Jakub Kicinski [Fri, 27 Feb 2026 00:33:59 +0000 (16:33 -0800)]
tcp: give up on stronger sk_rcvbuf checks (for now)
We hit another corner case which leads to TcpExtTCPRcvQDrop
Connections which send RPCs in the 20-80kB range over loopback
experience spurious drops. The exact conditions for most of
the drops I investigated are that:
- socket exchanged >1MB of data so its not completely fresh
- rcvbuf is around 128kB (default, hasn't grown)
- there is ~60kB of data in rcvq
- skb > 64kB arrives
The sum of skb->len (!) of both of the skbs (the one already
in rcvq and the arriving one) is larger than rwnd.
My suspicion is that this happens because __tcp_select_window()
rounds the rwnd up to (1 << wscale) if less than half of
the rwnd has been consumed.
Eric suggests that given the number of Fixes we already have
pointing to 1d2fbaad7cd8 it's probably time to give up on it,
until a bigger revamp of rmem management.
Also while we could risk tweaking the rwnd math, there are other
drops on workloads I investigated, after the commit in question,
not explained by this phenomenon.
udp: Unhash auto-bound connected sk from 4-tuple hash table when disconnected.
Let's say we bind() an UDP socket to the wildcard address with a
non-zero port, connect() it to an address, and disconnect it from
the address.
bind() sets SOCK_BINDPORT_LOCK on sk->sk_userlocks (but not
SOCK_BINDADDR_LOCK), and connect() calls udp_lib_hash4() to put
the socket into the 4-tuple hash table.
It computes a new hash based on the wildcard address and moves
the socket to a new slot in the 4-tuple hash table, leaving a
garbage in the chain that no packet hits.
Let's remove such a socket from 4-tuple hash table when disconnected.
Note that udp_sk(sk)->udp_portaddr_hash needs to be udpated after
udp_hash4_dec(hslot2) in udp_unhash4().
Fixes: 78c91ae2c6de ("ipv4/udp: Add 4-tuple hash for connected socket") Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20260227035547.3321327-1-kuniyu@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Long Li [Thu, 26 Feb 2026 19:28:33 +0000 (11:28 -0800)]
net: mana: Ring doorbell at 4 CQ wraparounds
MANA hardware requires at least one doorbell ring every 8 wraparounds
of the CQ. The driver rings the doorbell as a form of flow control to
inform hardware that CQEs have been consumed.
The NAPI poll functions mana_poll_tx_cq() and mana_poll_rx_cq() can
poll up to CQE_POLLING_BUFFER (512) completions per call. If the CQ
has fewer than 512 entries, a single poll call can process more than
4 wraparounds without ringing the doorbell. The doorbell threshold
check also uses ">" instead of ">=", delaying the ring by one extra
CQE beyond 4 wraparounds. Combined, these issues can cause the driver
to exceed the 8-wraparound hardware limit, leading to missed
completions and stalled queues.
Fix this by capping the number of CQEs polled per call to 4 wraparounds
of the CQ in both TX and RX paths. Also change the doorbell threshold
from ">" to ">=" so the doorbell is rung as soon as 4 wraparounds are
reached.
Cc: stable@vger.kernel.org Fixes: 58a63729c957 ("net: mana: Fix doorbell out of order violation and avoid unnecessary doorbell rings") Signed-off-by: Long Li <longli@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://patch.msgid.link/20260226192833.1050807-1-longli@microsoft.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Victor Nogueira [Wed, 25 Feb 2026 13:43:48 +0000 (10:43 -0300)]
net/sched: Only allow act_ct to bind to clsact/ingress qdiscs and shared blocks
As Paolo said earlier [1]:
"Since the blamed commit below, classify can return TC_ACT_CONSUMED while
the current skb being held by the defragmentation engine. As reported by
GangMin Kim, if such packet is that may cause a UaF when the defrag engine
later on tries to tuch again such packet."
act_ct was never meant to be used in the egress path, however some users
are attaching it to egress today [2]. Attempting to reach a middle
ground, we noticed that, while most qdiscs are not handling
TC_ACT_CONSUMED, clsact/ingress qdiscs are. With that in mind, we
address the issue by only allowing act_ct to bind to clsact/ingress
qdiscs and shared blocks. That way it's still possible to attach act_ct to
egress (albeit only with clsact).
Florian Westphal [Thu, 26 Feb 2026 16:19:17 +0000 (17:19 +0100)]
selftests: netfilter: nft_queue.sh: avoid flakes on debug kernels
Jakub reports test flakes on debug kernels:
FAIL: test_udp_gro_ct: Expected software segmentation to occur, had 23 and 17
This test assumes that the kernels nfnetlink_queue module sees N GSO
packets, segments them into M skbs and queues them to userspace for
reinjection.
Hence, if M >= N, no segmentation occurred.
However, its possible that this happens:
- nfnetlink_queue gets GSO packet
- segments that into n skbs
- userspace buffer is full, kernel drops the segmented skbs
-> "toqueue" counter incremented by 1, "fromqueue" is unchanged.
If this happens often enough in a single run, M >= N check triggers
incorrectly.
To solve this, allow the nf_queue.c test program to set the FAIL_OPEN
flag so that the segmented skbs bypass the queueing step in the kernel
if the receive buffer is full.
Also, reduce number of sending socat instances, decrease their priority
and increase nice value for the nf_queue program itself to reduce the
probability of overruns happening in the first place.
Fixes: 59ecffa3995e ("selftests: netfilter: nft_queue.sh: add udp fraglist gro test case") Reported-by: Jakub Kicinski <kuba@kernel.org> Closes: https://lore.kernel.org/netdev/20260218184114.0b405b72@kernel.org/ Signed-off-by: Florian Westphal <fw@strlen.de> Link: https://patch.msgid.link/20260226161920.1205-1-fw@strlen.de Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jonas Köppeler [Thu, 26 Feb 2026 11:40:16 +0000 (12:40 +0100)]
net/sched: sch_cake: fixup cake_mq rate adjustment for diffserv config
cake_mq's rate adjustment during the sync periods did not adjust the
rates for every tin in a diffserv config. This lead to inconsistencies
of rates between the tins. Fix this by setting the rates for all tins
during synchronization.
Paul Moses [Mon, 23 Feb 2026 15:05:44 +0000 (15:05 +0000)]
net/sched: act_gate: snapshot parameters with RCU on replace
The gate action can be replaced while the hrtimer callback or dump path is
walking the schedule list.
Convert the parameters to an RCU-protected snapshot and swap updates under
tcf_lock, freeing the previous snapshot via call_rcu(). When REPLACE omits
the entry list, preserve the existing schedule so the effective state is
unchanged.
Fixes: a51c328df310 ("net: qos: introduce a gate control flow action") Cc: stable@vger.kernel.org Signed-off-by: Paul Moses <p@1g4.org> Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Reviewed-by: Victor Nogueira <victor@mojatatu.com> Link: https://patch.msgid.link/20260223150512.2251594-2-p@1g4.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Chintan Vankar [Tue, 24 Feb 2026 18:13:59 +0000 (23:43 +0530)]
net: ethernet: ti: am65-cpsw-nuss/cpsw-ale: Fix multicast entry handling in ALE table
In the current implementation, flushing multicast entries in MAC mode
incorrectly deletes entries for all ports instead of only the target port,
disrupting multicast traffic on other ports. The cause is adding multicast
entries by setting only host port bit, and not setting the MAC port bits.
Fix this by setting the MAC port's bit in the port mask while adding the
multicast entry. Also fix the flush logic to preserve the host port bit
during removal of MAC port and free ALE entries when mask contains only
host port.
Fixes: 5c50a856d550 ("drivers: net: ethernet: cpsw: add multicast address to ALE table") Signed-off-by: Chintan Vankar <c-vankar@ti.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260224181359.2055322-1-c-vankar@ti.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
bridge: Check relevant options in VLAN range grouping
The br_vlan_opts_eq_range() function determines if consecutive VLANs can
be grouped together in a range for compact netlink notifications. It
currently checks state, tunnel info, and multicast router configuration,
but misses two categories of per-VLAN options that affect the output:
1. User-visible priv_flags (neigh_suppress, mcast_enabled)
2. Port multicast context options (mcast_max_groups, mcast_n_groups)
When VLANs have different settings for these options, they are incorrectly
grouped into ranges, causing netlink notifications to report only one
VLAN's settings for the entire range.
Fix by checking priv_flags equality, but only for flags that affect netlink
output (BR_VLFLAG_NEIGH_SUPPRESS_ENABLED and BR_VLFLAG_MCAST_ENABLED),
and comparing multicast context options (mcast_max_groups, mcast_n_groups).
Add a test with four test cases for each option, to ensure that VLANs with
different values are not grouped into ranges and VLANs with matching
values are properly grouped together.
====================
Danielle Ratson [Wed, 25 Feb 2026 14:39:56 +0000 (16:39 +0200)]
selftests: net: Add bridge VLAN range grouping tests
Add a new test file bridge_vlan_dump.sh with four test cases that verify
VLANs with different per-VLAN options are not incorrectly grouped into
ranges in the dump output.
The tests verify the kernel's br_vlan_opts_eq_range() function correctly
prevents VLAN range grouping when neigh_suppress, mcast_max_groups,
mcast_n_groups, or mcast_enabled options differ.
Each test verifies that VLANs with different option values appear as
individual entries rather than ranges, and that VLANs with matching
values are properly grouped together.
Example output:
$ ./bridge_vlan_dump.sh
TEST: VLAN range grouping with neigh_suppress [ OK ]
TEST: VLAN range grouping with mcast_max_groups [ OK ]
TEST: VLAN range grouping with mcast_n_groups [ OK ]
TEST: VLAN range grouping with mcast_enabled [ OK ]
Danielle Ratson [Wed, 25 Feb 2026 14:39:55 +0000 (16:39 +0200)]
bridge: Check relevant per-VLAN options in VLAN range grouping
The br_vlan_opts_eq_range() function determines if consecutive VLANs can
be grouped together in a range for compact netlink notifications. It
currently checks state, tunnel info, and multicast router configuration,
but misses two categories of per-VLAN options that affect the output:
1. User-visible priv_flags (neigh_suppress, mcast_enabled)
2. Port multicast context (mcast_max_groups, mcast_n_groups)
When VLANs have different settings for these options, they are incorrectly
grouped into ranges, causing netlink notifications to report only one
VLAN's settings for the entire range.
Fix by checking priv_flags equality, but only for flags that affect netlink
output (BR_VLFLAG_NEIGH_SUPPRESS_ENABLED and BR_VLFLAG_MCAST_ENABLED),
and comparing multicast context (mcast_max_groups and mcast_n_groups).
Example showing the bugs before the fix:
$ bridge vlan set vid 10 dev dummy1 neigh_suppress on
$ bridge vlan set vid 11 dev dummy1 neigh_suppress off
$ bridge -d vlan show dev dummy1
port vlan-id
dummy1 10-11
... neigh_suppress on
$ bridge vlan set vid 10 dev dummy1 mcast_max_groups 100
$ bridge vlan set vid 11 dev dummy1 mcast_max_groups 200
$ bridge -d vlan show dev dummy1
port vlan-id
dummy1 10-11
... mcast_max_groups 100
After the fix, VLANs 10 and 11 are shown as separate entries with their
correct individual settings.
Fixes: a1aee20d5db2 ("net: bridge: Add netlink knobs for number / maximum MDB entries") Fixes: 83f6d600796c ("bridge: vlan: Allow setting VLAN neighbor suppression state") Signed-off-by: Danielle Ratson <danieller@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Acked-by: Nikolay Aleksandrov <razor@blackwall.org> Link: https://patch.msgid.link/20260225143956.3995415-2-danieller@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Davide Caratti [Tue, 24 Feb 2026 20:28:32 +0000 (21:28 +0100)]
net/sched: ets: fix divide by zero in the offload path
Offloading ETS requires computing each class' WRR weight: this is done by
averaging over the sums of quanta as 'q_sum' and 'q_psum'. Using unsigned
int, the same integer size as the individual DRR quanta, can overflow and
even cause division by zero, like it happened in the following splat:
Linus Torvalds [Thu, 26 Feb 2026 16:00:13 +0000 (08:00 -0800)]
Merge tag 'net-7.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Paolo Abeni:
"Including fixes from IPsec, Bluetooth and netfilter
Current release - regressions:
- wifi: fix dev_alloc_name() return value check
- rds: fix recursive lock in rds_tcp_conn_slots_available
Current release - new code bugs:
- vsock: lock down child_ns_mode as write-once
Previous releases - regressions:
- core:
- do not pass flow_id to set_rps_cpu()
- consume xmit errors of GSO frames
- netconsole: avoid OOB reads, msg is not nul-terminated
- netfilter: h323: fix OOB read in decode_choice()
- tcp: re-enable acceptance of FIN packets when RWIN is 0
- udplite: fix null-ptr-deref in __udp_enqueue_schedule_skb().
- wifi: brcmfmac: fix potential kernel oops when probe fails
- phy: register phy led_triggers during probe to avoid AB-BA deadlock
- eth:
- bnxt_en: fix deleting of Ntuple filters
- wan: farsync: fix use-after-free bugs caused by unfinished tasklets
- xscale: check for PTP support properly
Previous releases - always broken:
- tcp: fix potential race in tcp_v6_syn_recv_sock()
- kcm: fix zero-frag skb in frag_list on partial sendmsg error
- xfrm:
- fix race condition in espintcp_close()
- always flush state and policy upon NETDEV_UNREGISTER event
- bluetooth:
- purge error queues in socket destructors
- fix response to L2CAP_ECRED_CONN_REQ
- eth:
- mlx5:
- fix circular locking dependency in dump
- fix "scheduling while atomic" in IPsec MAC address query
- gve: fix incorrect buffer cleanup for QPL
- team: avoid NETDEV_CHANGEMTU event when unregistering slave
- usb: validate USB endpoints"
* tag 'net-7.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (72 commits)
netfilter: nf_conntrack_h323: fix OOB read in decode_choice()
dpaa2-switch: validate num_ifs to prevent out-of-bounds write
net: consume xmit errors of GSO frames
vsock: document write-once behavior of the child_ns_mode sysctl
vsock: lock down child_ns_mode as write-once
selftests/vsock: change tests to respect write-once child ns mode
net/mlx5e: Fix "scheduling while atomic" in IPsec MAC address query
net/mlx5: Fix missing devlink lock in SRIOV enable error path
net/mlx5: E-switch, Clear legacy flag when moving to switchdev
net/mlx5: LAG, disable MPESW in lag_disable_change()
net/mlx5: DR, Fix circular locking dependency in dump
selftests: team: Add a reference count leak test
team: avoid NETDEV_CHANGEMTU event when unregistering slave
net: mana: Fix double destroy_workqueue on service rescan PCI path
MAINTAINERS: Update maintainer entry for QUALCOMM ETHQOS ETHERNET DRIVER
dpll: zl3073x: Remove redundant cleanup in devm_dpll_init()
selftests/net: packetdrill: Verify acceptance of FIN packets when RWIN is 0
tcp: re-enable acceptance of FIN packets when RWIN is 0
vsock: Use container_of() to get net namespace in sysctl handlers
net: usb: kaweth: validate USB endpoints
...
Vahagn Vardanian [Wed, 25 Feb 2026 13:06:18 +0000 (14:06 +0100)]
netfilter: nf_conntrack_h323: fix OOB read in decode_choice()
In decode_choice(), the boundary check before get_len() uses the
variable `len`, which is still 0 from its initialization at the top of
the function:
unsigned int type, ext, len = 0;
...
if (ext || (son->attr & OPEN)) {
BYTE_ALIGN(bs);
if (nf_h323_error_boundary(bs, len, 0)) /* len is 0 here */
return H323_ERROR_BOUND;
len = get_len(bs); /* OOB read */
When the bitstream is exactly consumed (bs->cur == bs->end), the check
nf_h323_error_boundary(bs, 0, 0) evaluates to (bs->cur + 0 > bs->end),
which is false. The subsequent get_len() call then dereferences
*bs->cur++, reading 1 byte past the end of the buffer. If that byte
has bit 7 set, get_len() reads a second byte as well.
This can be triggered remotely by sending a crafted Q.931 SETUP message
with a User-User Information Element containing exactly 2 bytes of
PER-encoded data ({0x08, 0x00}) to port 1720 through a firewall with
the nf_conntrack_h323 helper active. The decoder fully consumes the
PER buffer before reaching this code path, resulting in a 1-2 byte
heap-buffer-overflow read confirmed by AddressSanitizer.
Fix this by checking for 2 bytes (the maximum that get_len() may read)
instead of the uninitialized `len`. This matches the pattern used at
every other get_len() call site in the same file, where the caller
checks for 2 bytes of available data before calling get_len().
Fixes: ec8a8f3c31dd ("netfilter: nf_ct_h323: Extend nf_h323_error_boundary to work on bits as well") Signed-off-by: Vahagn Vardanian <vahagn@redrays.io> Signed-off-by: Florian Westphal <fw@strlen.de> Link: https://patch.msgid.link/20260225130619.1248-2-fw@strlen.de Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Junrui Luo [Tue, 24 Feb 2026 11:05:56 +0000 (19:05 +0800)]
dpaa2-switch: validate num_ifs to prevent out-of-bounds write
The driver obtains sw_attr.num_ifs from firmware via dpsw_get_attributes()
but never validates it against DPSW_MAX_IF (64). This value controls
iteration in dpaa2_switch_fdb_get_flood_cfg(), which writes port indices
into the fixed-size cfg->if_id[DPSW_MAX_IF] array. When firmware reports
num_ifs >= 64, the loop can write past the array bounds.
Add a bound check for num_ifs in dpaa2_switch_init().
dpaa2_switch_fdb_get_flood_cfg() appends the control interface (port
num_ifs) after all matched ports. When num_ifs == DPSW_MAX_IF and all
ports match the flood filter, the loop fills all 64 slots and the control
interface write overflows by one entry.
The check uses >= because num_ifs == DPSW_MAX_IF is also functionally
broken.
build_if_id_bitmap() silently drops any ID >= 64:
if (id[i] < DPSW_MAX_IF)
bmap[id[i] / 64] |= ...
Jakub Kicinski [Mon, 23 Feb 2026 23:51:00 +0000 (15:51 -0800)]
net: consume xmit errors of GSO frames
udpgro_frglist.sh and udpgro_bench.sh are the flakiest tests
currently in NIPA. They fail in the same exact way, TCP GRO
test stalls occasionally and the test gets killed after 10min.
These tests use veth to simulate GRO. They attach a trivial
("return XDP_PASS;") XDP program to the veth to force TSO off
and NAPI on.
Digging into the failure mode we can see that the connection
is completely stuck after a burst of drops. The sender's snd_nxt
is at sequence number N [1], but the receiver claims to have
received (rcv_nxt) up to N + 3 * MSS [2]. Last piece of the puzzle
is that senders rtx queue is not empty (let's say the block in
the rtx queue is at sequence number N - 4 * MSS [3]).
In this state, sender sends a retransmission from the rtx queue
with a single segment, and sequence numbers N-4*MSS:N-3*MSS [3].
Receiver sees it and responds with an ACK all the way up to
N + 3 * MSS [2]. But sender will reject this ack as TCP_ACK_UNSENT_DATA
because it has no recollection of ever sending data that far out [1].
And we are stuck.
The root cause is the mess of the xmit return codes. veth returns
an error when it can't xmit a frame. We end up with a loss event
like this:
-------------------------------------------------
| GSO super frame 1 | GSO super frame 2 |
|-----------------------------------------------|
| seg | seg | seg | seg | seg | seg | seg | seg |
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
-------------------------------------------------
x ok ok <ok>| ok ok ok <x>
\\
snd_nxt
"x" means packet lost by veth, and "ok" means it went thru.
Since veth has TSO disabled in this test it sees individual segments.
Segment 1 is on the retransmit queue and will be resent.
So why did the sender not advance snd_nxt even tho it clearly did
send up to seg 8? tcp_write_xmit() interprets the return code
from the core to mean that data has not been sent at all. Since
TCP deals with GSO super frames, not individual segment the crux
of the problem is that loss of a single segment can be interpreted
as loss of all. TCP only sees the last return code for the last
segment of the GSO frame (in <> brackets in the diagram above).
Of course for the problem to occur we need a setup or a device
without a Qdisc. Otherwise Qdisc layer disconnects the protocol
layer from the device errors completely.
We have multiple ways to fix this.
1) make veth not return an error when it lost a packet.
While this is what I think we did in the past, the issue keeps
reappearing and it's annoying to debug. The game of whack
a mole is not great.
2) fix the damn return codes
We only talk about NETDEV_TX_OK and NETDEV_TX_BUSY in the
documentation, so maybe we should make the return code from
ndo_start_xmit() a boolean. I like that the most, but perhaps
some ancient, not-really-networking protocol would suffer.
3) make TCP ignore the errors
It is not entirely clear to me what benefit TCP gets from
interpreting the result of ip_queue_xmit()? Specifically once
the connection is established and we're pushing data - packet
loss is just packet loss?
4) this fix
Ignore the rc in the Qdisc-less+GSO case, since it's unreliable.
We already always return OK in the TCQ_F_CAN_BYPASS case.
In the Qdisc-less case let's be a bit more conservative and only
mask the GSO errors. This path is taken by non-IP-"networks"
like CAN, MCTP etc, so we could regress some ancient thing.
This is the simplest, but also maybe the hackiest fix?
Similar fix has been proposed by Eric in the past but never committed
because original reporter was working with an OOT driver and wasn't
providing feedback (see Link).
====================
vsock: add write-once semantics to child_ns_mode
Two administrator processes may race when setting child_ns_mode: one
sets it to "local" and creates a namespace, but another changes it to
"global" in between. The first process ends up with a namespace in the
wrong mode. Make child_ns_mode write-once so that a namespace manager
can set it once, check the value, and be guaranteed it won't change
before creating its namespaces. Writing a different value after the
first write returns -EBUSY.
One patch for the implementation, one for docs, and one for tests.
Bobby Eshleman [Mon, 23 Feb 2026 22:38:34 +0000 (14:38 -0800)]
vsock: document write-once behavior of the child_ns_mode sysctl
Update the vsock child_ns_mode documentation to include the new
write-once semantics of setting child_ns_mode. The semantics are
implemented in a preceding patch in this series.
Bobby Eshleman [Mon, 23 Feb 2026 22:38:33 +0000 (14:38 -0800)]
vsock: lock down child_ns_mode as write-once
Two administrator processes may race when setting child_ns_mode as one
process sets child_ns_mode to "local" and then creates a namespace, but
another process changes child_ns_mode to "global" between the write and
the namespace creation. The first process ends up with a namespace in
"global" mode instead of "local". While this can be detected after the
fact by reading ns_mode and retrying, it is fragile and error-prone.
Make child_ns_mode write-once so that a namespace manager can set it
once and be sure it won't change. Writing a different value after the
first write returns -EBUSY. This applies to all namespaces, including
init_net, where an init process can write "local" to lock all future
namespaces into local mode.
Fixes: eafb64f40ca4 ("vsock: add netns to vsock core") Suggested-by: Daan De Meyer <daan.j.demeyer@gmail.com> Suggested-by: Stefano Garzarella <sgarzare@redhat.com> Co-developed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Link: https://patch.msgid.link/20260223-vsock-ns-write-once-v3-2-c0cde6959923@meta.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Bobby Eshleman [Mon, 23 Feb 2026 22:38:32 +0000 (14:38 -0800)]
selftests/vsock: change tests to respect write-once child ns mode
The child_ns_mode sysctl parameter becomes write-once in a future patch
in this series, which breaks existing tests. This patch updates the
tests to respect this new policy. No additional tests are added.
Add "global-parent" and "local-parent" namespaces as intermediaries to
spawn namespaces in the given modes. This avoids the need to change
"child_ns_mode" in the init_ns. nsenter must be used because ip netns
unshares the mount namespace so nested "ip netns add" breaks exec calls
from the init ns. Adds nsenter to the deps check.
Jianbo Liu [Tue, 24 Feb 2026 11:46:52 +0000 (13:46 +0200)]
net/mlx5e: Fix "scheduling while atomic" in IPsec MAC address query
Fix a "scheduling while atomic" bug in mlx5e_ipsec_init_macs() by
replacing mlx5_query_mac_address() with ether_addr_copy() to get the
local MAC address directly from netdev->dev_addr.
The issue occurs because mlx5_query_mac_address() queries the hardware
which involves mlx5_cmd_exec() that can sleep, but it is called from
the mlx5e_ipsec_handle_event workqueue which runs in atomic context.
The MAC address is already available in netdev->dev_addr, so no need
to query hardware. This avoids the sleeping call and resolves the bug.
Shay Drory [Tue, 24 Feb 2026 11:46:51 +0000 (13:46 +0200)]
net/mlx5: Fix missing devlink lock in SRIOV enable error path
The cited commit miss to add locking in the error path of
mlx5_sriov_enable(). When pci_enable_sriov() fails,
mlx5_device_disable_sriov() is called to clean up. This cleanup function
now expects to be called with the devlink instance lock held.
Add the missing devl_lock(devlink) and devl_unlock(devlink)
Fixes: 84a433a40d0e ("net/mlx5: Lock mlx5 devlink reload callbacks") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260224114652.1787431-5-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Shay Drory [Tue, 24 Feb 2026 11:46:50 +0000 (13:46 +0200)]
net/mlx5: E-switch, Clear legacy flag when moving to switchdev
The cited commit introduced MLX5_PRIV_FLAGS_SWITCH_LEGACY to identify
when a transition to legacy mode is requested via devlink. However, the
logic failed to clear this flag if the mode was subsequently changed
back to MLX5_ESWITCH_OFFLOADS (switchdev). Consequently, if a user
toggled from legacy to switchdev, the flag remained set, leaving the
driver with wrong state indicating
Fix this by explicitly clearing the MLX5_PRIV_FLAGS_SWITCH_LEGACY bit
when the requested mode is MLX5_ESWITCH_OFFLOADS.
Fixes: 2a4f56fbcc47 ("net/mlx5e: Keep netdev when leave switchdev for devlink set legacy only") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260224114652.1787431-4-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Shay Drory [Tue, 24 Feb 2026 11:46:49 +0000 (13:46 +0200)]
net/mlx5: LAG, disable MPESW in lag_disable_change()
mlx5_lag_disable_change() unconditionally called mlx5_disable_lag() when
LAG was active, which is incorrect for MLX5_LAG_MODE_MPESW.
Hnece, call mlx5_disable_mpesw() when running in MPESW mode.
Fixes: a32327a3a02c ("net/mlx5: Lag, Control MultiPort E-Switch single FDB mode") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260224114652.1787431-3-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Shay Drory [Tue, 24 Feb 2026 11:46:48 +0000 (13:46 +0200)]
net/mlx5: DR, Fix circular locking dependency in dump
Fix a circular locking dependency between dbg_mutex and the domain
rx/tx mutexes that could lead to a deadlock.
The dump path in dr_dump_domain_all() was acquiring locks in the order:
dbg_mutex -> rx.mutex -> tx.mutex
While the table/matcher creation paths acquire locks in the order:
rx.mutex -> tx.mutex -> dbg_mutex
This inverted lock ordering creates a circular dependency. Fix this by
changing dr_dump_domain_all() to acquire the domain lock before
dbg_mutex, matching the order used in mlx5dr_table_create() and
mlx5dr_matcher_create().
Lockdep splat:
======================================================
WARNING: possible circular locking dependency detected
6.19.0-rc6net_next_e817c4e #1 Not tainted
------------------------------------------------------
sos/30721 is trying to acquire lock: ffff888102df5900 (&dmn->info.rx.mutex){+.+.}-{4:4}, at:
dr_dump_start+0x131/0x450 [mlx5_core]
but task is already holding lock: ffff888102df5bc0 (&dmn->dump_info.dbg_mutex){+.+.}-{4:4}, at:
dr_dump_start+0x10b/0x450 [mlx5_core]
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
Jakub Kicinski [Thu, 26 Feb 2026 03:54:28 +0000 (19:54 -0800)]
Merge tag 'wireless-2026-02-25' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless
Johannes Berg says:
====================
A good number of fixes:
- cfg80211:
- cancel rfkill work appropriately
- fix radiotap parsing to correctly reject field 18
- fix wext (yes...) off-by-one for IGTK key ID
- mac80211:
- fix for mesh NULL pointer dereference
- fix for stack out-of-bounds (2 bytes) write on
specific multi-link action frames
- set default WMM parameters for all links
- mwifiex: check dev_alloc_name() return value correctly
- libertas: fix potential timer use-after-free
- brcmfmac: fix crash on probe failure
* tag 'wireless-2026-02-25' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
wifi: mac80211: fix NULL pointer dereference in mesh_rx_csa_frame()
wifi: mac80211: bounds-check link_id in ieee80211_ml_reconfiguration
wifi: mac80211: set default WMM parameters on all links
wifi: libertas: fix use-after-free in lbs_free_adapter()
wifi: mwifiex: Fix dev_alloc_name() return value check
wifi: brcmfmac: Fix potential kernel oops when probe fails
wifi: radiotap: reject radiotap with unknown bits
wifi: cfg80211: cancel rfkill_block work in wiphy_unregister()
wifi: cfg80211: wext: fix IGTK key ID off-by-one
====================
====================
team: Fix reference count leak when changing port netns
Patch #1 fixes a reference count leak that was reported by syzkaller.
The leak happens when a net device that is member in a team is changing
netns. The fix is to align the team driver with the bond driver and have
it suppress NETDEV_CHANGEMTU events for a net device that is being
unregistered.
Without this change, the NETDEV_CHANGEMTU event causes inetdev_event()
to recreate an inet device for this net device in its original netns,
after it was previously destroyed upon NETDEV_UNREGISTER. Later on, when
inetdev_event() receives a NETDEV_REGISTER event for this net device in
the new nents, it simply leaks the reference:
case NETDEV_REGISTER:
pr_debug("%s: bug\n", __func__);
RCU_INIT_POINTER(dev->ip_ptr, NULL);
break;
addrconf_notify() handles this differently and reuses the existing inet6
device if one exists when a NETDEV_REGISTER event is received. This
creates a different problem where it is possible for a net device to
reference an inet6 device that was created in a previous netns.
A more generic fix that we can try in net-next is to revert the changes
in the bond and team drivers and instead have IPv4 and IPv6 destroy and
recreate an inet device if one already exists upon NETDEV_REGISTER.
Patch #2 adds a selftest that passes with the fix and hangs without it.
====================
Ido Schimmel [Tue, 24 Feb 2026 12:57:09 +0000 (14:57 +0200)]
selftests: team: Add a reference count leak test
Add a test for the issue that was fixed in "team: avoid NETDEV_CHANGEMTU
event when unregistering slave".
The test hangs due to a reference count leak without the fix:
# make -C tools/testing/selftests TARGETS="drivers/net/team" TEST_PROGS=refleak.sh TEST_GEN_PROGS="" run_tests
[...]
TAP version 13
1..1
# timeout set to 45
# selftests: drivers/net/team: refleak.sh
[ 50.681299][ T496] unregister_netdevice: waiting for dummy1 to become free. Usage count = 3
[ 71.185325][ T496] unregister_netdevice: waiting for dummy1 to become free. Usage count = 3
And passes with the fix:
# make -C tools/testing/selftests TARGETS="drivers/net/team" TEST_PROGS=refleak.sh TEST_GEN_PROGS="" run_tests
[...]
TAP version 13
1..1
# timeout set to 45
# selftests: drivers/net/team: refleak.sh
ok 1 selftests: drivers/net/team: refleak.sh
ip link add name team1 type team
ip link add name dummy1 mtu 1499 master team1 type dummy
ip netns add ns1
ip link set dev dummy1 netns ns1
ip -n ns1 link del dev dummy1
and also found that the same issue was fixed in the bond driver in
commit f51048c3e07b ("bonding: avoid NETDEV_CHANGEMTU event when
unregistering slave").
Let's do similar thing for the team driver, with commit ad7c7b2172c3 ("net:
hold netdev instance lock during sysfs operations") and commit 303a8487a657
("net: s/__dev_set_mtu/__netif_set_mtu/") also applied.
Dipayaan Roy [Tue, 24 Feb 2026 12:38:36 +0000 (04:38 -0800)]
net: mana: Fix double destroy_workqueue on service rescan PCI path
While testing corner cases in the driver, a use-after-free crash
was found on the service rescan PCI path.
When mana_serv_reset() calls mana_gd_suspend(), mana_gd_cleanup()
destroys gc->service_wq. If the subsequent mana_gd_resume() fails
with -ETIMEDOUT or -EPROTO, the code falls through to
mana_serv_rescan() which triggers pci_stop_and_remove_bus_device().
This invokes the PCI .remove callback (mana_gd_remove), which calls
mana_gd_cleanup() a second time, attempting to destroy the already-
freed workqueue. Fix this by NULL-checking gc->service_wq in
mana_gd_cleanup() and setting it to NULL after destruction.
Call stack of issue for reference:
[Sat Feb 21 18:53:48 2026] Call Trace:
[Sat Feb 21 18:53:48 2026] <TASK>
[Sat Feb 21 18:53:48 2026] mana_gd_cleanup+0x33/0x70 [mana]
[Sat Feb 21 18:53:48 2026] mana_gd_remove+0x3a/0xc0 [mana]
[Sat Feb 21 18:53:48 2026] pci_device_remove+0x41/0xb0
[Sat Feb 21 18:53:48 2026] device_remove+0x46/0x70
[Sat Feb 21 18:53:48 2026] device_release_driver_internal+0x1e3/0x250
[Sat Feb 21 18:53:48 2026] device_release_driver+0x12/0x20
[Sat Feb 21 18:53:48 2026] pci_stop_bus_device+0x6a/0x90
[Sat Feb 21 18:53:48 2026] pci_stop_and_remove_bus_device+0x13/0x30
[Sat Feb 21 18:53:48 2026] mana_do_service+0x180/0x290 [mana]
[Sat Feb 21 18:53:48 2026] mana_serv_func+0x24/0x50 [mana]
[Sat Feb 21 18:53:48 2026] process_one_work+0x190/0x3d0
[Sat Feb 21 18:53:48 2026] worker_thread+0x16e/0x2e0
[Sat Feb 21 18:53:48 2026] kthread+0xf7/0x130
[Sat Feb 21 18:53:48 2026] ? __pfx_worker_thread+0x10/0x10
[Sat Feb 21 18:53:48 2026] ? __pfx_kthread+0x10/0x10
[Sat Feb 21 18:53:48 2026] ret_from_fork+0x269/0x350
[Sat Feb 21 18:53:48 2026] ? __pfx_kthread+0x10/0x10
[Sat Feb 21 18:53:48 2026] ret_from_fork_asm+0x1a/0x30
[Sat Feb 21 18:53:48 2026] </TASK>
Mohd Ayaan Anwar [Tue, 24 Feb 2026 12:28:47 +0000 (17:58 +0530)]
MAINTAINERS: Update maintainer entry for QUALCOMM ETHQOS ETHERNET DRIVER
Replace Vinod Koul with Mohd Ayaan Anwar as the maintainer of the
QUALCOMM ETHQOS ETHERNET DRIVER. Vinod confirmed he is no longer
active in this area and agreed to be removed.
Acked-by: Vinod Koul <vkoul@kernel.org> Suggested-by: Russell King (Oracle) <linux@armlinux.org.uk> Signed-off-by: Mohd Ayaan Anwar <mohd.anwar@oss.qualcomm.com> Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/20260224-qcom_ethqos_maintainer-v1-1-24e02701ea52@oss.qualcomm.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Felix Gu [Tue, 24 Feb 2026 11:04:04 +0000 (19:04 +0800)]
dpll: zl3073x: Remove redundant cleanup in devm_dpll_init()
The devm_add_action_or_reset() function already executes the cleanup
action on failure before returning an error, so the explicit goto error
and subsequent zl3073x_dev_dpll_fini() call causes double cleanup.
Fixes: ebb1031c5137 ("dpll: zl3073x: Refactor DPLL initialization") Reviewed-by: Ivan Vecera <ivecera@redhat.com> Signed-off-by: Felix Gu <ustc.gu@gmail.com> Link: https://patch.msgid.link/20260224-dpll-v2-1-d7786414a830@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
tcp: re-enable acceptance of FIN packets when RWIN is 0
this series restores the ability to accept in‑sequence FIN packets
even when the advertised receive window is zero, and adds a
packetdrill test to guard the behavior.
====================
Simon Baatz [Tue, 24 Feb 2026 08:20:12 +0000 (09:20 +0100)]
tcp: re-enable acceptance of FIN packets when RWIN is 0
Commit 2bd99aef1b19 ("tcp: accept bare FIN packets under memory
pressure") allowed accepting FIN packets in tcp_data_queue() even when
the receive window was closed, to prevent ACK/FIN loops with broken
clients.
Such a FIN packet is in sequence, but because the FIN consumes a
sequence number, it extends beyond the window. Before commit 9ca48d616ed7 ("tcp: do not accept packets beyond window"),
tcp_sequence() only required the seq to be within the window. After
that change, the entire packet (including the FIN) must fit within the
window. As a result, such FIN packets are now dropped and the handling
path is no longer reached.
Be more lenient by not counting the sequence number consumed by the
FIN when calling tcp_sequence(), restoring the previous behavior for
cases where only the FIN extends beyond the window.
Fixes: 9ca48d616ed7 ("tcp: do not accept packets beyond window") Signed-off-by: Simon Baatz <gmbnomis@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com> Link: https://patch.msgid.link/20260224-fix_zero_wnd_fin-v2-1-a16677ea7cea@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
vsock: Use container_of() to get net namespace in sysctl handlers
current->nsproxy is should not be accessed directly as syzbot has found
that it could be NULL at times, causing crashes. Fix up the af_vsock
sysctl handlers to use container_of() to deal with the current net
namespace instead of attempting to rely on current.
This is the same type of change done in commit 7f5611cbc487 ("rds:
sysctl: rds_tcp_{rcv,snd}buf: avoid using current->nsproxy")
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Bobby Eshleman <bobbyeshleman@meta.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Fixes: eafb64f40ca4 ("vsock: add netns to vsock core") Link: https://patch.msgid.link/2026022318-rearview-gallery-ae13@gregkh Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The kaweth driver should validate that the device it is probing has the
proper number and types of USB endpoints it is expecting before it binds
to it. If a malicious device were to not have the same urbs the driver
will crash later on when it blindly accesses these endpoints.
Cc: stable <stable@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Simon Horman <horms@kernel.org> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Link: https://patch.msgid.link/2026022305-substance-virtual-c728@gregkh Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The kalmia driver should validate that the device it is probing has the
proper number and types of USB endpoints it is expecting before it binds
to it. If a malicious device were to not have the same urbs the driver
will crash later on when it blindly accesses these endpoints.
Cc: stable <stable@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Simon Horman <horms@kernel.org> Fixes: d40261236e8e ("net/usb: Add Samsung Kalmia driver for Samsung GT-B3730") Link: https://patch.msgid.link/2026022326-shack-headstone-ef6f@gregkh Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The pegasus driver should validate that the device it is probing has the
proper number and types of USB endpoints it is expecting before it binds
to it. If a malicious device were to not have the same urbs the driver
will crash later on when it blindly accesses these endpoints.
nfc: pn533: properly drop the usb interface reference on disconnect
When the device is disconnected from the driver, there is a "dangling"
reference count on the usb interface that was grabbed in the probe
callback. Fix this up by properly dropping the reference after we are
done with it.
Linus Torvalds [Thu, 26 Feb 2026 00:39:25 +0000 (16:39 -0800)]
Merge tag 'erofs-for-7.0-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs
Pull erofs fixes from Gao Xiang:
- Do not share the page cache if the real @aops differs
- Fix the incomplete condition for interlaced plain extents
- Get rid of more unnecessary #ifdefs
* tag 'erofs-for-7.0-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
erofs: fix interlaced plain identification for encoded extents
erofs: remove more unnecessary #ifdefs
erofs: allow sharing page cache with the same aops only
Linus Torvalds [Wed, 25 Feb 2026 18:41:14 +0000 (10:41 -0800)]
Merge tag 'ata-7.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux
Pull ata fixes from Niklas Cassel:
- The newly introduced feature that issues a deferred (non-NCQ) command
from a workqueue, forgot to consider the case where the deferred QC
times out. Fix the code to take timeouts into consideration, which
avoids a use after free (Damien)
- The newly introduced feature that issues a deferred (non-NCQ) command
from a workqueue, when unloading the module, calls cancel_work_sync(),
a function that can sleep, while holding a spin lock. Move the function
call outside the lock (Damien)
* tag 'ata-7.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux:
ata: libata-core: fix cancellation of a port deferred qc work
ata: libata-eh: correctly handle deferred qc timeouts
Linus Torvalds [Wed, 25 Feb 2026 18:34:23 +0000 (10:34 -0800)]
Merge tag 'vfs-7.0-rc2.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull vfs fixes from Christian Brauner:
- Fix an uninitialized variable in file_getattr().
The flags_valid field wasn't initialized before calling
vfs_fileattr_get(), triggering KMSAN uninit-value reports in fuse
- Fix writeback wakeup and logging timeouts when DETECT_HUNG_TASK is
not enabled.
sysctl_hung_task_timeout_secs is 0 in that case causing spurious
"waiting for writeback completion for more than 1 seconds" warnings
- Fix a null-ptr-deref in do_statmount() when the mount is internal
- Add missing kernel-doc description for the @private parameter in
iomap_readahead()
- Fix mount namespace creation to hold namespace_sem across the mount
copy in create_new_namespace().
The previous drop-and-reacquire pattern was fragile and failed to
clean up mount propagation links if the real rootfs was a shared or
dependent mount
- Fix /proc mount iteration where m->index wasn't updated when
m->show() overflows, causing a restart to repeatedly show the same
mount entry in a rapidly expanding mount table
- Return EFSCORRUPTED instead of ENOSPC in minix_new_inode() when the
inode number is out of range
- Fix unshare(2) when CLONE_NEWNS is set and current->fs isn't shared.
copy_mnt_ns() received the live fs_struct so if a subsequent
namespace creation failed the rollback would leave pwd and root
pointing to detached mounts. Always allocate a new fs_struct when
CLONE_NEWNS is requested
- fserror bug fixes:
- Remove the unused fsnotify_sb_error() helper now that all callers
have been converted to fserror_report_metadata
- Fix a lockdep splat in fserror_report() where igrab() takes
inode::i_lock which can be held in IRQ context.
Replace igrab() with a direct i_count bump since filesystems
should not report inodes that are about to be freed or not yet
exposed
- Handle error pointer in procfs for try_lookup_noperm()
- Fix an integer overflow in ep_loop_check_proc() where recursive calls
returning INT_MAX would overflow when +1 is added, breaking the
recursion depth check
- Fix a misleading break in pidfs
* tag 'vfs-7.0-rc2.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
pidfs: avoid misleading break
eventpoll: Fix integer overflow in ep_loop_check_proc()
proc: Fix pointer error dereference
fserror: fix lockdep complaint when igrabbing inode
fsnotify: drop unused helper
unshare: fix unshare_fs() handling
minix: Correct errno in minix_new_inode
namespace: fix proc mount iteration
mount: hold namespace_sem across copy in create_new_namespace()
iomap: Describe @private in iomap_readahead()
statmount: Fix the null-ptr-deref in do_statmount()
writeback: Fix wakeup and logging timeouts for !DETECT_HUNG_TASK
fs: init flags_valid before calling vfs_fileattr_get
Gao Xiang [Tue, 24 Feb 2026 10:31:25 +0000 (18:31 +0800)]
erofs: fix interlaced plain identification for encoded extents
Only plain data whose start position and on-disk physical length are
both aligned to the block size should be classified as interlaced
plain extents. Otherwise, it must be treated as shifted plain extents.
This issue was found by syzbot using a crafted compressed image
containing plain extents with unaligned physical lengths, which can
cause OOB read in z_erofs_transform_plain().
net: stmmac: fix timestamping configuration after suspend/resume
When stmmac_init_timestamping() is called, it clears the receive and
transmit path booleans that allow timestamps to be read. These are
never re-initialised until after userspace requests timestamping
features to be enabled.
However, our copy of the timestamp configuration is not cleared, which
means we return the old configuration to userspace when requested.
This is inconsistent. Fix this by clearing the timestamp configuration.
Fixes: d6228b7cdd6e ("net: stmmac: implement the SIOCGHWTSTAMP ioctl") Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/E1vuUu4-0000000Afea-0j9B@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jens Axboe [Tue, 24 Feb 2026 18:51:16 +0000 (11:51 -0700)]
media: dvb-core: fix wrong reinitialization of ringbuffer on reopen
dvb_dvr_open() calls dvb_ringbuffer_init() when a new reader opens the
DVR device. dvb_ringbuffer_init() calls init_waitqueue_head(), which
reinitializes the waitqueue list head to empty.
Since dmxdev->dvr_buffer.queue is a shared waitqueue (all opens of the
same DVR device share it), this orphans any existing waitqueue entries
from io_uring poll or epoll, leaving them with stale prev/next pointers
while the list head is reset to {self, self}.
The waitqueue and spinlock in dvr_buffer are already properly
initialized once in dvb_dmxdev_init(). The open path only needs to
reset the buffer data pointer, size, and read/write positions.
Replace the dvb_ringbuffer_init() call in dvb_dvr_open() with direct
assignment of data/size and a call to dvb_ringbuffer_reset(), which
properly resets pread, pwrite, and error with correct memory ordering
without touching the waitqueue or spinlock.
Paolo Abeni [Tue, 24 Feb 2026 14:03:07 +0000 (15:03 +0100)]
Merge tag 'for-net-2026-02-23' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth
Luiz Augusto von Dentz says:
====================
bluetooth pull request for net:
- purge error queues in socket destructors
- hci_sync: Fix CIS host feature condition
- L2CAP: Fix invalid response to L2CAP_ECRED_RECONF_REQ
- L2CAP: Fix result of L2CAP_ECRED_CONN_RSP when MTU is too short
- L2CAP: Fix response to L2CAP_ECRED_CONN_REQ
- L2CAP: Fix not checking output MTU is acceptable on L2CAP_ECRED_CONN_REQ
- L2CAP: Fix missing key size check for L2CAP_LE_CONN_REQ
- hci_qca: Cleanup on all setup failures
* tag 'for-net-2026-02-23' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth:
Bluetooth: L2CAP: Fix missing key size check for L2CAP_LE_CONN_REQ
Bluetooth: L2CAP: Fix not checking output MTU is acceptable on L2CAP_ECRED_CONN_REQ
Bluetooth: Fix CIS host feature condition
Bluetooth: L2CAP: Fix response to L2CAP_ECRED_CONN_REQ
Bluetooth: hci_qca: Cleanup on all setup failures
Bluetooth: purge error queues in socket destructors
Bluetooth: L2CAP: Fix result of L2CAP_ECRED_CONN_RSP when MTU is too short
Bluetooth: L2CAP: Fix invalid response to L2CAP_ECRED_RECONF_REQ
====================
Shyam Sundar S K [Mon, 23 Feb 2026 07:40:20 +0000 (13:10 +0530)]
MAINTAINERS: Update AMD XGBE driver maintainers
Due to additional responsibilities, Shyam Sundar S K will no longer be
supporting the AMD XGBE driver. Maintenance will be handled by
Raju Rangoju going forward.
Here LED_TRIGGER_PHY is registering LED triggers during phy_attach
while holding RTNL and then taking triggers_list_lock.
[ 1362.191101] [<806c2640>] register_netdevice_notifier+0x60/0x168 <-- Trying to get lock "rtnl_mutex" via rtnl_lock();
[ 1362.197073] [<805504ac>] netdev_trig_activate+0x194/0x1e4
[ 1362.202490] [<8054e28c>] led_trigger_set+0x1d4/0x360 <-- Hold lock "triggers_list_lock" by down_read(&triggers_list_lock);
[ 1362.207511] [<8054eb38>] led_trigger_write+0xd8/0x14c
[ 1362.212566] [<80381d98>] sysfs_kf_bin_write+0x80/0xbc
[ 1362.217688] [<8037fcd8>] kernfs_fop_write_iter+0x17c/0x28c
[ 1362.223174] [<802cbd70>] vfs_write+0x21c/0x3c4
[ 1362.227712] [<802cc0c4>] ksys_write+0x78/0x12c
[ 1362.232164] [<80014504>] syscall_common+0x34/0x58
Here LEDS_TRIGGER_NETDEV is being enabled on an LED. It first takes
triggers_list_lock and then RTNL. A classical AB-BA deadlock.
phy_led_triggers_registers() does not require the RTNL, it does not
make any calls into the network stack which require protection. There
is also no requirement the PHY has been attached to a MAC, the
triggers only make use of phydev state. This allows the call to
phy_led_triggers_registers() to be placed elsewhere. PHY probe() and
release() don't hold RTNL, so solving the AB-BA deadlock.
Reported-by: Shiji Yang <yangshiji66@outlook.com> Closes: https://lore.kernel.org/all/OS7PR01MB13602B128BA1AD3FA38B6D1FFBC69A@OS7PR01MB13602.jpnprd01.prod.outlook.com/ Fixes: 06f502f57d0d ("leds: trigger: Introduce a NETDEV trigger") Cc: stable@vger.kernel.org Signed-off-by: Andrew Lunn <andrew@lunn.ch> Tested-by: Shiji Yang <yangshiji66@outlook.com> Link: https://patch.msgid.link/20260222152601.1978655-1-andrew@lunn.ch Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The break would only break out of the scoped_guard() loop, not the
switch statement. It still works correct as is ofc but let's avoid the
confusion.
Reported-by: David Lechner <dlechner@baylibre.com>
Link:: https://lore.kernel.org/cd2153f1-098b-463c-bbc1-5c6ca9ef1f12@baylibre.com Signed-off-by: Christian Brauner <brauner@kernel.org>
Ziyi Guo [Sun, 22 Feb 2026 05:06:33 +0000 (05:06 +0000)]
net: usb: pegasus: enable basic endpoint checking
pegasus_probe() fills URBs with hardcoded endpoint pipes without
verifying the endpoint descriptors:
- usb_rcvbulkpipe(dev, 1) for RX data
- usb_sndbulkpipe(dev, 2) for TX data
- usb_rcvintpipe(dev, 3) for status interrupts
A malformed USB device can present these endpoints with transfer types
that differ from what the driver assumes.
Add a pegasus_usb_ep enum for endpoint numbers, replacing magic
constants throughout. Add usb_check_bulk_endpoints() and
usb_check_int_endpoints() calls before any resource allocation to
verify endpoint types before use, rejecting devices with mismatched
descriptors at probe time, and avoid triggering assertion.
skb_may_tx_timestamp() may acquire sock::sk_callback_lock. The lock must
not be taken in IRQ context, only softirq is okay. A few drivers receive
the timestamp via a dedicated interrupt and complete the TX timestamp
from that handler. This will lead to a deadlock if the lock is already
write-locked on the same CPU.
Taking the lock can be avoided. The socket (pointed by the skb) will
remain valid until the skb is released. The ->sk_socket and ->file
member will be set to NULL once the user closes the socket which may
happen before the timestamp arrives.
If we happen to observe the pointer while the socket is closing but
before the pointer is set to NULL then we may use it because both
pointer (and the file's cred member) are RCU freed.
Drop the lock. Use READ_ONCE() to obtain the individual pointer. Add a
matching WRITE_ONCE() where the pointer are cleared.
Jakub Kicinski [Thu, 19 Feb 2026 19:50:21 +0000 (11:50 -0800)]
netconsole: avoid OOB reads, msg is not nul-terminated
msg passed to netconsole from the console subsystem is not guaranteed
to be nul-terminated. Before recent
commit 7eab73b18630 ("netconsole: convert to NBCON console infrastructure")
the message would be placed in printk_shared_pbufs, a static global
buffer, so KASAN had harder time catching OOB accesses. Now we see:
printk: console [netcon_ext0] enabled
BUG: KASAN: slab-out-of-bounds in string+0x1f7/0x240
Read of size 1 at addr ffff88813b6d4c00 by task pr/netcon_ext0/594
Allocated by task 1:
nbcon_alloc+0x1ea/0x450
register_console+0x26b/0xe10
init_netconsole+0xbb0/0xda0
The buggy address belongs to the object at ffff88813b6d4000
which belongs to the cache kmalloc-4k of size 4096
The buggy address is located 0 bytes to the right of
allocated 3072-byte region [ffff88813b6d4000, ffff88813b6d4c00)
Fixes: c62c0a17f9b7 ("netconsole: Append kernel version to message") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260219195021.2099699-1-kuba@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Duoming Zhou [Thu, 19 Feb 2026 12:46:37 +0000 (20:46 +0800)]
net: wan: farsync: Fix use-after-free bugs caused by unfinished tasklets
When the FarSync T-series card is being detached, the fst_card_info is
deallocated in fst_remove_one(). However, the fst_tx_task or fst_int_task
may still be running or pending, leading to use-after-free bugs when the
already freed fst_card_info is accessed in fst_process_tx_work_q() or
fst_process_int_work_q().
Memory state around the buggy address: ffff88800aad0f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff88800aad0f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff88800aad1000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^ ffff88800aad1080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88800aad1100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================
Fix this by ensuring that both fst_tx_task and fst_int_task are properly
canceled before the fst_card_info is released. Add tasklet_kill() in
fst_remove_one() to synchronize with any pending or running tasklets.
Since unregister_hdlc_device() stops data transmission and reception,
and fst_disable_intr() prevents further interrupts, it is appropriate
to place tasklet_kill() after these calls.
The bugs were identified through static analysis. To reproduce the issue
and validate the fix, a FarSync T-series card was simulated in QEMU and
delays(e.g., mdelay()) were introduced within the tasklet handler to
increase the likelihood of triggering the race condition.
Jann Horn [Mon, 23 Feb 2026 19:59:33 +0000 (20:59 +0100)]
eventpoll: Fix integer overflow in ep_loop_check_proc()
If a recursive call to ep_loop_check_proc() hits the `result = INT_MAX`,
an integer overflow will occur in the calling ep_loop_check_proc() at
`result = max(result, ep_loop_check_proc(ep_tovisit, depth + 1) + 1)`,
breaking the recursion depth check.
Fix it by using a different placeholder value that can't lead to an
overflow.
net/rds: fix recursive lock in rds_tcp_conn_slots_available
syzbot reported a recursive lock warning in rds_tcp_get_peer_sport() as
it calls inet6_getname() which acquires the socket lock that was already
held by __release_sock().
kworker/u8:6/2985 is trying to acquire lock: ffff88807a07aa20 (k-sk_lock-AF_INET6){+.+.}-{0:0}, at: lock_sock include/net/sock.h:1709 [inline] ffff88807a07aa20 (k-sk_lock-AF_INET6){+.+.}-{0:0}, at: inet6_getname+0x15d/0x650 net/ipv6/af_inet6.c:533
Reading from the socket struct directly is safe from possible paths. For
rds_tcp_accept_one(), the socket has just been accepted and is not yet
exposed to concurrent access. For rds_tcp_conn_slots_available(), direct
access avoids the recursive deadlock seen during backlog processing
where the socket lock is already held from the __release_sock().
However, rds_tcp_conn_slots_available() is also called from the normal
softirq path via tcp_data_ready() where the lock is not held. This is
also safe because inet_dport is a stable 16 bits field. A READ_ONCE()
annotation as the value might be accessed lockless in a concurrent
access context.
Note that it is also safe to call rds_tcp_conn_slots_available() from
rds_conn_shutdown() because the fan-out is disabled.
Fixes: 9d27a0fb122f ("net/rds: Trigger rds_send_ping() more than once") Reported-by: syzbot+5efae91f60932839f0a5@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=5efae91f60932839f0a5 Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de> Reviewed-by: Allison Henderson <achender@kernel.org> Link: https://patch.msgid.link/20260219075738.4403-1-fmancera@suse.de Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The mesh_matches_local() check above only validates the Mesh ID,
Mesh Configuration, and Supported Rates IEs. It does not verify the
presence of the Mesh Channel Switch Parameters IE (element ID 118).
When a received CSA action frame omits that IE, ieee802_11_parse_elems()
leaves elems->mesh_chansw_params_ie as NULL, and the unconditional
dereference causes a kernel NULL pointer dereference.
A remote mesh peer with an established peer link (PLINK_ESTAB) can
trigger this by sending a crafted SPECTRUM_MGMT/CHL_SWITCH action frame
that includes a matching Mesh ID and Mesh Configuration IE but omits the
Mesh Channel Switch Parameters IE. No authentication beyond the default
open mesh peering is required.
Crash confirmed on kernel 6.17.0-5-generic via mac80211_hwsim:
Fix by adding a NULL check for mesh_chansw_params_ie after
mesh_matches_local() returns, consistent with how other optional IEs
are guarded throughout the mesh code.
The bug has been present since v3.13 (released 2014-01-19).
Fixes: 8f2535b92d68 ("mac80211: process the CSA frame for mesh accordingly") Cc: stable@vger.kernel.org Signed-off-by: Vahagn Vardanian <vahagn@redrays.io> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Tung Nguyen [Fri, 20 Feb 2026 05:05:41 +0000 (05:05 +0000)]
tipc: fix duplicate publication key in tipc_service_insert_publ()
TIPC uses named table to store TIPC services represented by type and
instance. Each time an application calls TIPC API bind() to bind a
type/instance to a socket, an entry is created and inserted into the
named table. It looks like this:
In the above table, each entry represents a route for sending data
from one socket to the other. For all publications originated from
the same node, the key is UNIQUE to identify each entry.
It is calculated by this formula:
key = socket portid + number of bindings + 1 (1)
where:
- socket portid: unique and calculated by using linux kernel function
get_random_u32_below(). So, the value is randomized.
- number of bindings: the number of times a type/instance pair is bound
to a socket. This number is linearly increased,
starting from 0.
While the socket portid is unique and randomized by linux kernel, the
linear increment of "number of bindings" in formula (1) makes "key" not
unique anymore. For example:
- Socket 1 is created with its associated port number 20062001. Type 1000,
instance 1 is bound to socket 1:
key1: 20062001 + 0 + 1 = 20062002
Then, bind() is called a second time on Socket 1 to by the same type 1000,
instance 1:
key2: 20062001 + 1 + 1 = 20062003
- Socket 2 is created with its associated port number 20062002. Type 1000,
instance 1 is bound to socket 2:
key3: 20062002 + 0 + 1 = 20062003
TIPC looks up the named table and finds out that key2 with the same value
already exists and rejects the insertion into the named table.
This leads to failure of bind() call from application on Socket 2 with error
message EINVAL "Invalid argument".
This commit fixes this issue by adding more port id checking to make sure
that the key is unique to publications originated from the same port id
and node.
Fixes: 218527fe27ad ("tipc: replace name table service range array with rb tree") Signed-off-by: Tung Nguyen <tung.quang.nguyen@est.tech> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260220050541.237962-1-tung.quang.nguyen@est.tech Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ivan Vecera [Fri, 20 Feb 2026 15:57:54 +0000 (16:57 +0100)]
dpll: zl3073x: fix REF_PHASE_OFFSET_COMP register width for some chip IDs
The REF_PHASE_OFFSET_COMP register is 48-bit wide on most zl3073x chip
variants, but only 32-bit wide on chip IDs 0x0E30, 0x0E93..0x0E97 and
0x1F60. The driver unconditionally uses 48-bit read/write operations,
which on 32-bit variants causes reading 2 bytes past the register
boundary (corrupting the value) and writing 2 bytes into the adjacent
register.
Fix this by storing the chip ID in the device structure during probe
and adding a helper to detect the affected variants. Use the correct
register width for read/write operations and the matching sign extension
bit (31 vs 47) when interpreting the phase compensation value.
Fixes: 6287262f761e ("dpll: zl3073x: Add support to adjust phase") Signed-off-by: Ivan Vecera <ivecera@redhat.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260220155755.448185-1-ivecera@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jiayuan Chen [Thu, 19 Feb 2026 01:42:51 +0000 (09:42 +0800)]
kcm: fix zero-frag skb in frag_list on partial sendmsg error
Syzkaller reported a warning in kcm_write_msgs() when processing a
message with a zero-fragment skb in the frag_list.
When kcm_sendmsg() fills MAX_SKB_FRAGS fragments in the current skb,
it allocates a new skb (tskb) and links it into the frag_list before
copying data. If the copy subsequently fails (e.g. -EFAULT from
user memory), tskb remains in the frag_list with zero fragments:
head skb (msg being assembled, NOT yet in sk_write_queue)
+-----------+
| frags[17] | (MAX_SKB_FRAGS, all filled with data)
| frag_list-+--> tskb
+-----------+ +----------+
| frags[0] | (empty! copy failed before filling)
+----------+
For SOCK_SEQPACKET with partial data already copied, the error path
saves this message via partial_message for later completion. For
SOCK_SEQPACKET, sock_write_iter() automatically sets MSG_EOR, so a
subsequent zero-length write(fd, NULL, 0) completes the message and
queues it to sk_write_queue. kcm_write_msgs() then walks the
frag_list and hits:
WARN_ON(!skb_shinfo(skb)->nr_frags)
TCP has a similar pattern where skbs are enqueued before data copy
and cleaned up on failure via tcp_remove_empty_skb(). KCM was
missing the equivalent cleanup.
Fix this by tracking the predecessor skb (frag_prev) when allocating
a new frag_list entry. On error, if the tail skb has zero frags,
use frag_prev to unlink and free it in O(1) without walking the
singly-linked frag_list. frag_prev is safe to dereference because
the entire message chain is only held locally (or in kcm->seq_skb)
and is not added to sk_write_queue until MSG_EOR, so the send path
cannot free it underneath us.
Also change the WARN_ON to WARN_ON_ONCE to avoid flooding the log
if the condition is somehow hit repeatedly.
There are currently no KCM selftests in the kernel tree; a simple
reproducer is available at [1].
Ankit Garg [Fri, 20 Feb 2026 21:53:24 +0000 (13:53 -0800)]
gve: fix incorrect buffer cleanup in gve_tx_clean_pending_packets for QPL
In DQ-QPL mode, gve_tx_clean_pending_packets() incorrectly uses the RDA
buffer cleanup path. It iterates num_bufs times and attempts to unmap
entries in the dma array.
This leads to two issues:
1. The dma array shares storage with tx_qpl_buf_ids (union).
Interpreting buffer IDs as DMA addresses results in attempting to
unmap incorrect memory locations.
2. num_bufs in QPL mode (counting 2K chunks) can significantly exceed
the size of the dma array, causing out-of-bounds access warnings
(trace below is how we noticed this issue).
UBSAN: array-index-out-of-bounds in
drivers/net/ethernet/drivers/net/ethernet/google/gve/gve_tx_dqo.c:178:5 index 18 is out of
range for type 'dma_addr_t[18]' (aka 'unsigned long long[18]')
Workqueue: gve gve_service_task [gve]
Call Trace:
<TASK>
dump_stack_lvl+0x33/0xa0
__ubsan_handle_out_of_bounds+0xdc/0x110
gve_tx_stop_ring_dqo+0x182/0x200 [gve]
gve_close+0x1be/0x450 [gve]
gve_reset+0x99/0x120 [gve]
gve_service_task+0x61/0x100 [gve]
process_scheduled_works+0x1e9/0x380
Fix this by properly checking for QPL mode and delegating to
gve_free_tx_qpl_bufs() to reclaim the buffers.
Cc: stable@vger.kernel.org Fixes: a6fb8d5a8b69 ("gve: Tx path for DQO-QPL") Signed-off-by: Ankit Garg <nktgrg@google.com> Reviewed-by: Jordan Rhee <jordanrhee@google.com> Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com> Signed-off-by: Joshua Washington <joshwash@google.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20260220215324.1631350-1-joshwash@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Hyunwoo Kim [Fri, 20 Feb 2026 09:40:36 +0000 (18:40 +0900)]
tls: Fix race condition in tls_sw_cancel_work_tx()
This issue was discovered during a code audit.
After cancel_delayed_work_sync() is called from tls_sk_proto_close(),
tx_work_handler() can still be scheduled from paths such as the
Delayed ACK handler or ksoftirqd.
As a result, the tx_work_handler() worker may dereference a freed
TLS object.
To prevent this race condition, cancel_delayed_work_sync() is
replaced with disable_delayed_work_sync().
Fixes: f87e62d45e51 ("net/tls: remove close callback sock unlock/lock around TX work flush") Signed-off-by: Hyunwoo Kim <imv4bel@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Link: https://patch.msgid.link/aZgsFO6nfylfvLE7@v4bel Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Sat, 21 Feb 2026 01:18:58 +0000 (17:18 -0800)]
MAINTAINERS: include all of framer under pef2256
The "framer" infrastructure only has one driver - pef2256
and is not covered by any MAINTAINERS entry of its own.
This leads to author not being CCed on patches.
Let's include all of framer/ under the pef2256 entry.
We can split it in the very unlikely event of another
driver appearing.
Damien Le Moal [Fri, 20 Feb 2026 03:09:12 +0000 (12:09 +0900)]
ata: libata-core: fix cancellation of a port deferred qc work
cancel_work_sync() is a sleeping function so it cannot be called with
the spin lock of a port being held. Move the call to this function in
ata_port_detach() after EH completes, with the port lock released,
together with other work cancellation calls.
Fixes: 0ea84089dbf6 ("ata: libata-scsi: avoid Non-NCQ command starvation") Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
A deferred qc may timeout while waiting for the device queue to drain
to be submitted. In such case, since the qc is not active,
ata_scsi_cmd_error_handler() ends up calling scsi_eh_finish_cmd(),
which frees the qc. But as the port deferred_qc field still references
this finished/freed qc, the deferred qc work may eventually attempt to
call ata_qc_issue() against this invalid qc, leading to errors such as
reported by UBSAN (syzbot run):
UBSAN: shift-out-of-bounds in drivers/ata/libata-core.c:5166:24
shift exponent 4210818301 is too large for 64-bit type 'long long unsigned int'
...
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:94 [inline]
dump_stack_lvl+0x100/0x190 lib/dump_stack.c:120
ubsan_epilogue+0xa/0x30 lib/ubsan.c:233
__ubsan_handle_shift_out_of_bounds+0x279/0x2a0 lib/ubsan.c:494
ata_qc_issue.cold+0x38/0x9f drivers/ata/libata-core.c:5166
ata_scsi_deferred_qc_work+0x154/0x1f0 drivers/ata/libata-scsi.c:1679
process_one_work+0x9d7/0x1920 kernel/workqueue.c:3275
process_scheduled_works kernel/workqueue.c:3358 [inline]
worker_thread+0x5da/0xe40 kernel/workqueue.c:3439
kthread+0x370/0x450 kernel/kthread.c:467
ret_from_fork+0x754/0xd80 arch/x86/kernel/process.c:158
ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
</TASK>
Fix this by checking if the qc of a timed out SCSI command is a deferred
one, and in such case, clear the port deferred_qc field and finish the
SCSI command with DID_TIME_OUT.
Reported-by: syzbot+1f77b8ca15336fff21ff@syzkaller.appspotmail.com Fixes: 0ea84089dbf6 ("ata: libata-scsi: avoid Non-NCQ command starvation") Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
Bluetooth: L2CAP: Fix missing key size check for L2CAP_LE_CONN_REQ
This adds a check for encryption key size upon receiving
L2CAP_LE_CONN_REQ which is required by L2CAP/LE/CFC/BV-15-C which
expects L2CAP_CR_LE_BAD_KEY_SIZE.
Link: https://lore.kernel.org/linux-bluetooth/5782243.rdbgypaU67@n9w6sw14/ Fixes: 27e2d4c8d28b ("Bluetooth: Add basic LE L2CAP connect request receiving support") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Tested-by: Christian Eggers <ceggers@arri.de>
Bluetooth: L2CAP: Fix not checking output MTU is acceptable on L2CAP_ECRED_CONN_REQ
Upon receiving L2CAP_ECRED_CONN_REQ the given MTU shall be checked
against the suggested MTU of the listening socket as that is required
by the likes of PTS L2CAP/ECFC/BV-27-C test which expects
L2CAP_CR_LE_UNACCEPT_PARAMS if the MTU is lowers than socket omtu.
In order to be able to set chan->omtu the code now allows setting
setsockopt(BT_SNDMTU), but it is only allowed when connection has not
been stablished since there is no procedure to reconfigure the output
MTU.
Link: https://github.com/bluez/bluez/issues/1895 Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Mariusz Skamra [Thu, 12 Feb 2026 13:46:46 +0000 (14:46 +0100)]
Bluetooth: Fix CIS host feature condition
This fixes the condition for sending the LE Set Host Feature command.
The command is sent to indicate host support for Connected Isochronous
Streams in this case. It has been observed that the system could not
initialize BIS-only capable controllers because the controllers do not
support the command.
As per Core v6.2 | Vol 4, Part E, Table 3.1 the command shall be
supported if CIS Central or CIS Peripheral is supported; otherwise,
the command is optional.
Fixes: 709788b154ca ("Bluetooth: hci_core: Fix using {cis,bis}_capable for current settings") Cc: stable@vger.kernel.org Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Bluetooth: L2CAP: Fix response to L2CAP_ECRED_CONN_REQ
Similar to 03dba9cea72f ("Bluetooth: L2CAP: Fix not responding with
L2CAP_CR_LE_ENCRYPTION") the result code L2CAP_CR_LE_ENCRYPTION shall
be used when BT_SECURITY_MEDIUM is set since that means security mode 2
which mean it doesn't require authentication which results in
qualification test L2CAP/ECFC/BV-32-C failing.
Link: https://github.com/bluez/bluez/issues/1871 Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Jinwang Li [Thu, 5 Feb 2026 06:26:00 +0000 (14:26 +0800)]
Bluetooth: hci_qca: Cleanup on all setup failures
The setup process previously combined error handling and retry gating
under one condition. As a result, the final failed attempt exited
without performing cleanup.
Update the failure path to always perform power and port cleanup on
setup failure, and reopen the port only when retrying.
Fixes: 9e80587aba4c ("Bluetooth: hci_qca: Enhance retry logic in qca_setup") Signed-off-by: Jinwang Li <jinwang.li@oss.qualcomm.com> Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Bluetooth: purge error queues in socket destructors
When TX timestamping is enabled via SO_TIMESTAMPING, SKBs may be queued
into sk_error_queue and will stay there until consumed. If userspace never
gets to read the timestamps, or if the controller is removed unexpectedly,
these SKBs will leak.
Fix by adding skb_queue_purge() calls for sk_error_queue in affected
bluetooth destructors. RFCOMM does not currently use sk_error_queue.
Fixes: 134f4b39df7b ("Bluetooth: add support for skb TX SND/COMPLETION timestamping") Reported-by: syzbot+7ff4013eabad1407b70a@syzkaller.appspotmail.com Closes: https://syzbot.org/bug?extid=7ff4013eabad1407b70a Cc: stable@vger.kernel.org Signed-off-by: Heitor Alves de Siqueira <halves@igalia.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Bluetooth: L2CAP: Fix result of L2CAP_ECRED_CONN_RSP when MTU is too short
Test L2CAP/ECFC/BV-26-C expect the response to L2CAP_ECRED_CONN_REQ with
and MTU value < L2CAP_ECRED_MIN_MTU (64) to be L2CAP_CR_LE_INVALID_PARAMS
rather than L2CAP_CR_LE_UNACCEPT_PARAMS.
Also fix not including the correct number of CIDs in the response since
the spec requires all CIDs being rejected to be included in the
response.
Link: https://github.com/bluez/bluez/issues/1868 Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Bluetooth: L2CAP: Fix invalid response to L2CAP_ECRED_RECONF_REQ
This fixes responding with an invalid result caused by checking the
wrong size of CID which should have been (cmd_len - sizeof(*req)) and
on top of it the wrong result was use L2CAP_CR_LE_INVALID_PARAMS which
is invalid/reserved for reconf when running test like L2CAP/ECFC/BI-03-C:
> ACL Data RX: Handle 64 flags 0x02 dlen 14
LE L2CAP: Enhanced Credit Reconfigure Request (0x19) ident 2 len 6
MTU: 64
MPS: 64
Source CID: 64
< ACL Data TX: Handle 64 flags 0x00 dlen 10
LE L2CAP: Enhanced Credit Reconfigure Respond (0x1a) ident 2 len 2
! Result: Reserved (0x000c)
Result: Reconfiguration failed - one or more Destination CIDs invalid (0x0003)
Fiix L2CAP/ECFC/BI-04-C which expects L2CAP_RECONF_INVALID_MPS (0x0002)
when more than one channel gets its MPS reduced:
> ACL Data RX: Handle 64 flags 0x02 dlen 16
LE L2CAP: Enhanced Credit Reconfigure Request (0x19) ident 2 len 8
MTU: 264
MPS: 99
Source CID: 64
! Source CID: 65
< ACL Data TX: Handle 64 flags 0x00 dlen 10
LE L2CAP: Enhanced Credit Reconfigure Respond (0x1a) ident 2 len 2
! Result: Reconfiguration successful (0x0000)
Result: Reconfiguration failed - reduction in size of MPS not allowed for more than one channel at a time (0x0002)
Fix L2CAP/ECFC/BI-05-C when SCID is invalid (85 unconnected):
> ACL Data RX: Handle 64 flags 0x02 dlen 14
LE L2CAP: Enhanced Credit Reconfigure Request (0x19) ident 2 len 6
MTU: 65
MPS: 64
! Source CID: 85
< ACL Data TX: Handle 64 flags 0x00 dlen 10
LE L2CAP: Enhanced Credit Reconfigure Respond (0x1a) ident 2 len 2
! Result: Reconfiguration successful (0x0000)
Result: Reconfiguration failed - one or more Destination CIDs invalid (0x0003)
Fix L2CAP/ECFC/BI-06-C when MPS < L2CAP_ECRED_MIN_MPS (64):
> ACL Data RX: Handle 64 flags 0x02 dlen 14
LE L2CAP: Enhanced Credit Reconfigure Request (0x19) ident 2 len 6
MTU: 672
! MPS: 63
Source CID: 64
< ACL Data TX: Handle 64 flags 0x00 dlen 10
LE L2CAP: Enhanced Credit Reconfigure Respond (0x1a) ident 2 len 2
! Result: Reconfiguration failed - reduction in size of MPS not allowed for more than one channel at a time (0x0002)
Result: Reconfiguration failed - other unacceptable parameters (0x0004)
Fix L2CAP/ECFC/BI-07-C when MPS reduced for more than one channel:
> ACL Data RX: Handle 64 flags 0x02 dlen 16
LE L2CAP: Enhanced Credit Reconfigure Request (0x19) ident 3 len 8
MTU: 84
! MPS: 71
Source CID: 64
! Source CID: 65
< ACL Data TX: Handle 64 flags 0x00 dlen 10
LE L2CAP: Enhanced Credit Reconfigure Respond (0x1a) ident 2 len 2
! Result: Reconfiguration successful (0x0000)
Result: Reconfiguration failed - reduction in size of MPS not allowed for more than one channel at a time (0x0002)
Link: https://github.com/bluez/bluez/issues/1865 Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
This config option goes way back - it used to be an internal debug
option to random.c (at that point called DEBUG_RANDOM_BOOT), then was
renamed and exposed as a config option as CONFIG_WARN_UNSEEDED_RANDOM,
and then further renamed to the current CONFIG_WARN_ALL_UNSEEDED_RANDOM.
It was all done with the best of intentions: the more limited
rate-limited reports were reporting some cases, but if you wanted to see
all the gory details, you'd enable this "ALL" option.
However, it turns out - perhaps not surprisingly - that when people
don't care about and fix the first rate-limited cases, they most
certainly don't care about any others either, and so warning about all
of them isn't actually helping anything.
And the non-ratelimited reporting causes problems, where well-meaning
people enable debug options, but the excessive flood of messages that
nobody cares about will hide actual real information when things go
wrong.
I just got a kernel bug report (which had nothing to do with randomness)
where two thirds of the the truncated dmesg was just variations of
random: get_random_u32 called from __get_random_u32_below+0x10/0x70 with crng_init=0
and in the process early boot messages had been lost (in addition to
making the messages that _hadn't_ been lost harder to read).
The proper way to find these things for the hypothetical developer that
cares - if such a person exists - is almost certainly with boot time
tracing. That gives you the option to get call graphs etc too, which is
likely a requirement for fixing any problems anyway.
See Documentation/trace/boottime-trace.rst for that option.
And if we for some reason do want to re-introduce actual printing of
these things, it will need to have some uniqueness filtering rather than
this "just print it all" model.
Fixes: cc1e127bfa95 ("random: remove ratelimiting for in-kernel unseeded randomness") Acked-by: Jason Donenfeld <Jason@zx2c4.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Mon, 23 Feb 2026 17:33:08 +0000 (09:33 -0800)]
default_gfp(): avoid using the "newfangled" __VA_OPT__ trick
The default_gfp() helper that I added is not wrong, but it turns out
that it causes unnecessary headaches for 'sparse' which doesn't support
the use of __VA_OPT__ (introduced in C++20 and C23, and supported by gcc
and clang for a long time).
We do already use __VA_OPT__ in some other cases in the kernel (drm/xe
and btrfs), but it has been fairly limited. Now it triggers for pretty
much everything, and sparse ends up not working at all.
We can use the traditional gcc ',##__VA_ARGS__' syntax instead: it may
not be the "C standard" way and is slightly less natural in this
context, but it is the traditional model for this and avoids the sparse
problem.
Reported-and-tested-by: Ricardo Ribalda <ribalda@chromium.org> Reported-and-tested-by: Richard Fitzgerald <rf@opensource.cirrus.com> Reported-by: Ben Dooks <ben.dooks@codethink.co.uk> Fixes: e19e1b480ac7 ("add default_gfp() helper macro and use it in the new *alloc_obj() helpers") Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ariel Silver [Fri, 20 Feb 2026 10:11:29 +0000 (10:11 +0000)]
wifi: mac80211: bounds-check link_id in ieee80211_ml_reconfiguration
link_id is taken from the ML Reconfiguration element (control & 0x000f),
so it can be 0..15. link_removal_timeout[] has IEEE80211_MLD_MAX_NUM_LINKS
(15) elements, so index 15 is out-of-bounds. Skip subelements with
link_id >= IEEE80211_MLD_MAX_NUM_LINKS to avoid a stack out-of-bounds
write.
Fixes: 8eb8dd2ffbbb ("wifi: mac80211: Support link removal using Reconfiguration ML element") Reported-by: Ariel Silver <arielsilver77@gmail.com> Signed-off-by: Ariel Silver <arielsilver77@gmail.com> Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260220101129.1202657-1-Ariel.Silver@cybereason.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Hongbo Li [Sat, 14 Feb 2026 03:02:48 +0000 (03:02 +0000)]
erofs: allow sharing page cache with the same aops only
Inode with identical data but different @aops cannot be mixed
because the page cache is managed by different subsystems (e.g.,
@aops for compressed on-disk inodes cannot handle plain on-disk
inodes).
In this patch, we never allow inodes to share the page cache
among plain, compressed, and fileio cases. When a shared inode
is created, we initialize @aops that is the same as the initial
real inode, and subsequent inodes cannot share the page cache
if the inferred @aops differ from the corresponding shared inode.
This is reasonable as a first step because, in typical use cases,
if an inode is compressible, it will fall into compressed
inodes across different filesystem images unless users use plain
filesystems. However, in that cases, users will use plain
filesystems all the time.
wifi: mac80211: set default WMM parameters on all links
Currently, mac80211 only initializes default WMM parameters
on the deflink during do_open(). For MLO cases, this
leaves the additional links without proper WMM defaults
if hostapd does not supply per-link WMM parameters, leading
to inconsistent QoS behavior across links.
Set default WMM parameters for each link during
ieee80211_vif_update_links(), because this ensures all
individual links in an MLD have valid WMM settings during
bring-up and behave consistently across different BSS.
Daniel Hodges [Fri, 6 Feb 2026 19:53:56 +0000 (14:53 -0500)]
wifi: libertas: fix use-after-free in lbs_free_adapter()
The lbs_free_adapter() function uses timer_delete() (non-synchronous)
for both command_timer and tx_lockup_timer before the structure is
freed. This is incorrect because timer_delete() does not wait for
any running timer callback to complete.
If a timer callback is executing when lbs_free_adapter() is called,
the callback will access freed memory since lbs_cfg_free() frees the
containing structure immediately after lbs_free_adapter() returns.
Both timer callbacks (lbs_cmd_timeout_handler and lbs_tx_lockup_handler)
access priv->driver_lock, priv->cur_cmd, priv->dev, and other fields,
which would all be use-after-free violations.
Use timer_delete_sync() instead to ensure any running timer callback
has completed before returning.
This bug was introduced in commit 8f641d93c38a ("libertas: detect TX
lockups and reset hardware") where del_timer() was used instead of
del_timer_sync() in the cleanup path. The command_timer has had the
same issue since the driver was first written.
Fixes: 8f641d93c38a ("libertas: detect TX lockups and reset hardware") Fixes: 954ee164f4f4 ("[PATCH] libertas: reorganize and simplify init sequence") Cc: stable@vger.kernel.org Signed-off-by: Daniel Hodges <git@danielhodges.dev> Link: https://patch.msgid.link/20260206195356.15647-1-git@danielhodges.dev Signed-off-by: Johannes Berg <johannes.berg@intel.com>
wifi: brcmfmac: Fix potential kernel oops when probe fails
When probe of the sdio brcmfmac device fails for some reasons (i.e.
missing firmware), the sdiodev->bus is set to error instead of NULL, thus
the cleanup later in brcmf_sdio_remove() tries to free resources via
invalid bus pointer. This happens because sdiodev->bus is set 2 times:
first in brcmf_sdio_probe() and second time in brcmf_sdiod_probe(). Fix
this by chaning the brcmf_sdio_probe() function to return the error code
and set sdio->bus only there.
Fixes: 0ff0843310b7 ("wifi: brcmfmac: Add optional lpo clock enable support") Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Acked-by: Arend van Spriel<arend.vanspriel@broadcom.com> Link: https://patch.msgid.link/20260203102133.1478331-1-m.szyprowski@samsung.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Johannes Berg [Tue, 17 Feb 2026 12:05:26 +0000 (13:05 +0100)]
wifi: radiotap: reject radiotap with unknown bits
The radiotap parser is currently only used with the radiotap
namespace (not with vendor namespaces), but if the undefined
field 18 is used, the alignment/size is unknown as well. In
this case, iterator->_next_ns_data isn't initialized (it's
only set for skipping vendor namespaces), and syzbot points
out that we later compare against this uninitialized value.
Fix this by moving the rejection of unknown radiotap fields
down to after the in-namespace lookup, so it will really use
iterator->_next_ns_data only for vendor namespaces, even in
case undefined fields are present.
Daniil Dulov [Wed, 11 Feb 2026 08:20:24 +0000 (11:20 +0300)]
wifi: cfg80211: cancel rfkill_block work in wiphy_unregister()
There is a use-after-free error in cfg80211_shutdown_all_interfaces found
by syzkaller:
BUG: KASAN: use-after-free in cfg80211_shutdown_all_interfaces+0x213/0x220
Read of size 8 at addr ffff888112a78d98 by task kworker/0:5/5326
CPU: 0 UID: 0 PID: 5326 Comm: kworker/0:5 Not tainted 6.19.0-rc2 #2 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Workqueue: events cfg80211_rfkill_block_work
Call Trace:
<TASK>
dump_stack_lvl+0x116/0x1f0
print_report+0xcd/0x630
kasan_report+0xe0/0x110
cfg80211_shutdown_all_interfaces+0x213/0x220
cfg80211_rfkill_block_work+0x1e/0x30
process_one_work+0x9cf/0x1b70
worker_thread+0x6c8/0xf10
kthread+0x3c5/0x780
ret_from_fork+0x56d/0x700
ret_from_fork_asm+0x1a/0x30
</TASK>
The problem arises due to the rfkill_block work is not cancelled when wiphy
is being unregistered. In order to fix the issue cancel the corresponding
work in wiphy_unregister().
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
Linus Torvalds [Sun, 22 Feb 2026 21:09:33 +0000 (13:09 -0800)]
Merge tag 'libcrypto-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux
Pull crypto library fix from Eric Biggers:
"Fix a big endian specific issue in the PPC64-optimized AES code"
* tag 'libcrypto-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:
lib/crypto: powerpc/aes: Fix rndkey_from_vsx() on big endian CPUs
Mark Brown [Tue, 17 Feb 2026 13:10:46 +0000 (13:10 +0000)]
CREDITS: Add -next to Stephen Rothwell's entry
Stephen retired and stepped back from -next maintainership, update his
entry in CREDITS to recognise his 18 years of hard work making it what
it is today and all the impact it's had on our development process.
Also update to his current GnuPG key while we're here.
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au> Acked-by: SeongJae Park <sj@kernel.org> Reviewed-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Mark Brown <broonie@kernel.org> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Arnd Bergmann [Tue, 17 Feb 2026 08:26:49 +0000 (08:26 +0000)]
x509: select CONFIG_CRYPTO_LIB_SHA256
The x509 public key code gained a dependency on the sha256 hash
implementation, causing a rare link time failure in randconfig
builds:
arm-linux-gnueabi-ld: crypto/asymmetric_keys/x509_public_key.o: in function `x509_get_sig_params':
x509_public_key.c:(.text.x509_get_sig_params+0x12): undefined reference to `sha256'
arm-linux-gnueabi-ld: (sha256): Unknown destination type (ARM/Thumb) in crypto/asymmetric_keys/x509_public_key.o
x509_public_key.c:(.text.x509_get_sig_params+0x12): dangerous relocation: unsupported relocation
Select the necessary library code from Kconfig.
Fixes: 2c62068ac86b ("x509: Separately calculate sha256 for blacklist") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Haiyue Wang [Sun, 22 Feb 2026 12:11:00 +0000 (20:11 +0800)]
xz: fix arm fdt compile error for kmalloc replacement
Align to the commit bf4afc53b77a ("Convert 'alloc_obj' family to use the
new default GFP_KERNEL argument") update the 'kmalloc_obj' declaration
for userspace to fix below compile error:
In file included from arch/arm/boot/compressed/../../../../lib/decompress_unxz.c:241,
from arch/arm/boot/compressed/decompress.c:56:
arch/arm/boot/compressed/../../../../lib/xz/xz_dec_stream.c: In function 'xz_dec_init':
arch/arm/boot/compressed/../../../../lib/xz/xz_dec_stream.c:787:28: error: implicit declaration of function 'kmalloc_obj'; did you mean 'kmalloc'? [-Wimplicit-function-declaration]
787 | struct xz_dec *s = kmalloc_obj(*s);
| ^~~~~~~~~~~
| kmalloc
Signed-off-by: Haiyue Wang <haiyuewa@163.com> Fixes: 69050f8d6d07 ("treewide: Replace kmalloc with kmalloc_obj for non-scalar types") Fixes: bf4afc53b77a ("Convert 'alloc_obj' family to use the new default GFP_KERNEL argument") Reviewed-by: Kees Cook <kees@kernel.org> Acked-by: Lasse Collin <lasse.collin@tukaani.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / thirdparty / kernel / linux.git / log
