoded
to false. Users may want to enable prefer_busy_poll to be used in
conjunction with gro_flush_timeout and defer_hard_irqs_count to keep device
IRQs masked.

Other busy poll methods allow enabling or disabling prefer busy poll via
SO_PREFER_BUSY_POLL, but epoll-based busy polling uses a hardcoded value.

Fix this edge case by adding support for a per-epoll context
prefer_busy_poll option. The default is false, as it was hardcoded before
this change.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoeventpoll: Add per-epoll busy poll packet budget
Joe Damato [Tue, 13 Feb 2024 06:16:43 +0000 (06:16 +0000)]
eventpoll: Add per-epoll busy poll packet budget

When using epoll-based busy poll, the packet budget is hardcoded to
BUSY_POLL_BUDGET (8). Users may desire larger busy poll budgets, which
can potentially increase throughput when busy polling under high network
load.

Other busy poll methods allow setting the busy poll budget via
SO_BUSY_POLL_BUDGET, but epoll-based busy polling uses a hardcoded
value.

Fix this edge case by adding support for a per-epoll context busy poll
packet budget. If not specified, the default value (BUSY_POLL_BUDGET) is
used.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoeventpoll: support busy poll per epoll instance
Joe Damato [Tue, 13 Feb 2024 06:16:42 +0000 (06:16 +0000)]
eventpoll: support busy poll per epoll instance

Allow busy polling on a per-epoll context basis. The per-epoll context
usec timeout value is preferred, but the pre-existing system wide sysctl
value is still supported if it specified.

busy_poll_usecs is a u32, but in a follow up patch the ioctl provided to
the user only allows setting a value from 0 to S32_MAX.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: ena: Remove redundant assignment
Kamal Heib [Tue, 13 Feb 2024 03:17:18 +0000 (22:17 -0500)]
net: ena: Remove redundant assignment

There is no point in initializing an ndo to NULL, therefore the
assignment is redundant and can be removed.

Signed-off-by: Kamal Heib <kheib@redhat.com>
Reviewed-by: Brett Creeley <brett.creeley@amd.com>
Acked-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agonet: smc: fix spurious error message from __sock_release()
Dmitry Antipov [Mon, 12 Feb 2024 14:34:02 +0000 (17:34 +0300)]
net: smc: fix spurious error message from __sock_release()

Commit 67f562e3e147 ("net/smc: transfer fasync_list in case of fallback")
leaves the socket's fasync list pointer within a container socket as well.
When the latter is destroyed, '__sock_release()' warns about its non-empty
fasync list, which is a dangling pointer to previously freed fasync list
of an underlying TCP socket. Fix this spurious warning by nullifying
fasync list of a container socket.

Fixes: 67f562e3e147 ("net/smc: transfer fasync_list in case of fallback")
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
David S. Miller [Wed, 14 Feb 2024 10:37:33 +0000 (10:37 +0000)]
Merge branch '40GbE' of git://git./linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2024-02-12 (i40e)

This series contains updates to i40e driver only.

Ivan Vecera corrects the looping value used while waiting for queues to
be disabled as well as an incorrect mask being used for DCB
configuration.

Maciej resolves an issue related to XDP traffic; removing a double call to
i40e_pf_rxq_wait() and accounting for XDP rings when stopping rings.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoOcteontx2-af: Fetch MAC channel info from firmware
Hariprasad Kelam [Mon, 12 Feb 2024 13:15:23 +0000 (18:45 +0530)]
Octeontx2-af: Fetch MAC channel info from firmware

Packet ingress and egress MAC/serdes channel numbers are configurable
on CN10K series of silicons. These channel numbers inturn used while
installing MCAM rules to match ingress/egress port. Fetch these channel
numbers from firmware at driver init time.

Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next...
David S. Miller [Wed, 14 Feb 2024 10:25:27 +0000 (10:25 +0000)]
Merge branch '100GbE' of git://git./linux/kernel/git/tnguy/next-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2024-02-12 (ice)

This series contains updates to ice driver only.

Grzegorz adds support for E825-C devices.

Wojciech reworks devlink reload to fulfill expected conditions (remove
and readd).
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge tag 'linux-can-next-for-6.9-20240213' of git://git.kernel.org/pub/scm/linux...
David S. Miller [Wed, 14 Feb 2024 10:00:35 +0000 (10:00 +0000)]
Merge tag 'linux-can-next-for-6.9-20240213' of git://git./linux/kernel/git/mkl/linux-can-next

Marc Kleine-Budde says:

====================
linux-can-next-for-6.9-20240213

this is a pull request of 23 patches for net-next/master.

The first patch is by Nicolas Maier and targets the CAN Broadcast
Manager (bcm), it adds message flags to distinguish between own local
and remote traffic.

Oliver Hartkopp contributes a patch for the CAN ISOTP protocol that
adds dynamic flow control parameters.

Stefan Mätje's patch series add support for the esd PCIe/402 CAN
interface family.

Markus Schneider-Pargmann contributes 14 patches for the m_can to
optimize for the SPI attached tcan4x5x controller.

A patch by Vincent Mailhol replaces Wolfgang Grandegger by Vincent
Mailhol as the CAN drivers Co-Maintainer.

Jimmy Assarsson's patch add support for the Kvaser M.2 PCIe 4xCAN
adapter.

A patch by Daniil Dulov removed a redundant NULL check in the softing
driver.

Oliver Hartkopp contributes a patch to add CANXL virtual CAN network
identifier support.

A patch by myself removes Naga Sureshkumar Relli as the maintainer of
the xilinx_can driver, as their email bounces.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoocteontx2-af: Remove the PF_FUNC validation for NPC transmit rules
Subbaraya Sundeep [Sun, 11 Feb 2024 19:00:38 +0000 (00:30 +0530)]
octeontx2-af: Remove the PF_FUNC validation for NPC transmit rules

NPC transmit side mcam rules can use the pcifunc (in packet metadata
added by hardware) of transmitting device for mcam lookup similar to
the channel of receiving device at receive side.
The commit 18603683d766 ("octeontx2-af: Remove channel verification
while installing MCAM rules") removed the receive side channel
verification to save hardware MCAM filters while switching packets
across interfaces but missed removing transmit side checks.
This patch removes transmit side rules validation.

Fixes: 18603683d766 ("octeontx2-af: Remove channel verification while installing MCAM rules")
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 months agoMerge branch 'add-multi-buff-support-for-xdp-running-in-generic-mode'
Jakub Kicinski [Wed, 14 Feb 2024 03:22:35 +0000 (19:22 -0800)]
Merge branch 'add-multi-buff-support-for-xdp-running-in-generic-mode'

Lorenzo Bianconi says:

====================
add multi-buff support for xdp running in generic mode

Introduce multi-buffer support for xdp running in generic mode not always
linearizing the skb in netif_receive_generic_xdp routine.
Introduce generic percpu page_pools allocator.
====================

Link: https://lore.kernel.org/r/cover.1707729884.git.lorenzo@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>