vlan->flags, vlan->ingress_priority_map[] and vlan->nr_*_mappings
are written under RTNL, but read from lockless contexts:
- the RX path reads vlan->flags (VLAN_FLAG_REORDER_HDR) and
the ingress priority map,
- the TX path reads vlan->flags,
- /proc/net/vlan/<dev> reads vlan->flags and the ingress priority
map,
- vlan_get_size() reads vlan->nr_*_mappings,
- vlan_fill_info() will soon run without RTNL.
Add the missing READ_ONCE()/WRITE_ONCE() annotations.
vlan_dev_change_flags() is also made a bit more readable by using
a local new_flags variable, avoiding the repeated re-reads of
vlan->flags after it has been published.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260916132532.1785657-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dpaa2_ptp_probe() gets a referenced device node with
of_find_compatible_node() and stores it in dev->of_node.
The probe failure paths release this reference with of_node_put().
However, after a successful probe, dpaa2_ptp_remove() tears down the
PTP and MC resources without dropping the device node reference.
Release dev->of_node during driver removal and clear the pointer after
dropping the reference.
This issue was found by manual code inspection.
Cc: stable+noautosel@kernel.org # untested fix to a driver init path
Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
Link: https://patch.msgid.link/20260916030015.2818545-1-lgs201920130244@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Breno Leitao says:
====================
net: a sockopt_t quirk for the options that write past optlen
This series continues the migration of our protocols to sockopt_t, as
described in [1].
There are some protocols that use optlen as the header size, and the real
buffer size comes from a field inside the header. This means a bug, given
that optlen should be the full buffer size, but there are indications [2]
that we have programs that use the bad behaviour above, and we want to
avoid breaking them (or, honestly, avoid being cursed by Linus).
That said, create a quirk helper that preserves the same behaviour, even
using sockopt_t. The way to do it is simple:
1) Only do it for userspace callers (ubuf), otherwise a bug here will
corrupt the kernel instead of a simple SIGSEGV.
2) Expand optval mid-air based on the header field.
IP_MSFILTER is the first user, and the smallest one: a single caller, no
compat variant, and a reply written front to back. MCAST_MSFILTER on ipv4
and ipv6 comes next, and TCP_AO_GET_KEYS has the same shape. Do this
quirk on IP_MSFILTER to make sure the dynamic is ok, so, we can expand
it later.
None of this is meant to change what userspace sees.
====================
Link: https://patch.msgid.link/20260914-getsockopt_phase6-v2-0-e48befc9602e@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
IP_MSFILTER reads its reply through ip_mc_msfget(), reached from
do_ip_getsockopt() and from nowhere else. Convert it, and build the
sockopt_t at the call site for as long as the caller still carries a
sockptr_t pair.
optlen here only has to cover the header, and the real reply size comes
from the imsf_numsrc field inside it. This is nasty, but userspace
relies on it, so sockopt_expand_out() preserves the same mechanism: it
grows optval only for a user address, and assumes the caller left room
for the size its own header asked for.
The *optlen store moves out of ip_mc_msfget() and into the call site,
guarded by !err so the -EINVAL, -ENODEV and -EADDRNOTAVAIL returns still
leave the caller's optlen word untouched.
The source list also moves from copy_to_sockptr_offset() to a sequential
copy_to_iter(). IP_MSFILTER_SIZE(0) and offsetof(struct ip_msfilter,
imsf_slist_flex) are both 16, so the bytes land where they did.
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20260914-getsockopt_phase6-v2-2-e48befc9602e@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Some getsockopt options size their reply from a count the caller left in
optval rather than from optlen, and so write past the optlen the caller
declared. Userspace relies on that, so the sockopt_t conversion has to
keep doing it.
IP_MSFILTER is the first one to convert: its reply covers the
imsf_numsrc sources the caller asked for, while optlen only has to cover
the fixed header.
Add sockopt_expand_out() to grow opt->iter_out mid-air, so the quirk
sits in one place instead of each protocol assuming it implicitly. It is
a no-op unless the reply outruns optlen. Growing re-anchors the iterator
at the head of optval, so it has to be called before anything is written
through iter_out.
Only a user buffer can be longer than optlen says. A kernel-backed
optval keeps the bounded iterator, and a callback that asks to grow one
gets a WARN_ON_ONCE() and -EINVAL.
Nothing in tree can trip that WARN: the only kernel-backed path into
do_ip_getsockopt() is sol_ip_sockopt(), which takes IP_TOS and
IP_TRANSPARENT only. It is an assert for the in-kernel callers BPF and
io_uring gain once the conversion is done, so they declare an optlen
covering the whole buffer instead of repeating the userspace mistake.
The conversion in the next patch calls sockptr_to_sockopt() from
net/ipv4/, so drop its static and declare it in net.h.
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20260914-getsockopt_phase6-v2-1-e48befc9602e@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Dumazet says:
====================
net: rmnet: lockless rmnet_fill_info()
rmnet_fill_info() currently relies on RTNL being held, because it
uses rmnet_get_port_rtnl() to reach the rmnet_port attached to the
underlying real device.
While auditing the fields exposed there, it appears that both
priv->mux_id and port->data_format are written under RTNL but read
from the data path without any lock. Note that rmnet_newlink() can
update port->data_format of an already active port, and that
rmnet_changelink() can change both fields while traffic is flowing.
- Patch 1 annotates the data-races around port->data_format. It also
samples the field only once per packet and passes the value down,
so that both ends of a packet transformation agree on the format
(TX used to size the MAP headroom and set MAP_NEXT_HEADER_FLAG
from two separate reads), and prevents rmnet_changelink() from
publishing an intermediate value to the data path.
- Patch 2 annotates the data-races around priv->mux_id and
ep->mux_id. It also publishes ep->mux_id before the endpoint is
inserted into its new bucket, so that a concurrent lookup walking
that bucket can no longer miss it and drop the packet.
- Patch 3 converts rmnet_fill_info() to RCU. rmnet_get_port_rcu()
was only called from the data path and thus used
rcu_dereference_bh(); its lockdep condition is relaxed so that it
can also be called from process context.
The lockless lookup is safe because rmnet_get_port_rcu() checks
real_dev->rx_handler before returning rx_handler_data, while
rmnet_unregister_real_device() clears rx_handler and waits for a
grace period (in netdev_rx_handler_unregister()) before freeing the
port.
This is part of a larger effort to remove the RTNL dependency from
rtnl_link_ops->fill_info().
====================
Link: https://patch.msgid.link/20260916132231.1780091-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rmnet_fill_info() used rmnet_get_port_rtnl(), and thus required
RTNL.
Now that priv->mux_id and port->data_format are properly
annotated, rmnet_fill_info() can simply fetch the port under
rcu_read_lock().
rmnet_get_port_rcu() was only used from the data path and thus
used rcu_dereference_bh(). Relax its lockdep condition so that it
can also be called from process context under a plain
rcu_read_lock(), and use it from rmnet_fill_info().
Note that rmnet_get_port_rcu() checks real_dev->rx_handler before
returning rx_handler_data: this is what makes the lockless lookup
safe against rmnet_unregister_real_device(), which clears
rx_handler, waits for a grace period in
netdev_rx_handler_unregister(), and only then frees the port.
While at it, add missing const qualifiers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Sean Tranchetti <sean.tranchetti@oss.qualcomm.com>
Reviewed-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260916132231.1780091-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
priv->mux_id is written under RTNL from rmnet_vnd_newlink() and
rmnet_changelink(), but rmnet_egress_handler() reads it from the
transmit path without any lock.
Likewise ep->mux_id is written under RTNL from rmnet_changelink(),
while rmnet_get_endpoint() reads it from the receive path under
RCU only.
Add the missing READ_ONCE()/WRITE_ONCE() annotations.
While at it, publish ep->mux_id before inserting the endpoint in
its new bucket in rmnet_changelink(). Otherwise a concurrent
rmnet_get_endpoint() walking the new bucket could find the
endpoint still carrying its old mux_id, and drop the packet.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Sean Tranchetti <sean.tranchetti@oss.qualcomm.com>
Reviewed-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260916132231.1780091-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
port->data_format is written under RTNL from rmnet_newlink() and
rmnet_changelink(), but it is read from the data path without any
lock, both from RX (rmnet_rx_handler() context) and TX
(rmnet_egress_handler() context).
Note that rmnet_newlink() can be called for a real_dev which is
already hooked to rmnet: the rx_handler is live and traffic can
already be flowing when port->data_format is overwritten.
Add the missing READ_ONCE()/WRITE_ONCE() annotations.
While at it, sample port->data_format only once per packet and pass
the value down, so that all the decisions taken for a given packet
are based on a single consistent value.
Otherwise the two ends of a packet transformation could disagree.
On TX, rmnet_map_egress_handler() sized the headroom from one read
while rmnet_map_add_map_header() decided on MAP_NEXT_HEADER_FLAG
from another one, so a concurrent rmnet_changelink() could produce
a MAP header announcing a v5 csum header that was neither reserved
nor written. On RX, rmnet_map_validate_packet_len() and
__rmnet_map_ingress_handler() could likewise disagree on the
expected layout, and rmnet_map_send_ack() could trim a dl csum
trailer that the ingress path never accounted for.
rmnet_map_add_map_header(), rmnet_map_command(),
rmnet_map_deaggregate(), rmnet_map_send_ack(),
rmnet_map_validate_packet_len() and __rmnet_map_ingress_handler()
therefore get the value from their caller instead of re-reading it.
rmnet_changelink() now computes the new value in a local variable
and publishes it with a single store, instead of letting the data
path observe the intermediate (old_data_format & ~flags->mask)
value.
rmnet_vnd_headroom() is left alone, all its callers hold RTNL.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Sean Tranchetti <sean.tranchetti@oss.qualcomm.com>
Reviewed-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260916132231.1780091-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
fs_enet_probe() allocates the platform information structure with
kzalloc_obj() and stores it in fep->fpi.
The probe failure paths release this allocation with kfree(). However,
after a successful probe, fs_enet_remove() tears down the network
device without freeing fep->fpi, leaving the platform information
structure allocated after driver removal.
Free fep->fpi before releasing the network device.
This issue was found by manual code inspection.
Cc: stable+noautosel@kernel.org # untested fix to a driver init path
Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260916031414.2875022-1-lgs201920130244@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Arria10 SoCFPGA DWMAC nodes use reset-names = "stmmaceth",
"stmmaceth-ocp", which matches what dwmac-socfpga looks up for the
OCP reset. The generic snps,dwmac binding only allowed "ahb" as the
second reset name, so Arria10 DTS failed schema validation.
Allow "stmmaceth-ocp" as an alternative second reset name and drop
the related TODO from the Altera SoCFPGA DWMAC binding.
Signed-off-by: Nazim Amirul <muhammad.nazim.amirul.nazle.asmade@altera.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20260909080559.2028-1-muhammad.nazim.amirul.nazle.asmade@altera.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The `strcpy()` function is deprecated and moving towards code-tree
elimination. Replacing it with `strscpy()` fixes potential buffer
overflow vectors by ensuring safe NULL-termination based on the
destination buffer size limit [1][2].
To make the interface safer and more robust for future callers, refactor
`tipc_bearer_get_name()` to accept a destination buffer length parameter.
Replace `strcpy()` with `strscpy()` and pass through any potential
`-E2BIG` truncation error code up to the caller. Update the existing
caller in `net/tipc/monitor.c` to pass its array size.
Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy [1]
Link: https://github.com/KSPP/linux/issues/88 [2]
Signed-off-by: Ajith Parappalliyalil Venugopal <ajithpv.linux@gmail.com>
Reviewed-by: Tung Nguyen <tung.quang.nguyen@est.tech>
Link: https://patch.msgid.link/20260915145641.2434403-1-ajithpv.linux@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
port-index is the only way rate-set can address a leaf (port) rate object,
but it was never listed in the request, so the generated policy has no
entry for it. The op declares .maxattr = DEVLINK_ATTR_PARENT_DEV, so the
attribute still reaches info->attrs[], validated against a zeroed slot -
NLA_UNSPEC, length 0 - which GENL_DONT_VALIDATE_STRICT accepts at any
length.
devlink_port_get_from_attrs() then runs nla_get_u32() on it. Handed a
zero-length port-index the kernel reads the four bytes past the payload,
which are the next attribute's header, and acts on the port index those
spell out. Since we're reading a linear skb the OOB read is still
within the same memory allocation, it's just garbage. We also do not
echo the garbage back to the user so it's not an info leak either.
Hence not treating this is a real bug fix.
rate-new is left alone on purpose. It creates rate nodes, resolved by
name through devlink_rate_node_get_from_attrs(), and never looks at
port-index.
Link: https://patch.msgid.link/20260915161341.1053476-10-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We are missing the pad attribute in a number of subsets.
Worse, dl-attr-stats which is a separate attr space doesn't
define pad, but reuses the value from the root attr set
(DEVLINK_ATTR_PAD). This would break parsing stats on
an arch without CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS.
Link: https://patch.msgid.link/20260915161341.1053476-9-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
devlink_nl_rate_parent_fill() emits DEVLINK_ATTR_PARENT_DEV via
devlink_nl_put_nested_handle(), which adds DEVLINK_ATTR_NETNS_ID
whenever the parent rate node lives on a devlink instance in another
netns. Now that we defined dl-nested-devlink subset, which unlike
dl-parent-dev includes netns-id, let's switch to it. The rate APIs
input genuinely does not accept the netns-id, so we need to subset
them to stick with dl-parent-dev.
Fixes: b5f90fd458 ("devlink: Add parent dev to devlink API")
Link: https://patch.msgid.link/20260915161341.1053476-8-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Three attributes are emitted repeatedly inside their nest:
- devlink_nl_region_read_fill() calls
devlink_nl_cmd_region_read_chunk_fill() once per 256 byte unit
- devlink_nl_region_snapshots_id_put() emits one
DEVLINK_ATTR_REGION_SNAPSHOT sub-nest per entry on the region's
snapshot list
- devlink_nl_linecard_fill() emits one DEVLINK_ATTR_LINECARD_TYPE per
linecard->types_count entry
In all three cases we're missing the multi-attr properties, and the
attributes themselves are missing from the attr list used by C code gen.
Link: https://patch.msgid.link/20260915161341.1053476-7-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
devlink has a one huge root attribute set for the whole family,
we haven't taken the time to properly define the sub-sets for
each command. Do it for port-set so that we can drop the hand
written policy used by devlink_port_function_set().
We need this subsetting because within the DEVLINK_ATTR_PORT_FUNCTION
nest DEVLINK_PORT_FN_ATTR_OPSTATE and DEVLINK_PORT_FN_ATTR_DEVLINK
are output-only so we have to filter them out of the input set.
Link: https://patch.msgid.link/20260915161341.1053476-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dl-port-function stops at caps, but the nest also carries
DEVLINK_PORT_FN_ATTR_DEVLINK (5) and DEVLINK_PORT_FN_ATTR_MAX_IO_EQS (6)
both put by devlink_nl_port_function_attrs_put() on every port-get
do and dump. YNL raises
Space 'dl-port-function' has no attribute with value '6'
for any port reporting max_io_eqs or a nested devlink handle,
i.e. for mlx5 SFs and VFs.
Commit 5af3e3876d ("devlink: Support setting max_io_eqs") added the
uAPI value and the hand written policy but never touched the spec.
Add the missing attributes, subsequent commit reworks the code
to use the YNL-generated policy.
Link: https://patch.msgid.link/20260915161341.1053476-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The "# TODO: fill in the attributes in between" gap before selftests
(176) hides exactly one attribute, DEVLINK_ATTR_NESTED_DEVLINK (175).
devlink_nl_fill() calls devlink_nl_nested_fill() unconditionally, which
emits one of these per entry in devlink->nested_rels (mlx5, ice).
Unlike the other TODO gaps this is a single attribute of an already
modelled type, so just fill it in.
Link: https://patch.msgid.link/20260915161341.1053476-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The bug the comment describes was fixed by
commit 61c43780e9 ("devlink: fix port dump cmd type") two years ago.
devlink_nl_port_get_dump_one() fills DEVLINK_CMD_PORT_NEW (7), not
DEVLINK_CMD_NEW (3) now.
Note that this change is likely a noop. _dictify_ops_directional()
picks the "do" section's value when an op has both "do" and "dump".
Link: https://patch.msgid.link/20260915161341.1053476-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
DEVLINK_ATTR_RELOAD_LIMITS carries enum devlink_reload_limit, the spec
says enum devlink_reload_action. The two are unrelated, and have
a different set of values (bits 1, 2 vs bits 0, 1).
AFAICT this is a cosmetic change - both DEVLINK_RELOAD_LIMIT_UNSPEC
and the out of bounds bit 2 will be rejected either way because drivers
don't declare them as supported. User will see either a policy
validation failure or "Requested limit is not supported by the driver".
While at it annotate the right enum for reload-stats-limit
Link: https://patch.msgid.link/20260915161341.1053476-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
stmmac_setup_ptp() returns void and swallows both PTP setup errors:
the PTP reference clock enable and stmmac_init_timestamping()
failures are logged but never propagated. When they fail, the MAC
system time counter is left in its post-reset, non-running state,
while the driver keeps operating as if timestamping were up.
This matters for TAPRIO/EST qdisc offloading, which derives the EST
base time from the hardware timestamp counter: arming the gate list
against a non-advancing time base would leave the schedule permanently
stuck.
Make stmmac_setup_ptp() return an error code and propagate the
failure in __stmmac_open(), stopping the DMA engines when PTP setup
fails. In stmmac_resume(), re-initialise timestamping only when the
PTP clock was registered before suspend, re-apply the platform clock
rate configuration, and stop the DMA engines if re-initialisation
fails.
While at it, factor the hardware timestamping capability check into a
stmmac_check_timestamp_cap() helper and apply it to the hwtstamp get
path and the ethtool ts_info path. stmmac_setup_ptp() now enables the
PTP reference clock and runs the platform ptp_clk_freq_config()
callback before validating the resulting clock rate; if no valid rate
is available, it disables the clock and returns success without
registering the PTP clock. This keeps the interface operational on
platforms with PTP-capable silicon but an unconfigured PTP clock,
where timestamping cannot be enabled: instead of failing, the driver
proceeds without setting up timestamping. Extend the timestamping gate
on the set path (which keeps testing the core's basic/advanced
timestamping capability) and the devlink registration with the PTP
reference clock rate requirement.
Gate the PTP reference clock enable/disable in the platform noirq
suspend/resume callbacks on priv->ptp_enabled, keeping the clock
reference balanced against the new early return in
stmmac_release_ptp().
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@oss.qualcomm.com>
Link: https://patch.msgid.link/20260914-stmmac-ptp-error-propagate-v5-1-81149897e65d@oss.qualcomm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Replace the kernel's own `transmute::FromBytes` and `AsBytes` traits
with their zerocopy equivalents. Specifically, this updates
`GenlMsg::put` to rely on `IntoBytes` for converting attributes into
byte slices.
Also, add `Immutable` trait bound on `GenlMsg::put` as zerocopy
splits the `no interior mutability` guarantee that `AsBytes` bundled
together.
Link: https://github.com/Rust-for-Linux/linux/issues/1241
Signed-off-by: Sagar Taunk <sagartaunk@proton.me>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/20260915144231.31932-1-sagartaunk@proton.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ctrl_policy_family[] accepts CTRL_ATTR_FAMILY_ID and ctrl_getfamily()
looks the family up by it, but the spec lists only family-name in the
request. C YNL generates struct members and setters from that list, so
struct nlctrl_getfamily_req has no way to carry an id and resolving a
family id - the one a client gets handed in nlmsg_type - back to a name
needs a full dump.
Python YNL happens to encode any attribute of the op's attribute set,
which is why this went unnoticed.
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20260916021516.1701760-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mask and pad are declared the wrong way round, giving mask 11 and
pad 12, while enum netlink_policy_type_attr has
NL_POLICY_TYPE_ATTR_PAD at 11 and NL_POLICY_TYPE_ATTR_MASK at 12.
The kernel emits the mask for every NLA_POLICY_MASK() attribute, so a
Python YNL client parsing a policy dump maps it onto pad and throws it
away - 30 of them in ethtool's policy alone. In the other direction the
64-bit alignment padding, which the kernel does emit on architectures
without efficient unaligned access, is decoded as a u64 mask and blows
up the whole dump. But those architectures are rare.
The generated C parser is not affected, it keys the attributes off the
uAPI symbol names rather than off the values the spec derives.
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20260916021516.1701760-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Donggeun Yoo says:
====================
net: phy: qca808x: fix the active-high LED polarity mode
A QCA8081 with an 'active-high' LED node does not work today, in two
separate ways, so this is now two patches: qca808x_led_polarity_set()
rejects the mode outright, and once it stops doing that
qca808x_config_init() still will not put the bit back after the reset
phy_init_hw() runs.
Tested in QEMU against a synthetic MDIO bus that answers as a QCA8081 and
emulates MMD7 0x901a, with the reset clearing BIT(6) as f203c8c77c
describes. One kernel, one harness, three arms, x86_64, 7.3.0-rc2+:
DT node base 1/2 only 1/2 + 2/2
active-high -EINVAL inverted correct
active-low correct correct correct
no polarity node correct correct correct
high-impedance -EINVAL -EINVAL -EINVAL
Still no QCA808x hardware here; the harness emulates the one register,
so a confirmation on a real board would be welcome.
====================
Link: https://patch.msgid.link/20260915223138.321307-1-donggeunyoo.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
With the previous patch an 'active-high' LED node is accepted, so
led_polarity_mode can now hold 0. qca808x_config_init() only re-asserts
QCA808X_LED_ACTIVE_HIGH when the mode is -1, the value that means device
tree asked for nothing, so an explicit active-high does not reach the
register.
That matters because the bit does not survive a reset. phy_init_hw() runs
.soft_reset before .config_init on every attach and resume, and
commit f203c8c77c ("net: phy: qcom: qca808x: default to LED active High if not set")
records why: "on PHY reset, the Active High bit is not set resulting in
the LED driven as active-low". The polarity written from device tree
during phy_probe() is therefore gone by the time the link comes up, and
the LED runs inverted.
Re-assert the bit for anything other than an explicit active-low, which
is the one case that wants it clear and gets that from the reset for
free. Name the unset value while here: led_polarity_mode otherwise
holds a PHY_LED_ACTIVE_* value, so spelling the comparison
PHY_LED_ACTIVE_LOW says what it means where -1, 0 and 1 did not.
Cc: stable+noautosel@kernel.org # never worked
Closes: https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260908105959.70453-1-donggeunyoo.kernel%40gmail.com
Signed-off-by: Donggeun Yoo <donggeunyoo.kernel@gmail.com>
Reviewed-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20260915223138.321307-3-donggeunyoo.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
commit a274465cc3 ("net: phy: support 'active-high' property for PHY LEDs")
added PHY_LED_ACTIVE_HIGH, but qca808x_led_polarity_set() still rejects
everything except PHY_LED_ACTIVE_LOW. An LED node carrying 'active-high'
therefore falls into the default case and returns -EINVAL, which
phy_probe() propagates: the mdio device is left unbound and
phy_attach_direct() falls back to the genphy driver, so the PHY loses
every qca808x-specific feature, not just the LED.
Accept the mode. The register write below already handles both
polarities.
Cc: stable+noautosel@kernel.org # never worked
Signed-off-by: Donggeun Yoo <donggeunyoo.kernel@gmail.com>
Reviewed-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://patch.msgid.link/20260915223138.321307-2-donggeunyoo.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
netkit_fill_info() used rtnl_dereference() to fetch nk->peer,
and thus required RTNL.
nk->peer is already an RCU protected pointer, updated with
rcu_assign_pointer() and read from the fast path with
rcu_dereference(). Simply use rcu_dereference() under
rcu_read_lock() instead.
While at it, use READ_ONCE() when reading nk->policy, because
netkit_change_link() can change it at any time (it already uses
WRITE_ONCE()).
Other fields (primary, mode, pair, scrub) are only set from
netkit_new_link(), before the device is visible to dumps.
Also add missing const qualifiers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/20260916111953.1611574-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Minor touch ups to the XSK-related doc strings in the netdev spec.
Tx checksum is (obviously?) Layer 4 (TCP/UDP), not Layer 3 (IP).
The xdp_zc_max_segs change is more subtle, I guess, but to kernel devs
saying "frags" may sound like the number of skb frags. The value includes
the head in this case. The default of 1 means single buffer, XDP_USE_SG
sockets are refused when it is 1. Let's use the word "buffer" instead
of "frag".
Link: https://patch.msgid.link/20260916021608.1702731-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Previous releases - regressions:
- netfilter: hold reference on ct until flow is released
- bridge:
- move switchdev call outside rcu
- vlan: fix bugs caused by switchdev deletion errors
- wifi:
- mac80211: reset state when starting AP fails
- cfg80211: don't free driver-owned scan requests
- tcp: don't call skb_clone_and_charge_r() for close()d listener in tcp_v6_do_rcv().
- mptcp: return sk_wait_data() errors from recvmsg()
- xfrm: serialize state GC with device state flush
- drop_monitor: synchronize tracepoint unregistration on error path
- bluetooth:
- eir: validate service data length before reading UUID
- hci_sync: serialize local codec list cleanup
- RFCOMM: avoid socket lock inversion in listener cleanup
- eth: lan743x: fix RX checksum use-after-free
- eth: mvpp2: prevent buffer overflow in page_pool allocation
Previous releases - always broken:
- core: lock the socket in sock_gettstamp()
- neighbour: enforce min/max to NDTPA_INTERVAL_PROBE_TIME_MS.
- sched: codel: bound the dropping loop per dequeue call
- wifi: mac80211: include TIM bitmap control for buffered S1G mcast traffic
- psp: avoid conflicts with skb->decrypted and sk_validate_xmit_skb()
- xfrm: fix stack OOB read in iptfs_skb_reset_frag_walk()
- bluetooth: hci_qca: do not write to the serial port after it is closed
- dsa: mxl862xx: disable the stats poll on teardown
- eth: stmmac: fix TSO header length truncation
- eth: ip_tunnel: initialize `options_len` before referencing options
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=Gw4S
-----END PGP SIGNATURE-----
Merge tag 'net-7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Paolo Abeni:
"Including fixes from Netfilter, Bluetooth, IPSec and WiFi.
Previous releases - regressions:
- netfilter: hold reference on ct until flow is released
- bridge:
- move switchdev call outside rcu
- vlan: fix bugs caused by switchdev deletion errors
- wifi:
- mac80211: reset state when starting AP fails
- cfg80211: don't free driver-owned scan requests
- tcp: don't call skb_clone_and_charge_r() for close()d listener in
tcp_v6_do_rcv()
- mptcp: return sk_wait_data() errors from recvmsg()
- xfrm: serialize state GC with device state flush
- drop_monitor: synchronize tracepoint unregistration on error path
- bluetooth:
- eir: validate service data length before reading UUID
- hci_sync: serialize local codec list cleanup
- RFCOMM: avoid socket lock inversion in listener cleanup
- eth:
- lan743x: fix RX checksum use-after-free
- mvpp2: prevent buffer overflow in page_pool allocation
Previous releases - always broken:
- core: lock the socket in sock_gettstamp()
- neighbour: enforce min/max to NDTPA_INTERVAL_PROBE_TIME_MS.
- sched: codel: bound the dropping loop per dequeue call
- wifi: mac80211: include TIM bitmap control for buffered S1G mcast
traffic
- psp: avoid conflicts with skb->decrypted and sk_validate_xmit_skb()
- xfrm: fix stack OOB read in iptfs_skb_reset_frag_walk()
- bluetooth: hci_qca: do not write to the serial port after it is
closed
- dsa: mxl862xx: disable the stats poll on teardown
- eth:
- stmmac: fix TSO header length truncation
- ip_tunnel: initialize `options_len` before referencing options"
* tag 'net-7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (159 commits)
mptcp: fix bad accounting in __mptcp_subflow_push_pending()
mptcp: close race between scheduler and state change
mptcp: avoid unneeded actions on subflow reset
net: skbuff: do not leave stale header offsets after pskb_carve()
selftests: net: packetdrill: test exclusion of old ACK from TCP fast path
tcp: exclude old ACKs from tcp fast path
dpll: reject a reference sync pin which is not on the pin's dpll
net: mvpp2: prevent buffer overflow in page_pool allocation
net: macb: fix ordering around PTP timestamp read
selftests: drv-net: psp: test PSP and TCP ULP mutual exclusion
net: psp: avoid conflicts with skb->decrypted and sk_validate_xmit_skb()
net: stmmac: preserve real_num_tx_queues on mqprio setup failure
net: stmmac: propagate FPE preemption-class mapping errors
net: wwan: t7xx: validate the netif index in t7xx_ccmni_recv_skb()
net: wwan: mhi_wwan_mbim: check skb_copy_bits() return value
net: wwan: mhi_wwan_mbim: guard against a cyclic NDP chain
net: ethernet: cortina: Ack RX overrun interrupt correctly
net: lock the socket in sock_gettstamp()
eth: fbnic: ring the doorbell if a burst ends in a drop
net: netsec: fix device_node reference leak on phy_np
...
A collection of small fixes. Most of them are device-specific fixes
while there are a few core fixes. The continued flux, but not too
scaring yet. Some highlights below.
ALSA Core:
- Fix potential UAF after asynchronous card release
- Fix a race condition in PCM timer initialization order
USB-Audio:
- Hardening fixes for issues reported by fuzzer for 6fire, bcd2000,
and implicit FB packets
- Fix double list addition in implicit FB handling
- Quirks for AVerMedia GC553Pro and Behringer FCA1616
HD-Audio:
- Quirks / fixes for HP OmniBook 7, OMEN 15, and Victus 15 laptops
ASoC:
- Support for DAI link codec channel mask to avoid mismatches
- Fix HDMI-codec channel status change report
- Fixes for various codecs and platforms: Realtek rt712/rt721
(calibration, reset fixes), Cirrus Logic (empty EFI variable
validation, capture channel fixup), AMD ACP SoundWire (bounds
checks, refactorings), ADAU1977 (OF match table support, SPI
cleanups), ES8336 (Huawei Matebook B3-420 quirk), UX500 (macro fix)
-----BEGIN PGP SIGNATURE-----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=PnHM
-----END PGP SIGNATURE-----
Merge tag 'sound-7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of small fixes. Most of them are device-specific fixes
while there are a few core fixes. The continued flux, but not too
scaring yet. Some highlights below.
ALSA Core:
- Fix potential UAF after asynchronous card release
- Fix a race condition in PCM timer initialization order
USB-Audio:
- Hardening fixes for issues reported by fuzzer for 6fire, bcd2000,
and implicit FB packets
- Fix double list addition in implicit FB handling
- Quirks for AVerMedia GC553Pro and Behringer FCA1616
HD-Audio:
- Quirks / fixes for HP OmniBook 7, OMEN 15, and Victus 15 laptops
ASoC:
- Support for DAI link codec channel mask to avoid mismatches
- Fix HDMI-codec channel status change report
- Fixes for various codecs and platforms: Realtek rt712/rt721
(calibration, reset fixes), Cirrus Logic (empty EFI variable
validation, capture channel fixup), AMD ACP SoundWire (bounds
checks, refactorings), ADAU1977 (OF match table support, SPI
cleanups), ES8336 (Huawei Matebook B3-420 quirk), UX500 (macro
fix)"
* tag 'sound-7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (33 commits)
ASoC: adau1977-i2c: add OF match table for I2C
ASoC: adau1977-spi: drop __maybe_unused and of_match_ptr()
ASoC: adau1977: make the Kconfig symbols user selectable
ASoC: amd: acp: fix card name length warning in SOF SoundWire machine driver
ASoC: amd: acp: fix ffs() operator precedence for SoundWire link ID
ASoC: amd: acp: refactor codec config count in SOF SoundWire machine driver
ASoC: amd: acp: bounds-check SoundWire link ID in machine drivers
ASoC: cs-amp-lib: Prevent NULL pointer if efi variable is zero length
ASoC: codecs: rt712-sdca-dmic: fix uninitialized stream_config->type
ASoC: hdmi-codec: Report a change when the channel status moves
ASoC: ux500: Parenthesize MSP_{RX,TX}_CLKPOL_BIT() arguments
ASoC: rt721: Reset codec to fix abnormal sound
ALSA: usb-audio: fix list_add double-add in push_back_to_ready_list
ALSA: hda: trace PCM open only after assigning a stream
ALSA: usb-audio: skip the broken mute control on AVerMedia GC553Pro
ALSA: hda/realtek: Enable mute LEDs on HP OmniBook 7 17-dc0xxx
ALSA: 6fire: fix OOB write from device-reported iso length
ALSA: usb-audio: Add capture quirk for Behringer FCA1616
ALSA: hda/realtek: Add mute LED quirk for HP OMEN 15-ax
ASoC: Intel: sof_es8336: Add a quirk for Huawei Matebook B3-420
...
- fix fwnode reference leak on failure in shared GPIO handling
- fix regression in OF_POPULATED logic after the unification of GPIO hog
handling between OF, ACPI and machine variants
- don't call free_irq() if no IRQ is installed in gpio-virtuser
-----BEGIN PGP SIGNATURE-----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=NiRo
-----END PGP SIGNATURE-----
Merge tag 'gpio-fixes-for-v7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux
Pull gpio fixes from Bartosz Golaszewski:
- fix fwnode reference leak on failure in shared GPIO handling
- fix regression in OF_POPULATED logic after the unification of GPIO
hog handling between OF, ACPI and machine variants
- don't call free_irq() if no IRQ is installed in gpio-virtuser
* tag 'gpio-fixes-for-v7.3-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: virtuser: skip free_irq when no IRQ is installed
gpiolib: of: don't mark hog nodes OF_POPULATED before a chip is found
gpiolib: Put fwnode reference on failure
Matthieu Baerts says:
====================
mptcp: misc fixes for v7.3-rc4
Here are two unrelated fixes:
- Patch 1: avoid unneeded actions on subflow reset. A fix for another
fix introduced in v6.12 and targeting a commit from v5.7.
- Patch 2: close a possible race when scheduling a closing path. A fix
for another fix introduced in v6.0 and targeting v5.10.
- Patch 3: fix bad accounting when __subflow_push_pending returns an
error. A fix for v6.6.
====================
Link: https://patch.msgid.link/20260917-net-mptcp-misc-fixes-7-3-rc4-v2-0-0cf5c72667c8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If __subflow_push_pending() errors out we should avoid updating the
copied byte counters, to avoid mismatch push call later on.
Fixes: 0fa1b3783a ("mptcp: use get_send wrapper")
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260917-net-mptcp-misc-fixes-7-3-rc4-v2-3-0cf5c72667c8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The mptcp scheduler may race with subflow sockets state change: data
transmission on the selected socket may fail and a later release could
try to use mss_now reset to 0 for a divide operation.
Address the issue by explicitly checking for the critical scenario.
Fixes: c886d70286 ("mptcp: do not queue data on closed subflows")
Cc: stable@vger.kernel.org
Reported-by: Shardul Bankar <shardul.b@mpiricsoftware.com>
Reported-by: Xinyang Ge <xinyang@anthropic.com>
Closes: https://lore.kernel.org/20260525194828.1137119-1-shardul.b@mpiricsoftware.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260917-net-mptcp-misc-fixes-7-3-rc4-v2-2-0cf5c72667c8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Once in a blue moon, the mptcp receive path can recursively call
mptcp_data_ready() via state change under unlucky error conditions, and
then try to hold the data lock again.
Break the recursion loop explicitly checking for the exceptional
condition.
Add a new flag instead of using an existing one like 'closing', to exit
early in subflow_state_change(), and explicitly flush the RX queue at
reset time.
This avoids unneeded processing to check for available data -- calling
get_mapping_status() and more on a dying subflow -- but also in error
reporting and worker scheduling.
Note that we must consume the currently peeked skb before invoking
mptcp_dss_corruption to avoid consuming it again after the eventual
reset has freed it.
Fixes: e32d262c89 ("mptcp: handle consistently DSS corruption")
Cc: stable@vger.kernel.org
Reported-by: Xinyang Ge <xinyang@anthropic.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260917-net-mptcp-misc-fixes-7-3-rc4-v2-1-0cf5c72667c8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few fixes for the DMA-mapping code:
- resolved regression in accessing encrypted memory by IOMMU-backed
devices (Aneesh Kumar K.V),
- improved failure handling and removed rare bug in swiotlb/highmem
(Donggeun Yoo).
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQSrngzkoBtlA8uaaJ+Jp1EFxbsSRAUCaquwMwAKCRCJp1EFxbsS
RMT6AP0elpdaZXNY0KwUBTwU95H604J+donqriepHABIBhIDEQD9GWZqNf/m1gEI
tR5lHQ3+NGs0Q7Vd2ed1vSe82HQSsgU=
=QxUC
-----END PGP SIGNATURE-----
Merge tag 'dma-mapping-7.3-2026-09-17' of git://git.kernel.org/pub/scm/linux/kernel/git/mszyprowski/linux
Pull dma-mapping fixes from Marek Szyprowski:
"A few fixes for the DMA-mapping code:
- resolved regression in accessing encrypted memory by IOMMU-backed
devices (Aneesh Kumar K.V)
- improved failure handling and removed rare bug in swiotlb/highmem
(Donggeun Yoo)"
* tag 'dma-mapping-7.3-2026-09-17' of git://git.kernel.org/pub/scm/linux/kernel/git/mszyprowski/linux:
x86/mm: Don't force unencrypted DMA for IOMMU-backed devices
dma-mapping: don't trace the DMA address when the allocation fails
swiotlb: use the adjusted address for the highmem page lookup
dma-coherent: report a failed reserved memory assignment
An rmnet device may be created with its real device in a different
netns than the rmnet device itself (rmnet_newlink() resolves it in
link_net), and the config paths below only check CAP_NET_ADMIN against
dev_net(dev), while mutating rmnet port state attached to the real
device:
- rmnet_changelink() rewrites the endpoint mux table and
port->data_format and, via rmnet_vnd_update_dev_mtu(), can shrink the
MTU of the rmnet endpoint netdevs.
- rmnet_add_bridge() and rmnet_del_bridge(), reachable via
ndo_add_slave/ndo_del_slave through RTM_SETLINK IFLA_MASTER, flip
port->rmnet_mode and port->bridge_ep on the real device's port; with
bridge_ep pointing at a caller-owned device, rmnet_rx_handler() then
forwards real-device ingress frames to it.
- rmnet_set_coalesce() rewrites the port aggregation parameters via
ETHTOOL_SCOALESCE (ioctl) or ETHTOOL_MSG_COALESCE_SET (netlink),
whose capability checks likewise only cover dev's netns.
A caller privileged only in the rmnet device's netns can therefore
rewrite the shared cellular data-path state owned by another netns, and
steer its ingress traffic.
Gate the rtnl paths with rtnl_dev_link_net_capable(), matching the
"require CAP_NET_ADMIN in the device netns for changelink" series
(vxlan/geneve, CVE-2026-68432), and gate the ethtool setter with
ns_capable() in the real device netns, mirroring the check dev_ethtool()
already applies to dev's netns. Report the new rejections through
extack where one is available.
The gates cover the configuration paths only. Teardown (RTM_DELLINK
reaching rmnet_dellink(), and rmnet_config_notify_cb() unregistering
the bridge when the slave device is deleted) is intentionally left
ungated: link deletion is normal netdev lifecycle behaviour, and the
changelink series this matches (vxlan/geneve, CVE-2026-68432) gated
only the configuration paths.
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Abdifatah Suruur <suruurism@gmail.com>
Reviewed-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260915195642.1912-1-suruurism@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently netlink-raw.yaml contains two if keys and two then keys in a
single mapping that enforces a "len" for "pad" members and a "len" or
"struct" for binary members.
During validation PyYAML resolves duplicate keys last-wins, so only the
binary rule survives. Pad has not been validated since commit
bf08f32c8c ("tools/net/ynl: Add support for nested structs") added
the second if/then pair in January 2024.
None of the current specs violate this rule, but this validation should
not be parser dependent and unspecified. Strict YAML validators such as
Red Hat's VS Code YAML extension and Adrien Verge's yamllint will
reject the netlink-raw.yaml schema:
Command:
$ yamllint Documentation/netlink/netlink-raw.yaml
Output:
185:13 error duplication of key "if" in mapping (key-duplicates)
189:13 error duplication of key "then" in mapping (key-duplicates)
The following invalid netlink family spec will pass validation in the
current ynl tooling:
# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
---
name: minimal-raw
doc: Minimal netlink-raw family for schema validation testing.
protocol: netlink-raw
protonum: 0
definitions:
-
name: test-struct
type: struct
members:
-
name: reserved
type: pad
# len intentionally omitted
attribute-sets: []
operations:
list: []
Signed-off-by: Taylor Bates <tmbates12@gmail.com>
Reviewed-by: Hangbin Liu <liuhangbin@kylinos.cn>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20260914-ynl-robustness-v2-1-c6125003b21e@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
pskb_carve_inside_header() and pskb_carve_inside_nonlinear() remove
the first bytes of a packet and reallocate skb->head.
All the headers that were present before the operation are gone,
but both functions call skb_headers_offset_update(skb, 0), which
is a no-op : skb->mac_header, skb->network_header,
skb->transport_header and skb->csum_start keep their old values and
now describe bytes which are no longer there.
Both helpers size the new head from the old skb_end_offset(), so the
stale offsets still land inside the new allocation. They point past
skb_tail_pointer() though, to bytes that were never initialized.
pskb_carve_inside_nonlinear() is the worst case, because it leaves a
zombie skb with an empty linear part (skb->data ==
skb_tail_pointer(skb), skb_headlen(skb) == 0), while
skb_mac_header_was_set() is still true and skb->mac_header is way
ahead of skb->data.
The only user of pskb_extract() is rds_tcp_data_recv(), and the
carved skb is queued on tinc->ti_skb_list. When the RDS incoming
message is released, rds_tcp_inc_free() calls skb_queue_purge(),
which frees the skbs with SKB_DROP_REASON_QUEUE_PURGE. This is
visible from drop_monitor, which then tries to pull back to the
(bogus) mac header :
skbuff: __skb_pull(len=234)
skb len=6968 data_len=6968 headroom=0 headlen=0 tailroom=0
end-tail=384 mac=(234,14) mac_len=14 net=(248,40) trans=288
shinfo(txflags=0 nr_frags=1 gso(size=1428 type=16 segs=5))
csum(0x100120 start=288 offset=16 ip_summed=3 complete_sw=0 valid=1 level=0)
hash(0x7b446c6c sw=0 l4=1) proto=0x86dd pkttype=0 iif=60
kernel BUG at ./include/linux/skbuff.h:2847!
Add skb_carve_reset_headers() to mark the mac and transport headers
as not set, reset the network header, clear skb->mac_len, and drop
a now meaningless CHECKSUM_PARTIAL (csum_start no longer describes
anything).
Invalidate the inner offsets as well. Unlike mac_header and
transport_header they have no "unset" sentinel, so a leftover
non-zero value still looks like a real header. Zero
skb->inner_mac_header, skb->inner_network_header,
skb->inner_transport_header, skb->inner_protocol and
skb->encapsulation, so that all the header state is invalidated in
one place.
v2: fixed an inaccurate changelog. The stale offsets stay inside the
new skb->head, which is never smaller than the old one, they
simply point past skb_tail_pointer() to bytes that are gone.
Thanks to Xuanqiang Luo for insisting on this.
Also invalidate the inner header state, as suggested by the
netdev AI review :
https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260911114922.621937-1-edumazet%40google.com
Fixes: 6fa01ccd88 ("skbuff: Add pskb_extract() helper function")
Reported-by: syzbot+586af68eb819833c2d91@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/6aa3e9d3.f2639fcc.29487d.0028.GAE@google.com/
Cc: Xuanqiang Luo <xuanqiang.luo@linux.dev>
Cc: Allison Henderson <achender@kernel.org>
Cc: rds-devel@oss.oracle.com
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Xuanqiang Luo <luoxuanqiang@kylinos.cn>
Link: https://patch.msgid.link/20260915130423.3956471-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Introduce the following new tests: `ipv4_verify_same_scope_addr_order`,
`ipv4_verify_inter_scope_addr_order`, `ipv6_verify_same_scope_addr_order`,
`ipv6_verify_inter_scope_addr_order` to check the ordering of a
set of IP addresses after being inserted.
The implementations of these protocols are inconsistent
in regard to address ordering. IPv4 addresses stay in the
same order as inserted while IPv6 addresses appear in
reverse order. This inconsistency has prompted attempts
to unify the ordering, so both protocols act the same
(as IPv4). This however caused user-space regressions in
certain applications, which relied on the order as it
was prior to the change (particularly NetworkManager).
A similar inconsistency occurs when inserting different
scope addresses, where IPv4 puts link local ones before global
and IPv6 does the opposite. Attempts were already made
to unify this also, risking further regressions.
The addition of these tests aims to consolidate current
behaviour to prevent regressions in the future.
The expected behaviour is the initial one, where each
protocol acts differently. The same goes for inter scope
addresses.
Conversations detailing the decision processes for creating
these tests are linked below.
Link: https://lore.kernel.org/netdev/20260521135310.GC977@cmadams.net/
Link: https://lore.kernel.org/netdev/20260529112357.5079-1-fmancera@suse.de/
Link: https://lore.kernel.org/netdev/20260721090114.GA2510713@shredder/
Suggested-by: Fernando Fernandez Mancera <fmancera@suse.de>
Signed-off-by: Martin Jabůrek <martin.jaburek@suse.com>
Reviewed-by: Fernando Fernandez Mancera <fmancera@suse.de>
Link: https://patch.msgid.link/20260914065848.10518-1-martin.jaburek@suse.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
IPV6_FREEBIND and IPV6_TRANSPARENT still take the socket lock despite
using inet_assign_bit() to update the flags shared with IP_FREEBIND
and IP_TRANSPARENT. IPv6 TCP and UDP sockets can already update these
flags without the lock through the corresponding SOL_IP options.
Move both options to the lockless setsockopt path.
Signed-off-by: Xuanqiang Luo <luoxuanqiang@kylinos.cn>
Reviewed-by: Hangbin Liu <liuhangbin@kylinos.cn>
Link: https://patch.msgid.link/20260912134040.215664-1-xuanqiang.luo@linux.dev
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
